Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author. A COMPARISON OF THE FATE OF ELEMENTAL SULPHUR AND SULPHATE SULPHUR BASED FERTILIZERS IN PASTURE SOILS A thesis presented in partial fulfi lment of the requirements for the degree of Doctor of Philosophy in Soil Science at Massey University Sathien Phimsarn 1991 Massey University Library Thesis Copyright Form T.tl f th ? ,... "A Comparison Of The Fate Of 1 e o es1s: Elemental Sulphur And Sulphate SSruis. . Sulphur Based Fertilizers In P?tvre (1) (a) I gxve pe'rmiSSJon for my thesis to be made ava.:t?a-I:He-to readers in Massey Uruvers1ty Library under conclit-iG.ru det?lrmined by the Librarian. (b) I do not wish my thesis to be made available to readers without my written consent for 3>. months. (2) (a) I agree that my thesis, or a copy, may be sent to?another institution under conditions deter mineG-.hy the_ Libr ar ia:n. (b) I do not wish my thesis, or a copy, to be sent to another institution without my written consent for J.. months. (3) (a) !-agree that my tll?lsi.s may-be copied fur-Library ..u.s.e..... (b) I do not wish my thesis to be copied for Library use for 3-. months. frr\?\e\"\ ?-i IY"lB:)fT\ Signed ?? ? Date The copyright of this thesis belongs to the author. Readers must sign their name in the space below to show that they recognise this. They are asked to add their permanent address. NAME AND ADDRESS DATE 15 __ \ ABSTRACT Nitrogen fixation by legumes has a particular requirement fo r adequate soi l sulphu r status. Sulphur (S) is a mobile nutrient and is easily leached from aquic soil environments, therefore regular topdressing with S fertil izer is required to maintain legume vigor and pasture production in most New Zealand pasture soils. Escalating fertil izer costs have focused attention on the efficiency of use of S fertilizers, particularly superphosphate (SSP) and alternative elemental S (S0) based fertilizers less liable to leaching loss in this aquic environment. Field and glasshouse trials, using the resident clover/ryegrass sward on undisturbed soil cores (150 mm diameter, 100 mm depth) , were undertaken to determine the comparative short-term fate of SSP and different particle sizes of S0 . Methods for manufacturing radioactively labelled (35s) ferti l izers were developed. I n addition, the effect of sheep dung on the short-term immobi lization of soil and fertil izer S was also investigated. A simple computer s imulation model explain ing the observed transformation of soi l su lphur and 35s labelled fertilizer was developed. Initially, the effect of sheep dung on the short-term immobilization of soil and fertilizer S was investigated. Very small amounts (about 2-5%) of plant (clover/ryegrass pasture) S and P , within 1 00 mm of the area surrounding the dung pellet, were derived from the dung. Under the experimental conditions that prevailed, dung S behaved as a slow release S form causing neither greater immobilization of soil or fertilizer S nor mineralization of soil organic S. lt was concluded that the impact of dung return on short-term (< one year) S fertilizer fate need not be considered. An initial field trial comparing the fate of microfine S0 (< 0.010 mm) relative to sulphate-based SSP was undertaken on Tokomaru silt loam, a New Zealand yellow-grey earth (Fragiaqualf) . The microfine S0 oxidized within 30 days of application but initially (up to 60 days) was slightly less effective than SSP in terms of plant uptake. Over longer periods of time (150 days) their pe rformances were comparable. Final cumulative plant uptake at 150 days accounted for 13.6% of microfine S0 and 16.3% of the SSP-sulphate. The major transformation of 35s from microfine S0 and 35s ?belied gypsum In SSP to soil organic 35s forms occurred in the first 30 days after application. The organic 35s activity formed from 35s0 was twice that formed from sulphate-based fertilizer and was mainly carbon- i i i bonded 35s in the top 33 mm of the pasture soil profile. The amount of organic 35s remaining as carbon-bonded 35s decreased with soil depth and the reverse occurred for the ester? sulphate 35s. By 1 50 days, greater activity from the microfine 35s0 remained in the soil organic s fraction than from the sulphate-35s ferti l izer, indicating that more soil o rganic S reserves may be formed through the use of f ine S0 fertilizer than from the sulphate-based fertilizer. This also indicated the advantage of using S0 in minimizing the S leaching losses in this aquic environment. An inverse d ilution technique using an isotope injector developed at Massey University to uniformly label undisturbed soil cores with carrier-free 35so4= solution was used to measure the impact of S0 and sulphate-based fertilizers on the fate of so il S. Results were consistent with the labelled fertilizer technique and both techniques indicated rapid incorporation of 35s into soil organic S and that the carbon-bonded S formed was likely to be a subsequent source of mineralized S available to plants. Soil samples from the prel iminary field study were used to evaluate soil preparat ion and extraction techniques. Soil sampling and preparation techniques were evaluated on the basis that an extract sampling the plant available S pool in soil shou ld have the same 35s specific activity as plant growing on that soil. The average 35s specific activity in a calcium dihydrogen phosphate (CaP-S) (0.04 M) extract from a freeze-dried sample of the top 60 mm of a pasture soil was most closely related to the 35s specific activity of plants growing on that soil . CaP-S extracts from field-moist soil or 0 .01 M CaCI2 extracts from fie ld-moist or freeze-dried soils had higher specific activities than plants. lt was concluded that plants were able to extract soil S from soils which was not exchangeable with added 35so4= fertilizers during e ither the field experiment or extraction with 0.01 M CaCI2. The second series of field and glasshouse trials were conducted to investigate the fate of 35s labelled SSP, gypsum and S0 of varying particle s izes (<0 . 1 50 mm, 0 . 1 50-0 .250 mm and 0.250-0.500 mm, in granulated and non-granulated forms) in two pasture soils contrasting in mineralogy and fertility status. Under glasshouse conditions, 50 mm of simulated rainfall was applied to each of the undisturbed soil cores during the first 56 days after fertilizer application. For the remainder of the period, cores were watered from below using a saucer. Field cores remained subject to the local cl imate. Both the rate of oxidation in soil and the efficiency of plant use of S0 was improved by decreasing its particle size . Relative to soluble so4=-s applied as gypsum or SSP, the plant utilization of oxidized SJ was similar. Granulation of finer S0 with or without finely ground phosphate rock had l ittle effect on the iv long-term ( 1 80 days) oxidation rate or the efficiency with which , after oxidation, finely ground S0 was taken up plants. Apart from S0 of large particle size (>0 . 150 mm) which had not oxidized, the major fate of ferti l izer 35s, e ither under glasshouse or field conditions, was again in soil organic matter mostly formed in the top 33 mm of the soil . Applications of gypsum and SSP caused 35s to move to the 33-1 00 mm soil depths but there was no additional influence of P on the depth to which so4= was leached. A pre l iminary computer simu lation model describing the fate of 35so4 =-s fe rti l izer was developed. The model provided a very accurate method of predicting plant uptake of S from both SSP fertilized and unfertilized soil cores. The model also indicated that, at any particular soi l depth , on average, actual rates of mineralization and immobilization may exceed root uptake of S by 1 .5 to 2 fold (mg S turned over per unit of S taken up by plants) . The accuracy of the est imated turnover rate could not be validated because the model gave re lative ly inaccurate predictions of the measured movement and transformations of 35s tracer added to the soil as SSP. There was, however, relative similarity between the observed and predicted proportional distribution of 35s between soil and plant S forms. Such a distribution supported the concept of using root activity as a modifie r of mineralization and immobil ization rates in order to describe the extent of these processes at different soil depths. The study has emphasized the greater importance of the surface few mil l imeters of pasture soil in S transformations, important in the fate of fertilizer and pasture plant nutrit ion . There appears to be scope in manipulating S0 particle size to improve the efficiency of the S fertilizer used. V ACKNOWLEDGEMENTS The author g rateful ly acknowledges the gu idance, u nderstanding , constructive crit icism , encouragement, enthusiasm and endless patience of Dr. M. J . Hedley, my chief supervisor. My gratitude and appreciation is also extended to the following people for their contributions toward the completion of this thesis Drs. P .E.H. G regg and S.K. Saggar for their readiness to help, constructive criticism and valuable discussion. Mr. R.W. Tillman; Drs. N.S. Bolan; L.K. Heng; D.R. Scotter; W. Chatupote; K. Sakadevan and P. Loganathan for their advice, comments and assistance throughout the research and during the preparation of this manuscript. Mrs. M . Horne (nee Wallace) ; Messrs L. Currie ; M. Eggles; I .A. Painter and M . Bretherton for technical assistance. Past and p re se nt members of t he So i l Science D epartment fo r t he i r fr iendship and encouragement. Fellow postgraduate students, Mr. A. Hammond, M iss K. Hodgson , M r. D. Tambunan and Mrs C. Hedley for their friendship and proof-reading. The Thai M inistry of Agriculture and Cooperatives for providing study leave, the New Zealand Overseas Deve lopment Aid Programme for f inancial assistance, the Ferti l izer and Lime Research Centre for facilities. My family, my brothers and sisters for their love and support. Finally, Suwaree and Nisanart, my wife and my daughter, for their pat ience, support and encouragement throughout. vi TABLE OF CONTENTS ABSTRACT ................................................................................................................................. ii ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi LIST OF APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix CHAPTER 1 I NTRODUCTION ........................................................................................................................ 1 CHAPTER 2 A REVIEW OF LITERATURE ON ASPECTS OF SULPHUR CYCLING IN GRAZED PASTURE SYSTEM 2.1 INTRODUCTION ........................................................................................................... 4 2.2 NATURE, FORMS AND DISTRIBUTION OF SOIL SULPHUR ............. ..... ....... ............ 4 2.2.1 Total sulphur ................................................................................................. 4 2.2.2 Inorganic sulphur ........................................................... .......... ..................... 5 2.2.2. 1 Readily soluble sulphate-S . . . ....... .... . . . . . ... .. .. . . . . . ...... ......... . . . . . .... 5 2.2.2.2 Adsorbed sulphate-S .. ... . . . . ... . . . . . ... ... ..... . ...... . . . . . ... ... ......... . . ....... 6 2.2.2.3 Other inorganic S forms . . . . . . . ... .. .... . . . . . . . . ... . . . . . . . . ..... . . . . . . . . . .. .. .... . 10 2.2.3 Organic sulphur .................................................................................. ........ 11 2.2.4 Other methods for characterizing soil organic sulphur .......... .. ...... . ........... . 13 2.2.5 Microbial su lphur ............................................. . .......... . . . ........ ... . ........ ......... 14 2.3 B IOLOGICAL TRANSFORMATIONS OF S IN SOIL. .............. ......... . ........... ....... . . . . . ... 16 2.3.1 Mineralization ............................................................................................. 18 2.3.2 Immobil ization ............................................................................................ 21 2.4 PLANT REQUIREMENTS ........................................................................................... 26 2.4.1 Assessing soil sulphur availability .............................................................. 27 2.4. 1 . 1 Plant analyses .. . . .... . . . . ....... . ..... . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . ...... . ... . .. 28 2.4. 1.2 Soil analyses . ... . . . . . . . . ..... . ... . . . . .. . . . . .. .. .... . . .... . . . . . . .. . . . . ... ... . ... . . . . .... 30 2.4. 1.2. 1 Depth of sampling and sample preparations . . . . . . ..... . . . . ... ........ . 30 2.4. 1 .2.2 Extractants . . . . . . . . . . . . . . ... . . . . . . . . ... ....... .. . . ... . . .. . . . . . . . . . ... ..... ... . . . . . . . . . . . . 32 a. Readily soluble sulphate-S . ... . ... . . . . . . ... . . . . ...... . . . . . . . . . ... . . . . . . . . .. . . . .. . . . . . . . . . ... . . . 33 b. Adsorbed plus soluble sulphate-S . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 c. Readily available and adsorbed sulphate and portions of organic S .. .... 34 d. Other methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . 34 e. Use of radioactive 35 S .............. . . . . . . .... . . . . . . . . ... . . . . . . . . . . . . ... . . . . ...... . .. ....... ..... 34 2.4. 1 .2.3 Interpretation of soil tests . ...................................... ................. 35 2.4.2 Fertilizer S requirements ............................... ................ ......... . ........ .. ..... .... 37 2.4.2. 1 S recommendation for pastures in New Zealand .. . . ........ ... . .... 37 vii 2.4.2.2 2.4.2.3 Form of fertilizers . . ................. . ........ . ... . . ........ ..... . ... ................. 39 Efficient use of fertilizer S . . . . ....... . . . . .. ............... ................ ....... 4 1 2.5 THE FATE OF FERTILIZER S IN SOIL. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 2.5.1 Oxidation of S0 in soi l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 2.5.2 The fate of su lphate-S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 2.5.2. 1 Uptake by pasture ...... . ......... ....... ........................... ......... . ....... 47 2.5.2.2 L eaching losses ........ ......... ..... . . . ........ . . . . . ..... . . ...... ........ . . . . . . . . ... 49 2.5.2.3 Immobilization of applied fertilizer sulphur ....... . ... ..... . . . .... ....... 53 2.6 SULPHUR CYCLING IN GRAZED PASTURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 2.6.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 2.6.2 Uptake of S by pasture plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 2.6.3 Return and decomposition of plant litter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 2.6.4 Ingestion and excreta return by the grazing animal. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 2.6.4. 1 Sulphur in dung and urine .. . ..... . . . .......... . . . . ..... .... . . . . . . . . ...... . ... . . . 60 2.6.4.2 Distribution of dung and urine . . . ......... . ....... ... ....... ...... . . ..... . ..... 62 2.6. 4.3 Excreta decomposition .... ... ..... . .. ..... . . . .................... . . ..... . ......... 64 2.7 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 CHAPTER 3 EXPERIMENTAL SITES AND TECHNIQUES 3 . 1 METHODS FOR RADIOACTIVELY LABELLING FERTILIZER SULPHUR . . . . . . . . . . . . . . . . 68 3. 1 . 1 Introduction ................................................................................................. 68 3 . 1 .2 Calculating the required 35s enrichment in fertilizer materials . . . . . . . . . . . . . . . . . . 68 3 . 1 .3 Labelling S0 fertilizer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 3. 1.3. 1 Labelling microfine EP . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . 69 3. 1 . 3.2 Labelling EP of different particle sizes ......... ...... . ......... ...... ..... 69 3. 1 .4 Labelling sulphate containing fertilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 3.2 EXPERIMENTAL SITES, SOIL AND PLANT PREPARATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 3.2.1 Experimental sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 3.2.2 Isolating undisturbed soil cores for field and glasshouse experiments . . . . . . . 70 3 .2.3 Sample preparations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . 72 3.2.3. 1 Sheep dung sample preparation . ....... ........ ....... ............. .. ....... 72 3.2.3.2 Soil sample preparation .... . . . . .... . . . . ...... ......... ....... .......... .......... 72 3.2.3.3 Herbage sample preparation .... . . ....... . ..... .. . .............. . .......... . .. 72 3.3 CHEMICAL ANALYSES OF SAMPLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 3.3.1 Total S in plant material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 3.3. 1 . 1 Alkaline hypobromite oxidation method . ... . . ...... . . ...... .... . . ... ... . . 72 3.3. 1 . 2 LECO sulphur analyzer .... . ................... ...... . ...... ....... ...... . . .. . .... 73 3.3.2 Soluble and extractable S in soil samples (CaCI-S and CaP-S) (Ca(H2P04)2 and CaCI2 extractable S) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.3.3 Total S in soil (TT-S) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.3.4 Total S0 in soil samples (TT-els) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . 72 3.3.5 Organic S (Org-S) in soil samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 3.3.5. 1 HI-reducible S (HI-S) .... . .... .... .............. ..... . ...... . ....... ................ 75 3.3.5.2 Ester-sulphate (Est-S) ... . . ..... . . ..... .... . ................ .. . . . .......... ... ..... 75 3.3.5.3 Carbon-bonded S (Cb-S) ...... . . . .. ... .... ... . . . ...... ...... ..... . . ...... . . . . . . . 76 3.3.6 HI-reducible S in digested and extracted samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.3.7 Determination of total S in fertilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.3. 7. 1 EP containing fertilizers ..... . . . . .................... . .... ........ . . ..... . . . . . ... . . 76 3.3.8 viii 3.3. 7.2 Su[flhate containing fertilizers ................................................ 77 Radioassay of Ss and 32 P ....................................................................... 77 3.3.8. 1 Liquid scintillation counting ..................................................... 77 3.3.8.2 Establishing quench curves .................................................... 77 CHAPTER 4 EVALUATION OF THE ROLE OF SHEEP DUNG IN THE SHORT TERM IMMOBILIZATION OF SOIL AND FERTILIZER SULPHUR 4. 1 INTRODUCTION ......................................................................................................... 79 4.2 EXPERIMENTAL OBJECTIVES .................................................................................. 81 4.3 MATER IALS AND METHODS ............................................................................. ........ 81 4.3 .1 EXPERIMENT 1, AREA OF PASTURE INFL UENCED B Y SHEEP DUNG ............................................................................................ 81 4.3. 1. 1 Design of the experimen!;?? ................... ................................... 81 4.3. 1.2 Labelling of dung with 3 S and 32 P ..................................... 83 4.3.1.3 Total and extractable S and P in dung .................................... 83 4.3.2 EXPERIMENT 2, EFFECT OF SHEEP DUNG ON GROWTH AND YIELD OF PASTURE ........................................................................ 85 4.3.2. 1 Design of the experiment ................. ....................................... 85 4.3 .3 Chemical Analyses ..................................................................................... 8S 4.3.3. 1 Total S and P in pasture samples ........................................... 85 4.3.3.2 Total S and P in dung samples ............................................... 85 4.3.3.3 Extractable S in dung samples ............................................... 86 4.3.3.4 Exchangeable P in dung samples ........................................... 86 4.3.3.5 Radioassay of 3S S and 32 P activities ................................... 87 4.3.4 Presentation of resu lts (method of calculation) .......................................... 87 4.3 .5 Statistical analyses ..................................................................................... 88 4.4 RESULTS .................................................................................................................... 89 4.4. 1 EXPERIMENT 1, AREA OF PASTURE INFL UENCED BY SHEEP DUNG ............................................................................................ 89 4.4. 1. 1 Pasture yield ........................................................................... 89 4.4. 1.2 S and P congentration? in and their uptake by pasture ........... 90 4.4.1.3 Uptake of ;ss and 3 P by pasture ....................................... 90 4.4. 1.3. 1 Pattern of S uptake ............................................................ 90 4.4. 1.3.2 35 S specific activity and percent of plant S ctf!ived from dung (%SDFD) .................................. 90 4.4.1.3.3 Pattern of P uptake ............................................................ 91 4.4. 1.3.4 32 P specific activity and percent of plant P derived from dung (%PDFD) .................................. 92 4.4.2 DISCUSSION, EXPERIMENT 1 ................................................................. 92 4.4.3 EXPERIMENT 2, EFFECT OF SHEEP DUNG ON GROWTH AND YIELD OF PASTURE ........................................................ ................ 94 4.4.3. 1 Pasture yield ........................................................................... 94 4.4.3.2 Plant S concentration .............................................................. 94 4.4.3.3 Plant S uptake ....................................................................... 102 4.4.4 DISCUSSION, EXPERIMENT 2 ............................................................... 102 4.4.5 Area covered by dung and influenced by dung sulphur ........................... 1 05 4.S CONCLUSIONS ......................................... ............................................................... 1 05 ix CHAPTER 5 EVALUATI NG F I ELD EXPER IM ENTATION TECHN IQUES US I NG 35s LABELLED FERTILIZERS TO TRACE THE FATE OF SULPHATE AND ELEMENTAL S APPLIED TO PASTURE SOILS 5.1 I NTRODUCTION ....................................................................................................... 107 5.2 OBJECTIVES ............................................................................................................ 108 5.3 MATERIALS AND METHODS35 ................................................................................ 109 5.3.1 EXPERIMENT 1, S LABELLED FERTILIZER .................................... 109 5.3. 1. 1 Preparation of 35 S labelled EP and SSP .......................... ... 109 5.3. 1. 1. 1 35 EP labelled EP.. ..... .... .. ..... ........ . . .. ... .. ... .. ..... . . ....... ... .. .. ...... 109 5.3.1.1.2 35so4= labelled SSP ........................................................... 109 5.3. 1.2 Trial method ....................................... ...................... ............. 110 5.3. 1.3 Sampling and preparation of samples ................................... 110 5.3.2 EXPERIMENT 2, INVERSE ISOT OPIC DILUTI ON EXPERIMENT ........ 112 5.3.2. 1 Carrier-free 35 S preparation and injection technique .......... 112 5.3.2.2 Trial method .......................................................................... 112 5.3.2.3 Sampling and preparation of samples ................................... 112 5.3.3 Experimental site ...................................................................................... 113 5.3.4 Chemical analyses ................................................................................... 115 5.3.5 Statistical analyses .................................................................................. 115 5.4 RESULTS .................................................................................................................. 117 5.4.1 In itial soil S propertigs and weather conditions during experiments ......... 117 5.4.2 EXPERIMENT 1, 3 S LABELLED FERTILIZERS ................................... 117 5.4.2. 1 Accumulated herbage dry matter yield and total S uptake and comparisons betw?gn microplot and small plot experiments. 117 5.4.2.2 Total recovery of S activity in soil and plants ................... 121 5.4.2.3 Extractable Sin soil (CaP-S) ................................................ 121 5.4.2.4 Extractable 35 S in soil (CaP-35 S) ........................................ 122 5.4.2.5 Total organic 35s ............................ 35 .................. ............... 124 5.4.2.6 Carbon bonded and ester sulphate S .............. ................ 124 5.4.2.7 Prediction of the extent of S transformation .......................... 135 5.4.3 EXPERIMENT 2, INVERSE ISOT OPIC DILUTI ON ................................. 137 5.4.3.1 Pasture dry matter :gield and total sulphur uptake ................. 137 5.4.3.2 Total recovery of 3 Sin soil and plant .. ............................... 137 5.4.3.3 Recovery oyW,ected 3 5 Sin soil S fractions .......................... 138 5.4.3.4 Extractable S (CaP-S) ....................................................... 138 5.4.3.5 Total organic 35s .......... .................................................. ..... 140 5.5 DISCUSSION ............................................................................................................ 145 5.5.1 Pasture yield and S uptake ...................................................................... 145 5.5.2 Short term fate of the fernlizer sulphur ..................................................... 145 5.5.2. 1 Uptake of 3 S labelled fertilizers .................................... ..... 145 5.5.2.2 Recovery of 35 labelled fertilizer in the top 10 cm of soil .............................................................. 146 5.5.2.3 Transformation of 35 Sin the top 10 cm of soil . ..................... 146 5.5.3 Comparison between labelled fertilizer and inverse di lution techniques ....................................................................... 148 5.5.4 The microplot technique (undisturbed soil core) ....................................... 148 5.6 CONCLUSIONS ........................................................................................................ 149 X CHAPTER 6 THE MEASUREMENT OF PLANT AVAILABLE SOIL SULPHUR 6.1 I NTRODUCTION ........................................................................ ................... ............ 1 51 6.2 OBJECTIVES ............................................................................................................ 1 52 6.3 MATERIALS AND METHODS ............ ................. ...................................................... 1 54 6.3.1 Soil and herbage analyses .............. .................................... ..................... 1 54 6.3. 1. 1 Extractable S . . . . .. .. . . ...... . . .. . . . . ... . . . . .. . ..... . ..... . . . . . . ... . . . .. . . . . . . . . . . . . . . 154 6.3. 1.2 Herbage analyses . ...... . . . . .. . . . . ... . . .... . ..... . . .... . .. . . . ... . ....... . . . . . ... . . 154 6.3.2 Statistical analyses .................. ............................ ..................................... 1 55 6.4 RESULTS AND D ISCUSSION ...................... ........................... . .................. .............. 1 56 6.4 .1 Effect of soil preparation and extractant . . . .......................... .............. ........ 1 56 6.4. 1. 1 Effect of soil preparations . . .. . . . ... . .. . . ..... . .. . . .. . . . . . . . . .. .... . . . . . . . . . .. 156 6.4. 1. 1. 1 Effect of soil preparation on extractable S ... . .. . . . . .... .... . . . . . . . . .. 156 6.4. 1. 1.2 Effect of soil preparation on extractable 35 S activity . . . ... .. . .. 157 6.4. 1.2 Effect of extract ants .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . 158 6.4. 1.2. 1 Effect of extractant on extractable S ... .... . . . .... ... .. . . . ...... . .. . . . .. . 158 6.4. 1.2.2 Effect of extractant on extractable 35 S activity . . . ... . . . . . . . . .. . . .. 158 6.4 . 2 Specific activity of 35s i n soil extracts and herbage ............. ........ .......... . 1 62 6.4 .3 Relationships bgtween the specific activity of 35s i n herbage and 3 S in soil extractable S ......... .............. .......... .......... .......... 1 62 6.5 CONCLUSIONS ........................................................... ..................................... ........ 1 70 CHAPTER 7 THE INFLUENCE OF FERTILIZER FORM ON THE FATE OF SULPHUR IN SOILS 7.1 INTRODUCTION ...................................................................... ............................ ..... 1 71 7.2 OBJECTIVES ....................................................... ................... .................................. 1 72 7.3 MATERIALS AND METHODS ... ......... . .. ......... ............... .... .................. . ... . ...... ... ........ 1 72 7.3.1 Soils ..................................................................................... .......... ........... 1 72 7.3.2 Preparation of soil cores ............. ................................................ .............. 1 73 7.3 .3 Design of the experiments ....................................... ................................. 1 73 7.3.4 Labelling fertilizer S ........... ....................................................................... 1 75 7.3.4. 1 Labelling sO containing fertilizers .. . . . . .. . . .. . . . . . . ... . . . . . . ... . . . . .. . . . . . 175 7.3.4.2 Labelling sulphate containing fertilizers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 7.3.4.3 Granulation of sO and sO !phosphate rock .. . . ... . . . .. . ... . . . . . . . . . . . . 176 7.3.5 Experimental conduct ...... .......... ......... . . ... ...... . . ................... .......... .. .. . . ...... 1 78 7.3.5. 1 General . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . ... . . . . . . .. . . . . .. . . . . . 178 7.3.5.2 L eaching events . . . . . . . . . ..... . ... ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . .. . . . . . . . 178 7.3.5.3 Soil and herbage sampling . . ... . . . . . ... . . . . .. . . . . . ... . ... . . ... . . . . . . .... . .... 178 7.3.6 Chemical analyses . ..... ....... . . ........... . . .. ........ . ........... . . ......... . .. . .................. 181 7.3.6.1 Soil, plant and fertilizer samples . . . . . . . . . . .. . . . . . . . ..... . . . . . .. . . . . . . . . . . .. 181 7.3.6.2 Sulphate retention ... .... . . . .... . . .. . . . . . . . ... .... . . . ..... . . . ... . . . . . .... . . . .. . . . .. 181 7.3.6.3 Recovery o f added fertilizer 35 sO activity from soil cores at day 0 ... . . . . . . . . . . . . . . . . . . . ... .. . . . . . . . . . . . . . . . . .. . ... . . . . . . . . . .... . . . 182 7 .3. 7 Statistical analyses .......... . .... ............... . ............ .................. .................... .. 18 2 7 .3.8 Calculation of specific rates (K) of S0 oxidation ...... . . . . . ... .. . . . .. . . . . . . . . .. .. . . . ... 183 7.4 7.5 xi RESULTS AND DISCUSSION ............... ................................... .................... . . ..... ..... 184 7.4 . 1 The percentage recovery of 35so and %SDFF in pastures as influenced by fertilizer specific activity ..................... ........... ........ . ........ 184 7.4 .2 THE INFL UENCE OF PAR TICLE SIZE AND FERTILIZER FORM ON THE FA TE OF SJ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 7.4.2.1 Recovery of 35 s in soils_ and plants ... . . ...... . .. . . . ... . ...... ... . . ... . . . 186 7.4.2.2 Plant uptake of S and 35 S and percentage of plant S derived from fertilizer (%SDFF) . . . . . . . . . . . .. . . . . .... . . . . . . .... . 189 7.4.2.3 Residua/ 35 EfJ activity and EfJ (acetone extracts) and extractable S and 35 S activity (CaP-S) in soil . . . .. . . . . . . .. . . . . . ..... 192 7.4.2.4 Immobilization into soil organic forms ........ . ...... . . . . . .... . . . . ..... .. 205 7.4.2.5 Movement of 35 S down the profile and leaching losses . . .. ... 206 7.4.2.6 Summary . . . ....... . . . . . . . . ......... . . . .... . . . . . . . . . . . .... . . . .... .. . . . . . . . .. .. . . . . . . . ... 218 7 .4 .3 FA TE OF SULPHA TE-BASED FERTILIZERS ... . . . .. . . .... .. .. . . . . . ....... . . . . . .... . . 219 7.4.3. 1 Recovery of 35 S in soils_ and plants .... .. . . .. .. .... . . . . ... ...... . ... .. .. . 219 7.4.3.2 Plant uptake of S and 35s and percentage of plant sulphur derived from fertilizers (%SDFF) . .. . . .. . . . .. . . . . ..... 220 7.4.3.3 Immobilization into organic matter . . .. . . . . . ...... .... . . .. .. .... . . . . . ...... 220 7.4.3.4 Movement of 35 s down the profile and leaching losses . . ..... 221 7.4.3.5 Summary . . . . . . ... ... . . . . ... . . . . . . . . . ....... . .... . . . ... . . . ...... .. ... . . ... . . . . . . . . . . . ... 222 7.4 .4 Comparison of sulphate-based fertilizers and elemental S ......... . ............ 223 7.4 .5 Comparison of glasshouse and field trials ................... ..................... . ....... 224 CONCLUSIONS ........................................................................................................ 231 CHAPTERS MODELLING THE FATE OF SULPHUR IN THE SOIL-PLANT SYSTEM IN GRAZED PASTURE 8 . 1 INTRODUCTION .............. ............................................................... ......... ... . ... .. ........ 232 8 .2 MATERIALS AND METHODS .................................... ..... .... . ......... .......... ........ . ......... 234 8 .3 MODEL DEVELOPMENT .......................................................................................... 234 8 .3 . 1 Water balance ............................ ................ . ......... ............................. . ...... 234 8.3.1.1 Calculating drainage volumes .... . . .. . . . . . . . . . . . . . . .. ..... . . . . . . . . . ... . . . . . . 234 8.3. 1.2 Estimating actual daily evapotranspiration . . . . . . .. . . . . . . . . . . . . . . . . . . .. 235 8 .3. 2 A single su lphate pool for plant uptake, immobilization and leached sulphate . ......................................................................... ..... 236 8 .3 .3 Modell ing the fluxes of S between various pools ................................ ...... 237 8.3.3. 1 Predicting the soil solution sulphate concentration . . . . ... .. . . . . . . . . . . . . .. . . . .. . . ... . . . .. .. . . ... . . . . . .. . . . . . . . . . . . . . . . . . . .. 239 8.3.3.2 Accounting for leaching of sulphate . .. . . . . . . . . . . . .... . . . . . . .... . . . . . . . . . . 241 8.3.3.2. 1 Layered drainage model with mobile and immobile water phases . . . . .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . 241 8.3.3.3 Plant uptake of sulphur . . . . . .. . . . . . . . . . . .. . . . .. . . . . . . . ... .. . . . . . . . . . . . . . . . .. .. . . 244 8.3.3.4 Accounting for immobilization and mineralization of soil organic sulphur . .... . . . . . . . . . .. . . . .. . ...... . . ...... . . .... . .. . . . .. . . . . . . ... 245 8.3.3.5 Executing the model . .. . . . . . . .... . . .. . . .... . . . ... . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . 246 8 .4 RESULTS AND D ISCUSSION .................................................................................. 247 8 .4 . 1 Prediction of plant S uptake on unfertilized soil . ....................................... 247 8 .4 .2 Prediction of plant S uptake on superphosphate fertilized soil ................. 248 8 .4 .3 xii 8.4.2. 1 Movement of sulphate from superphosphate into soil... ....... 248 8.4.2.2 Prediction of plant uptake on SSP fertilized plots .......... ....... 250 Prediction of 35s movement and transformation in undisturbed soil cores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 8.5 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 CHAPTER 9 SUMMARY 9 . 1 REVIEW OF LITERATURE ....................................................................................... 260 9 . 2 LABELLING TECHNIQUES ............................................................ ........................... 2 6 1 9.3 THE EFFECT OF SHEEP DUNG ON THE SHORT-TERM IMMOBILIZATION OF SOIL AND FERTILIZER S ........................... ........................................................ 261 9 .4 THE FATE OF S FROM S0 AND SSP IN SOIL ....................................................... 2 62 9.5 MEASUREMENT OF PLANT AVAILABLE SOIL SULPHUR .................................... 263 9 . 6 INFLUENCE OF FERTILIZER FORM ON THE FATE OF S IN SOIL ....................... 2 64 9 .7 MODELLING THE SHORT-TERM FATE OF FERTILIZER S IN SOIL .............. . ...... 264 9.8 SUGGESTIONS FOR FURTHER RESEARCH ......................................................... 265 BIBLIOGRAPHY ............................. ................... ..................................................................... 267 APPENDICES ........................................................................................................................ 298 Figure 2.1 Figure 2.2 Figure 2.3 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 5.1 Figure 5.2 Figure 5.3 Figure 5.4 Figure 5.5 Figure 5.6 xiii LIST OF FIGURES A conceptual flow diagram of the main forms and transformations of sulphur in the soil-plant-animal system (MJ. Hedley, 1990. Lecture notes from Soil Fertility and the Environment, Soil Science Department, Massey University, New Zealand) . ............. 17 Simplified sulphur cycle showing gains and losses in the CFAS model (Sinclair and Saunders, 1984) ............................................................................................................ 38 Map of New Zealand showing broad climate zones which affect recommended fertilizer S0 particle size ranges (Boswell and Swanney, 1988) .................................. 44 A schematic illustration of nutrient transfers in a grazed hill-country pasture (Saggar et al., 1990a). ???????????????????????????????????????????????????????????????????????????????????????????????????????????????????? 82 Radioactivity per unit area in harvested pasture at four radial distances (R1, R2, R3 and R4) from radioactively labelled dung applied at three rates (D1, D2 and D3); A= 32p and B = 35s at the first (30 days) harvest and C = 35s at the second harvest (60 days) . ............................................................................................................................ 95 Total percent recovery of 32p and 35s in pasture harvested at four radial distances (R1, R2, R3 and R4) from radioactively labelled dung applied at three rates (D1, D2 and D3); A = 32p and B = 35s at the first harvest (30 days) and C = 35s at the second harvest (60 days) . ......................................................................................................... 96 Percentage of pasture P and S derived from dung at four radial distances (R1, R2, R3 and R4) from dung applied at three rates (Dl , D2 and D3); A= %PDFD and B = - %SDFD at the first harvest (30 days) and C = %SDFD at the second harvest (60 days) (calculated using specific activity of dung extracts) . ................................................... 97 Percentage of pasture P and S derived from dung (%PDFD and %SDFD) at increasing distances from dung (RI, R2, R3 and R4); %SDFD-1 and %PDFD were calculated from the first harvest data (30 days) and %SDFD-2 was calculated from the second harvest data (60 days) (calculated using the specific activity of the (A) exchangeable P and S and (B) the total dung P and S ............................................................................ 98 A cross sectional diagram of the injector system used to inject soil cores with radioisotope in the inverse isotopic dilution experiment (Hedley and Tillman, personal communication) .......................................................................................................... 114 Cumulative plant S uptake (fertilizer plus soil S), amount of plant S cumulatively derived from 35s labelled fertilizer (SDFF), extractable CaP-S at three soil depths and amount of S applied (microplots) . ....................................................................... 120 Percentage of applied 35s labelled fertilizer recovered as total soil S in three soil layers and cumulative 35s uptake by pasture . .......................................................... 126 Percentage of apQlied 35s present as extractable S (CaP-35s) in three soil layers, cumulative plant 35s uptake and 35s taken up by pasture at each of five harvests. 127 Percentage recovery of 35s applied as fertilizer in organic S in three soil layers . ... 128 Distributions of 35s ester sulphate (Ai: and carbon bonded 35s (B) in three soil layers (value expressed as percentage of Ss applied) . ............................................ 129 Figure 5.7 Figure 5.8 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 7.1 Figure 7.2 Figure 7.3 Figure 7.4 Figure 7.5 Figure 7.6 Figure 7.7 Figure 7.8 Figure 7.9 Figure 7.10 xiv Distribution of 35s ester sulphate (A) and carbon-bonded 35s (B) in three soil layers (value expressed as percentage of total 35s remaining in soil core, 0-10 cm) . ........ 130 Cumulative S (A) and 35s (B) taken up by plants at five harvesting times; observed vs. predicted ................................................................................................................ 131 Specific activity of 35s in plants and soil extracts (CaCl-S and CaP-S) from superphosphate treated cores at five harvests for three soil depths . .......................... 166 Relationships between specific activity of 35s in plants and HI-reducible 35s in soil extracts from the !QI2 layer; (A)= CaCl-S and (B)= CaP-S . ..................................... 167 Relationships between specific activity of 35s in plants and HI-reducible 35s in soil extracts from the top plus middle layers; (A) = CaCl-S and (B) = CaP-S . ................ 168 Relationships between specific activity of 35s in plants and HI-reducible 35s in soil extracts from the top plus middle and bottom layers; (A) = CaCl-S and (B) = CaP- S .................................................................................................................................. 169 The effect of S0 particle size on the cumulative S taken up by pasture grown on Ramiha (A) and Tokomaru (B) soils .......................................................................... 193 The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative S taken up by pasture on Ramiha (A) and Tokomaru (B) soils ..................................................................................................... 194 The effect of S0 particle size on the cumulative dry matter yield of pasture on Ramiha (A) and Tokomaru (B) soils . ...................................................................................... 195 The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative dry matter yield of pasture on Raimiha (A) and Tokomaru (B) soils ..................................................................................................... 196 The effect of S0 particle size on the cumulative percentage 35s uptake by pasture on Ramiha (A) and Tokomaru (B) soils . ......................................................................... 197 The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative 35s taken up by pasture on Ramiha (A) and Tokomaru (B) soils ..................................................................................................... 198 The effect of S0 particle size on the percentage of cumulative plant S derived from fertilizer (%SDFF) on Ramiha (A) and Tokomaru (B) soils . .................................... 200 The effect of granulation of S0 with or without phosphate rock on the percentage of cumulative plant S derived from fertilizer (%SDFF) on Ramiha (A) and Tokomaru (B) soils ............................................................................................................................. 201 The effect of 35so particle size on the percentage of oxidized 35so cumulatively taken up by pasture on Ramiha (A) and Tokomaru (B) soils . ................................... 201 The effect of granulation of S0 with or without phosphate rock on the percentage of oxidized 35s0 cumulatively taken up by pasture (SS/PR and SS/gr compared with SS) . ............................................................................................................................ 202 Figure 7.11 Figure 7.12 Figure 7.13 Figure 7.14 Figure 7.15 Figure 7.16 Figure 7.17 Figure 7.18 Figure 7.19 Figure 8.1 Figure 8.2 Figure 8.3 Figure 8.4 Figure 8.5 Figure 8.6 Figure 8.7 Figure 8.8 XV The effect of S0 particle size on the recovery of extractable 35s (CaP_35s) in soil cores (0-10 cm) in Ramiha (A) and Tokomaru (B) soils . .......................................... 208 The effect of granulation of S0 with or without phosphate rock on the recovery of extractable 35s (CaP_35s) in soil cores (0-10 cm) (SS/PR and SS/gr compared with SS) in Ramiha (A) and Tokomaru (B) soils . .............................................................. 209 Total recoveries of 3Ss in three layers of Ramiha soil 180 days after fertilization with two particle sizes of 35s0 ........??.?......???..?.??......?.....?...???.??.....?.????..????.????..?...?..?.?..?.??.. 213 Total recoveries of 35s in three soil layers of Tokomaru soils 180 days after fertilization with three particle sizes of 35s0 ..........?.........?.......?...............?......????...... 214 Total recoveries of 35s in three soil layers of Ramiha soil 180 days after fertilization with fine 35so (SS) granulated with and without phosphate rock. ............................ 215 Total recoveries of 35s in three soil layers of Tokomaru soil 180 days after fertilization with fine 35so (SS) granulated with and without phosphate rock . ........ 216 Cumulative dry matter yield (A) and S uptake (B) of pasture on Ramiha and Tokomaru soils; GP and SSP vs. SS . ......................................................................... 225 Cumulative 35s u?take (value expressed as percentage of 35s applied) by pasture (A) and %SDFF (B); 5s labelled GP and SSP vs. 35so labelled SS . ............................ 226 Simple relationships between cumulative 35s uptake (A) (value expressed as percentage of 35s applied) and S uptake (B) by plants against time (a comparison between sulphate-S and elemental S fertilizers) ........................................................ 227 A simple conceptual dynamic sulphur model for grazed pasture .............................. 233 Hierarchical order of the processes in the S-cycle ..................................................... 238 Comparison of predicted and observed plant S uptake on (A) control (unfertilized) and (B) SSP fertilized soil cores (experimental details Chapter 5). The coefficient of determination (R2) represents the variation in observed data accounted for by the model prediction (PV stands for the less mobile fraction of soil water; 1-a) ............ 253 The measured rainfall and predicted drainage occurring from the top 100 mm of Tokomaru soil during the experiment conducted in Chapter 5. The dotted line shows the disappearance of surface applied superphosphate (30 kg S ha-1) as it dissolves and moves into soil. The solid line shows the predicted accumulated leaching loss of S from top 100 mm ........................................................................................................ 254 Comparison of predicted (output every 5 days) and observed amounts of CaP-S in each soil depth in SSP fertilized cores .. ..................................................................... 255 Comparison of predicted and observed percentages of added 35s recovered in CaP-S fractions from different soil depths . ........................................................................... 256 Comparison of predicted (output every 5 days) and observed percentage of added 35s recovered in pasture plants ......................................................................................... 257 Comparison of predicted and observed percentage of added 35s recovered as organic S in different soil depths ............................................................................................. 258 Table 2.1 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 5 . 1 Table 5 .2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 5 . 10 xvi LIST OF T ABLES Some standard and alternative S fertilizers in New Zealand (Boswell and Swanney, 1986) . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . 40 Mean S and P concentrations, total radioisotope activity and specific activity (SA) of 35s or 32p in dung and extracts of dung (results expressed on dung dry weight) . ..... 84 Pasture dry matter yield as influenced by dung application rates and radial distances of pasture from dung . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Pasture S and P concentration, S and P uptake of pasture and cumulative S uptake as influenced by dung application and radial distances of pasture from dung . .. . . . . . . . . . . . . 100 The percent distribution of 35s and 32p taken up by pasture at different radial distances from 35s and 32P labelled dung applied at various rates . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Pasture dry matter yield (PART A), sulphur concentration (PART B) and total pasture S uptake (PART C) as influenced by dung and fertilizer applications at five harvests . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . ....... . . . . . . . . . . . . . 103 Calculated percentage of a paddock annually influenced by dung considering camping and noncarnping behaviour of sheep . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Characteristics of 35s labelled fertilizers used in this study .... . . . . . . . . . . . . . . . . . ... . . . . . . . . . .. . . . 1 1 1 Harvesting schedules of herbage from micro-plots at each harvest and number of soil cores destructively sampled . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 1 Chemical properties of sulphur in three soil layers collected before the experiment. 1 13 Chemical analyses for soil, plant and fertilizer samples .. . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . 1 15 Cumulative dry matter yield and total pasture S uptake (soil plus fertilizer S) in microplots and small plot experiments during spring-summer seasons (October, 1985 - February, 1986) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. 1 19 The percentage of total 35s labelled fertilizers appearing in soil organic and inorganic 35s fractions recovered from 0-1 0 cm depth of undisturbed field soil cores at five harvesting times . .. . . . .. . . . . . . . . . . . . . . . . . ... . . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 1 32 Average extractable 35s (CaP)Ss) expressed as a percentage of total 35s activity applied (PART A) and as a percentage of total 35s remaining (PART B)in three soil layers at five harvests . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . 133 Proportion of 35s labelled fertilizers appearing in soil organic and inorganic fractions; amount expressed as percent of total 35s recovered from the 0-10 cm depth of undisturbed soil cores at five harvesting times . .. . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 Observed plant S uptake and predicted values using relationships in equation 4 above . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 1 36 Dry matter yield (DM), total S uptake, cumulative dry matter yield and S uptake and specific activity of 35s in herbage from two harvests . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . 141 Table 5 . 1 1 Table 5.12 Table 5.13 Table 6.1 Table 6.2 Table 6 .3 Table 6.4 Table 6.5 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Table 7.6 Table 7.7 Table 7.8 Table 7.9 Table 7.10 Table 7.11 xvii Mean recovery of 35s from herbage and soils (0-10 cm depth) at two harvests, 30 and 60 days after injection of 35so 4 = into soil... . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . ............ . . .. . . . . . 142 Mean recovery of CaP-35s and specific activity and total amount of CaP-S in the 0-10 cm soil depth . .. . . . . . . . . . . . . . . . ............ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . .. . . . . . .. .. . . . . 143 Mean recovery of soil 35s organic S fractions from the 0-10 cm soil depth . . . . . .. ..... 144 Concentrations of HI-reducible S in CaCl-S (Part A) and CaP-S (Part B) extracts from three soil layers using two soil preparations . ............................................................. 159 Concentration of HI-reducible 35s in CaCl-S (Part A) and CaP-S (Part B) extracts from three soil layers using two soil preparation techniques . .................................... 160 Concentration of HI-reducible S (Part A) and 35s activity (Part B) in CaCl-S and CaP- S extracts from freeze-dried soils taken from three soil layers . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 The specific activity of 35s in CaCl-S and CaP-S extracts of field-moist and freeze- dried soils and plants . . .. . . . . . . . . . . . . . . . . . . . . . . ...... ......... . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ........ . . .. . . . . . 164 Simple linear relationshi?s, y=bx, between specific activity of total 35s in plants (y) and specific activity of 3 S in extracts from different soil layers (x) . ....................... 165 General properties of the soils used in the studies . .................................................... 174 The forms and distribution of soil sulphura in three soil layers collected from field sites before the experiment ......................................................................................... 175 Characteristics of 35so labelled fertilizer used in this study . .................................... 177 Characteristic of 35so4 = labelled superphosphate and gypsum used in this study . . 177 Treatments , soil sampling and herbage harvesting schedule for treatments with different particle sizes of elemental sulphur in the glasshouse trial . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Treatments, soil sampling and herbage harvesting schedule for treatments with different particle sizes, S0 in the field trials. .. . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . .. . . . .. . . . . . . . . . . . . . . . . . . . 180 Treatments and soil sampling and herbage harvesting schedule for treatments with 35s labelled gypsum and superphosphate fertilizers in the glasshouse trial. . .................. 180 Analyses for S and related 35s activities in soil, plant and fertilizer materials . .. . . . . . . 181 Effect of fertilizer specific activities on 35s recovery in pastures and percent of plant S derived from fertilizers (%SDFF) . .......................................................................... 185 Recovery of labelled 35s fertilizers in pasture, soil and total recovery in Ramiha soil cores in glasshouse and field trials (average of three replications from individual microplots) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . .......... . . . . . . . . . . 187 Recovery of labelled 35s fertilizers in pasture, soil and total recovery in Tokomaru soil cores in glasshouse and field trials (average of three replications from individual microplots) . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . .............. . . . . . .. . . .. 188 Table 7.12 Table 7.13 Table 7.14 Table 7.15 Table 7.16 Table 7.17 Table 7.18 Table 7.19 Table 8.1 Table 8.2 xviii Percentage applied 35so recovered as residual 35s0 (acetone extracts), in 0-3 cm soil depth (PART A) and extractable 35s (CaP-S) (PART B) in soil cores (0-10 cm) in glasshouse trial s and after 1 80 days in the field trials (average of three replications) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 210 Percentage applied S0 recovered as residual S0 in 0-3 cm layers (PART A) and total amounts of extractable S, CaP-S, (PART B) in soil cores (0-10 cm) in glasshouse trials and after 180 days in the field trials (average of three replications) . .. . . .. . . . . . . . . . . . . . . . . . . . 2 1 1 Percentage recovery o f total 35s (PART A) and organic 35s (PART B) i n soil cores (0-10 cm) at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . ..... 212 Percentage recovery of 35s in leachates (PART A) and total amounts of sulphur and HI-reducible S (PART B) in leachates at five leaching events in glasshouse trials . . 217 Total recovery of 35s in soil, plant in two soils resulting from applications of 35s labelled gypsum and superphosphate after 90 days . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228 Fractions of 35s in soils (% recovery) 90 days after applying 35s labelled gypsum (GP) and superphosphate (SSP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . ....... 229 Distribution (% recovered) of total 35s (TT), organic 35s (Org) and phosphate extractable 35s (CaP-S) in three soil depths of two soils 90 days after application of 35 S labelled gypsum (GP) and superphosphate (SSP) . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . ..... 229 Percentage reco?ery of 35s labelled fertilizer sulphur in leachates (PART A) and m? reducible S in leachates (PART B) at five leaching events after 35s labelled gypsum (GP) and superphosphate (SSP) applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . .. . . . . . . . . . . . .. ..... 230 Freundlich type equation fitted to laboratory and field measured isotherms . . . . . . . ...... 240 The simulated time taken for 30 kg S ha-1 to diffuse into Tokomaru silt loam and depth of sulphate movement (radius of outer hemispherical shell from center of granule) . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ... . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 Appendix 3. 1 Appendix 3 .2 Appendix 3.3 Appendix 4 .1 Appendix 4.2 Appendix 4.3 Appendix 4.4 Appendix 4.5 Appendix 4.6 Appendix 4. 7 Appendix 5. 1 Appendix 5.2 Appendix 5.3 Appendix 5.4 Appendix 5.5 Appendix 5.6 Appendix 5.7 xix LIST OF APPENDICES Results from NaOBr digestions of plant material compared with median values published by the International Plant-Analytical Exchange, Wageningen Agricultural University, Netherlands . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. . . . . 298 Calculation for 35s enrichment of fertilizer materials ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 The manufacturing of 35s labelled superphosphate ... . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . .. . . . . . 301 Activities of 35s and 32p in pasture per unit area (square centimetre) as influenced by different application rates of 35s and 32P labelled dung and distances of pasture from dung . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . 302 Influence of dung application rates and distances of pasture from 35s and 32p 35 32 labelled dung on the percent recovery of S and P by pasture ... . . . . . . .. . . . . . . . . . . . . . 303 Specific activity of 35s or 32p in pastures (S.A.) and the percentage of plant S and P derived from dung (%SDFD and %PDFD) calculated using the S .A. of 35s and 32P in the total pool (TP) and extractable pool (EP), CaP-S and Olsen-P, of dung samples . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 304 Percentage of plant S (PART A and B) and P (PART C) derived from dung applied at different rates (%SDFD and %PDFD)a in pasture at different radial distances away from dung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . .. . . . . . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Diffusion of sulphate and phosphate in soil . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . ... 306 Calculation for area of a paddock influenced by sheep dung . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Weekly rainfall (Rain), drainage water (Drainage), average maximum (Max) and minimum (Min) temperature during the field trial period (June 27, 1985 - August 29, 1985) . .. . . . ? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 308 Recovery of 35s from 35s labelled gypsum and microfine S0 fertilizers after application onto the surface of soil cores . . .. . . . . . . . . . . . . . .. .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Cumulative S uptake by pasture and the percentage and amount of plant S derived from fertilizers (%SDFF) with time . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . 3 10 Total CaP-S in three soil layers; CaP-S as percentage of S derived, PART A, from fertilizers (number in parentheses) and PART B, native extractable S and PART C, total extractable S in 0- 10 cm depth of soil cores . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . 3 1 1 Cumulative 35s uptake in herbage and total amount of 35s in 0-10 cm soil depth at five harvesting times . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3 12 Recovery of total 35 S in three layers of soils at five harvest times . . .. . . . . . . . . . . . . . .. . . 3 13 Organic 35s expressed as a?rcentage of total 35s activity applied (PART A) and as a percentage of total 55s remaining (PART B) in three soil layers at five harvests . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. ..... . . . . . .. . . .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 14 Carbon-bonded 35s expressed as apercentage of total 35s activity applied (PART A) and as a percentage of total 35s remaining (PART B) in three soil layers at five harvests . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 15 Appendix 5.8 Appendix 5.9 Appendix 6. 1 Appendix 7. 1 Appendix 7.2 Appendix 7.3 Appendix 7.4 Appendix 7.5 Appendix 7.6 Appendix 7. 7 Appendix 7.8 Appendix 7.9 Appendix 7.10 Appendix 7. 1 1 Appendix 7.12 Appendix 7 .13 XX 35s Ester sulphate expressed as a percent of total 35s activity applied (PART A) and as a percentage of total 35s remaining (PART B) in three soil layers at 5 harvests . .. . . . . . ..... . . .......... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . 316 Weekly amount of rainfall (Rain) and drainage water (Drainage) , average maximum (Max) and minimum (Min) temperature during the field trial period from October 1985-March 1986 . ...................................................................... ......... .... 3 17 Total S and 35s activity taken up by herbage from 35s labelled superphosphate and microfine S0 treated plots in five harvests . ................ ...................... ..................... 318 Cumulative sulphur taken up by pastures (PART A) and cumulative dry matter yield, DM, (PART B) in glasshouse trials (average of seven replications) . . ........ 3 19 Cumulative sulphur taken up (PART A) and cumulative dry matter yield, DM, (PART B) of pastures in Field trials; (average of three replications) . .................. 320 Sulphur concentration of pastures in glasshouse (PART A) and field (PART B) trials as influenced by different particle size of elemental S . .............. ... ............... 321 Cumulative percentage of 35s taken up by pastures in glasshouse (PART A) and field (PART B) trials at six sampling times . ......................................................... 322 Percentage of plant S cumulatively derived from fertilizers (%SDFF) in glasshouse (PART A) and field (PART B) trials ........................................... .............. . . . . . . .. . . .. 323 Percentage of oxidized elemental 35s0 cumulatively taken up by pastures in glasshouse and field trials; (average of three replications) . ................... ............... 324 Percentage recovery of extractable 3Ss (CaP-S) in three soil layers of Ramiha and Tokomaru soils at six sampling times in glasshouse trials and after 1 80 days in the field trials (average of three replications) . ................................................ . . . . . . . ...... 325 Concentration of soil extractable S (CaP-S) in three layers of Ramiha soil at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications) . ....................................... ... ......................................... . . . . . .. ....... 326 Concentration of soil extractable S (CaP-S) in three layers of Tokomaru soil at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications) . ............................................... .................................... . . . . . . . ....... 327 Percentage recovery of total 35s in three soil layers in glasshouse trials and after 180 days in the field trials (average of three replications) . ...................... . . . . . . . .. .. . . 328 Sulphur concentration, cumulative sulphur uptake and dry matter yield of pastures on two soils at three samplings in glasshouse trials . .... ............................ . . . . . . . . ..... 329 Cumulative 35s taken up by pasture and percentage of]Jlant sulphur derived from labelled fertilizers (%SDFF) after application of 35s labelled gypsum and superphosphate in glasshouse trials . . ..................................................................... 330 Phosphate extractable S concentration and amounts present in three soil depths of two soils 90 days after application of gypsum and superphosphate ...................... 331 Appendix 7.14 Appendix 7.15 Appendix 7.16 xxi Sulphate retention (%) and simple relationships between amounts of adsorbed sulphate (mg S kg-1) and sulphate concentrations in solution (mg S r1) for three layers of Ramiha and Tokomaru soils . . . . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . 332 Recovery of 35so from two sizes of elemental S at diD' 0 through NaHco3 dry digestion (total 35s) and acetone extraction (total 3:>so activity and S0) after application onto the surface of soil cores . .............................. ..................... . . . ....... 333 Weekly rainfall (Rain) and drainage water (Drainage), average maximum (Max) and minimum (Min) temperature during November 1987 - June 1988 . . .............. 334 1 CHAPTER 1 INTRODUCTION Sulphur is an essential e lement for plant growth and is required by plants mainly for protein and enzyme synthesis. lt is an important component of the nitrogenase enzymes which are involved in nitrogen fixation (Lehninger, 1 982) . Pasture production in New Zealand depends on atmospheric nitrogen fixation by legumes. Traditionally, fertilizer use in New Zealand pasture soils was determined mainly by phosphorus ( P) requirements. Single superphosphate (SSP) , 9% P and 1 1 % S, has been the major fertilizer used to correct P deficiency and it is fortuitous that in the process sufficient sulphur (S) was applied (Boswell, 1 985, Boswell and Swanney, 1 986; Till et al., 1 987) . Currently, New Zealand imports S worth approximately NZ$40 million every year mainly for the SSP fertilizer industry. Recent studies on the mass balance of the S cycle in grazed hil l country pastures, in the North Island of New Zealand, have shown that in areas where S is applied annually as SSP, large S leaching losses can occur (up to 70% of applied S per annum) which are mostly dependent upon the rate of application (Saggar et al. , 1 990a, 1 990b). I n other areas of New Zealand, particularly i n high rainfall areas, elemental S (S0) fertilizers which slowly oxidize to release sulphate-S have been recommended to reduce S leaching foss (Sincfair and Saunders, 1 984). ft has long been established that elemental S (S0) is a su itable S fert i l izer form for New Zealand pastoral soils (Ludecke , 1 965 ; Sincfair et al., 1 985; Wif f iams and Morton, 1 985 ; Swanney et al. , 1 988; Boswell and Swanney, 1 99 1 ) . Elemental S is an important S source because of its high analysis and slow rate of release of plant-available sulphate-S, which can be control led by its particle size distribution. Under severe leaching conditions the greater efficiency of S0 over so4= is clearly obvious and general recommendations (Sinclair and Saunders , 1 984) for S0 use are avai lable based on slow release (S0 oxidation) concepts (Swanney et al., 1 988) . Apart from studies measuring the amounts of S0 residues in soil (Lee et al., 1 987), no studies have been conducted to examine the fate of S0 in soils and whether this fate is the same as s from so4= based fertilizers. In contrast to excessive use of S fertilizer, in the economic climate of the past six years many farms in New Zealand have now not received fert il ize r or have received S-free phosphatic fert i lizers , such as diammonium phosphate, reactive phosphate rocks and phosphoric acid 2 acidulated reactive phosphate rocks. This reduced S input, combined with the potential of New Zealand soils to become S-deficient by leaching, has created some striking responses to reapplied gypsum (M.J. Hedley, personal communication) and a general need for re-assessing agronomic requ irements for sulphur (Sinclair et al. , 1 985) . Very little information o n the S status of New Zealand soil is available , however, with current knowledge it is difficult to predict accurately the S fertilizer requirements of pastoral systems varying widely in soil fertility status. Many questions remain to be answered prior to determining the efficiency with which soil and ferti l izer S is uti l ized in pastoral systems; e .g . what are the rates of mineral ization of soil organic S and immobilization of ferti l izer S into o rganic S reserves in soils under d ifferent environmental conditions and fa rming systems, and i n the long term can the u se of the insoluble slow release S0 lead to more efficient S fertilizer use ? I n general , efficient use of fert i l izer S occurs when the rate at which S is cycl ing in the so i l-plant-animal system is maximized and when the number of cycles completed are large compared to the rate of non? product losses from the cycle. The fate of S applied to pastoral soils in sulphate-S fertilizer form (gypsum or superphosphate) was studied in drier areas of New Zealand (Gregg, 1 976) and Australia (Till and May, 1 970a , 1 970b) but there remains a lack of understanding of the fate ( i .e . movement in soil profi le , leaching losses, incorporation into soil o rganic S and the forms of available soil S utilized by pasture plants) of applied S0 fertilizers in pastoral soils in moisture regimes l ike those of the central North Island of New Zealand. Such research is important as S becomes an increasingly costly input into New Zealand pastoral farming. The objective of this thesis is to develop radioisotope techniques to study the fate of sulphate? based and S0 fertilizers in field soils and having developed these techniques to i nvestigate the factors which influence the fate and plant availability of these fertil ize rs , in order to provide information that can be used to formulate and improve fertilizer recommendations for pastures. This thesis comprises n ine chapters. Following this introduction are a review of literature on aspects of su lphur cycling in grazed pasture systems and a chapter on the materials and methods used. Methods for radioactively labelling fertilizer sulphur were developed and were employed to produce fert i l izers for studies of the short-term fate of ferti l izer S which are discussed in Chapter 5 and 7. The main emphasis of this thesis is focused on the short-term fate of different particle s izes of S0 applied onto undisturbed soil cores cut from permanent ryegrass/clover pasture of contrasti ng fert i l i ty status (Chapte r 5-7) . These particle size diameters range from less than 0 .0 1 0 mm to 0 .500 mm. Sulphate-based S, app l i ed as gypsum and superphosphate, was also used as a refere nce in the studies. The effect of sheep dung on the short-term immobilization of soil and fertilizer S is discussed In Chapter 4. 3 A preliminary computer simulation model describing S cycling was developed (Chapter 8) in an effort to explain the obse rved transformation of S and 35s in o rder to calcu lat e actual mineralization and immobilization rates in an environment where sulphate leaching occurs. General conclusions and implications are g iven in Chapter 9 . 4 CHAPTER 2 A REV IEW OF LITERATURE ON ASPECTS OF SULPHUR CYCL ING IN GRAZED PASTURE SYSTEMS 2.1 INTRODUCTION A pasture system revolves around the functioning of several components: soils , plants, grazing animals and their residues and soil micro- and macro- flora and fauna. These components are responsible for recycling nutrients from senescing materials to plant available forms again. An understanding of the cycling of the specific plant nutrient, sulphur, requires an understanding of the functioning of the individua l components. Tracing su lphur's movement through soils, plants and animals i nvolves a multiplicity of complex reactions, transfers and transformations (Whitaker, 1 970) . Naturally, there is no one single pathway for the cycling of an element in a grazed pasture system. Several pathways exist. The objectives of this chapter are to review and develop an understanding of the nature of S and factors affecting its circulation through and losses from soil-plant-animal systems with particular emphasis on grazed pasture soils. 2.2 NATURE, FORMS AND DISTRIBUTION OF SOIL SULPHUR I n the late 1 970s, a greater awareness of the importance of S in crop production, particularly protein synthesis , and of soil as a source of S for plants resulted in several studies of the nature and forms of S in soils (Metson, 1 969, 1 979a, 1 979b, 1 979c; Williams, 1 975 ; Halstead and Rennie, 1 977; Biederbeck, 1 978). A knowledge of the forms and amounts of soil S is essential to ou r understandi ng and interpretation of S cycling. lt should be stated at the outset that our understanding of the forms and amounts of soil S is l imited by current analytical and separation techniques. Most specific S compounds associated with both organic and inorganic soil components cannot be identified without chemical treatment and separation or extraction from soil. The soil S cycle can only be rep resented by the dominant f ract ions that can be measured and will change as new techniques provide more information on the natural chemical structures of soil organic S . 2.2.1 Total sulphur The total S content of soils has been recently reviewed by Metson ( 1 969, 1 979a, 1 979b, 1 979c) , B lair ( 1 979) and Freney and Wil l iams ( 1 983) . S, the th irteenth most abundant 5 element, comprises about 0.052% of the Earths crust (Day, 1 963) . In nature, this element with six valance electrons can exist in a variety of o rganic and inorganic combinations, in so lid, solution or gaseous forms and in various states of oxidation ranging from -2 (sulphide) to +6 (sulphate) . The total S content of soils ranges from approximately 0 .002 to 5% (w/w). High levels are found in tidal marsh soils, where sulphide accumulates, in the soils in arid areas containing ino rganic sulphates such as gypsum, and soils subjected to severe i ndustrial pollution (Syers and Curtin, 1 987) . In organic soils (e.g. peats) total S may exceed 0 .5% dry weight (Halstead and Rennie, 1 977) . Most agricultural soils or mineral soils have S contents ranging from 50 to 1 000 mg kg-1 in the surface 1 5 cm (Freney and Williams, 1 983; Syers and Curtin, 1 987) . The proportions of S in organic or inorganic form vary according to so il type, depth in profi le (Wil l iams, 1 97 4), climate and cultural conditions (Bettany et al. , 1 979, 1 980). Sulphate-S (SO 4 =) is the form most available to plants. 2.2.2 Inorganic sulphur In the su rface horizons of most agricultu ral soils, i norganic S occurs almost ent i rely as sulphate but represents a relatively small proportion of total S, less than 25% of total S in most agricultural soils (Halstead and Rennie , 1 977) and more commonly less than 5%. lt is derived from wet and dry deposition of mainly sulphates and sulphur dioxide, weathering of soil parent rocks (ox idation of reduced inorganic fo rms of S, e .g . sulphide) and oxidation (termed mineralization) of organic S (Roy and Trudinger, 1 970) . Weathering reactions are thought to be a minor input of S in current topsails (Metson, 1 979a) . This is mainly because mineral sulphides are quickly weathered in aerobic, topsoil environments. In general , inorganic soil S includes easily soluble sulphate in soil solution , sulphate adsorbed on positively charged surfaces of soil particles, insoluble sulphate e.g. insoluble sulphate eo? precipitated with CaC03 and reduced inorganic compounds e.g. sulphides (Brown, 1 982) . 2.2.2.1 Readily soluble sulphate-S General ly, the surface of most agricultural soils contain only small amounts of this fraction although in semi-arid areas where evapotranspiration exceeds drainage, or in poorly drained conditions, high levels of sulphate may accumulate ( Freney and Williams, 1 983; Blakemore et al. , 1 968) . 6 The occurrence of only small amounts of soluble su lphate in soils has been explained by Metson (1 979a) to result from: 1 . Retention on soil colloids as adsorbed su lphate in anion retentive soils; 2. Leaching downward or laterally out of the soil profile; 3. Utilization for nutrition of plants and microorganisms; 4. Precipitation as insoluble sulphate. Soil sulphate levels are often subject to seasonal fluctuation depending upon the net balance between addition from rainfal l , irrigat ion water, m ineral izat ion of o rgan ic matter, applied fertilizers , and losses from leaching and plant and micro-o rganisms uptake (Wil l iams, 1 968 ; Staunes, 1 985). Data from Nguyen ( 1 982) , Cornforth et al., ( 1 983) , Nguyen et al., ( 1 989a, 1 989b) and Ghani et al. ( 1 990) confirm the importance of these processes. Nguyen ( 1 982) , Nguyen et al. , ( 1 989a, 1 989b) and Comforth et al. ( 1 983) found that in the North Island (New Zealand) the amounts of extractable soil sulphate present in spring are lower than in autumn, possibly due to the increase in leaching loss of sulphate and the slow rate of mineral izat ion during the winter time. Ghani et al. ( 1 990) found that amounts of so il sulphate can decrease in short spaces of time, particularly, after rainfall events causing drainage. 2.2.2.2 Adsorbed sulphate-S As mentioned above, the concentration of sulphate in the soil solution, an important factor in influencing plant uptake and leaching losses , is affected by several factors i ncluding sulphate adsorption-desorption and S mineralization-immobil ization reactions resulting from m icrobial activity and p lant uptake. Sulphate adsorption-desorption is therefore an impo rtant process affecting the rate of S cycling (Johnson and Todd, 1 983; Fu ller et al. , 1 985). Soils vary widely in their capacity to adsorb sulphate . Sulphate adsorption is primarily related to soil aluminium and i ron sesquiox ide content (Harward and Reisenauer, 1 966; Tabatabai, 1 987) . Heavily weathered iron- and aluminium- rich soils usually have greater adsorption capacities than less weathered soils (Biakemore et al., 1 968) . Tabatabai ( 1 987) outlined the following factors that influence sulphate adsorption : 1 . Clay content and type of clay mineral. Adsorption of soil sulphate increases with the clay content of the soils. Capacities of hydrogen saturated clays for sulphate adsorption are kaolinite > illite > bentonite. 2. Hydrous oxides. Hydrous oxides of AI and to a lesser extent of Fe, show marked 7 tendencies to retain su lphate . These compounds are probably responsible for most of the sulphate adsorption in many New Zealand soils. 3. Soil depth. Sulphate adsorption is often greater in subsoils due to the presence of more clay, AI and Fe. Subsoils commonly have lower concentrations of other anions (e.g. H2Po4- competing for sorption sites (see point 7, below)) . 4 . Effect of pH. Sulphate adsorption is favoured by strongly acidic conditions which protonate hydrous oxides into positively charged (M-OH2) + groups. lt becomes almost negligible at pH >6 when hydrous oxides lose their net positive charge . 5 . Su lphate concentration and temperature. The amount of sulphate adsorbed is concent rat ion and temperature dependent. Adsorbed su lphate is in kinetic equilibrium with su lphate in solution. Temperature has a relatively small effect on sulphate adsorption by soil. 6. Effect of time . Sulphate retention increases with the length of time it is in contact with adsorbing substances. 7. Presence of other anions. Sulphate is generally considered to be weakly held with the strength of ret ention decreasing in t he o rde r hydroxyl > phosphate > su lphate=acetate > n it rate=chloride . Phosphate wil l d isplace o r reduce the adsorption of sulphate but su lphate has little effect on phosphate (Bolan et al. , 1 986) . 8 . Effect of cations. The amount of sulphate retained is affected by the aff inity of the associated cation of the salt or by the exchangeable cations on the soil surface. I ncreased surface positive charge resu lts in a greater amount of sulphate in the diffuse double layer of cations and anions at clay and organic matter surfaces (Curtin and Syers , 1 990) . This effect follows the lyotropic series: H+ > sr2+ > Ba2+ > ca2+ > Mg2+ > Rb+ > K+ > NH4 + > Na+ > u+, which is essential ly ranking cations in order of charge per unit surface area of their hydrated ion. Our current understanding of adsorption mechanisms have been reviewed by Barrow ( 1 985) and Bohn et al. ( 1 986) . In general , the mechanisms of adsorption in soils are usually divided into two reactions : nonspecif ic , where the adsorbate is he ld as a counter-ion in the d iffuse double layer next to a positively charged col loid surface, and specific or chemi-sorption or ligand (all define the same reaction) , where the ions enter into coordination with the oxides of metals and become bonded to the metal in the structure as well as displacing other l igand ions (Rajan, 1 978) . Both mechanisms occur simultaneously in a given soil horizon. All major ions are considered to be involved in nonspecific adsorption , but only a few are subject to specific adsorpt ion o r l igand exchange (H ingston et al. , 1 967; Bohn et al. , 1 979 , 1 986) . Both mechanisms are mainly associated with surface oxides or hydrous oxides of AJ+3 and Fe+3 as well as some clay minerals ( Parfitt and Smart, 1 978) . 8 Nonspecific adsorption is pH dependent. Oxide surfaces may become protonated and require anions to balance the charge (Stumm and Morgan, 1 970) . At high pH, these same surfaces may lose protons and have net negative charge (Harward and Reisenauer, 1 966; H ingston et a/. , 1 967) . The pH at which the surface has no charge is the zero point of charge (ZPC) . The ZPC of Al+3 and Fe+3 oxides is near a pH of 9 , so that these sesquioxides would have a net positive charge in most acid soils and can retain large amounts of soil sulphate or phosphate. Specific adsorption, which is also pH dependent, occurs when anions exchange with hydroxyl ions of hydrous oxides at weathered mineral edges and enter into coordination with Al+3 and Fe+3 ions (Bohn et al. , 1 979) . This reaction will make the surface of the oxide more negative and in some cases cause a release of OH- ions (Stumm and Morgan, 1 970) . Sulphate adsorption capacity frequently increases with soil depth and this plays an important role in retain ing su lphate against leaching (Wil l iams, 1 975 ; Gregg, 1 976; Gregg and Goh, 1 978, 1 979, 1 982; Goh and Gregg, 1 982a, 1 982b) and minimizes the luxury uptake of su lphate (Barrow, 1 975) . lt is commonly observed that subsoils show an accumulation of adsorbed sulphate , and higher affinities for sulphate than topsoils ( Biakemore et al. , 1 968; Gregg and Goh, 1 978 ; Metson, 1 979b, 1 979c) . This is not only because of changes in mineralogy but because of decreases in pH (Ensminger, 1 954; Kamprath et al. , 1 956 and Harward and Riesenauer, 1 966) and lack of competition from the more strongly adsorbed phosphate ion (Ensminger, 1 954; Barrow, 1 967a) . Some important features of sulphate accretion by soil were identified by Fox et al. ( 1 983) . These are: 1 . The capacity of su lphate sorption increases with soil weathering. In the case of the Andepts , the order of increasing sorption is Typ ic Dystrandept < Hydric Dystrandept < Typic Hydrandept ; and in the case of so i ls deve loped from crystalline minerals the order is Alfisol < Ultisol < Oxisol . 2. Well-dra ined soils usually contain little sorbed sulphate, even if such soils have developed considerable sorption capacity, unless sulphate accrues in the soi l from an outside source, such as fertilizer. 3. Amounts of soluble sulphate and its availability do not follow the quantity of total S in the soil. S deficiency is associated with low sulphate saturation. According to Fox et al. ( 1 983) , considerable uncertainty exists about the nature and plant availability of adsorbed sulphate. For example, although plants do utilize adsorbed sulphate, it 9 has been observed that S deficiency may develop in crops on soils of the tropics which contain several thousand kg ha- 1 of so4=-s within the root zone. The capacity of a soi l layer to absorb sulphate (Barrow, 1 969a, 1 970) influences the rate at which it is taken up by plants. Results conducted by Barrow ( 1 969a) indicated that soil with a low ability to adsorb sulphate released sulphate into soil solution more rapidly than do high sulphate sorbing soils. In addition to the adsorbed sulphate in the surface horizons, the agronomic value of the subsoil sulphate to pasture and deep-rooting crops such as lucerne has been recognized by various workers, ( Ensminger, 1 958; Blakemore et al. , 1 969; Gregg et al. , 1 977) . For example, in a New Zealand yellow-brown pumice soil, Toxopeus ( 1 970) found no relationship between the available sulphate in topsoil (0-7 .5 cm) alone and yield response to S fertilizer. Including the amount of available sulphate in the subsoils, to a depth of 60 cm, improved the prediction of the pasture response. The availability of subsoil S was shown in a greenhouse study; corn plants became S deficient when g rown on Ap horizons alone, but had adequate S in their t issues when placed over B horizons with large amounts of adsorbed sulphate (Camberato and Kamprath , 1 986) . Fox and colleagues (Hasan et al., 1 970; Fox, 1 976; Fox et al. , 1 977) , Barber ( 1 984) , Hue et al. ( 1 984) and Camberato and Kamprath ( 1 986) , have determined approximate soil solution concentrations of su lphate S required for 95% maximum plant growth. These results indicated a soil solution sulphate concentration of 0 . 1 56 mM L - 1 SO 4 = -S for fine textured soils and 0.025 mM L-1 so4=-s for coarse textured soils were optimum for growth . Fox ( 1 980) showed that a number of subsoils maintained insufficient soil solution sulphate concentration despite large amounts of adsorbed sulphate. In general , it would appear that although the subsoil su lphate can provide S to plants, the s ize of its contribution depends upon soil chemical and physical factors (Gregg et al. , 1 977; Metson, 1 979a) as well as plant rooting depths. In strongly weathered soils of low pH, where adsorbed su lphate is not easi ly desorbed by soi l collo ids , this sulphate pool may not be available to plants (Metson, 1 979a; Fox, 1 980) because the sulphate solution concentration is too low or plant root development may be limited by aluminium toxicity (Ensminger, 1 954) . In contrast, as mentioned by Rennenberg ( 1 984) , it was considered that the acidification of soils enhances the availability of Al+3 to roots (Rorison, 1 973) ; polyvalent cations such as Al+3 and ca+2 may under some circumstances stimulate the uptake of sulphate by root cells ( Franklin, 1 971 ; Skjelbreid and N issen, 1 980; Jones and Smith, 1 98 1 ) . 1 0 Isotopic studies on the uptake of S by pasture plants (Gregg et al. , 1 977) have indicated that soil physical factors such as the availability of moisture within the soil profile and soil porosity, can affect the penetration of plant roots and consequently influence the magnitude of S uptake. Although it is difficult to judge the benefit of subsoil S for plant nutrition {Bohn et al. , 1 986) , an i ncreased ability of soils to retain sulphate is considered to reduce fertilizer S requ irements (Sinclair, 1 983) . To calcu late the ferti lizer requirements of pasture soils, the anion retention characteristics of the soil are required (Sinclair and Saunders, 1 984). The phosphate retention capacity (Saunders and Hogg, 1 971 ) is normally used as the anion retention index in New Zealand soils but specific sulphate retention soil tests have been developed e lsewhere; e.g. Barrow, 1 967a; During and Martin , 1 968 ; Saunders and Hogg, 1 971 . The New Zealand M inistry of Agriculture and Fisheries use the P retention value in their calculations as an index of S leaching potential. Typically, soi ls derived from sil iceous parent materials have low S retention values, while soils derived from volcanic parent materials tend to have a higher ability to retain sulphate (Saunders and Hogg, 1 971 ) . 2.2.2.3 Other inorganic S forms I n some calcareous soi ls a sign ificant fraction of total S is present as insoluble calcium su lphate eo-precipitated with Caco3. These compounds are generally unavailable to plants particu larly if associated with coarse particles of carbonate mineral. Williams and Steinbergs ( 1 962) measured the amounts in calcareous soils ranging from 25 to 3000 ppm S. High concentrations of sulphate transported to root surfaces by mass flow, in excess of plant demand, may accumulate at root surfaces (Mengel and Kirkby, 1 978) . In this manner gypsum may sometimes precipitate in rhizopheres of actively growing plants particularly in glasshouse soils heavily fertilized with superphosphate (Barber et al., 1 963) . Sulphides and some reduced S forms and even elemental S {8?) can occur under waterlogged conditions or poorly drained subsoils {Brown, 1 985 , Zucker and Zech, 1 985) . In well-drained, well-aerated soils the amount of mineral S occurring as compounds of lower oxidation states are negligible {Metson, 1 979a) . 2.2.3 Organic sulphur 1 1 Both plants and microbes incorporate su lphate into various organic compounds. These organic S forms enter soils as plant litter, animal excreta and corpses or are formed In situ by the soil biota. In most well drained and non-calcareous agricultural soils, the major proportion of the total S in surface soils is in organic form (Fitzgerald, 1 976, 1 978; Biederbeck, 1 978 ; Freney and Wil l iams, 1 983 ; Stevenson , 1 986) . This is mainly because the majority of soil organic S is insoluble in water and not susceptible to leaching losses. The forms of organic S, mainly deposited in soils, include the S containing amino acids and the sulfonates in which S Is directly bonded to carbon, and the true o rganic esters of sulphuric acid, in which S Is bonded to oxygen in the form of C-O-so3- linkages. In sulphamates, S occurs in the form of N-O-So3- and N-S03-. Little is known of the macro-molecular nature of organic S in soils ( Freney and Stevenson, 1 966; Freney, 1 967; Freney and Williams, 1 983) . Freney ( 1 967) and Lowe ( 1 969a, 1 969b) have shown a wide variety of S compounds were produced by organisms either in o r on soils. Most of these were susceptible to decomposition , did not accumulate in their mono-molecular form , and were not readily identified in the soils. The amount of soil organic S is closely associated with the amounts of carbon and nitrogen. Over a large sample of soi ls , the mean C :N:S ratio was 1 30 :1 0 :1 .3 (Freney and Williams, 1 983) , however, the amount of S in relation to carbon and nitrogen varies between soils. The d ifference may be attributed to differences in parent material , soil forming factors (climate, vegetat ion , leaching intensity, d rainage , and temperature) , cu ltivation and management practices (Freney and Williams, 1 983). The work of Freney and eo-workers (Freney, 1 961 , 1 967; Freney et al. , 1 969, 1 970, 1 971 , 1 975) suggested that most of the organic S in soils can be separated into the following forms by the use of chemical analytical techniques: 1 . Organic S which is reduced to H2S by a mixture of hydriodic acid, formic acid and hypophosphorous acid is known as hydriodic acid reducible S (H I-reducible S) and S in this fraction is not bonded directly to carbon. This fraction is believed to be main ly est e r-so4= , c-o-so3 - (a ry lsu lfatase , cho l ine su lphate , su lphate polysaccharides, etc.) and N-o-so3 -, (Freney, 1 961 ) . 2. Organic S which is not reduced by the above mixture is believed to comprise all of the S which is bonded directly to carbon, C-S, ( Freney, 1 961 ) . 1 2 3. Organic S which is reduced to inorganic sulphide by Raney-Nickle/NaOH consists almost entirely of S in the form of amino acids (Lowe and De Long, 1 963; Freney et al. , 1 975) and is known as Raney-Nickel reducible S. Since the HI-reduction method will also convert S0 and sulphate-S to H2s (Freney, 1 958) , the true amount of ester-so4= in soil is usually obtained by subtracting the amount of S0 and su lphate-S from the amount of H I - reducible S ( unless su lphate-S was extracted before reduction) . Details on ester-so4= formation, transformation, and its significance in the S cycle were reviewed by Fitzgerald (1 976, 1 978) . lt has been considered that carbon-bonded S may be derived from both leaf litter and root inputs, as well as microbial synthesis. Ester-So4= are generated predominantly by soi l microbial populations ( David et al. , 1 984 and F itzgerald , 1 978) . Laboratory studies have shown that the relative proportions of carbon bonded and ester-so4= formed depend on the avai labi l ity of carbon o r sulphate o r the C :S ratio of the o rganic substrate for microbial growth (Saggar, 1 980; Saggar et al. , 1 98 1 b; Ghani et al. , 1 988, 1 99 1 ) . I n general H I - reducible S accou nts for 30-70% of the total so i l o rganic S (Wi l l iams and Steinbergs, 1 959; Freney, 1 961 ; Lowe and De Long, 1 963; Tabatabai and Bremner, 1 972b; Bettany et al. , 1 973; Neptune et al. , 1 975) . In some soils the percentage of soil S present as H I-reducible S remains constant with depth (Wil liams, 1 975) , but in others the percentage increases with depth (Tabatabai and Bremner, 1 972b; Williams, 1 975) . This probably reflects differences in the carbon and S availability in d ifferent soils and soil depths. Upper horizons, ramified more intensely by roots, would have higher carbon input rates. The carbon-bonded S fraction includes amino acids and protein S such as meth ionine and cysteine. De Long and Lowe ( 1 962) proposed a procedure for the determination of carbon? bonded S in soils which claimed to recover all forms of organic S other than covalent sulphate and a lkyl sulphones. This proposed Raney-Nickel reduction method , however, does not reduce the carbon-bonded S in aliphatic sulphones or sulphonic acids and is subject to serious interference from iron and manganese (Freney et al. , 1 970) . I n addition, Freney et al. ( 1 970 , 1 972) found that, even under optimal conditions, the amount of Raney-Nickel reducible S was 44% less than the value of carbon-bonded S, calculated by subtracting the HI-reducible S from the total S content of the soi l . Thus most workers prefer to estimate the amount of carbon? bonded S by subtracting the H I-reducible S from the total S content of a soil ( Freney et al. , 1 970, 1 975; Bettany et al. , 1 973; Tabatabai, 1 982; Landers et al., 1 983) . 1 3 There i s confl icting evidence with regard to the agronomic s ign ificance of differentiating between carbon-bonded S and HI-reducible S. In mineralization studies using Canadian soils, Lowe ( 1 964 , 1 9 65) considered carbon-bonded S to be of l it t le value as a sou rce of minera lizable S, in contrast, Ghani et al. ( 1 988, 1 991 ) , using New Zealand soils found that most of the su lphate generated from soil incubation studies originated from carbon-bonded S. Freney et al. ( 1 975) also showed that over n ine months most of the available S ( 60%) , removed by plants in a pot experiment, came from carbon-bonded S although there were changes in all organic fractions. Sulphur in the HI-reducible S fraction can be easily hydrolyzed into inorganic sulphate by acid or alkali (F reney, 1 961 ) . lt was considered to be the most labile fraction of soil o rganic S and/or serve as an immediate reservoir of S that can be mineralized in response to biological demand (Mclaren et al. , 1 985 ; Strickland et al. , 1 987) . Widdowson ( 1 970) , however, found that neither fraction was related to plant S uptake on a range of Iowa soils. Due to this conflict in evidence it was concluded that, separately, these two fractions are unl ikely to be of any value for predicting the S requirement of plants. Cycling between the H I -reducib le S and carbon-bonded S has been observed by many researchers (e.g., Freney et al. , 1 971 ; Mclachlan and De Marco, 1 975 ; Saggar et al. , 1 98 1 b; Ghani et al. , 1 988, 1 99 1 ) . Mclachlan and De Marco ( 1 975) considered that rather than employ a soi l test, determining the rate of change between these two fractions may provide a more appropriate method for determining the soil S status than measuring the actual quantities found in either fraction at any one time. So far, no further work has been done on this aspect. Although most soil S may be present in the organic form, only a small percentage of this fraction (as little as 1 0%) may enter the active S cycling pool on an annual basis (Till and May, 1 971 ; Goh and Gregg, 1 982a, 1 982b; Chapman, 1 987a, 1 987b) . 2.2.4 Other methods for characterizing soil organic sulphur. As mentioned in Section 2.2.3, it has been observed that characterization of organic S i n soil , based on bonding relationships with carbon, does not always provide biologically meaningful fractions (Bettany et al., 1 974, 1 979; Goh, 1 988; Swift et al., 1 988) . This may be because soil o rganic matter comprises a complex heterogeneous mixtu re of l iving biomass ( micro? organisms plus fine root material) , partially decomposed and completely transformed plant and animal residues and some exists in intimate association with the mineral components of soil. Some workers have developed extraction techniques with the aim of separating the more labi le S from the more inert components irrespective of its bonding relationships to carbon. 1 4 Many approaches i n the study o f organic matter fractionation are as follows: (a) isolation, ident ification and measurement of individual compounds (Schreiner and Shorey, and eo? workers ( 1 908- 1 938) , according to Kononova, 1 966) ; (b) chemical extraction and physico? chemical separation and characterization of various fractions (Hayes and Swift, 1 978 ; Swift et al. , 1 988) ; (c) physical separation into s ize fractions and/or densimetric fractions without significant alteration of organo-mineral complexes (Saggar, 1 980) ; and (d) biological approach aided by radio and stable tracer techniques (Jenkinson, 1 976; Jenkinson et al. , 1 976; Bettany et al., 1 979 ; Saggar, 1 980; Ghani et al. , 1 988) . Chemical extraction and physico-chemical separation is most commonly and widely used, where in much effort has been concentrated on f ind ing solvents that wi l l ext ract h igh proportions of organic materials without significantly altering their composition. So far, few have attempted the fractionation of S in organic matter. Bettany et al. , ( 1 979) and Saggar ( 1 98 0) used an alkal i-pyrophosphate extract ion-separation technique o n some Canadian soils. Their resu lts showed that the percent of total S as HI-reducible S was found to be greatest in the fu lvic acid (FA-A) fraction (alkali soluble and acid soluble) and least in the humic acid fraction (HA-A) (alkali soluble and acid insoluble). Most of these S fractions were associated with the clay size fraction of soils. These resu lts were unable to be interpreted with respect to plant availability of different soil S forms. Swift et al. ( 1 988) proposed a mild fractionation method for organic S using 2 M acetylacetone (pH 8) in combination with u ltrasonic d ispers ion , but the significance of this procedure is unknown. More research on these various approaches is still needed before it can be stated that they are useful in characterizing the availability of so il organic S to plants. 2.2.5 Microbial sulphur As mentioned earl ier, microbial processes, associated with o rganic matter decomposition, create plant ava ilable S in the so il system. The plant avai lability of S in soi ls is largely dependent on the nature and amounts of S present as we ll as the dynamics of the soil microbial population. The microbial turnover of S may have a great effect on the short term supply of S in soils where most of the S is in organic forms. Thus a knowledge of the amount and nutrient composition of the soil microbial population (microbial biomass) , and further, the 15 response of the microbial biomass to the addition of decomposable carbon energy sources and fertilizer S is necessary to understand the nature of S transformations in soil . The S content of most microorganisms ranges from less than 0 . 1 to 1 % of dry weight (Alexander, 1 977) . The most conspicuous cellular constituents in bacteria containing S are amino acids, cyste ine and methionine, whereas several fungi can store intra-cellular S as choline sulphate (Harada and Spencer, 1 960; Takebe, 1 960) . Coughenour ( 1 978) , calculated on the basis of the quantity of S amino acids isolated from various bacteria (Laskin and Lechvalier, 1 973) , that the S content of bacteria ranged from 0.52 to 1 .02% and that of bacterial protein between 0.9 to 0.97%. The majority of S amino acids are associated with intercellular or metabolic components of bacteria. The contribution of ester-S04 = from within bacterial cells is l imited but the abil ity of bacteria to synthesize these compounds and subsequently re lease them extrace l lu lar ly cou ld be of cons iderable importance in soils ( Fitzgerald, 1 976). The S content of fungi , based on the S amino acid composition of various fungal proteins (Coughenour, 1 978) , ranges between 0 . 18 and 0.33%. Apart from S amino acids the ability of fung i t o synthes ize and accumu late cho l i ne su lphates and other este r-so4= i s wel l documented (Spencer and Harada, 1 960; Takebe, 1 960). Based on the microbial biomass data reported by Clark and Paul ( 1 970) , Kowalenko ( 1 978) estimated that bacteria and fungi in the surface of a grassland soil account for between 0 .3 and 1% of total soil organic S respectively. Although the microbial population contains a very small proportion of soil S, this fraction is extremely mobile and is considered to be the key driving force for S turnover in soil (Biederbeck, 1 978) . The latter statement is in agreement with tracer studies which showed that not more than 2-3% of the total organic S in Brown forest soils was in the active phase (van Praag, 1 973) . Saggar ( 1 980) and Saggar et al. ( 1 98 1 a) d eveloped a method for measuring microbial biomass S in soil. lt was found that microbial biomass S in Canadian prairie soils represented about 2 .8% of the total S in soil. This was in agreement with the values, measured by other authors, of 2-3% (van Praag, 1 973) , 1 .3% (Kowalenko, 1 978) , 1 .2 and 2.2% (Strick and Nakas, 1 984) , 0 .9 and 2.6% (Chapman, 1 987a) and 0.4-1 .8% (Ghani et a/. , 1 990) . Mclaren et al. ( 1 985) estimated the amount of so il microbial biomass S, measured by Saggars technique (Saggar, 1 980 and Saggar et al. , 1 981 a), was the same 'active' organic cycling pool as that derived by calculation from their incubation study. Goh and Gregg ( 1 982a, 1 982b) 1 6 found the active cycling pool of S ranged between 4 to 8 percent of the total organic S in a range of New Zealand soils. Furthermore, Chapman ( 1 987a, 1 987b) suggested that microbial biomass S can form a significant proportion of the organic S pool which is involved in S cycling and is potentially available to plants. Chapman ( 1 987a) also found that in soils of very low S status and known to respond to S fertilizers, biomass S exceeded 'plant-available' S extracted with phosphate and may thus be an important pool of S in these soils. Recently, Ghani et al. ( 1 990) showed that the amounts of microbial biomass S in New Zealand soils were fluctuating within seasons according to soil moisture and temperature changes. Pasture soils had larger amounts of soil microbial biomass S than cu ltivated and fallow soils. In general, it was also found t hat so i l su lphate l eve ls in so i ls we re i nter-re lated with m icrobial b iomass S concentration. Useful relationships have been found between soil organic matter, various S constituents (total S, carbon-bonded S , ester-so4= , organic S , enzymes S) , and microbial activity (Lee and Speir, 1 979; Lee et al. , 1 985) . David et al. ( 1 982) considered that these relationships indicate the potential fo r microbial S transfo rmat ions and the importance of microbial activity in dynamics of S in soil systems. 2.3 BIOLOGICAL TRANSFORMATIONS OF SULPHUR IN SOIL Processes i nvolved in the decomposition of plant litter, roots and organic matter in excreta, by soi l microflo ra and fauna remain the key processes in maintain ing nutrient cycl ing and, consequently, animal production in grazed pastures. lt is generally believed that most of the decomposition reactions releasing S in soils are carried out by micro-organisms and some perhaps by plant roots, although some abiotic chemical reactions are also possible. Immobilization and mineral ization , which are essentially enzymatically catalysed S oxidation and reduction reactions, are the main transformations occurring in soils and are concerned with the internal cycling of S from one soi l S form (pool) to another ( Freney and Williams, 1 983) . A d iagram represent ing the S cycle is shown in Figure 2 . 1 . A more complete discussion of the mechanisms and metabo lic pathways of various S t ransformat ions is available e lsewhere ( Freney, 1 967; Postgate , 1 968 ; Roy and Trudinger, 1 970 ; Fitzgerald, 1 976, 1 978; Scott , 1 985) . A summary of these processes is given below. In this review mineralization and immobil ization process are discussed. Off farm produce losses / Atmosphere so2 Plants I I I Animals S04=.c-O-S . C-S >- so4=. C-0-S ,C- Ul CD 0 tl) c m ttl U) O Ul - CD 0 0 >_ ? !!!. c. CD ?::J ttl - ?= ?u; tli Ul ::l o -.:J ? E Fertifizer so4=. s<> Primary minerals s=. so --..- inputs to farm I i1.1mojilization M mine?alization R redudion 0 oxidation Figure 2.1 r f j Roots so4=. C-0-S ,C- S Residues + Excreta so4=. c-o-s. c-s C:he:11::a! w?2!henil;i ' .t. "o,., ?(0 '(> 200) than in micro-organisms (<50) (Barrow, 1 960c; Saggar et al., 1 981 a; Chapman, 1 987a, 1 987b) . In other words, the rate of immobilization is governed by the potential of plants to fix carbon. The C:S ratio or S content of the residues will, however, influence the speed at which mineralization occurs and the final form of the S immobilized . lt is clear that more detailed study of S forms in field soils, particularly around plant litter, roots and dung patches is required before we have a full understanding of these processes. 2.4 PLANT REQUIREMENTS 26 Sulphur is requ ired by plants mainly for the synthesis of certain essential amino acids and proteins, a component of e nzymes involved in photosynthesis and production of vitamins. Su lphide is a particu larly i mportant component of n it rogenase enzymes which catalyse nitrogen fixation of Rhizobium species (Lehninger, 1 982), Flavours, odours and toxic agents in some plants can be linked to a range of S containing compounds. Sulphur is also important in improving the uniformity and quality of vegetables (Wainwright, 1 984; Schnug, 1 990) . Details of the sign ificance of sulphur in plant nutrition and forage quality and ruminant nutrit ion have been reviewed by Duke and Reiseneaur ( 1 986) and Tisdale ( 1 977) . A crop requirement for a specific nutrition is commonly defined as "the minimum content of that nutri ent associated with max imum y ie ld or the m in imum rate of i ntake of the nutrient associated with the maximum growth rate (Spencer, 1 975) . However, in practice, the term requirement can be expressed in a number of ways, for example for a pasture, the amount of nutrient requ ired for desirable pasture establishment , o r, the long-term suNival of pere nnial pastures, o r, the maintenance of appropriate levels of yield, or, the production of suitable quality forage, and lastly, to sustain the fixation of maximum amounts of nitrogen in legume based pastures (Smith and Siregar, 1 983) . Feed qual ity or the percentage of S containing amino acids of pastures may be important i n wool production because wool contains about 3.5 to 5% S, mainly as cysteine. In general , plant S requirements can be d ivided into two parts: (a) internal requirement, which refers to the relationship between the concentration of e lements in the plant and plant growth ; (b) external requirements, a relationship between the concentration of an element in the soil or culture solution and plant growth. Within these two requ irements, a concept of 'critical' value , (a minimum amount of nutrients per un it of soi l or plant associated with 80-90% of maximum plant growth) has been established by many workers. These 'critical' values are the basis for ranking soi l tests and plant analyses. I n recent years, inte rest in assess ing plant S requ i rements has increased considerably because deficiencies of S have been reported with increasing frequency throughout the World (Walker and Gregg, 1 975; Blair, 1 979; Fox et al. , 1 983; Tisdale et al. , 1 986; Chaiwanakupt et al. , 1 987; Syers and Curtin , 1 987; Murphy and Borgan, 1 988). The increase in S deficiencies may be generally classified into two categories: ( 1 ) decreased accretion and/or (2) increased depletion of soil reserves. The former may include the increased use of low S containing fertil izers and pesticides and less S input from atmospheric sources (Syers and Curtin , 1 987) . 27 The latter includes removal resulting from increases in crop yield, leaching and erosion losses. In addition, environmental and site factors may influence deficiency and sufficiency. These are distances from the sea or industry, temperature, crops and crop management, and inherent soil S status. 2.4.1 Assessing soil sulphur availabi l ity Assessing plant S requirements for nutrients can be achieved by soil testing, glasshouse trials, plant analyses, field fe rt i l izer trials and to a lesser extent by biological testing and visual deficiency symptoms (Reisenauer et al. , 1 973; Reisenauer, 1 975) . The technique employed wil l depend on the resources avai lable and state of knowledge concern ing a p articular combination of crops and soils and the nature of crop response to S. Attempts have been made to establish environmental factors [e.g. distance from the sea and S content in the rainfall (Walker and Gregg, 1 975 ; Ledgard and Upsdell, 1 99 1 ) , irrigatio n water, pH and S responses] for identifying S deficiency (Biair, 1 979 ; Sinclair and Saunders, 1 984). More data, however, are needed on the influence of these factors on the cycling of S before meaningful relationships can be developed to predict S deficiency which are not site specific. Currently, both soil test and plant analyses remain the most common techniques being used to monitor the changes in soil and plant status occurring in the soiVplant environment as a result of applied cultu ral practices. Soil tests are a relative index of the fertility status of the soil, while plant analyses provide a means of determining whether the soil plus cultural practices are providing the plant with sufficient nutrient. I n general , data for assess ing plant fert i l izer requirements can be derived from e ither glasshouse or f ield fertilizer trials. Essentially, field trials using a range of S application rates remain the most reliable technique for assessing plant requirements although the cost of trial operation is rather high. G lasshouse trials, although a relatively low cost method, have many l imitations (Smith and Siregar, 1 983; Jones, 1 986) . The most serious of these l imitations is that glasshouse trials often assess only the topsoil. In contrast, in the field trial some crops may be able to utilise the subsoil sulphate more effectively. Besides this, the low soil volume tor the plant root system and the favourable moisture and temperature regimes maintained during the trial period tend to accentuate the deficiencies and thus overestimate the apparent S requirements. Alternatively, the preparation of some soils for potting can also increase the mineralization (Williams and Steinbergs, 1 964; Sparling et al. , 1 985) of soil organic S and thus mask the symptom of S deficiency. Therefore, whenever possible, it is desirable that the 28 results from pot trials be confirmed under field conditions for their reliability in making fertilizer recommendations (Asher et al. , 1 983) . However both pot and field trials are site specific. Careful consideration is needed in extrapolating results to other environments. 2.4. 1 . 1 Plant analyses I nternal requi rements are most commonly assessed by plant tissue analyses and to some lesser extent by visual symptoms of plant parts. Evaluation and interpretation, i ncluding factors affecting critical values in plant analyses for S have been reviewed by Oijkshoorn et al. , 1 960; Eaton, 1 966; Oijkshoorn and van Wijk, 1 967; Bates, 1 971 ; Metson, 1 973; Jones, 1 975; Andrew , 1 975; Spencer, 1 975 ; Smith and Siragar, 1 983. A more detailed discussion of this area is not within the scope of this review. I n genera l , plant analyses used for diagnosing the S requirement of crops have i ncluded mostly determination of the following indices: total S, sulphate-S, the ratio of total N to total S and of HI-reducible S to total S (Metson, 1 973 ; Spencer et al. , 1 977a, 1 977b; Maynard et al., 1 983b) . The most commonly used indices are total S, N :S ratio and extractable S. Another approach towards interpreting plant analyses is the use of ORIS (diagnostic and recommendation integrating system). The ratios of S and other nutrients are cal ibrated according to ORIS of Beaufi ls ( 1 973). This method has been successful in diagnosing plant requirements for other nutrients but has not been tested extensively with S. Recently, Sinclair and Jones ( 1 99 1 ) have shown that OR IS appears to ho ld promise as a too l fo r the interpretation o f chemical analyses of clover samples from clover-based pastures in New Zealand. However, the authors considered that ORIS requires high quality data in terms of both the measurement of dry matter yield responses and chemical analysis of clover samples. lt is essential that proper evaluation and interpretation of plant test results take into account the crop environment. Crop environments include : the plant itself (species, age, parts analysed) , environmental conditions (deficiency o r toxicities of other e lements, drought, temperature , insects, grazing animal, disease, plant defoliation and type of crop culture, i .e . , field vs. potted soil vs. nutrient solution) and sample preparations (Jones, 1 975; Spencer et al. , 1 977a, 1 977b; Smith and Serigar, 1 983). For example, lower critical values have been often obtained in pot experiments than in the field (Jones, 1 975; Smith and Siregar, 1 983) . Evidence for the need to identify a target plant species for indicating l imited soi l S supplies was provided in a field investigation by Walker ( 1 957) and Walker and Adams (1 958) . lt was 29 suggested that the S status of clovers revealed the S requirement of the pastures better than that obtained by grasses. This is particularly relevant to legume based pastures where good legume vigour is required for nitrogen input. The stage of plant growth can markedly influence S concentrations in plant tissue (McNaught and Christoffels, 1 961 ; McNaught, 1 970; Spencer et al. , 1 977a, 1 977b; Cornforth et al. , 1 983). Such effects may make it difficult to interpret analytical data unless the approximate age of a plant is known. Sulphur content may vary within plant parts (Jones et al. , 1 972b; Spencer et al. , 1 977a, 1 977b) . For instance, Jones et al. ( 1 972b) reported the critical values of 0.23 and 0.08%, respectively, in the leaves and stems of Bur clover. The deficiency of other e lements may influence the magnitude of the critical values of plant S indices. For example, the correction of phosphorus and nitrogen deficiencies has been shown to lower the critical values of total S and su lphate-S in clover plants (Bouma et al. , 1 969; Jones et al. , 1 970 ; Spencer et al. , 1 977a, 1 977b; Cornforth et al. , 1 983) . Interpretation of plant analysis is based on a 'standard value' , 'critical level ' , or 'sufficiency level'. There are numerous sources for most of these values for a wide range of crops; e.g. Eaton ( 1 966) ; Metson (1 973) ; Maynard et al., ( 1 983b) ; Sinclair and Saunders ( 1 984) ; Martin? Prevel and Gauther ( 1 986) ; Reuter and Robinson ( 1 986). To be reliable, it is essential that the material sampled from the f ield for plant analysis matches material used for the o rig inal derivation of the critical concentration. lt is this aspect where plant t issue analysis is most commonly misused (Asher et al. , 1 983; Blair, 1 979) . lt was suggested (Biair , 1 979) that the tabulation of data, showing critical levels without specifying the methods of analysis and determination, is of little value for setting levels of adequacy. Errors in analytical procedures must be overcome before use of such indices can be made (Biair, 1 979) . From personal experience gained in analys ing herbage samples from t he Wageningen I nternat ional Exchange Programme, i t is c lear that there are large variations in laboratory resu lts from different countries. This supports the above statements of Blair ( 1 979) which reflects the need for standardized laborato ry techniques before establ ish ing 'critical ' , 'sufficiency' or 'standard' levels. 30 2.4. 1.2 Soil analyses Methods in soil S analyses and evaluation and interpretation of results have been reviewed by a number of authors e .g . , Beaton et al. , ( 1 968) ; Reisenauer et al. , ( 1 973) ; Andrew ( 1 975) ; Re isenaue r ( 1 975 ) ; Tabatabai ( 1 982) ; Landers et al. ( 1 983) ; Randall and Sakai ( 1 983) ; Blanchar (1 986) . Soils contain S in forms that range in their availabil ity to plants ; solution sulphate, adsorbed su lphate and o rgan ic S, which varies in !abi l ity from eas i ly decomposed to extremely recalcitrant. lt is probable that the !abil ity of soil organic S occupies a continuous spectrum. However in soils it has been convenient to partition S into two pools; an active cycling pool and inert pool (Till and May; 1 970a, 1 971 ; Goh and Gregg, 1 982a, 1 982b; Mclaren et al. , 1 985) . lt is probable that much of the active pool consists of l iving o rganic matter such as m icro? organisms and plant roots and litter. The size of the active pool can be quantified by using 35so4= as a radiotracer but it is not widely employed probably due to its technical complexity and high cost. Additionally the size of the cycling S pool may not relate to S availability. The immediately-available pool , includes soil sulphate-S which is believed to be the main form taken up by plants and is frequently measured in assessing S requirements for plants because of its s implicity of measurement and low operation cost. Most soil testing procedures measure only this immediate sulphate pool. However the sulphate pool may represent only 20% or less of the S that becomes plant available in a growing season and it has been recognized that levels of soil sulphate are not always well correlated to crop yields (this is described in more detail in this Section) . Soi l testing involves sampling, analysis and correlat ion of analyses with crop yield. The immediate goal of soi l testing is to predict the likely crop growth response to S fertil izer in field situations. I n the case of S nutrition, soi l parameters such as total S and extractable S, have sometimes been successfully corre lated with yield responses. However, most researchers have examined the value of extractable S (soluble su lphate and adsorbed sulphate) as a predictor of soil S status and plant growth responses to ferti l izer application. 2.4. 1 . 2. 1 Depth of sampling and sample preparations The amount of nutrient extracted may be influenced by the method of soil sampling and soil depth, soil preparation, soil-to-extractant ratio, extraction time and technique. 31 Depth of soil sampling may alter the relevance of a soil test result. Very acid soils may contain toxic concentrations of a luminium which l imit root development and sulphate avai labil ity to plants (Metson , 1 979a). However, there is evidence that appreciable amounts of S are obtained from subsoils (Bardsley and Kilmer, 1 963; Blakemore et al. , 1 969; Rorison, 1 973; Gregg et al. , 1 977; Risk and Boswell , 1 988) . Where soil properties do not l imit root growth it can be advocated that soil samples should include these soil depths. By using 35s in their field studies, Gregg et al. (1 977) showed uptake of subsoil S by pasture species. In a sandy soil and silt loam these authors calculated relative root activity using 35s carrier-free isotope. Relative to 7.5 and 22.5 cm depths, pasture plants obtained 24% and 1 1 %, respectively, of their S from subsoi l be low 22.5 cm depth in spring evaluation. The extent of the uptake zone reflected the depth of pastu re rooting. In other experiments they showed S uptake occurred from greater depths (1 00 cm) in a sandy loam than in a silt loam (60 cm) . However, like other authors using isotopically labelled nutrients to measure relative root activity, Gregg et al. ( 1 977) failed to account for the fact that mineralization/turnover rates of native organic S and therefore the sulphate pool turnover rate and isotopic dilution would decrease with soil depth. Thus lower relative percentages of plant S were derived from lower depths than estimated by these authors . Saunders and Cooper ( 1 975) and Nguyen et al. ( 1 989a, 1 989b) found no advantage in sampling to 1 5 cm relative to the normal 7.5 cm. Hoeft et al. ( 1 973) and Westerman ( 1 974) also reported no improvement in prediction of S response for pasture or for deep-rooted crops such as lucerne with the inclusion of soil samples below a 30 cm depth. The importance of subsoil S in plant nutrition appears to vary with climate, plant species and soil type. Thus no general ru le can be applied to the importance of soil testing depth which must be evaluated depending upon local knowledge of soils, climate and crops. Time of soil sampling has been shown to have a significant effect on the amount of extractable S. Ghani et al. ( 1 990) showed that sulphate S concentration fluctuated from values as low as 1 mg S kg soil to 8 mg S kg soil during a one year study on Wakanui silt loam soil at Lincoln, Canterbury, New Zealand. In particular, they found that sulphate concentrations appeared to be decreased by substantial rainfall events which occurred a week before sampling. However, when rainfall was sufficient to cause su lphate leaching, occurred more than 1 week before sampl ing , and subsequent warm temperature facil itated S mineralization and returned sulphate-S concentrations to pre-rainfall events. There is conflicting evidence with regard to the effect of air drying on the amount of extractable soil sulphate. Williams and Steinbergs ( 1 964) , Tabatabai and Bremner ( 1 972a) and Searle and Sparl ing ( 1 987) found that air-drying had no effect o n the amount of extractable S. In 32 contrast , some researchers had reported increases in the amounts of extractable sulphate after drying soil (Freney, 1 958; Williams and Steinbergs, 1 959 ; Barrow, 1 960c; Williams 1 967; Kowalanko and Lowe, 1 975) . Oven drying of soils at high temperature ( 1 05 ?C) markedly increased the amount of extractable S (Barrow, 1 96 1a ; Peveri l l et al. , 1 975; David et al., 1 982) . These i ncreases have been attributed to the m ineral izat ion of soi l organ ic S by microorganisms (Chaudhry and Cornfield, 1 967b) and the physical breakdown of complex organic S into simple organic compounds and sulphate (Barrow, 1 961 a; Kowalenko and Lowe, 1 975) . David et al. (1 982) found that drying soil increased the amount of extractable S only in the organic horizon soils. In some mineral horizons, a decrease in the amount of extractable S was found and this was attributed to the lower extractability of sulphate due to drying and possibly the crystallization of Fe and AI sesquioxides in B horizons which may have caused sulphate to be held in an unextractable form. Storage of soil samples may have some effect on amount of extractable S. Searle and Spariing ( 1 987) showed that the storage of moist or air-dried soil samples for up to 84 weeks resu lted in considerable increases in the amount of sulphate. Probert and Jones ( 1 977) , in contrast, found no effect of soil storage. In summary, to classify soil S availability a so il sampling depth appropriate to soil, crop and climatic conditions should be chosen and soil preparation and storage techniques should be standardized. 2.4. 1 .2.2 Extractants The immediately available pool of S is believed to consist of sulphate ions in the so il solution plus adsorbed sulphate which is in exchange equilibrium with sulphate in solution. Apart from the addition of SO 4 = th rough fert i l izers , an imal excret ion , rainfal l and pesticides , this immediately available pool is also generated through the mineralization of soil organic S (see Section 2.3.1 ). A large number of methods have been proposed to estimate the s ize of the immediately available soil sulphate pool (Beaton et al. , 1 966; Reisenauer et al., 1 973; Reisenauer, 1 975; Tabatabai, 1 982) . For routine soil analysis most authors have resorted to s imple chemical extractants involving anions that exchange with sulphate adsorbed to soil surfaces (Searle and Speir, 1 988) . The choice of chemical extractant is very much dependent on the properties of the soil to be analysed. Extractants can be divided into three groups: 1 . Those removing readily soluble su lphate-S; 2. Those removing readily soluble plus portions of adsorbed sulphate-S; 33 3. Those removing readily soluble and adsorbed su lphate, plus portions of labile organic S. a . Readily soluble sulphate-S When the amount of adsorbed sulphate is negligible, an extractant which employs an anion that is not adsorbed strongly on the soil su lphate sorption s ites, such as chloride (as either 0.01 M cac12 or LiCI) is adequate (Williams and Steinbergs, 1 959, 1 962; and Bettany et al., 1 974). Calcium chloride has an advantage; not only are clear solutions easily obtained but extraction of organic S which occurs with water extraction is prevented ( Fox et al. , 1 964) . Recent work by Bolan e t al. ( 1 991 , personal communication, N.S. Bolan) suggested that CaCI2 extraction of soils with low Ca exchange capacity may increase the surface positive charge on these soils resulting in slightly increased su lphate sorption. Amounts of sulphate i n CaCI2 extracts were found to correlate well with the amounts of S mineralized during soil incubations (Barrow, 1 961a ; Kowalenko and Lowe, 1 975 ; Tsuji and Goh, 1 979 ;) , but correlate less well to plant uptake (Tsuji and Goh, 1 975) and yield response in f ield conditions (Nguyen , 1 982) where soils were so4= retentive. b. Adsorbed plus soluble sulphate-S Extraction of adsorbed sulphate requires the use of solutions containing anions such as OW, H2Po4 - or Hco3 - which can effectively compete with, and displace, su lphate from positively charged sorption sites on the soil surfaces. A disadvantage of using OH- and HC03 - solutions is that large amounts of organic matter are extracted and this can cause interferences when su lphate is determined turbidimetrically or the organic S extracted gives rise to overestimation of su lphate values when it is determined by the reduction-distillation method. In this regard, Ca(H2P04)2 seems to be one of the best extractants, because phosphate ions displace the adsorbed sulphate , and Ca ions flocculate the soi l col lo ids and clear extracts are readi ly obtained. The calcium ion also l imits the extraction of organic matter (Ensminger 1 954; Hesse, 1 957; Fox et al. , 1 964 ; Barrow, 1 967a; Searle, 1 979 ; and Tabatabai, 1 982) . S ign ificant relationships between phosphate extractable S and plant S uptake have been reported for a wide range of soils (Fox et al. , 1 964; Barrow, 1 969a; Jones et al. , 1 972a; Hoeft et al. , 1 973 ; Tsuji and Goh, 1 979 ; Saunders et al. , 1 988) . c. Readily available and adsorbed sulphate and portions of organic S 34 The extractants that remove readily available and adsorbed sulphate and significant portions of organic S include acetic acid-phosphate (Cooper, 1 968; Hoeft et al. , 1 973; Saunders and Cooper, 1 975) , NaHC03 at pH 8 .5 (Bardsley and Kilmer, 1 963) and 0 .25 M KCI at 40 ?C (Biair et al. , 1 991 ) . These extractants are not widely employed by researchers involved in relating soil S availability to S responses. Regardless of the particular extractant used in assessing plant-available S, there is sti l l little known about the chemical nature of the organic S in soil extracts (Beaton et al. , 1 968) and because only empirical relationships between soil test values and crop yields or response are determined, there is no causal evidence to suggest that the forms of organic S extracted bear any relation to that part of the organic pool which supplies S to the growing plants (Hoque et al. , 1 987) . d. Other methods Apart from chemical extractants, anion exchange resin beads and anion exchange membrane have been employed in extracting soil sulphate-S or even soil phosphate (Kurmarohita, 1 973; Searle and Speir 1 988; Saggar et al. , 1 99 1 ) . Amounts of anion membrane extractable so4= agree well with those obtained by 0.01 M Ca(H2P04) extraction (Searle and Speir, 1 988) . Its s impl icity is of part icular interest. The extraction provides a non-contam inating, non? destructive method to extract the soi l sulphate without apparently affecting the microbial biomass, hence its appl ication in S t ransformation studies may be more worthwhile. For example, resins were employed by Hedley and Stewart. ( 1 982) to measure labile soil P prior to measuring microbial P in a soil P fractionation procedure. Hoque et al. ( 1 987) suggested that certain microbial assays could give meaningful indications of S mineral ization rate. M icrobial b ioassay test values were s ignificantly correlated with pasture response to applied S but it was concluded that much more research is needed to evaluate this approach to soil testing. e. Use of radioactive 35 S Although not suitable for routine soil analysis, there have been a few attempts to measure the pools of S avai lable to plants by isotopic di lution techniques. Many of these studies were carried out in laboratory or glasshouse conditions (Nearpass et al. , 1 961 ; Harward et al. , 1 962 ; 35 Bettany et al. , 1 974; Probert, 1 976; Tsuji and Goh, 1 979) . Generally, 'A', 'E' and 'L' values (see review by IAEA, 1 976; Vose, 1 980 and Manzel and Smith, 1 984; for definition of 'A' , 'E' and 'L') have been correlated well with plant S uptake, growth response and other estimates of plant available S. The 'E' and 'L' value, which are estimates of the exchangeable pool of S in soil, normally increase with increasing applied S and time. Gregg ( 1 976) , however, fai led to correlate the 'A' value and pasture yield response in field studies in New Zealand soils. By means of radioactive studies, Till and eo-workers (Till and May, 1 970a, 1 970b, 1 971 ) and Goh and Gregg ( 1 982a, 1 982b) were able to determine the size of the S 'cycling pool ' ('active' and 'inert' pool) in pasture soil . The 'active' cycling pool included S in entire living plants , S in residues, available soil S and S in animals. The 'inert' pool included S that does not enter the cycle ( recalcitrant material, less mineralizable) . The active cycling pools were variable among soils . However, further research is needed to verify the usefu lness of radioisotopes in measuring cycling pools of nutrients in soils as an indication of soil S status. 2.4.1.2.3 Interpretation of soil tests Strong corre lat ions between soi l tests using phosphate as an extractant and crop yield responses to applied S have been observed by investigators in both glasshouse and field trials (Spencer and Freney, 1 960; Hoeft et al. , 1 973; Probert and Jones, 1 977; Lee and Speir, 1 979; Saunder et al. , 1 981 ; Scott , 1 981 ; Lee et al. , 1 985). Empirical relationships only exist between the amount of S extracted and plant yield or uptake in the glasshouse. As reviewed by Re isenauer et al. ( 1 973) , no one procedure has proved consistently superior in predicting responses to S fert i l ization. Whereas, resu lts of other studies (Spencer and Glendinning, 1 980; Cornforth et al. , 1 983 ; Raymant, 1 983; Jones, 1 986; Hoque et al. 1 987; Skinner, 1 987 as quoted by Syers and Curtin, 1 987; Vaughn, 1 987; Nguyen and Rickard, 1 988; Saunders et al. , 1 988; Murphy, 1 990 and Nguyen and Goh, 1 991 ) have been far less encouraging. Apart from the need to ensure the soil sample taken reflects the plants effective rooting depth, the validity of empirical soil so4= tests for making S fertilization recommendations is l imited due to the need to quantify several seasonally dependent sources of S input and loss which influence soil test values, particularly input of plant available S from mineralization of organic matter and S loss by leaching (Reisenauer, 1 975; Saunders and Cooper, 1 975; Spencer and Glendinning, 1 980; Cornforth et al. , 1 983), (see later discussion in Section 2.3.1 and 2.5.2.2) . I n a temperate environment, like New Zealand with dramatic short-term changes in climate, G regg ( 1 976) suggested that adequate knowledge of the factors influencing the response patterns at each location along with a long term knowledge of fate of fertilizer S may be more 36 useful. In support of Gregg's statement, Cornforth et al. , ( 1 983) have shown large fluctuations in soil test results in New Zealand soils. This variability was attributed to seasonal changes in leaching. lt was suggested that soil su lphate test resu lts cannot be used to predict S fertilizer requirements on their own. Interpretation depends on pasture age and fertil izer application history (which reflects the reserves of organic S in a soil) , S leaching index and the season when the sample was taken. This test is best used to modify S ferti l izer requi rements estimated from a balance of S gained and losses from the system using a model by Sinclair and Saunders ( 1984) , (see discussion in Section 2.4.2.1 ). The l imitations of various empirical methods of soil analysis have also been discussed by Reisenauer et at .? 1 973 ; Reisenauer, 1 975 and Oertli 1 990. The authors again high light the substantial difficulties in interpretation of such tests. One problem with these empirical tests is that they are not based on proven mechanisms of plant nutrition. In conclusion it appears that results from empirical soil tests must be applied only to the type of soils or crops for which calibration of the method has been careful ly tested (Oertli, 1 990) . Oertli ( 1 990) proposed new approaches in the use of soi l tests in soi l fertility management. Soils which react similarly to nutrients should be combined in classification u nits. For these units, detailed information on the reaction behaviour must be obtained. A few simple analyses might show how a specific field fits into this reaction pattern. lt is difficult to see, however, how this 'new' approach differs to the current practice in New Zealand where clear patterns of growth and response to fert il izer have been established for different soil groups and types ( During, 1 984) . I n this context soil tests provide very useful i nformation. Such i nformation could probably be improved if one considered the question asked of soil tests. For example, for mobile nutrients like so4= does the test result indicate that during periods of plant growth the nutrient level will not meet plant demand. For perennial crops such as pastures it would seem logical to conduct a soil S test either immediately after or prior to the major spring? summer growth period. After is more suitable than before because climate effects o n growth cannot be predicted. Adjustments to fertilizer application rate for the next year can be based on these soil test values. A l imitation of this strategy is that dry summer soils are difficu lt to soil sample . Obviously, t he re are many factors other than so i l test level that need to be considered in arriving at an accurate estimate of the S fe rtil izer requ irement for a particular crop and soil type. Important among these are the effectiveness, cost and the availability of the fertilizer material. 2.4.2 Fertilizer S requirements 37 The amounts and types of S to be applied will depend on plant demand for S, the status of the plant environment; i .e. rate of mineralization of organic S, the contribution of S from subsoils, rooting habits, atmospheric and irrigation inputs, rainfall and leaching, temperature, solubility or availability of the fertilizer S , and interactions with other elements (Reisenauer, 1 975 ; Asher et al. , 1 983) . 2. 4.2. 1 S recommendations for pastures in New Zealand The practice for recommending S fert i l izer for New Zealand pasture is d iscussed as an examp le of how so i l test i ng procedu res can p rov ide i nfo rmation to assist fe rt i l izer recommendation. The estimation of fertilizer S requirements for pastures in New Zealand are at present , based on the mass balance nutrient cycling concept developed by Sinclair and Saunders (1 984). The Computerized Fertilizer Advisory Schemes (CFAS) model (Sinclair and Saunders, 1 984) which is an 'external model' as shown in Figure 2.2 takes account of external inputs such as atmospheric S and losses by leaching and through anima l products. The estimate of leaching loss depends on annual rainfall and the soi l d rainage (free or slow draining) and anion retention characteristics . Animal losses are estimated from product removal from the farm and the uneven return of nutrients in dung and urine to soil surface. A negative balance requires an input of S from fert i l izer. The s ize of ferti lizer input can be modified by considering the abil ity of so i l to provide S from o rganic S mineralization and current available status. This abil ity is estimated as an empirical index calculated from pasture age and previous soi l product ivity ( i .e . animal stocking rate) , and the current amount of phosphate extractable S (0.01 M Ca(H2P04)2 at pH 4.0 , 1 :5 soil solution ratio, and 30 minutes extraction time of samples from the top 7.5 cm of pasture soil). Based on this model, fertilizer S required for maintenance of pasture production for the North Island, New Zealand, ranged from 1 3 to 28 kg ha- 1 (Sinclair et al. , 1 985) Figure 2.2 38 s i n i r r il}45? slope) , respectively. Boswell ( 1 983) showed that at the campsites total soil S throughout the soil profile and leaching loss of S from soil were generally higher than that in the non-campsites. However, more than 50% of urine S was retained in soil by immobilization processes rather than redistributed by plants grown on campsites. The spatial distribution of urine is unknown but it is expected to follow dung distribution (Till, 1 975) . According to Till ( 1 975) "An estimate of the likely importance of redistribution can be made by considering an improved pasture stocked at 10 sheep ha-1 and fertilized with S at 25 kg ha-1 year-1 . In such a system the return of S in excreta would be about 4 kg ha -1 year- 1 in dung and 6 kg ha-1 year-1 in urine. If the dung distribution is 34% on 10% of area (Hilder, 1964), this would give deposition (and perhaps effective loss) of 1 .4 kg dung S or 5.6 percent of the annual application" 64 If dung and u rine are similarly distributed the loss would be 2.4 kg S and equivalent to 9 .6% of the annual S application. 2.6.4.3 Excreta decomposition With the exception of pioneering work by Barrow ( 1 961 b) and recent work by Boswell ( 1 983) , there has been litt le published i nformation on the rate of release of S from dung. Barrow ( 1 961 b) showed that the amount of S mineralized was closely related to the S content of the dung and the proportion of S mineralized was less than with plant material. Dung was more resistant to decomposition than was plant material (Barrow, 1 961 b; Boswell , 1 983) . Barrow ( 1 961 b) also reported that there was likely to be net immobilization below a dung S content of 0 .22% while similar results were likely to occur with plant litter with S content below about 0 . 1 2%. Dung mineralization rates are more rapid in crushed samples than pad samples (Bromfield and Jones, 1 970; Rowarth , 1 987) and in the presence of arthropods such as earthworms ( Barley, 1 964) and dung beetles (Bomemissza and Williams, 1 970) which can be expected to enhance the mineralization rate through the action of communition of material and transporting it to the vicinity of microorganisms (MacFadyen, 1 978) . Similarly dung decomposition rates are slow in dry condit ions as compared to cold moist conditions. For example , Hilder ( 1 966) reported sheep pellets remaining for 1 0 weeks in cold moist conditions in Armidale, Australia. However in New Zealand, Rowarth ( 1 987) reported that dung samples (by physical disappearance) decomposed within 20 days in cool moist winter periods, while longer periods (up to 60 days) were observed during summer. The fate of 35s labelled urine and dung and decomposition of unlabelled dung was studied by Boswel l ( 1 983) in contro lled conditions in the field by the litter bag technique. Results can be summarized as follows: 65 1 . Uptake of S from urine in camp site soil was initially higher than from non-camp site soil. 2. The mean rate of release of S from dung (0-37 days) was about 4.5 mg S g-1 S day-1 which was much lower than from litter. 3 . There was immobilization of soil S after S had been re leased from dung. This agreed with the results reported by Barrow ( 1 961 b) . 4. Movement of S from dung and u rine to soi l organic S was as rapid as from plant l itter. I n part icular, about 68% of u rine S was in organic S within 6 days. In general , more than 80% of the S released was converted into organic forms. Unfortunately, there was no partitioning of organic S fractions in the study. The uptake of 35s labelled sheep excreta was also investigated by Kennedy and Till ( 1 981 b) under field conditions. Results showed that the percentage of plant S derived form u rine and dung (Excreta, %SDFE) were as high as 94% and 54% respectively. About 20% and 1 0% of the S from the urine and dung were taken up by plants. Recovery of S by pasture plants from both sources dropped rapidly with t ime , especially from ur ine. l t was calcu lated that the excreta from 20 sheep ha- 1 could provide 20% of the S requirement of the above ground plants. Furthermore, it was found less than 40% of total activity were recovered by pastures after 384 days. Leaching loss and lateral spread were suggested as responsible for the low recovery. Both dung and urine have also been shown to affect soil properties where they were returned, particu larly at campsites. They have been shown to increase cation exchange capacity, o rganic matter , N , P, exchangeable potassium, calcium and magnesium (MacDiarmid and Watkin, 1 972a; During and Weeda, 1 973) and soi l pH (Doak, 1 952; Watson and Lapins, 1 969; During and Weeda, 1 973). As the soil pH increases, greater desorption of sulphate may occur (Ensminger, 1 954; Kamprath et al. , 1 956; During and Weeda, 1 973 ; Bolan et al. , 1 988) and mineralization of soil organic S may increase (White , 1 959; Barrow, 1 960b, 1 960c; Wil liams, 1 967) . This favours accelerated local losses of soil S by leaching (Boswel l , 1 983; Sinclair and Saunders , 1 984; Saggar et al. , 1 990a, 1 990b) and leads many researchers to consider that nutrient cycling through grazing animals is an inefficient process resulting in considerable loss of effective nutrients from productive areas of paddocks (Hedley et al. , 1 990; Saggar et al., 1 990a, 1 990b; Goh and Nguyen, 1 990) To summarize, the general fate of S in a paddock being grazed by animals can be represented by litter, dung, urine and amounts retained in animal tissue. Of the total S in pasture plants, about 1 0-30% are in litter, 25-30% as dung, 30-40% as urine and about 1 0- 15% held In animal 66 products. Of the total S i n l itter , dung and u rine retu rned to the soi l , it appears that approximately 5-1 0%, less than 5% and 20-30%, respectively, wil l be re-utilized in the short term by pasture plants. The remainder (about 60-75% of total S returned to the soil) is mostly immobi l ized into so il organic matter. From this review, the urine deposition is expected to stimulate the greatest increase i n plant S uptake or leaching in the short term. Deposition of S in carbon rich, plant litter and particularly dung is expected to have low plant recoveries and in the short term may be responsible for promoting immobilization of soil sulphate. 2.7 SUMMARY AND CONCLUSIONS Sulphur p lays an important ro le in the nutrit ion of h igher plants , particu larly in protein synthesis. The primary n itrogen fixation step in legume based pastures is dependent on adequate S nutrition. I n an aquic temperate climate (USDA taxonomy) such as most of New Zealand, S fertilizers are required to maintain adequate soil S status for pasture production . Su lphur can be a highly mob i le nutrient in soi l as we l l as very bio logical ly active . The persistence of plant available S in soils is therefore dependent on rainfal l , drainage, sulphate retention characteristics of soils and the extent to which added S is converted to organic S reserves. Although considerable laborato ry and g lasshouse research has studied the influence of soi l properties on S uptake, immobilization and retention and the mineralization of organic S, few of the findings can be extrapolated with confidence to effectively explain the cycling of S under field soil conditions on naturally structured soils. M ine ra l izat ion and immobi l izat ion are the key processes that govern the amount of immediately plant available S and the fate of fertilizer S in soils. Leaching or surface run-off under field conditions lead to inefficient use of sulphate based ferti lizers e .g. superphosphate. Incorporation of ferti l izer S i nto o rganic S may conserve S against leaching losses. Even though the use of slow release forms of S such as elemental S (S0) is considered to reduce these initial leaching losses in some climatic situations in New Zealand. Little information is available on the comparative fate of sulphate and S0 based fertilizers in soils. Radio isotope techniques have proved to be useful for t racing the fate of applied S and measuring the flux of S between various soil pools. Their use in tracing the fate of S0 is a comparatively recent development . S0 labelling techniques require further development so that uniformly labelled S0 in various particle sizes can be conveniently produced and blended with P sources to simulate commercially available fertilizers. 67 The general objective of this thesis is to develop radioisotope techniques to study the fate of sulphate based and S0 ferti l izers in field soils and having developed these techniques to investigate the factors which influence the fate and plant availabil ity of these fertilizers, in order to provide information that can be used to formulate and improve fertilizer recommendations for pastures. CHAPTER 3 EXPERIMENTAL SITES AND TECHNIQUES 3.1 METHODS FOR RADIOACTIVEL V LABELLING FERTILIZER SULPHUR 3.1 .1 INTRODUCTION 68 The radioactive isotope of sulphur-35s has been used extensively to measure the agronomic efficiency of S fertilizers, to follow the fate of applied fertilizer S (Gregg and Goh, 1 978, 1 979, 1 982; Shedley et al. , 1 979; Goh and Gregg, 1 982a, 1 982b; Boswell , 1 983) and to determine the nature of the su lphur cycle in soils (Till and May, 1 970a, 1 970b, 1 971 ; Goh and Gregg 1 982a, 1 982b; David and Mitchell , 1 987). The 35s isotope is particularly well suited for use in longer term soil/plant studies because it has a long half-life (87.4 days) and upon decay emits a low energy p particle (p max 0 . 1 67 MeV) which is easily detected by l iquid scintillation counting systems and is safe to use. Most studies have used 35s labelled so4=-s fert i l izers but few studies have used 35s labelled e lemental S (S0) . Except for Shedley ( 1 982) , no technique for manufacturing labelled S0 fert i l izer has been published . However, Shedley ( 1 982) labelled S0 by dissolving 35s labelled and unlabelled S0 in carbon disulphide. Labelled S0 crystals were obtained after carbon disulphide was evaporated. However, it is hazardous to work with carbon disulphide. A safer labelling technique needed to be developed. 3.1 .2 Calculating the required 35s enrichment In fertilizer materials The amount of 35s to be added to the fertilizer material depends on the fate of S in the soil and plant system being studied, length of study, dilution of the label by soi l and herbage S, sample size taken for analysis and the efficiency of the radioactivity counting system ( IAEA 1 976) . Previous research with a particular soil and plant system is helpfu l so that both the amount of S that is l ikely to exchange with the isotope and the rate of exchange can be estimated. Based on the above, the optimum amount of radioisotope to be mixed with fertilizer material can be estimated and will be expected to g ive a satisfactory count rate in samples taken from the various plant and soil S pools. Calculations, in detail, are shown in Appendix 3.2. A minimum count rate between 500-1 000 cpm mr1 for 35s in a 0.04 M Ca(H2P04)2.H2o soil extract ( 1 : 1 0 ; soil :extraction solution ratio) was used for calculating the total amount of activity required to label superphosphate or S0 for 1 50 day experiments. 3.1 .3 Labelling S0 fertilizer 69 Both forms of radioact ive su lphur , 35so4 = and 3 5s0 (carrier-free) were suppl ied by Amersham International Plc . , England. All subsequent operations were carried out in a fume hood. 3. 1.3. 1 Labelling microfine EfJ Microfine particles ( less than 0 .01 0 mm) of 35s0 labelled S0 were prepared by dissolving agricultural grade S0 (< 0.500 mm particle size) in toluene (commercial grade) and mixing this with carrier-free 35so . Aliquots were then pipetted onto a layer of 1 5 g of finely ground soil in a glass-petri dish and evaporated to leave crystalline S0 on the surface of the soil particles. This method of labelling produced microfine S0 (< 0.01 0 mm particle size) which was used in the field experiment in Chapter 5. 3. 1 .3.2 Labelling EfJ of different particle sizes To produce labelled S0 of different particle sizes (< 0.075 mm to 1 .00 mm or greater) , aliquots of the carrier-free 35s0 dissolved in toluene were added to samples of unlabelled S0 (< 0.500 mm particle size and agricu ltural g rade) p laced on th in alumin ium foil in flat-bottomed porcelain crucibles (approximately 1 .00 ml of toluene was added per 3 .5 g ground S0) . The mixture was carefully melted on an e lectric hot plate (with remote switching) at 1 1 5?- 120?C. Five to e ight minutes of heating were required to melt all the S0 . To reduce volatilization of S0 during melting , a watch-glass containing 2.00 ml acetone was placed on top of the crucible for cooling and condensing the remaining toluene. The melt was agitated carefully to ensure even mixing, removed from the hot plate and immediately cooled in the air draught of a fume hood. After overnight cooling in a fume cupboard, the solidified S0 was crushed with a pestle and mortar and sieved to different particle sizes using nylon sieve cloth. There is loss of some isotope in S0 of the incorrect particle s ize . This can be reme lted and crushed again if necessary. This 35s0 was used in field and glasshouse experiments in Chapter 7 which were conducted for 1 80 days. 3.1 .4 Labelling sulphate containing fertil izers 70 Single superphosphate (SSP) containing sulphate labelled with 35so4= was prepared by adding carrier-free 35so4= to the sulphuric acid used to acidulate phosphate rock (Nunn and Dee 1 952; Siems, 1 953; Young et al., 1 985; Bolan et al. , 1 987). Calculat ions are shown in detail i n Appendix 3.3. The stoichiometric equation for the reaction between apatite rock and sulphuric acid is shown i n equation 1 . A mixture of 1 : 1 Nauru lsland:Christmas Island A grade phosphate rock was acidulated with 65% (w/w) sulphuric acid labelled with 35so4= in a food mixer at an acid :rock ratio of 0.62. The diluted sulphuric acid was preheated to 60 ?C before mixing (Bolan et al. , 1 987) . Labelled 35s gypsum was prepared by reacting a slight excess amount of Caco3 with an equivalent amount of 35so4= labelled 50% W/W H2so4 as shown in equation 2. 2caco3 + 2H2 35so 4 (2) The labelled ca35so4.2H2o precipitate and superphosphate were dried at room temperature to avoid formation of the hemihydrite (Caso4.H20, plaster of Paris) . In the production of superphosphate and gypsum al l the radioisotope was conserved in the product. 3.2 EXPERIMENTAL SITES, SOIL AND PLANT PREPARATIONS 3.2.1 Experimental sites The research p rogramme undertaken involved both f ie ld and g lasshouse t rials and was designed to examine the short-term fate of fert i l izer S in natura lly structured samples of pasture so i l . A l l f ie ld experiments were conducted on permanent pasture o n Massey University farms located within 1 5 km of the University campus. Site and soil descriptions are presented in the Material and Methods Sections of the appropriate chapters. 3.2.2 Isolating undisturbed soli cores for field and glasshouse experiments Galvanized steel cylinders ( 1 5 cm diameter, 1 5 cm long, wall thickness 0.5 mm) were used to cut and contain soil cores for field and glasshouse experiments. The fifteen centimetre depth was chosen as the appropriate depth for the cores because many investigators (Gil l irigham et al. , 1 980; Will iams, 1 988; Williams et al. , 1 990a, 1 990b) have shown that ryegrasstwhite clover pastures, particularly in yellow-grey earth soils, utilize insignificant amounts of nutrients from soil below a depth of 1 0 cm. The majority of pasture roots (85% of total weight) were found in 71 this zone and routine soil samples for diagnostic purposes are commonly taken from the upper 7.5 cm portion of the soil profile (Saunders et al. , 1 988; Nguyen et al. , 1 989a, 1 989b) . The diameter of the soil cores was 1 5 cm which was considered to be adequate to provide a representative sample of the pattern of soil macropores, i .e . naturally structured samples, in the soil used (Bouma, 1 980; Williams et al., 1 990a, 1 990b) . The cores were driven into areas of uniform ryegrass/white clover pasture at each f ield site . Each cylinder was pushed into a depth of about 1 2 .5 cm; a 2 .5 cm high rim was allowed to avoid the loss of rain water or liquids that were later applied to the surface during experiments. Care was taken to avoid areas of sward which had obviously been recently affected by urine o r dung patches. For field experiments, cylinders i n each replicate were about 10 cm apart and were fenced to prevent access by cattle. For glasshouse experiments these soil cores were dug out and carefully transported to the laboratory. 72 3.2.3 Sample preparations 3.2.3. 1 Sheep dung sample preparation All dung samples were dried and mil led in a s imi lar manner to herbage samples which is described below. 3.2.3.2 Soil sample preparation Soil samples from non-radioactive experiments were air-dried at room temperatures up to 25 0c . To stop microbial activity during drying and to prevent lysis of soil micro-organisms, all soil samples from radioactive experiments were frozen and freeze-dried before mil ling. All soil samples were ground in a laboratory hammer mil l . To prevent contamination between samples, the hammer mil l was cleaned after each sample. After mill ing at least 95% of the ground soil passed through a 0.5 mm sieve. 3.2.3.3 Herbage sample preparation Herbage was dried in a forced-draught oven at 65 ?C for 24 hours and weighed to constant weight . Al l herbage and root material was ground in a small coffee grinder and stored in sealed plastic bags awaiting analyses. 3.3 CHEMICAL ANALYSES OF SAMPLES 3.3.1 Total S In plant material 3.3. 1 . 1 Alkaline hypobromite oxidation method Finely ground samples of dried herbage (0.02-0.05 g) , were digested with sodium hypobromite (NaOBr) solution in digestion tubes (2.5 cm in diameter and 20 cm long) on a thermostatically controlled electric aluminium digestion block. The method was based on that published by Tabatabai and Bremner ( 1 970a) . The amount of freshly prepared sodium hypobromite (NaOBr) used in this digestion varied between 4.00 ml and 8 .00 ml depending on the amount of sample. In the original method 3.00 ml of NaOBr was recommended for digestion of plant sample, containing 0.01 0-0 .050 mg of S, in a 50 ml round bottomed, digestion-distillation flask on a sand bath at 250-260 ?C. Many incomplete digestions were observed, however, and the results were h ighly variable when using 3.00 ml of NaOBr in digestion tubes in the aluminium block. Modifications to the method were made to make it suitable for use on an aluminium block digester. A greater volume of NaOBr was added stepwise and the digestion temperature was gradually increased as described below. 73 1 . A 1 .00 ml aliquot of NaOBr was slowly added to 0.02 g of finely ground sample, containing 0 .04 to 0 .08 mg of S, in a dry digestion tube . The tube and contents were stored overnight in a fume hood. 2. A fu rther 1 .00 ml of NaOBr was slowly added and the tube placed into a cool d igest ion block and the tempe rature raised to 1 50 ?C (over a pe riod of approximately 20 minutes) . The digestion was continued reducing the volume of the digest by two thirds. 3 . The tubes were briefly removed from the hot plate and a further 1 .00 ml of NaOBr was added, the tube contents were mixed for a few seconds using a vortex mixer. After replacing the tube in the block, the temperature was raised to 200 ?c and the tube heated until one third of the volume in the tube remains. The tube was again briefly removed from the block, al lowed to cool for about 5 minutes and a further 1 .00 ml of NaOBr added and again the contents were mixed thoroughly. 4. When the temperature reached 250-260 ?C, the digesting tube is placed into the digestion block and heated unt i l the ir contents were evaporated to d ryness. Heating was continued for an additional 25-30 minutes after taking the tube contents to dryness. The tubes were removed from the digestion block and allowed to cool, and 1 .00 ml of formic acid (98-1 00%) added. Then , the tubes' contents were d iluted to a desired volume with deionized water, m ixed thoroughly with a vortex mixe r and left until a l l the sediment had settled. Al iquots of the clear supernatant were then removed for S and 35s analysis as described in Section 3.3.6 and Section 3.3 .8, respectively. A 0 .20 ml aliquot was taken for measuring 35s radioactivity. This modification of the d igestion method was evaluated using test herbage samples supplied f rom the I nternat ional Plant Analytical Exchange , Wagen ingen Agricultural Univers ity, Netherlands. The result are shown in Appendix 3. 1 . The relationship between the analytical results, (average 36 to 229 mmoles kg-1 ) obtained using this method 'Y' and the median value obtained by several international analytical laboratories 'X' was Y = 1 .7 + 1 .02X; R2=99.2, indicating that results from this modified method agreed well with other laboratories. Although the modif icat ion method i nvolves several handl ing steps, the benef it of this semimicro? digestion is that up to 1 00 to 1 50 digests can be performed in one working day using four digestion blocks, each taking 32 tubes. 3.3. 1 .2 LECO sulphur analyzer The method i nvolves the combustion of the sample in a stream of oxygen and so2 evolved is analysed by idiometric t itration. The LECO semi-automated total S analyzer consisted of an 74 induction furnace (model 765-000) and su lphur titrator (model 532-000) and gas purifying train (model 51 6-000) . Details were described by Bremner and Tabatabai (1 971 ) , Jones and lsaac ( 1 972) . This method was used for determination of total S in herbage from non-radioactive treated plots. Results of total S in herbage using the LECO analyzer were similar to those using the alkaline hypobromite method discussed above (data not shown) . 3.3.2 Soluble and extractable S In soli samples (CaCI-S and CaP-S) The soluble and extractable S was determined by the method described by Williams and Steinbergs ( 1 959) , Searle ( 1 979) , Landers et al. ( 1 983) , Tabatabai ( 1 982) . Soil samples (5.00 g) were weighed into 50 ml screw capped centrifuge tubes and shaken with 40 ml of 0 .04 M calcium dihydrogen phosphate [Ca(H2P04)2, CaP] at pH 4, for extractable S, CaP-S (Searle, 1 979) , and 0.01 M CaCI2 for soluble S, CaCI-S (Williams and Steinbergs, 1 959) . The samples were shaken on an end-over-end shaker for two hours. The suspension was centrifuged at 1 0 ,000 rpm to remove suspended part icu lates. The supernatant was then filtered through Whatman No.42 fi lter paper and kept frozen in a plastic vial before the concentration of H I reducible S and the 35s activity i n the filtrate were determined as described in Section 3.3.6 and Section 3 .3 .8 respect ively. Both the CaP-S and CaCI-S fractions may contain some soluble organic S. A 1 .00 ml aliquot of each extract was taken for measuring 35s radioactivity. 3.3.3 Total S In soil (TT-S) The total S content of so i l was determined by us ing a modification of the dry oxidation procedure, described Landers et al. ( 1 983) . The sample (0.50 g) was combined with 1 scoop (about 0 .50 g) of mixed oxidant (25 g NaHC03 and 1 g Ag20) in a Pyrex glass tube which was 1 .00 cm in diameter and 1 5 cm long. The contents were mixed thoroughly with a Vortex mixer. Another scoop of the mixed oxidant was layered on the top of the sample as a trap for so2 gas. The tube was then placed on sand material in a clay o r a lumin ium container. The container was placed in a cold muffle furnace and the temperature was raised to 550 ?c and heated for 3 hou rs . After coo l ing , the dry m ixtu re was d isso lved in 1 0 ml of 3 M HCI . Depending on the amount of radioact ivity remain ing, larger samples can be used with proportional increases in the amount of m ixed oxidant. The Pyrex tube diameter and the volume of acid to dissolve the oxidized sample should also be increased. The diluted digest mixture was mixed thoroughly and left to stand overnight before the concentration of 32s and the activity of 35s su lphate S were determined as described in Section 3 .3.6 and Section 3 .3 . 8 , respect ively . A 0 . 2 ml of a l iquot of the d i luted d igest was taken for measuring radioactivity. 3.3.4 Total 5? In soil samples (TT-els) 75 A method described by Chatupote ( 1 990) was used. Soil samples of about 40 g containing 20- 40 mg s0, were mixed with 200 ml acetone in glass bottles and were shaken on an e nd-over? end shaker overnight at approximately 20 ?C. The acetone-soil suspensions were allowed to sett le . Al iquots containing 1 0-50 micrograms of S0 were placed in d isti l lation tubes and evaporated to nearly dryness and the S0 content was determined as described by Nguyen ( 1 988) . In this method the S0 was converted to hydrogen su lphide by a tin and hydrochloric acid reducing mixture using the modified reduction and distillation procedure of Johnson and Nishita ( 1 952) and the resulting hydrogen su lphide was assayed colorimetrically as methylene blue manually. Aliquots of 1 -2 ml of acetone were evaporated in a scintillation vial (Chatupote , 1 990) before measuring 35so activity as described in Section 3.3.8. 3.3.5 Organic S (Org-S) In soli samples Total organic S was determined by the difference between total soil (TT-S) and inorganic soil S (CaP-S) as follows: Org-S = (TT-S) - (CaP-S) or (TT-S) - (CaP-S) - (TT-els) ; for S0 treated plot The error in this calculation caused by the organic content of the CaP-S extract (3-1 0 ppm S) is smal l compared to the amount of o rganic S (200-600 ppm S) and its natura l spatial variability. 3.3.5. 1 HI-reducible S (HI-S) This was determined by the method described by Landers et al. ( 1 983) by reacting a sample (usually 0 .2-0.5 g) with 6.00 ml of reducing mixture , consisting of hydriodic acid, formic acid and hypophosphorous acid in the ratio of 4:2:1 in the modified Johnson and Nishita apparatus (8 unit d istil lation apparatus). Each sample was refluxed at 1 1 0 ?C for 30 minutes. The reduced su lphur (H2S) was determined colorimetrically by the methy lene blue method . Radioassays of the 35s activitv in this fraction were determined after the methylene blue was deve loped as described in Section 3 .3 .6 . A 1 .0 ml al iquot of the H I-reducible S trapping solution was taken for measuring radioactivity (Section 3.3.8) . 3.3.5.2 Ester-SO 4 = (Est-S) Ester-so4= S was determined by subtracting the CaP-S from the H I- reducible fractions (H I? S) . For the samples from the S0 treated plots, the amount of s0 remaining In soli was also 76 subtracted from the total HI-reducible S fraction. Unoxidised S0 still remaining in the sample was also reduced to hydrogen sulphide. Hence 3.3.5.3 Est-S Est-S = (H I-S) - (CaP-S) or = (H I-S) - (CaP-S) - (TT-els) ; for S0 treated plots Carbon-bonded S (Cb-S) The carbon-bonded S was determined as the d ifference between the total S and the H I reducible S (HI-S). Cb-S (TT-S) - (H I-S) 3.3.6 HI-reducible s In digested and extracted samples The determinat ion of su lphate S in aliquots from herbage digestion (Section 3.3. 1 . 1 ) , soil extract ion (Section 3.3.2 and Section 3.3.3) and ferti l izer extracts (Section 3 .3 .7 . 2) was performed using a modification of the method of Johnson and N ishita ( 1 952) described earlier. Prior to the introduction of an automated reduction and d istillation procedure in 1 987, reduction and d istillation was achieved using an 8 unit distillation apparatus and the hydrogen sulphide produced was determined manually by the methylene blue method. Since 1 987, the total S in su lphate solutions has been analysed by using an automated reduction and dist illation unit (CS IRO Division of Forest Research, Method No. PS1 7) . The hydrogen su lphide produced after reduction in the acid mixture was determined colorimetrically and automatical ly by forming bismuth sulphide (Dean, 1 966) . 3.3.7 Determination of total S In fertilizers 3.3. 7. 1 S> containing fertilizers Solubility of S0 at 25 ?C in toluene was about 2.0 1 8 g S 1 00 g-1 of saturated solution and 2.084 g S 1 00 mr1 of 1 00% acetone solution , respectively. The toluene extract can be directly added into a scintillation cocktail which is also toluene based. Acetone, on the other hand, produces some quenching if it was not evaporated before adding to the cocktai l solution (Chatupote, 1 990). A method deve loped by Chatupote ( 1 990) was employed. Approximately 0.050 g of 35so labelled S0 fertilizer ( 1 0-1 00% S) was mixed with 200 ml acetone in a 250 ml glass bottle and shaken on an end-over-e nd shake r overnight . After d i lu t ing , a l iquots of 1 -2 ml were evaporated in scintillation vials (Chatupote, 1 990) before measuring 35s activity as described in Section 3.3.8. 77 A suitable aliquot of each extract, containing 20-50 microgram of S, was used for determining the total s0 using the direct determination of S0 as described by Nguyen ( 1 988) . 3.3. 7.2 Sulphate containing fertilizers Total S in sulphate fertilizer materials was determined using a modified method described by AOAC ( 1 984) . About 0.1 00 g of fertilizer was mixed with 200 ml of 1 : 1 HCI:H20 and shaken on an end-over-end shaker overnight. A 0 .1 ml aliquot was used for measuring 35s activity by mixing with 1 2 ml of the cocktail solution and 1 ml of deionized water (Section 3.3.8) . Aliquots of the HCI extract containing 1 0-50 microgram of S, were used for determining total S as described in Section 3.3 .6. 3.3.8 RADIOASSA V OF 32p AND 35s ACTIVITIES The radioassay of weak beta-emitters is best performed by liquid scintillation counting. A Beckman Liquid Scintillation System, model 3801 , Bench-top and microprocessor-contro lled, was used for the radioassay. The instrument is also equipped with an automatic quench correction programme. The automatic quench correction programme was used for single and dual label counting (Beckman Liquid Scintil lation System: Operation Manual) of 35s and 32p radioactivity in samples from experiments described in Chapter 4 and 5. 3.3.8. 1 Liquid scintillation counting The scinti l lat ion cocktail used in these studies was prepared by m ixing 8 g of P PO (2 ,5- diphenyloxazole) , 0.2 g of POPOP (1 ,4-di-[2-(5-phenyloxazole)]-benzene) , 1 340 ml of toluene and 600 ml of Triton-X 1 00 (Faires and Boswell, 1 98 1 ) . The mixtures were stirred overnight by a magnetic stirrer and kept in a dark glass bottle. Twenty ml g lass scintillation vials were used for counting. The various aliquots of digest (described above) were mixed with volumes of deionized water, where appropriate, to give aqueous phase:scintillation cocktail ratios of 1 :1 2 . Commonly total liquid volume was 1 3 ml except when larger sample aliquots were required to give sufficient counts. The mixtures were mixed thoroughly and left overnight in a dark room in order to l imit chemiluminescence before measurement of radioactivity. 3.3.8.2 Establishing quench curves The l iquid scintillation device is based on the direct relationship between the amount of light being emitted from the vial being l inearly related to the number of beta particle emissions from the samples. In practice, a number of factors (beta particle not activating the fluor, chemicals deactivating fluor molecules, or light from the fluor being adsorbed by other chemicals I n the 78 mixture) act to reduce the amount of light being emitted per beta emission. The phenomenon is referred to as "quenching" and a sample in which it occurs is said to be "quenched." All samples prepared in the laboratory are quenched to some degree . Therefore , the number of counts recorded must be converted to a value that correct ly reflects the number of beta emissions-d isintegrations that actually occurred in the sample. In these experiments, the quench curves were generated using the "H-number" method, for single and dual isotope counting (Beckman Liquid Scintillation System: Operation Manual) . Separate sets of quenched samples were prepared for each counting system (e .g . soil extracts , herbage extracts, etc.) and both isotopes 35s and 32P. The following quench curves were generated: 1 . Sulphate S extracts - after extraction with CaP-S and CaC12 extraction Y = 96.09 + 0.086H - 0.0006H2, 2. HI-reducible S after methylene blue development y 2 95.20 + 0.079H - 0.0004H , 3. Herbage S digests (NaOBr-digestion) y = 98.33 + o.o59H - o.ooo5u2, 4. Total S soil extracts (NaHC03 + Ag20 digestion) y = 2 97.46 + 0.082H - 0.0006H , 5. Herbage P digests y = 94.7 + o.o74H - o.ooo5H2, 6. Extractable P (Olsen P) Where y = 2 95.0 + 0.062H - 0.0008H , Y = Counting efficiency (%) H = H-number R2 = 0.93 R2 = 0.94 R2 = 0.95 R2 = 0.94 R2 = 0.97 R2 = 0.94 After counting, all data were normalized to the day when the S labelled fertilizers were applied to experimental plots. Details of calculations of relevance to 32p and 35s activities in soil and herbage samples are described in the appropriate sections. 79 CHAPTER 4 EVALUATION OF THE ROLE OF SHEEP DUNG IN THE SHORT TERM IMMOBILIZATION OF SOIL AND FERTILIZER SULPHUR 4.1 INTRODUCTION Any study which attempts to measure the immediate fate of ferti l izer u nder pastoral field conditions needs to consider the whole nutrient cycle and factors , which in the short term, may influence the direction of nutrient flow. Once sulphate enters the plant root and is assimilated into shoot S, the above ground nutrient flows (as modelled by Saggar et al. , 1 990a, 1 990b) are similar whether S is applied as s0 or sulphate-S fertilizer ( Figure 4. 1 ) . Thus factors which are likely to play important roles in determining whether S0 is conserved more efficiently in soil than sulphate-S must influence the S transformations which generate plant available sulphate in the soil or affect the rate at which the sulphate pool is depleted by microbial immobilization, leaching and plant uptake. Some of these factors , which have a large influence on short term nutrient flow, are : cl imate, plant species, soil type and structure and defoliation frequency (grazing interval) . These factors can be easily simulated and their effects studied in small plot or undisturbed soi l core experiments. The effects of excretal return from the grazing animal are less easy to s imu late . Some researchers , however (Smith R.G and M.J . Mclaughlin, personal communications) , have used the return of dung (or dung products) on field plots and glasshouse pots to simulate excretal return. In general, however, little information is available on the immediate effect of sheep dung on the uptake of S by pasture plants. The return of animal excreta is not easily simulated in small scale experiments. This is partly because on the field scale the grazing animal returns excreta , containing large amounts of growth-l imiting nutrients, unevenly across a field. The irregular pattern of return depends on topographical and cl imatic influences upon the animals' grazing and camping behaviour (Hi lder, 1 964; Gillingham, 1 980; Gil lingham et al. , 1 980; Rowarth, 1 987; Rowarth et al., 1 985, 1 988, 1 990; Saggar et al. , 1 990a, 1 990b) . Furthermore large nutrient (particularly nitrogen) returns in excreta applied to a small trial plot would have a major influe nce over nutrient availability and pasture production, probably confounding o r obscuring the influence of the rate of a fertilizer S treatment. The return of S to soils as animal excreta, however, remains a major pathway of S in the S cycle of g razed pastures. Experiments conducted with sheep to measure the effect of their excreta on pasture production have shown that the return of sheep dung gave an 1 8% increase in dry matter production (Sears and Goodall , 1 948) , whilst Watkin ( 1 954) found that sheep dung made no contribution to pasture production , and only observed an effect in combination with a h igh rate of n itrogen fertil izer. Site specific results such as these are expected because the responsiveness of pasture growth to dung return wil l partly reflect the availability of nutrients in the dung and current soil nutrient status. 80 In a study in New South Wales, Australia, using 35s label gypsum, it was calculated that the excreta of 20 sheep per hectare could provide 20% of the S requirement of pasture plants over a graz ing period of 1 20 days (Kennedy and Til l , 1 98 1 a) . Of the S excreted by sheep approximately 60% is excreted in the urine and 40% in dung (Till, 1 975) . Dung has a C : S ratio at >200 : 1 . The return of excreta to a pasture soil may influence the immediate fate of fertilizer S in the following manner: 1 . I ncreased pasture and root growth, stimulated mainly by the excretal nitrogen, could increase the plant uptake of soi l and fertilizer S, or accelerate the conversion of sulphate S to organic S forms (Barrow, 1 967b; Curll , 1 982; Boswell, 1 983; Goh and Nguyen, 1 990; Haynes and Williams, 1 99 1 ) . 2 . The carbon added to soil as undigested herbage in dung has been shown to decay slowly (Barrow, 1 961b ; Boswell , 1 983) in soil and therefore could act as a carbon source used by micro-organisms to immobilize free sulphate-S from the soil solution. Goh and Nguyen ( 1 990) have suggested that excretal returns could stimu late so i l m icro-and macro-f lo ra and fauna such that S0 oxidation may increase. Research on the spatial distribution of excreta has been undertaken for cattle (Petersen et al., 1 956; Richard and Wolton, 1 976) and for sheep (Tallis and Donald, 1 964; Donald and Leslie, 1 969 ; Gi l l ingham and During, 1 973 ; Gi l l ingham, 1 978 ; G il l ingham, 1 980; G illingham et al. , 1 980; Thorrold et al. , 1 985) . Recently, in New Zealand Morton and Baird ( 1 990) have shown that the spatial distribution of sheep dung in relation to stocking rates was best described by a negative binomial function. There was significantly more aggregation of dung patches at lower stocking densities than at higher stocking densities. These authors considered the returns of n itrogen from dung and u rine affected an insufficient area to influence pasture growth. I n addition , Rowarth, ( 1 987) and Rowarth e t al. ( 1 985, 1 988 and 1 990) have shown the major mechanisms contro lling the movement of P from sheep dung into soil was the rate of physical break-down of sheep dung rather than the leaching of P from the dung sample . During the winter/spring period, the physical break-down take place within a month, as a result of high rainfall and biological activity, whereas in the summer/autumn period the dung persisted for approximately three months. During a short period ( 1 7 and 8 weeks for autumn and spring, respectively) dung P was less available to pasture than monocalcium phosphate. There is no information describing the influence of sheep dung on the subsequent fate of soil and fertilizer S in the zone of excreta-affected soil. Although Kennedy and Till ( 1 981 a) and Boswell ( 1 983) have studied the fate of dung S alone. Thus, before proceeding with the main experiments to examine the immediate fate of fertilizer S in pastoral soils grazed by sheep, preliminary experiments have had to be carried out to determine : a. the likely area of pasture influenced by sheep dung return and b. the influence of dung return on the short term fate of fertilizer S. 81 These experiments also served to evaluate radioisotope handling and measuring techniques. 4.2 EXPERIMENTAL OBJECTIVES 1 . To determine the area of soil i nfluenced by sheep dung by observing the uptake of 35s and 32p isotopes by pasture at different radial distances from radioactively labelled sheep dung pellets. 2. To determine the influence of sheep dung on the growth of unfertilized pasture and pasture fertilized with S0 or sulphate-S fertilizer (superphosphate) . 4.3 MATERIALS AND METHODS 4.3. 1 EXPERIMENT 1, AREA OF PASTURE INFL UENCED BY SHEEP DUNG 4.3. 1. 1 Design of the experiment The exper iment was conducted on a permanent clover/ryegrass pasture growing on a Tokomaru si lt loam (Typic Fragiaqualf) at Keeble farm, thre e ki lometers southwest of the Massey Univers ity campus. Some soi l properties are presented in Table 7 . 1 and Section 7.3. 1 . The pasture was mown to approximately 3 cm height at the beginning of the experiment and the trial area was fenced off to prevent access by cattle. On June 27, 1 985, three rates of dung weighing 0.5 g(D1 ) , 1 .0 g(D2) and 2.0 g(D3) (dry weight basis) , labelled with 35s and 32p (as described below) , were randomly applied to the surface of the pasture soil. Stee l pins were used to mark the position of the dung. Each dung application was replicated five times. Each plot was approximately 1 .0 square meter. There were a total of 1 5 circular plots in this experiment. Plots were harvested twice at 30 and 60 days after dung application . A jig of concentric rings, 1 0 cm(R1 ) , 1 5 cm(R2), 25 cm(R3) and 30 cm(R4) in radius, was centred around the steel pins remaining in each plot . The pasture within each concentric ring was harvested separately using e lectric hand shears. All pasture samples were dried at 80 ?c, weighed and ground to pass through a 1 .0 mm sieve. Weekly rainfall, temperature data and sunshine hours were used to calculate the amount of drainage water (Scotter et al. , 1 979) over the period of the experiment (Appendix 4.7) . Figure 4. 1 L o w s l u p o ( \ - 1 2 ? ) M c c.J i um s l o p o \ \ :J - 2 5 ? ) S l c o p s l o ? o ( i!: 2 G o ) L M - FE ll r -----._ f:" -- ?f(r A r? 1 1..\ 1\ L U f' T /\ K E -- --? G a l l \ \ .I . . . llil []?I ? ? ? ;r_ .Y. :J ?< I-I -. W n. o: :J llll L o s s s L o s s \ ?? 82 A schematic illustration of nutrient transfers in a grazed hill-country pasture (Saggar et al., 1990a). 4.3. 1.2 Labelling of dung with 35 S and 32 P 83 Aqueous solutions of carrier free 35s sulphate and 32p orthophosphate were obtained from the Radiochemical Centre, Amersham, U .K. Three mill i l iters of the 35s su lphate solution (approximately 26.337 MBq mr1 ) were mixed with 8 .0 ml of 32p solution (containing approximately 7.225 MBq mr1 32P) and the mixture was then diluted with deionized water to 20.0 ml. Aliquots of the mixture (0.9 ml per gram of dung) were thoroughly mixed with pre-weighed samples of finely ground sheep dung (0.5, 1 .0 and 2.0 g) in glass capsules (0.5 g dung is equivalent to 2-3 sheep dung pe llets) . The labelled dung samples were incubated at room temperature for two weeks before being applied to the field plots. During the incubation period, deionized water was applied to the dung in the glass capsules to keep the dung at a 30-40% moisture content. 4.3. 1.3 Total and extractable S and P in dung The S and P fractions in the incubated dung samples were measured using three replicates of dung samples (Table 4.1 ) . Almost 90% of the 35s and 54% of the 32p in the dung were in a soluble form. Calcium phosphate extractable sulphate accounted for approximately 1 0% of the total dung S and Olsen extractable P accounted for 43% of total P in the dung. This proportion agreed with other values reported in the literature (Kennedy and Til l , 1 98 1 a; Bromfield , 1 961 and Barrow and Lamboume, 1 962) . The dung sample contained 0 .3% of total S and 1 .27% of total P. Based on the specific activity of the soluble fractions it was calcu lated that the 35s isotope had exchanged with about 1 0 .2% of the total S in the dung whereas the 32p had exchanged with 8 0% of total P i n the dung (Section 4 .3 .5) ) . Therefore the incubation technique resu lted in dung P being more evenly labelled with isotope than dung S. Losses of both radioactive tracers during the incubation were negligible. Table 4. 1 Mean S and P concentrations, total radioisotope activity and specific activity (SA) o f 35s or 32p i n dung and extracts of dung (results expressed on dung dry weight). Concentration of S or P in dung S forms mg kg-1 TOTAL POOL Total S 3000 (9.02/ Total P 12706 (12.87) EXCHANGEABLE POOL Cap-S 275 (4.22) Olsen-P 5454 (7.30) a calculated as described in Section 4.3.4c b calculated as described in Section 4.3.4a c numbers in parenthesis are standard errors of means Total % activity of of total 35s or 32p activity MBq g-1 dung % 3.570 100.0 (0.059) 2.588 100.0 (0.035) 3 . 195 89.94 (0.055) 1 .392 53.79 (0.035) S.Aa Exchan?eable pool as % of total S or P MBq mg-1 S or P % 1 . 190 0.204 1 1 .6 18 1 0.20 0.255 80.00 CXl ? 85 4.3.2 EXPERIMENT 2, EFFECT OF SHEEP DUNG ON GRO WTH AND YIELD OF PASTURE 4.3. 2. 1 Design of the experiment This experiment, designed to examine the effect of sheep dung on the growth of an unfertilized permanent pasture and a pasture fertilized with superphosphate and elemental S (S0) , was conducted at the same field site described in Chapter 5. Treatments consisted of two dung rates : 0 and 375 kg dung ha- 1 and thre e fert i l izer treatments : contro l ( no fe rt i l izer) , superphosphate (SSP, 1 1 .9% S) and very fine agricultural grade S0 (99% S, <0.5 mm in diameter) applied at 30 kg S ha-1 . Treatments were replicated four times. Fertilizers were broadcast evenly on the plots which were 2.5 m by 1 .5 m. The plots were then subdivided into two areas of 2.0 m by 1 .5 m and 1 .5 m by 0 .5 m. The smaller area received dung applications. Pasture on the plots was mown to a height of 2 .5 cm one week prior to dung and fertilizer applications. Ground sheep dung (total S = 2640 mg S kg-1 of dung and calcium phosphate extractable S = 244 mg kg-1 of dung) was applied at the rate of 375 kg ha- 1 at harvest (equivalent to an amount of dung return from 25 sheep ha-1 , assuming that there were 500 g of dry excretion per sheep per day (Cie land A., personal communication) . Sheep dung was reapplied every 30 days after herbage harvesting. Superphosphate, S0 and KCI (at 40 kg K ha-1 ) were applied on October 1 4, 1 985. Finely ground rnonocalcium phosphate was applied on the S0 plot at a rate (25 kg P ha-1 ) equivalent to the amount of P in the superphosphate treatment. Sheep dung was first applied on November 1 8 , 1 985, after the first harvest. Plots were harvested five times with a 30 day interval between mowings. Electric hand shears were used to cut the pasture at 2.5 cm above the ground level inside a quadrant (40 x 40 cm) which was randomly placed within each plot. The pasture in each plot was harvested separately. All pasture was dried at 80 ?C and weighed. A subsample (approximately 25 g) was taken for grinding and subsequent chemical analysis. 4.3.3 Chemical analyses 4.3.3. 1 Total S and P in pasture samples The determination of total S and 35s activity in pasture samples were described in Section 3 .3 . 1 . 1 and 3 .3 . 1 .2 (for experiment 1 and 2, respectively) The total P and 32P activity in plant samples were determined after a Kjeldahl-type digestion. The P concentration in the digested solution was measured using the method of Murphy and R i ley ( 1 962) . Abso rbances were measured at 7 1 2 nm us ing a Pye Un icam SP 1 800 spectrophotometer. A 1 .0 ml aliquot of the diluted digest solution was taken for measuring 32p activity using a dual label counting program as described in Section 3.3.8 86 4.3.3.2 Total S and P in dung samples Total S, P , 35s and 32P act ivit ies i n dung samples were determ ined using the d igestion method of Dick and Tabatabai ( 1 977) . Total 35s and 32P activity in the digest solution were measured simultaneously using a dual label counting program as described in Section 3.3.8. 4.3.3.3 Extractable S in dung samples Wet dung samples ( moistu re content between 30-35%) were m ixed with 1 0 m l 0 .04 M Ca(H2P04)2 in 50 ml polyethylene tubes. The samples were then shaken on an end-over-end shaker for 2 hours , centrifuged at 1 0 ,000 rpm for ten minutes (SS34 head, RC2B Sorvall centrifuge) and filtered through No. 42 Whatman filter paper. A 4.0 ml aliquot of the extract was used for determining su lphate-S concentration by a modified method of Johnson and N ishita ( 1 952) as described in Section 3.3.6. A 1 .0 ml aliquot of the extract solution was taken and its 35s activity was measured, as described in Section 3.3 .8. 87 4. 3.3.4 Exchangeable P in dung samples Wet dung samples (moistu re content between 30-35 %) were mixed with 1 0 m l of 0.5 M NaHco3 (Oisen et al. , 1 954) , shaken on an end-over-end shaker for 0 .5 hour , centrifuged at 1 0 ,000 rpm (as above) for 1 0 minutes and filtered through No. 42 filter paper. A 1 .0 m l aliquot was used for determining the concentration of P as described by Murphy and Ri ley ( 1 962) . Also 0 .1 m l of extract solution was used for determining the 32P activity as described in Section 3.3.8. The amount of exchangeable S and P calculated to be in the dung were reported on a dry weight basis. 4.3.3.5 Radioassay of 35 S and 32 P activities The method has been described in Chapter 3, Section 3.3.8. A dual counting program was used. 4.3.4 Presentation of results (method of calculation) All radioactivity data were normalized to the day when the dung treatments were applied (day O; June 27, 1 985) , using the formula: Ao = Atexp(t*0.693ff) Bq where Ao = radioactivity at day 0 At = radioactivity at time t (days) t = time from day 0 T = half life of the isotope (days): 35s=87.5 days; 32P=14.3 days a. Extractable plus recently immobilized pool of S or P in dung These are the percentage fractions of dung S or P that exchanged with each radioisotope added (i.e. the exchangeable S or P pool in dung plus that which was recently immobilized into an organic form). where El = El = TS = ES = TS/ES*lOO percent percent of activity that had been exchanged and recently immobilized specific activity of total pool in dung (total activity I total nutrient content, Bqlmg) specific activity of the extracted fraction (activity in extract I amount of nutrient in extract, Bqlmg; extractions were described above). b. Percentage recovery of radioactivity in plant materials Re = Ac*Dm!B* 100 percent where Re = recovery percentage Ac = activity of radioactive nutrient per gram of plant (Bq g-1 plant) Dm = total dry matter of plant (g) B = total radioactivity applied to dung (Bq) c. Specific activity of radioactive nutrients in pasture and dung where SA = SA = Ac = Se = Ad Se Bq mg-1 nutrient specific activity of the radioactive nutrient, 35 S or 32p activity of radioactive nutrient in plant or dung (Bq g-1 of plant) nutrient content (mg S or P g- 1 plant) d. Percentage of plant S and P derived from dung 88 Pasture plants can derive nutrients from the extractable pool of nutrients in the dung and an unknown amount of organic nutrient in the dung which will mineralize during the field trial. During the pre-incubation the dung S and P were incompletely labelled with 35s or 32P. Boswell (1983) has shown that the mineralization of organic S from radioactively labelled sheep dung in moist soil was extremely slow, about 4.5 mg S day-1 g-1 dung S during a period of 0-37 days. Two dung specific activity values were used to calculate the percentage of plant nutrients derived from the dung. This is because a degree of uncertainty exists about the specific activity of the plant available pool in dung. Firstly, the specific activity of the extractable pool of S and P in the dung after pre-incubation and before addition to the soil was used. This represents a situation where no mineralization of unlabelled dung S and P occurred after dung addition to the soil and the highest specific activity of the S and P pools available for plant uptake. Secondly, a specific activity calculated using the total amount of S and P in the dung was used which represents the highest amount of mineralization of organic nutrient and the lowest specific activity of the S and P pool available for plant uptake. %NDFD where %NDFD SAp SAd = = = = SAp/SAd*100 percent percent of plant S and P derived from dung specific activity of radioactive 35s or 32p in plant (Bq mg- 1 of S or P) specific activity of radioactive 35s or 32p in the extractable pool in the dung or the total nutrient specific activity of the dung (Bq mg-1 of S or P) 4.3.5 Statistical analyses Analysis of variance was performed on data obtained from the treatments (three rates of dung application with pasture sampled at four radial distances away from the dung) arranged in a sp l it plot des ign . The SAS (SAS I nstitute I nc . , 1 985) and M in itab (Minitab Inc . , 1 989) programmes were employed to process the data. 4.4 RESULTS 4.4. 1 EXPERIMENT 1, AREA OF PASTURE INFL UENCED BY SHEEP DUNG 89 An area of pasture significantly influenced by a dung spot should e ither have a higher pasture yield, higher S and P uptake than surrounding area or have derived a significant amount of S and P from the dung i nstead of the soi l . The fo llowing Section (Section 4.4. 1 . 1 ) presents results used to evaluate these effects. 4.4. 1. 1 Pasture yield There was no effect of dung weight nor the proximity of pasture to the dung on the yield of pasture (Table 4.2) . This indicates an adequate supply of native soil S , probably derived from the previous superphosphate applications made annually to the trial site. Pasture yields from the second harvest, however, were greater than those of the first harvest. This was probably due to the more favorable conditions of both temperature and moisture during the second growth period. Pasture yields at both harvests during the winter period were much lower than those of the second experiment (Section 4.4.2) which was carried out during the spring? summer season (Appendix 4.7). The slower winte r growth rate may have contributed to the lack of pasture response to the dung. 4.4. 1.2 S and P concentrations in and their uptake by pasture The S and P concentrations of mixed pasture from all plots ranged from 0.1 7% to 0 .30% and 0.28% to 0 .30%, respectively. At the first harvest, the highest dung rate (03) produced higher pasture S and P concentrations than the lower dung rates (01 and 02) , as shown in Table 4.3. At the second harvest , pasture from the lowest dung rate ( 0 1 ) had a lower mean S concentration than that of the higher rates. In general , S concentrations in the pasture at both harvests, and P concentrations in the pasture at the first harvest, increased with the increasing rate of dung appl ied. The concentration of both e lements in the pastu re at increasing distances from a dung patch (R1 , R2, R3 and R4) , however, showed no significant change. The S content in pasture from the second harvest was higher than that of the first harvest. Furthermore , as a consequence of the h igher pasture yield at the second harvest , the total uptake of S by pasture in the second harvest was more than twice that of the first harvest. Although the concentrations of S in the pasture at the first harvest were low, ranging from 0 . 17- 0.21%, the total uptake of both elements showed no effects of dung rate (D) or proximity (R) of pasture to the dung at e ither harvest (Table 4.3) . The appl ication of higher dung weights yie lded h igher cumulative (fi rst plus second harvest , 1 +2) u ptake of S. There were no sign ificant i nteractions between dung rate and total u ptake of both nutrients at d ifferent distances from the dung. 90 4.4. 1.3 Uptake of 35 S and 32 P by pasture The pattern of isotope uptake must be considered prio r to calculating the amount of plant S and P derived from the dung. 4.4. 1.3. 1 Pattern of 35 S uptake The total activity of 35s recovered per unit area of pastu re at both harvests increased with increasing amounts of dung added (Figures 4.28 and 4.2C, and Appendix 4. 1 ) , but increasing amounts of dung added had no significant effect on the percentage of added 35s recovered in pasture at both harvests ( Figures 4.38 and 4.3C , and Appendix 4 .2 ) . Even though the harvested surface area was smaller in the central ring, the total recovery of 35s and therefore the percentage of added 35s recovered in pasture of the central ring at both harvests were markedly higher than that recovered in each of the outer rings (as shown in Appendix 4.2). At the second harvest, larger percentages of 35s were recovered by the pasture in areas R2, R3 and R4 than at the first harvest (data in Appendix 4.2 and Figure 4.38 compared to Figure 4.3C) . This is partly due to the higher dry matter yield and S content of pastures i n these areas at the second harvest. There may also have been more lateral movement of 35so4= away from the dung by the second month. The soil used in this study has little capacity to retain sulphate-S ( P-retention = 20-30%, see Appendix 7. 1 4) . During and Martin ( 1 968) and Kennedy and Till ( 1 98 1 b) had shown that there was rapid lateral movement of 35s out of the center zone where 35s labelled gypsum was applied to pasture soil . The total recovery of isotope in the pasture expressed as the amount of activity per square centimeter of harvested pasture showed s ign ificant interactions between dung rates and distances of pastu re from the dung, as shown in Figures 4.28 and 4.2C, and Appendix 4. 1 . The percentages of added 35s which were recovered in the pasture , however, showed no significant interaction between the rate of dung added and the distances of the pasture from the dung , as shown in Figures 4.38 and 4.3C, and Appendix 4.2. 4.4. 1 . 3.2 35 S specific activity and percent of plant S derived from dung (%SDFD) The specific activity of 35s in the harvested pasture (Appendix 4.3) and the specific activity of either the extractable-S pool or the total pool of S i n the dung (Table 4. 1 ) were used to estimate the percent of plant S which was derived from the dung . The results presented in Figures 4.48 and 4.4C and Appendices 4.3 and 4.4 show that at both harvests the percentage of plant S derived from dung (%SDFD) increased with increasing dung application rate and also increased as proximity to the dung increased. The specific activity of the 35s in pasture and %SDFD were influenced significantly by an interaction between dung rate and distance (Appendices 4.3 and 4.4, Figures 4.48 and 4.4C). 91 The percentages o f plant S that were derived from dung in the second harvest were , i n general, about 50% of the first harvest (Figures 4.48, 4.4C and Appendix 4.4) . I n addition to the percent of plant S derived from dung calculated using the specific activity of the extractable sulphate poo l ( Figu re 4.4, Appendix 4.4) , the %SDFD was also calculated us ing the specific activity of the total pool (Appendix 4 .3) because some of t h e non? exchangeable pool remaining in the dung might be mineral ized in the field after application. Table 4.1 shows the activity of the 35s in the exchangeable pool represents almost 90% of the total activity and this fraction represents 1 0% of the total dung S. There were ten-fold differences between the val ues produced by using the different specific activit ies in the %SDFD calcu lations (Appendix 4.3) . 1 t appears un likely that much organic dung S would mineralize during the span of this experiment because only about 0.03-0.04% of sheep dung S mineralized during the first 1 00 days of an experiment reported by Boswell ( 1 983) . Therefore, the %SDFD, calcu lated using the specific activity of the extractable pool , may be a more valid estimate than that calculated using the specific activity of the total dung S. l t should be pointed out here that the percentages of S derived from dung shown in Appendix 4.4 were the average value for particular zones away from the dung (0- 1 0 cm(R 1 ) , 1 0-1 5 cm(R2) , 1 5-25 cm(R3) and 25-30 cm(R4)) . I n the 0-1 0 cm zone the value was less than 0.5%. In fact, the %SDFD might be greater at a closer proximity to the dung. Unfortunately, it was impossible to harvest pasture closer to the dung. A mathematical procedure, however , was used to predict the %SDFD at distances less than 1 0 cm away from the dung. Isotopic uptake data from the D3 treatment were used to derive a relationship for the cumulative amount of plant S derived from the dung as distance from the dung increased. This relationship was integrated for distance intervals of 1 cm instead of the original 1 0 , 1 5 , 25 and 30 cm distances. In each case, the predicted total cumulative 35s derived from the dung was within 1 % of the values measured at the 1 0, 1 5, 25 and 30 cm distances. The results are shown in Figure 4.5A and Figure 4.58. Using these figures, the average amount of pasture S derived from dung during the first 30 days in a zone 1 cm away from the dung is about 2%, if the specific activity of the extractable pool was used and 1 7% if the specific activity of the total pool was used. 4.4. 1 .3. 3 Pattern o f 32 P uptake Data on P uptake was available only from the first harvest because thereafter the 32P activity in the pasture was too low to determine with any accuracy. The total recovery per unit area and recovery percentage of 32p were lower than those of 35s (Appendices 4. 1 and 4.2, F igure 4.2A compared to Figuures 4. 1 8 and 4.1 C) . Most of the activity remained in the cent e r rings (R 1 ) and the amount recovered by pasture growing in the center ring increased with the dung rate (Appendix 4.2 and Figure 4.3A) . However, the dung rates had no effect o n the total pe rcentage recovery of 32P by p lants from al l rings . More activity was recovered in the second ring of the higher dung rate ; showing that a significant interaction occurred between dung rate and distance of pasture from dung ( Figure 4.2A) . 92 4.4. 1.3.4 32p specific activity and percent of plant P derived from dung (%PDFD) The specific activity of 32p in pasture (Appendix 4.3) was used in the same manner as 35s to calcu late percentages of plant P derived from dung (%POFO) . The calculation showed that the %POFO were much higher than percent of plant S derived from dung, %SOFO, (Appendix 4.4, Figures 4.4A, 4.48 and 4.5C) . The dung application 01 gave lower %POFO than higher dung rates (02 and 03) . The %POFO in the center ring (R1 ) was much higher than those of the outer rings, as shown in Figure 4.4A. There were only small differences between the %POFO calculated us ing either the specific activity of the exchangeable or total P pool, as shown in Appendix 4.3 and Figures 4.5A and 4.58. Approximately 42% of total P was in the extractable pool and 53% of the total 32p activity was in this pool (Table 4.1 ) . lt has been suggested that the fraction of organic S and P in dung contributes little to plant nutrition in the short term (McAuliffe et al. , 1 949a, 1 949b; 8romfield, 1 961 ; 8oswell , 1 983; Rowarth, 1 987; Rowarth et a!. , 1 985, 1 988, 1 990) . Thus, the %POFO calculated using the 32p specific activity of the extractable P pool was considered to be a more valid estimate. 4. 4.2 DISCUSSION, EXPERIMENT 1 Prior to conducting this experiment it was envisaged that all other conditions being equal , the area of pasture soil influenced by dung will mainly depend on the following factors : 1 . The average maximum radial distance that roots, effective in nutrient uptake, extend from pasture species. The radial distance that roots extend horizontally in a soil is a function of the pasture species and physical soil properties that influence root ramification through soil ( i .e . site dependent) . 2 . The radial rate o f movement o f nutrients away from the dung patch into the soil is influenced by the amount of soluble nutrient present in the dung or the weight of dung deposited . In this respect, nutrients which are not strongly adsorbed by soi l surfaces and have larger coefficients of d iffusion through soi l may be expected to move to g reater distances f rom the site of dung deposit ion . Therefore sulphate was initially expected to have a greater radial distance of diffusion than phosphate and lead to a greater recovery by plants. As outlined below this expected result did not occur. The average amount of35s and 32p activity in the pasture in each ring , 35s(R1 ) , 35s(R2), 35s(R3), 35S(R4), 32P(R1 ) , 32P(R2), 32P(R3) , and 32P (R4) , expressed as the percent of total activity taken up, I.35s or I.32P(R1 +R2+R3+R4) , from each dung application (01 , 02, and 03) is shown in Table 4.4. After the f irst 30 days, the largest amounts of 35s activity 93 taken up by the pasture were h ighest in the center zone (R1 ) . However, the percentage of 35s activity taken up (or the %SDFD at 30 days, Figure 4.48) in the outer zones, R3 and R4, were smaller than the percentage of 32p taken up (or %PDFD at 30 days , Figure 4.4A) . Unexpectedly, the plant uptake of 32P had less activity in the central zone but more in the outer zone than that of plant 35s activity. This suggested that lateral diffusion of the 35s was slower than that of 32p during the first 30 days. Despite the fact that the amount of 35s activity in the labelled dung present in an extractable S pool (CaP-S) was about 40% higher than that the activity of 32p in a similar pool (Oisen P) (Table 4 .1 ) . In addit ion , by the second harvest (60 days after application) the amounts of 35s in the pasture in the outer rings were about twice as much as the amount of 35s taken up in the first 30 days but were 30-50% less than the amount of 32p taken up after the first 30 days. This also provides more evidence for the unexpected slow lateral movement of labelled sulphate S (see Table 4.4) . This result was unexpected and requ i red the d iffusion theory, discussed above , to be examine closely (Appendix 4.5) . The calculation conducted in Appendix 4.5 , however, demonstrated that the initial rates of P and S diffusion away from the dung should be similar despite the smaller d iffusion coefficient for P in soil because of a much greater gradient in P concentration between dung and soil than S gradient. Thus the differential rates of diffusion did not explain the observed differences in 32p and 35s recovery by plants A large drainage event ( 1 40 mm) occurred during the experimental period (Appendix 4.7) . Some o f the exchangeable 35so 4 = i n the dung could have been leached beyond the 1 0 cm soil depth. As most of the root activity is in the top 1 0 cm of the soil profile (Hedley et al. , 1 992, personal communication) leaching would have reduced the recovery of su lphate by the plants. I n contrast , the 32p present as exchangeable P in the dung would be strong ly adsorbed by the soil surface and would not be susceptible to leaching. A greater amount of the added 32P may remain in the active root zone and this may have caused the 32p uptake pattern to have a slightly wider spread than the 35s uptake at the first harvest (Table 4.4) . Therefore, the resu lts suggest that the relatively lower recovery of 35s than 32p activity, taken up by plants at greater distances away from dung, may result from: (a) a greater leaching loss of 35s than 32P or (b) that roots at greater radial distances from the mother plant take up P more actively than S . Although S concentration i n pasture increased with dung application rates, there was no s ignif icant effect of dung rate on plant y ie ld nor on the radial d istribution of pasture S concentration around the dung. Consequently, total uptake by the pasture did not increase as a result of dung application . This agreed with the results from the second experiment (see Section 4.4.2). The %SDFD in the zone closest to the dung patch was less than 2% ( Figure 4 .5A) . Furthermore, the influence of dung S, measured as %SDFD decreased toward the outer rings. 94 As mentioned above, the lateral movement of sulphate S was slow, thus in the short term (30 days) the area of pasture significantly influenced by the dung is expected to be less t han 5.0 cm away from the dung. Although the percentage recovery of 35s in plant at both harvests was higher than that of 32P, the %SDFD was much lower compared to that of plant P. As d iscussed earl ier , larger amounts of soi l sulphate , including 35so4= were probably leached during the heavy rain events (Hogg and Cooper, 1 964; Hogg and Toxopeus, 1 966; Gregg and Goh, 1 978, 1 979) or was transformed into organic forms before being utilized and assimilated by pasture plants (Freney et al. , 1 971 ; Gregg, 1 976) . More soil S was taken up by plants in the innermost zone. This was probably caused by a synergistic effect of other elements particularly dung N. Generally, it can be seen that dung application did provide some S to plants but only in the zone closest to the dung and this did not have a significant effect on total S uptake. This also suggested that the soil used in this trial had sufficient available S. Extractable S in the top 3 cm was approximately 1 5 mg S kg-1 , a level at which pastures are not expected to show S deficiency (Nguyen et al., 1 989a, 1 989b) . 4.4.3 EXPERIMENT 2, EFFECT OF SHEEP DUNG ON GROWTH AND YIELD OF PASTURE 4. 4.3. 1 Pasture yield Pasture yields (group means) at four harvests for different dung and fertilizer applications are shown in Table 4.5A. Neither the application of dung nor fertilizers produced a significant yield response at any harvest, and there were no significant interactions between dung application and the type of ferti l izer used (S0 or superphosphate) . Compared to the first experiment carried out in winter, pasture yields were much higher, which was probably due to the more favourable temperature during the spring and summer seasons (Appendix 4.7). 4.4.3.2 Plant S concentration There was no s ign if icant effect of sheep dung on p lant S concentration (Table 4.58). Throughout the season, pasture on control plots had a significantly lower S concentration than that of the fertilized plots. At the second harvest only, superphosphate fertilized plots yielded significantly higher pasture S concentrations than elemental S fertilized plots. Thereafter, no s ignificant differences between the two treatments were observed, nor did any sign ificant interaction between dung and type of fertilizer occur. l :-?. ;:IJ - -< < n ::n (() ? Percent Percent 1 5 T Exchangeable poo l (A) * %SDFD-1 1 ol? * %SDFD-2 5 1 '-"?r-"1""""'"- 8 %PDFD 0 1 5 T Total pool (B) * %SDFD-1 1 0?? * %SDFD-2 5 ????@ 8 %PDFD ?????? 0 0 5 1 0 1 5 20 25 30 Distance Figure 4.5 Percentage of pasture P and S derived from dung (%PDFD and %SDFD) at increasing distances from dung (Rl , R2, R3 and R4); o/oSDFD-1 and %PDFD were calculated from the first harvest data (30 days) and o/oSDFD-2 was calculated from the second harvest data (60 days) (calculated using the specific activity of the (A) exchangeable P and S and (B) the total dung P and S . tO OJ 99 Table 4.2 Pasture dry matter yield as influenced by dung application rates and radial distances of pasture from dung. . . . Harvest . . . Treatments 1 2 -2 . . . . . gm m .. . . . Dung ratesa D1 27.8 49.2 D2 26.1 53.4 D3 27.4 53.7 Distancesb R 1 29.4 46.3 R2 27.9 49.0 R3 25.2 56.3 R4 25.8 56.8 Lsd 5% Dung ns ns Distances ns ns C.V. % 25.5 23.6 ns = not significant at 5% level a averaged over all four distances and five replications b averaged over all three dung rates and five replications Cumulative yield 77. 1 79.5 81 .2 75.8 76.9 8 1 .5 82.7 ns ns 15. 1 Table 4.3 Pasture S and P concentration, S and P uptake of pasture and cumulative S uptake as influenced by dung application and radial distances of pasture from dung. Harvest 1 Harvest 2 Cumulative Harvest 1 s s s s s p p concentration uptake concentration uptake uptake concentration uptake Treatments % mg m-2 % mg m-2 mg m-2 % mg m-2 Dung ratesa D1 0 . 18 50.3 0.25 123.9 174.2 0.26 73.3 D2 0.17 45.5 0.27 143.9 189.4 0.28 74.3 D3 0.21 58.9 0.28 150.0 209.0 0.29 78.5 Distancesb RI 0.21 60.9 0.26 122.6 183.4 0.27 79.4 R2 0.17 48.6 0.26 1 30.7 179.3 0.28 77.0 R3 0.18 47.3 0.27 1 50.7 197.8 0.28 7 1.2 R4 0.19 49.7 0.27 1 53 .2 202.8 0.29 73.5 Lsd 5% . Dung 0.03 ns 0.01 ns 23.0 0.02 ns Distances ns ns ns ns ns ns ns C.V. % 2 1 .8 35.3 5.8 26.4 18.9 9.9 26.3 nS=not significant at 5% level a averaged over all four distances and five replications b averaged over all three dung rates and five replications ..... 0 0 1 01 Table 4.4 The percent distribution of 35s and 32p taken up by pasture at different radial distances from 35s and 32P labelled dung applied at various rates. Dung Distances DI D2 D3 Means PART A (FIRST HARVEST) . . . . . % 358 .. . . . RI 83. I 88.7 80. I 83.4 R2 6.6 7.4 I3.5 10.6 R3 7.I 3.2 5 .I 4.7 R4 3.2 0.6 1 .3 1 .3 PART B (SECOND HARVEST) 35 .... . % s ..... Rl 55.7 68.9 64.8 64.9 R2 I3.4 I 1 .3 I7.9 I5.0 R3 I9.9 I4.3 13.5 I4.7 R4 10.8 5.5 3.7 5.3 PART C (FIRST HARVEST) 32 .... . % p .. . . . RI 40.6 45.6 48.2 46.6 R2 20.6 27.3 23. I 24 . I R3 20.3 18.8 21.6 20.7 R4 18.4 8.2 7.I 8.6 1 02 4.4.3.3 Plant S uptake Amounts of plant S uptake are shown in Table 4.5C. At the second harvest which was taken 30 days after dung application, the dung treatment resulted in significantly lower S uptake than the no dung treatment. This may be due to immobilization of the available S during the initial decomposition of dung by organisms. There was no significant effect of dung in the following harvests. Overal l , the contro l treatment had lower S uptake than the fertilizer treatments, although the differences were not s ignificant. There was no s ignificant interaction between dung and type of fertilizer. 4.4.4 DISCUSSION, EXPERIMENT 2 I n this experiment an amount of dung, similar to that returned by sheep after one grazing was applied uniformly to unfertilized pasture plots and plots fertilized with S0 and superphosphate. S imi lar to experiment 1 , dung appl icat ion had no effect on dry matter yield. Fert i l izer application did affect the pasture S concentration and uptake throughout the period but again did not influence pasture yield . The increase in S concentration and uptake without any increase in yield due to fertilizer application represents the luxury uptake of S. The crease in S content and total S uptake (Tables 4.5B and 4.5C) in fertilized plots above the control ranged from 6-41% and 1 2-36%, respectively, but this did not increase dry matter yield on fertilized plots (Table 4.5A) . As discussed in Section 4.4.2, the trial area had a sufficient soil S status. Considering the contribution of dung S to pasture plants in this experiment; amounts of less than 1 .0 kg S ha-1 were applied to the experimental plots (375 kg ha-1 of dung, total S = 0.26%) . The total available S added was probably less than 0 . 1 kg S ha-1 . Therefore, this amount should not have any effect on pasture as compared to the more readily available fertilizer forms, such as superphosphate and slowly released S0. Watkin ( 1 954) and Skrijka ( 1 979) have shown that the sheep dung had no effect on pasture yield. Recently, in New Zealand, Morton and Baird ( 1 990) have shown that the return of N from dung and urine (which contained more N than S) affected an insufficient area to influence pasture growth even at high stocking density ( 1 800 sheep ha-1 day-1 ) . Furthermore , Smith ( 1 976) and Ledgard et al., ( 1 991 ) have shown that, for well developed, regularly fertilized, permanent pastures, fertilizer S can be withheld for 2-3 years before there is a measurable S deficiency. Presumably a similar period for S responsiveness to occur is required after dung and u rine S is withheld. 1 03 Table 4.5 Pasture. dry matter yield (PART A), sulphur concentration (PART B) and total pasture S uptake (PART C) as influenced by dung and fertilizer applications at five harvests. Days after fertilizer application Treatments 30 60 90 120 150 PART A, DRY MATTER YIEW -2 .... . g m .. . . . Dung applicationa w/o dung ND 199.5 201 .2 1 19 . 1 178.8 w/ dung ND 194.5 194.6 131 .6 1 89.0 Fertilizersb Superphosphate 182.6 190.0 190.4 121 .2 185.8 Elemental S 203.0 208 .1 209.2 125.1 180.0 Control 228.9 193.0 193.7 129.7 177.9 Lsd 5% Dung ns ns ns ns Fertilizers ns ns ns ns ns C.V. % 14.2 9.9 9.5 18.5 19.0 PART B, S CONCENTRATION . . . . . % .... Dung applicationa w/o dung ND 0.47 0.41 0.48 0.44 w/ dung ND 0.44 0.4 1 0.46 0.40 Type of fertilizersb Superphosphate 0.46 0.54 0.44 0.49 0.45 Elemental S 0.44 0.42 0.44 0.49 0.44 Control 0.33 0.39 0.36 0.42 0.35 Lsd 5% Dung ns ns ns ns Fertilizer 0.09 0.04 0.06 0.03 0.06 C.V. % 14.4 9.9 15.0 7.5 14.0 PART C, S UPTAKE -2 .... . mg m ... . . Dung applicationa w/o dung ND 94.0 83.3 60.1 8 1 .4 w/ dung ND 84. 1 Type of fertilizersb 79.9 62.7 8 1 .8 Superphosphate 84.0 102.2 83.9 60.1 81 .4 Elemental S 87.3 87.9 91 .4 61 .5 86.1 Control 75.0 77.1 69.5 53.5 63.2 Lsd 5% Dung 8.9 ns ns ns Fertilizers ns 1 1 .4 15 .1 10.9 18.8 C.V. % 12.6 12.3 17.7 18.5 23.4 ns = not significant at 5% level; NO = not determined a averaged over three fertilizer types and four replications b averaged over two dung rates and four replications Table 4.6 1 04 Calculated percentage of a paddock annually influenced by dung considering camping and noncamping behaviour of sheep. noncamping campinga campsite noncampsite total . .. . . percent of paddock ..... Radius of the area of influence per ?atch Rp + Rd = Rt 6.4 + 10.0 = 16.4 cm 27. 76(2.3 1 )c 5.00 18.60(1.55) 23.60 6.4 + 8.0 = 14.4 cm 21 .40( 1 . 79) 5.00 14.34(1 . 19) 19.34 6.4 + 6.0 = 12.4 cm 15.87(1.32) 5.00 10.63(0.88) 15.63 6.4 + 4.0 = 10.4 cm 1 1 .16(0.93) 3 .68 7.48(0.62) 1 1 . 16 6.4 + 2.0 = 8.4 cm 7.28(0.60) 2.40 4.80(0.40) 7.20 6.4 + 0.0 = 6.4 cm 4.22(0.35) 1 .39 2.83(0.23) 4.22 Annual S return from dungd kg ha- 1 9.85 3.25 6.60 9.85 Readill plant available fractione kg ha- 0.985 0.325 0.66 0.985 lt is assumed that average radius of dung patch=6.4 cmf; 6 excretions/day/animal9; and 1 5 sheep/ha/year and dung patches were evenly distributed over the area a 33 percent of the total dung output located in 5% of paddock (Hilder, 1964) b Rp=radius of dung patch Rd=extended radius of influence Rt=total radius of influence c numbers in parenthesis are percent of paddock influenced monthly by dung d calculated from mass of dung of 0. 1 kg/excretion and 0.3% total S e calculated from mass of dung of 0. 1 kg/excretion and 0.03% Cap-S f Herriott and Wells (1963) g Sears and Newbold (1942); Herriott and Wells 1963 and Skrijka, 1987 4.4.5 Area covered by dung and Influenced by dung sulphur 1 05 A simple calculation of the percentage of a paddock annually and monthly covered by sheep dung and dung S when both camping and non-camping behavior of sheep are considered is presented in Table 4 .6 (calcu lat ions for the number of sheep and amount of dung are presented in Appendix 4.6). With 15 sheep per hectare, the total area covered by dung (radius = Rp), annually, is less than 5% unless there is an extended area of influence ( radius = Rd) on pasture growth around the dung. Results from the first experiment showed a very small contribution of S from sheep dung to the surrounding pasture within 5 cm of the dung. The maximum amount of plant S derived from the dung was about 2% for the pasture at the centre , where the labelled dung was deposited. This small amount , and the small increase of the total S uptake in the fertilized p lot of the second experiment, are considered to have insignificant effects on increasing plant dry yield. Thus, based on these results (percent of plant S derived from dung, pasture yield, and S and P uptake) , the extended radius of influence ( Rd) of dung S on pasture growth is small being <5 cm. This suggested that approximately 3-1 1% of paddock may be influenced by dung return each year (Table 4.6) 4.5 CONCLUSION Although about 5% of the paddock is covered by discrete sheep dung patches annually, in the short term the pastu re d ry matte r y ie ld in the a rea d i rectly adjacent to the dung was insignificantly influenced by the nutrients P and S contained in the dung. Furthermore , the presence of the dung d id not s ignificantly inf luence the immobil ization of so il o r ferti l izer derived phosphate and sulphate. Very small amounts of plant P and S (about 2-5%) within 1 0 cm of the dung were derived from the dung . The total available nutrients in the dung are very small compared to those of readily available chemical fertilizers or that of available forms already existing in the soil. The majority of plant P and S were derived from soi l sou rces, a large part of which will be derived from slowly mineralizing dung that has accumulated with time. The fertility of the soi ls at these experimental sites has been well maintained to ensure an optimum pasture production . Even when the dung at the normal rate of application (375 kg OM ha- 1 g razing-1 ) was evenly distributed over the whole pasture, there was no effect on pasture yield or nutrient content . 1 06 This study confirms Boswell 's ( 1 983) finding that the short term effect of dung on nutrient availability was small. Considerable amounts of S (approximately 34% of the S taken up from soil by pasture (Saggar et al. , 1 990a, 1 990b) are returned to the soil as dung. Therefore the role of dung is one of a slow S release component as it is decomposed by soil organisms and converted i nto soi l humus which continues to re lease S . If the C : S rat io of the non? exchangeable S in the dung (Table 4. 1 ) is approximately 200 :1 and the soil microflora have a C : S ratio of 45:1 then it would take 2 generations of organisms with a growth yield of 0.4 ( i .e . 60% of the carbon is evolved as C02 and al l S is conserved at the s ite of decomposition) before a substantia l amount of the dung S cou ld be re leased as so4=. The average generation time of microflora growing on the recalcitrant organic matter in dung may exceed 1 year ( Dr , K . Tate , perso nal commun icati on ) . Such s imp le calcu l at ions support the experimental evidence that dung S is a slow release S form. I n conducting future field trials designed to examine the short term fate of fertilizer S in well developed permanent pastures, it is not necessary to apply sheep dung to the experimental plots in order to simulate the effect of dung return because probably this will have little effect on the size of the 'bank' of the dung at various stages of decomposition in pasture soils. For long term (several years) experiments on permanent pasture o r experiments on less well developed pasture where soil fertil ity is very low, animal dung may be needed to be deposited to maintain soil organic S reserves. Many authors have accepted that the distribution of urine is similar to that of dung ( Petersen et al. , 1 956; Til l , 1 975; Taylor, 1 980) . Using data presented by Gil lingham, ( 1978) ; Gi l l ingham, ( 1 980) and Gill ingham et al., ( 1 980) it can be calculated that during a month , urine is returned to approximately 5% of the paddock if there is no overlap of urine spots. lt is well known that applications of sheep urine stimulate increased pasture growth (Sears and Goodal l , 1 948; Watkin, 1 954; Kennedy and Til l , 1 98 1a ; Boswell, 1 983; Williams et al. , 1 989; Morton and Baird, 1 990; Sakadevan, 1 99 1 ) mainly through the high concentrations of N, K and S that are returned to the soil through the u rine . At the t ime of th is study, other Ph .D . programs were i nvestigat ing u rine effects i nclud i ng S (Sakadevan, 1 99 1 ) and i t was considered unnecessary to duplicate these. 1 07 CHAPTER S EVALUATING F IE LD EXPER IMENTATION TECHN IQUES US ING 35s LABELLED FERTILIZERS TO TRACE THE FATE OF SULPHATE AND ELEMENTAL S APPLIED TO PASTURE SOILS 5.1 INTRODUCTION In many New Zealand soils the literature review indicates that sulphate-containing fertilizers such as gypsum, in superphosphate, may be prone to loss by leaching if applied in seasons when rainfall intensity and drainage are high. To some extent the amount of S loss may be reduced by spring rather than autumn application of superphosphate (Toxopeus and Gordon, 1 971 ; Nguyen et al. , 1 989a, 1 989b) . However, Saggar e t al. ( 1 990a, 1 990b) estimated that even when applied during the spring and summer, the long term leaching losses of S from North Island hill country pasture ranged from 40-77% of the fertilizer (superphosphate) applied. Sakadevan ( 1 991 ) showed that approximately 30% of this S loss was leached directly from the fertilizer application. Currently the use of S0 based fertilizers instead of sulphate-S fertilizers is recommended to reduce S leaching loss (Sinclair and Saunders, 1 984) . The mode by which it reduces leaching losses is either through slow release from S0 fertilizer maintaining lower soil sulphate concentrations during the period of intense leaching events or that relatively more S being incorporated into plants and soil organic matter from S0 than sulphate-based fertilizers. The relative importance of each process is unclear. D ifferent forms of S fertilizers, labelled with 35s, can be applied to pasture soils to test whether the fate of S0 and sulphate-S differ. I n studying the fate of 35s labelled fertilizers in soi l , two techniques are generally employed, direct labelling and inverse dilution ( IAEA, 1 976; Vose, 1 980; Manzel and Smith, 1 984) . I n the direct labelling technique the fertilizer materials are labelled with 35s radioisotope and added to soil systems. Then , after certain periods, plant and soil samples are harvested and determined for radioactivity in the samples. For example, Gregg and Goh ( 1 978, 1 979, 1 982) and Goh and Gregg (1 982a, 1 982b) used 35s labelled gypsum in the study of the fate of fertilizer S in pasture systems in South Island, New Zealand. In the inverse dilution technique the soil system is labelled with carrier-free 35s radioisotope. After equil ibrium is obtained unlabelled fertilizers are applied to the soil and plant systems. The S released from the fertilizer material will dilute the specific activity of 35s in the system. 1 08 Changes in the ratio of specific radioactivity (SR) of 35s in soi l o r plant grown on fertil izer treated cores are measured. The specific activity ratio is calculated as follows: Where SR = SR = SAT = SAC = SAT / SAC specific activity ratio specific activity of 35s in treated cores, Bq g-l S specific activity of 35s in control cores, Bq g-l S When there is little release of S from fertilizer the ratio (SR) is about 1 . Release of S from fertilizers will result in a reduction of this ratio (SAT is less than SAC) . This technique has the advantage that comparison can be made of materials which may be difficult to label (e .g. rock phosphate, dung and urine). For example, Boswell (1 983) employed this technique to study the fate of sheep dung and urine in pasture soil. However this technique has a drawback in that the nutrient re leased from ferti l izer may be i ncorporated i nto non-active forms hence resu lts may overestimate the plant availability of any nutrient released to soi l (Freney et al. , 1 971 , 1 975) . Alternatively addition of fert i l izer may stimulate root growth leading to higher isotope recovery. 5.2 OBJECTIVES The general objective was to use 35s as a tracer to study the fate of different forms of fertilizer S, namely sulphate and S0 , when applied to a pastoral soil. Two field trials were conducted to meet the general objective. The first experiment investigated the redistribution of fert i l izer-35s between forms of soil S, plant S and sulphate leached from the top ten centimetres of pasture soil during a period of 1 50 days after a late spring fertilizer application. Two fertilize r S sources were used , namely, superphosphate (SSP) and e lemental S (S0) ; both were labelled with 35so4= and 35s0, respectively. The second short term experiment (60 days) investigated the use of an inverse isotopic dilution technique to determine the influence of applied fertilizer S on the size of the labile soil 35so4= pool. I n the study a solution of carrier-free 35s was used to pre-label soil S prior to fertilizer applicat ion. 5.3 MATERIALS AND METHODS 5.3. 1 EXPERIMENT 1, 35s LABELLED FERTILIZER 1 09 Superphosphate (SSP) and S0 were chosen for this experiment since these two fertil izers are common S fertilizers applied to New Zealand pasture soils. A microplot ( Peverill and Douglas, 1 976; Shedley et al. , 1 979 ; Vose, 1 980; Martin , 1 985; Will iams, 1 988; Destain et al. , 1 989) technique was employed. 5.3. 1. 1 Preparation of 35 S labelled s:> and SSP An aqueous solution of 35so4= and a solution of 35s0 dissolved in toluene, were obtained from the Radiochemical Centre , Amersham, U.K. 5.3. 1. 1 . 1 35 s:> labelled s:> About 462 MBq of carrier-free 35s0 in toluene were mixed with 1 1 0 ml toluene containing 1 .22 g of dissolved unlabelled S0 (or 53 mg per 5 ml) . Five ml of the mixture were pipetted onto a 1 5 g layer of soil in a glass petri dish and dried overnight in a fume hood. This yielded 20.86 MBq per 53 mg of S with a 35s specific activity of 394 KBq mg- 1 S (Table 5 . 1 ) . Later observations using a light microscope indicated that the S0 produced using this process was microfine with no particle exceeding 0.01 0 mm in diameter. Details were described in Section 3 . 1 .3. 5.3. 1. 1.2 35 so4= labelled SSP SSP labelled with 35so4= was prepared by adding carrier-free 35so4= to the sulphuric acid used to acidulate phosphate rock (Nunn and Dee 1 952; Young et al. , 1 985; Bolan et al. , 1 987) . Details were described in Section 3 . 1 .4. About 9 g of a mixture of 50 :50 Christmas I sland 'A' grade :Nauru phosphate rock was acidulated with 8 .45 g of 65% H2so4 containing 585 MBq of 35s ( 1 5.8 rnci) at an acid :rock ratio of 0 .61 and then allowed to dry for 20 minutes. The product was dried overnight at 30 ?C and then ground i n a mortar and passed through a 1 00 mm sieve. The ground SSP was divided into 20 samples, 0 .45 g each. Each sample contained 1 8.24 MBq of 35s activity with 35s specific activity of 342.07 KBq mg-1 S and 1 1 .85% S (Table 5 . 1 ) . 1 1 0 5.3. 1.2 Trial method The two sources of fertilizer-S, 35s labelled SSP and S0 were applied to a 1 76.78 cm2 area contained 1 5 cm diameter and 1 0 cm deep galvanized steel cores (microplot) . Treatments were repl icated four t imes and arranged in a randomized complete block allowing for five separate harvest dates , provided by 40 separate soil cores. The galvanized cyl inders had been driven into the permanent pasture to isolate soil cores about 9 cm deep. On October 1 5, 1 985, both fertilizer S forms were broadcast evenly over the soil cores at the rate of 30 kg S ha-1 . Monocalcium phosphate, giving an equivalent amount of P to the SSP treatment was also applied to cores fertilized with S0. A basal application of muriate of potash (KCI) at the rate of 40 kg K ha-1 was also given to all cores. 5.3. 1.3 Sampling and preparation of samples At each harvest ( i .e . 30, 60, 90, 1 20 and 1 50 days after fertilizer application) herbage from cores (microplots) was cut to within 2.5 cm of the soil surface and four replicate cores of each treatme nt were destru ct ively sampled and transported to the laboratory. Schemes for sampling herbage and soil are shown in Table 5.2. After removing herbage stalks from the surface of each soi l core , the core was divided i nto t hree equal sections of about 3 cm thickness; top (0-33 mm) , middle (33-66 mm) and bottom (66- 1 00 mm) . The soil from each depth was mixed and approximately 5.00 g of the fresh wet soil was randomly sampled for analyses on fresh, f ield-moist soi l . The remaining soil samples were put into plastic bags, frozen and then freeze-dried for 7 days. The freeze-dried soils were ground by a hammer mill . Both root and herbage samples were dried in an oven at 80 ?c for 48 hours. A domestic coffee grinder was used for grinding herbage samples. Table 5.1 3Ss Labelled Fertilizers SSP so Table 5.2 Fertilizers 1 1 1 Characteristics of 35s labelled fertilizers used in this study. 3Ss 3Ss Fertilizer s Activity Specific applied applied applied activity mg core-1 mg core-1 MBq core-1 KBq mg-1 S 450 53.3 18.24 342.05 53 53.1 20.86 393.58 Harvesting schedules of herbage from microplots at each harvest and number of soil cores destructively sampled. Days after fertilizer application 30 60 90 120 150 Number of replications HERBAGE CUT SSP 20 16 12 8 4 so 20 16 12 8 4 SOIL CORES SSP 4 4 4 4 4 so 4 4 4 4 4 5.3.2 EXPERIMENT 2, INVERSE ISOTOPIC DIL UTION EXPERIMENT 1 1 2 A field experiment using carrier-free 35s was also carried out in conjunction with the previous experiment. The 35s carrier-free solution was injected into the top layer of fertilized pasture soil enclosed in micro plots. After injection and prior to fertilizer application, the 35s isotope is expected to exchange with the labile soil sulphate pool. The main objective was to investigate the influence of fertilizers on the fate of the labile soil S pool. 5.3.2. 1 Carrier-free 35 S preparation and injection technique An aqueous solution of 35so4= was obtained from the Radiochemical Centre, Amersham, U.K. A 702.08 MBq ( 18 .975 mci) carrier-free 35so4= solution was diluted with 200 ml of deionized water and the solution containing 3.51 MBq mr1 was used for labelling soil in microplots. A microplot injector (M.J . Hedley and R.W. Tillman, 1 988, personal communication) as shown in Figure 5.1 was used to inject S carrier-free solution to soil cores. 5.3.2.2 Trial method The microplot technique (as described in Section 3 .2 .2) was also used in this experiment. Twenty four galvanized stee l cylinde rs were pressed i nto plots of the mowing p lot trial (described in detail in Section 4.3.2) where plots having three ferti l izer treatments (SSP , S0 and control) arranged in RCB design with four replications were laid out. Two cylinders were put in each plot. On October 1 5, 1 985, about 6.8 ml of the carrier-free solution was injected into the top 1 -6 cm soil layer in each microplot using the microplot injector (Figure 5 . 1 ) . A week afte r the injection, granulated SSP and commercial grade S0 ( 1 00% <0.250 mm and 50%<0 . 150 mm particle size) were surface applied individually to these microplots at the rate of 30 kg S ha-1 . A basal application of monocalcium phosphate and muri ate of potash was also applied in the same manner as described in Section 4 .3.2. 5.3.2.3 Sampling and preparation of samples At 30 and 60 days after the injection, all herbage from the microplots was cut to within 2.5 cm of the soil surface and the four replicate cores from each treatment were destructively sampled and transported to the laboratory. After removing herbage stalk, the whole sol i cores and 1 1 3 herbage were dried in a forced-draught oven at 60 ?C. The dried soil was then f inely ground using a hammer mill. No attempt was made to subdivide soi l cores into 3 sections as in the previous experiment. The herbage samples were dried and ground in the same manner as the previous experiment. 5.3.3 Experimental site A permanent pasture site on the Tokomaru silt loam soil (yellow-grey earth , Fragiaqualf) was chosen for the study. lt was located on the Keeble farm, 4 km south of the Massey University campus. The pasture consisted main ly of perennial ryegrass, white clover, subterranean clover together with other less dominant species of grass weeds. Before this experiment was started, the paddock had received S at the rate of approximately 20-30 kg S ha-1 annum-1 as SSP. The average annual rainfall is about 1 000 mm, distributed throughout the year but normally peaking in the months of December and June. Cattle and sheep had rotationally grazed the trial area. After the experiment had been established the experimental area was fenced to prevent access by stock. Before the fert i l izer application, pasture on plots was mown to about 2 .5 cm i n he ight above ground level and discarded. This was done in conjunction with the small plot mowing trial described in Section 4.3.2. A general background of the soil used is given in Chapter 7, Table 7. 1 , and some chemical properties of the soil are shown in Table 5.3. Table 5.3 Chemical propertiesa of S in three soil layers collected before the experiment. Layer Depth Total Organic Ester Carbon Extractable S cm s s sulphate bonded s s CaP-S CaCI-S mg S kg- soil Top 0-3 540 525 184 343 15.1 9 .0 Middle 3-6 380 382 193 189 7.0 1 .2 Bottom 6-10 320 3 12 210 102 7.1 1 .4 a analytical methods were those of Landers et al., 1983 T rovo t l lno 1 --) o o c t l on ? 1 m t p l n n t l n ;:/;. n y r i n v o ( >< 2 0 ) ___ /'; ? -) ? ?--? l l ypo t l n nn l o q rn t ln- n l. 11 l n l n n n n l nn l /;, l n j n o t n r nnod l n // ( K l U ) '/:: ? ? ?; // ? ?? 1 1 4 - 'l'hr.,adod !l oo tlonn to movo t ravo l l i ny s oo t lons wh ioh n lmu l t o noouo l y f i l l o r omp ty t l ul n y r i ngoa and rn l s o o r lowor t 1 1o nood los i n to tho g round V T rnvoll lng a ac t lon Figure 5.1 A cross sectional diagram of the injector system used to inject soil cores with radioisotope in the inverse isotopic dilution experiment (Hedley and Tillman, personal communication). 5.3.4 Chemical analyses Chemical analyses of samples are listed in the following Table : Table 5.4 Chemical analyses of soil, plant and fertilizer samples Analyses Total S in soil sample S0 in soil samfle Extractable S Soil organic S Total S in herbage Total S in fertilizers HI-reducible S in extract Radioassay of 35s extractable S (CaP-S) performed on freeze-dried soil 5.3.5 Statistical analyses As described in Section 3.3.3 3 .3 .4 3.3.2 . 3.3.5 3.3 . 1 . 1 3.3.7 3 .3.6 3.3.8 Chapter 3 3 3 3 3 3 3 3 1 1 5 To establish whether treatment effects were significant the measurements were subjected to variance analysis using the SAS package. Least significant differences (lsd) at the 1 and 5% level of probability were used for comparison of means between treatments and result of the comparisons were presented in appropriate tables and appendices. The SAS (SAS I nstitute Inc. , 1 985) and M initab (Minitab Inc. , 1 989) computer programmes were employed. All 35s activity data were normalized to the day when the 35s fertil izers or injected solution were applied (October 15 , 1 985) using the relationship as described in Section 4.3.4. Some relevant herbage 35s and soil 35s calculations are presented as described in Section 4.3.4. Additional calculations are as follows: a. Proportion of 35s in different soil fractions in soil cores; amounts of 35s (CaP-35s, carbon? bonded 35s , ester- 35so4 =? organic 35s) expressed as a percentage of total 35s remaining in soil cores. = = = = % proportion of 35s in soil S fraction at time t amount of 3Ss activity appearing in that fraction at time t total 35s activity remaining in soil cores at time t b. Percentage of applied 35s cumulatively taken up by pastures = = = = recovery percentage of 3Ss in pasture at time t amounts of 35s activity cumulatively taken up by pasture at time t total amount of 3Ss activity in S labelled fertilizer applied at day 0 1 1 6 A short term experiment was also designed to examine the recovery from soil of total 35s activity from labelled fertilizers applied at day 0. Gypsum and microfine S0 contain ing 35s activity of approximately 2.4 and 2.7 MBq g-1 S, respectively, were applied to slices (0-3 cm) of surface soi ls which were sampled from the f ield using the same size of galvanized steel cylinders as described in Section 5.3 . 1 .2. There were three application rates for each fertil izer and the two treatments were replicated four times. After the label led ferti l izers were evenly applied to the soil surface, the whole soil cores were freeze-dried for 7 to 1 0 days. The dried samples were hammer milled before total 35s determination. Results are shown in Appendix 5 .1 . The recovery of 35s activity was consistently low, 65-78% of that applied in fertilizer. There was no significant difference between application rates. The mean, 73%, calculated across all application rates , was used as a recovery factor to calculate the total fertilizer 35s remaining in each soil core. The low recovery represents loss of fertil izer S during soi l preparation for count ing because spikes of 35so4= added to soi l prior to digestion gave on average 1 00% recovery of spike activity. 5.4 RESULTS 5.4.1 Initial soli S properties and weather conditions during experiments 1 1 7 In itial soil S and other related properties are shown in Table 5.3 and Table 7.2 (Chapter 7) . I n general, the available soil S as measured by CaP-S at 10 ppm was considered to be optimum (Sincla ir et al. , 1 985) for pasture growth . This results from a history of regular annual top dressing with SSP at the rate of approximately 250 kg ha-1 . Sulphate retention was low hence some sulphate in soil solution is expected to leach down the soi l profi le. Other properties such as soil Olsen-P and pH were also near opt imum for the study rate carried (Cornforth and Sinclair, 1 984) . During the trial period, the experimental site received rainfall of approximately 554 mm and had total drainage water of approximately 1 76 mm (calcu lated according to Scotter et al. , 1 979) as shown in Appendix 5.9. At the beginning of the trial about 5 mm of drainage water occurred. Larger amounts of drainage occurred during the second half of the trial period which may have influenced the leaching of sulphate. 5.4.2 EXPERIMENT 1, 35s LABELLED FERTIL IZERS I n this section the influence of different fertilizers on the redistribution of soil S and fert i l izer S and 35s are considered. 5.4.2. 1 Accumulated herbage dry matter yield and total S uptake and comparisons between microplot and small plot experiments. The accumu lated dry matter yield and total S (so i l S plus labelled fertilizer) u ptake for five harvests i n the microplot and small plot trials are presented in Table 5.5. The range of mean resu lts of these key parameters are simi lar for both microplot and small p lot tria ls . The microplot technique produced sl ightly more variable results (%C.V. for microplot range from 7 to 24% cf . smal l plot 5-3 1 %) . In experimenting with labe l l i ng fert il izers , the m icroplot technique is more preferable as it requ ires small amounts of costly isotope and the labelled fert i l izer can be more easily contained to meet safety regulations. On average the results indicated that measurements made using the field microplot technique are applicable to and real istical ly s imu late the results of the nearby small plot tr ial . This indicates that in both techniques, herbage S may be drawn from similar soil depths and volumes of soil. Thus, it appears probable that results from the microplot technique can be extrapolated to a larger field scale. 1 1 8 In general , total uptake of S by pasture from both fertilized treatments were higher than that of the unfertilized treatment ( in the small plot experiment) . Pasture growth, however, was not responsive to additional S fertil izer and increased S uptake did not generate h igher herbage dry matter production on the S fertilized plots. Therefore, the results indicate that there was luxury consumption of the S by herbage in the fertilized plots. Du ring a f ive month period (spring-summer seasons) , pasture production had a mean cumulative dry matter y ie ld of about 9.0 t ha-1 on ferti lized plots and a mean cumulative S uptake of about 41 kg S ha-1 (Table 5.5 and Figure 5.2) . However, in the control plot about 34 kg S ha-1 of native soil sulphate S were taken up by pasture plants. Apparently, about 7 kg S ha-1 ( 1 7.5%) might be derived from fertilizers (fertilized plot - control plot) but from the 35s recovery on the 0. 1 77 m2 cores the actual fertilizer uptake ranged between 5.3 and 4.3 kg ha- 1 with SSP and s0 respectively (see later discussion) . Throughout the experiment, patterns of the cumulative plant S uptake on plots fertil ized with SSP and S0 were similar. Cumulative plant S uptake in both fertil ized plots increased l inearly with time ( Figure 5 .2 and Appendix 5.2) while the amount of soil total CaP-S (F igure 5.2 and Appendix 5 .3) remained relatively unchanged after 60 days. The results indicated that during the trial period (October 1 4 , 1 985 to March 1 5 , 1 986) an average rate of S uptake from soil was about 0 .45-0.48 mg S core-1 day-1 , o r approximately 0.25 kg S ha-1 day-1 . Without mineralization of organic S or atmospheric deposition, the total amount of CaP-S extracted from soil wou ld be expected to decrease with time. As no significant decrease occurred after 60 days, then significant net minera l ization of soil organic S must be invoked to explain the continued linear increase in herbage S, particularly during the period 60-1 50 days. This is the period after which much of the fertil izer S and original soi l S (sulphate-S) had been rapidly immobilized by the soil o rganic cyc le , or removed by l eaching . The 35s plant u ptake and soi l 35s fract ionat ion data presented in the following Sections should support or refute the mineralization hypothesis. The contribution of S from atmospheric inputs within one month is considered to be minor (Smith 1 979; Ledgard and Upsdel l , 1 99 1 ) . During the trial period, there was about 554 mm rainfall and the S concentration in the rainfall would have been about 2 ppm (Smith, 1 979; Heng, 1 99 1 ) . Therefore , on average, the contribution of S from rainfall was less than 2 mg core-1 during the trial period. But the total S taken up by plant was about 70 mg core?1 . 1 1 9 Table 5.5 Cumulative dry matter yield and total pasture S uptake (soil plus fertilizer S) in microplots and small plot experiments during spring-summer seasons (October, 1985 - February, 1986). Days after fertilizer application Fertilizers 30 60 90 120 150 SMALL PLOTS -2 Dry matter yield .. . . . g m .. . . . Superphosphate 183 385 590 670 884 Elemental S 1 202 399 598 762 918 Control 229 428 628 725 908 Lsd 5% ns ns ns ns ns C.V. % 14.1 8.7 8 .1 6.9 8.2 Cumulative S uptake .. . . . g S m?2 . . . . . Superphosphate 0.84 1 .96 2.88 3.41 4 .12 Elemental S 1 0.73 1 .75 2.62 3 .23 4.13 Control 0.75 1 .56 2.28 2.76 3.38 lsd 5% ns 0.28 0.3 1 0.28 0.3 1 C.V. % 12.60 10.20 7.40 3 1 . 10 5.20 M/CROP LOTS Cumulative dry matter yield -2 .... . g m .. . . . Superphosphate 181 385 572 702 860 Elemental S 1 175 345 487 781 990 F-test ns ns ns ns ns C.V. % 24.0 9.9 7.3 12.2 19.5 Cumulative S uptake s -2 .. . . . g m . . . . . Superphosphate 0.80 1 .78 2.32 2.91 3.95 Elemental S 1 0.78 1 .65 2.01 3 .18 4 .18 F-test ns ns ns ns ns C.V. % 21 .7 13 .1 17.6 1 1 .7 19.2 Sulphur concentration .. . . . . . % .. . . . . . . Superphosphate 0.44 0.47 0.39 0.47 0.41 Elemental S 1 0.39 0.47 0.40 0.47 0.44 F-test * ns ns ns ns C.V. % 1 1 . 1 15.7 16.2 22.3 29.0 ? = significant at 5% level; ns = not significant 1 particle size = <10J,Lm SDFF and amount of S rrtg S/core 80 70 60 50 40 30 20 10 e Figure 5.2 e 30 60 90 12e 158 e 3B 6EJ 98 120 15B Superphosphate Days M icrof ine S Cumulative plant S uptake (fertilizer plus soil S) , amount of plant S cumulatively derived from 35s labelled fertilizer (SDFF), extractable CaP-S at three soil depths and amount of S applied (microplots). [Ill CaP-S top ? CaP-S m idd le 0 CaP-S bottom a;) S a pp I ied 53 rrtg S/core -?- SDFF -L:>-- S uptake mg S/core ..... 1\) 0 1 21 Of the total plant S uptake, the fractions of herbage S which were derived from labelled fertilizers (%SDFF) ranged from 19 to 29% at the 30 day harvest and decreased to 1 0-1 4% at the 1 50 day harvest (Appendix 5.2) . In total this was equivalent to 7.6-9.3 mg S core-1 (4.3- 5.3 kg ha-1 ) (application rate=53 mg S core-1 = 30 kg S ha-1 ) derived from fertilizer at the final harvest ( Figure 5 .2) . This value is lower than the increase in S uptake caused by fertilizer application (approximately 7 kg S ha- 1 , as mentioned above) and i ndicates that fert i l izer application induced greater soil S uptake - commonly called 'priming effect' . 5.4.2.2 Total recovery of 35 S activity in soil and plants Total recovery of 35s in plant and soil samples are shown in Appendix 5.4 and Figure 5.3. In general, total recovery of 35s was higher in the 35so treated core. I n the first 30 days considerable amounts of 35s activity were lost and on ly 51% and 65% of the 35s were recovered from total soil and plant analyses in the SSP and S0 cores, respectively. Thereafter the total 35s recovered decreased further; however, the decrease in soil 35s mainly occurred in the top 3 cm of soil in S0 fertilized core and in the top 6 cm in the SSP fertilized core and was mainly accounted for by plant uptake (Figure 5.3) . I n soil, more residual 35s from the fertilizer remained in the S0 treated core than that of SSP core. I rrespective of fertilizer treatment, the majority of total 35s remained in the top 3 cm of the soil cores (Appendix 5 .5 and Figure 5 .3) . On average the activity remaining in the S0 core was significantly larger (>2 fold) than in the SSP core , particularly in the top and m iddle soil layers (0-6 cm) . This may be agronomically significant in terms of fert i l izer S conservation since most pasture plant roots occur in the top 1 0 cm of soil (Williams, 1 988) . At the conclusion of the experiment ( 1 50 days) about 60.8% and 44.2% of S from SSP and S0, respectively, were not accounted for. The largest losses of 35s occurred in the first 30 days of the experiment. 5.4.2.3 Extractable S in soil (CaP-S) The total amount of CaP-S (soi l CaP-S plus amount derived from labelled fertilizer S) , the percentage of CaP-S derived from ferti l izer (numbers in parentheses) and the amounts of extractable S derived from soil in each layer, and total amounts of CaP-S in the whole core are presented in Appendices 5 .3A, 5 .38 and 5 .3C. The data presented I n Appendix 5 .3 are summarized in Figure 5 .2. 1 22 The total amounts of CaP-S in the top 3 cm layer of both treatments were significantly higher than those in lower depths, partly reflecting the recent fert i l izer application. At 30 days, the average CaP-S in the whole core fert i l ized with S0 was lower than that of SSP core but thereafter (60-1 50 days) the amount of CaP-S in the S0 treated core remained higher than in the SSP fertilized core. The differences are small and are not considered to be agronomically significant. The amounts of CaP-S derived from the soi l sulphate pool at each depth are presented in Appendix 5.38. lt was calcu lated as the difference between the total soil CaP-S and the amount of CaP-S derived from fertilizer S. The contribution from fertilizer was calculated by dividing the total CaP-35s activity in the extract by the specific activity of 35s in the applied fertilizers. For the first 30 days, the amount of CaP-S soil S in each of the three soi l depths for both fertilizer treatments were similar and were not significantly different between treatments. Thereafter (30-60 days) the amounts decreased sharply from 1 5-1 6 ppm to 2-8 ppm but did not decrease fu rther fo r the remai nder of the experiment . F rom 0-30 days much of the decrease in t he CaP-S cou ld be accounted for i n p lant uptake , h owever at 60 days accumulated plant S uptake could not account for the decrease in CaP-S from 0-60 days ( F igure 5 .2 ) . Du ring this period 3 drainage periods occurred (Appendix 5 .9) and some sulphate would have been leached. Further i t is expected that the addition of fertil izer S which increased CaP-S level at 0-30 days stimulated S immobilization ('priming effect') . This effect lasted only 60 days. The percentage contributions of fertilizer S to CaP-S (number in parenthesis in Appendix 5 .3A) was initial ly larger (26-27%) but decreased markedly. Notably, by 60 days a substantial percentage of the fertilizer S remaining as sulphate had moved to the lower soil depth. This evidence supports the view that leaching of su lphate had occurred beyond the 1 0 cm depth of the intact soil cores. 5.4.2.4 Extractable 35 S in soil (CaP-35 S) At time o, all the SSP 35s would have been in this fraction, by 30 days only 1 8% (Table 5 .6 and Figure 5 .4) of the applied 35s remained in this fraction. These results confirm the large decreases in the CaP-S that occurred during this period (Section 5.4.2.3) . Over the same time period (up to 30 days) increases in organic 35s only accounted for 25.4% of the added 35s (Table 5 .6) and plant uptake, 6.4% (Appendix 5.4, Figure 5.4) g iving a total 35s recovery of 50.9%. The 49. 1% not accounted for is presumed to have leached beyond 1 0 cm depth. 1 23 At time o in the s0 core, there is no 35s in soil sulphate form or any other soil S fractions. The oxidation of the 35s labelled microfine s0 was very rapid. Less than 2% of the 35s label remained as S0 (acetone extractable) in the top layer of the soil cores at the first 30 days (Table 5.6). Thereafter less than 1% of the 35s0 label remained as S0 . Negligible amounts of 35so were detected i n the lower layer of soi l . A calcu lation using a model developed by Chatupote ( 1 990) revealed that the oxidation of S0 having a particle size between 0 .005-0.01 0 mm would be nearly complete within 25 days. The amounts of S0 were also determined in the acetone extracts and the results were consistent with the 35s data (Table 5.6). At 30 days, 48% of the added 35s0 appeared as total organic soi l 35s, 3.6% in plant uptake, 9.7% as CaP-S and 1 .8% remaining as S0, giving a total recovery of 64.4%. At the end of the experiment 35.6% of the added S0 (Appendix 5.4) could not be accounted for and was presumed leached as so4=-s. The CaP-S fraction includes most of the total inorganic 35s in the soil system. At 30 days the largest inorganic 35s fractions were in the top soil layer (0-3 cm) (Table 5.7A). As with the CaP-S data shown in Figure 5.2, the 35s activity in the CaP-S extracts decreased rapidly from 30-60 days. By 1 50 days, only 0.8% and 1 .3% of S from the applied SSP and S0 respectively remained in this fraction in the top 0-3 cm layers. When expressed as the percentage of total 35s labelled fertilizers remaining in the soil core layers, the percent remaining as the CaP-S, generally increased with depth (Table 5 .78) . This may suggest that transformations of inorganic S in the lower soil layers were slower than in the upper layers and this might be associated with more microorganism activity and more available carbon sources in the top layer. The higher amount of the CaP-35s in the lower layers may be attributed to leaching of the 35so4= down the profile. Data in Table 5.8 shows the relative proportions of 35s fractions remaining in soil cores (0-1 0 cm) . I nitially , larger percentages, about 42% of 35s labelled SSP remained as cap_35s. After 60 days, however, less than 1 0% of both fertilizers remained as CaP-S. Up to 60 days initial incorporation of 35s labelled SSP into organic 35s appeared slower than t hat of 35s labelled S0 and might partly explain the larger cumulative 35s uptake of the SSP fertilized pastures at the beginning of the trial. In general , the amount of the added 35s represented by the CaP-S fraction was much smaller than the 35s organic form, as shown in Table 5.6. 1 24 5.4.2.5 Total organic 35 S After only 30 days the organic fraction contained the majority of the 35s other than that presumed leached. The percentage of applied 35s accounted for by organic 35s fractions is g iven i n Table 5 .6 , Appendix 5.6A (as percentage of applied fertilizer), Appendix 5 .68 (as percentage of 35s remaining in the soil) , Table 5 .8 and Figure 5 .5 . Significantly more organic 35s was formed from 35so fertilizer (Table 5 .6) but because more SSP-35s was leached the proportions of 35s remaining in the soil as organic 35s were similar for both fertilizers (Table 5.8 and Appendix 5.68) . In both fertilized cores, larger amounts of labelled organic 35s were formed in the top layers (0-3 cm) of the soils as shown in Appendix 5 .6A and Figure 5.5. I n the lower layers of the soil cores as shown in Appendix 5 .6A, there were no significant differences between the amounts of organic S that were derived from microfine S0 or SSP-S. Significant differences occurred only in the top layer (0-3 cm) . The major t ransformation of both labelled fert i l izers into o rganic forms was very fast and occurred duri ng the f irst 30 days (F igure 5 .5 ) and thereafter there was no measurable accumulation of 3Ss into the organic fraction ( Figure 5.5 , Appendix 5.6A and Table 5.6). More precisely, the percentage of fertilizer derived 35s in the organic fraction decreased from 60 to 1 50 days. During this period (90-150 days) mineralization of organic 35s probably occurred because plant 35s continued to accumu late slowly (Figure 5 .3 , Appendix 5 .4) despite o nly small decreases in CaP-35s (Table 5.6). Expressing the recovery of 35s in each fraction as a percentage of the total 35s remaining in the three layers of each soil core, it is possible to show that in itial conversion of s0 to organic S was more rapid than the S from SSP (Appendix 5 .68 and Table 5.8) . The more rapid conversion of S0 in this core may be associated with the additional microbial growth derived from oxidation of S0. After the first 30 days, 58% of the total remaining 35s in SSP and 8 1% in S0 fertilized soi l core were present i n organic forms (Table 5.8) . After 60 days, there were only small differences among the proportions of 35s remaining as organic 35s throughout the soil layers with greater than 90% of the 35s remaining in each layer having been incorporated into organic S. 5.4.2. 6 Carbon-bonded and ester sulphate 35 S By separating the organic S into carbon-bonded and ester-so4= fractions, it is evident that the majority of the organic S formed from fertilizer S during the first 30 days was carbon-bonded S (20% and 40% of the fertilizer 35s added in the SSP and S0 cores respectively) , Table 5 .6 and 1 25 Figure 5.6, Appendices 5.7 and 5.8. During the same period only 6% and 9% of the 35s of SSP and s0 respectively was t ransformed into ester-so4=. I n the S0 fertil ized core three times as much carbon-bonded 35s was formed compared to the SSP core . Between 60 and 1 50 days there was little effect of fertil izer form on the percentage of applied 35s in the ester? SO 4 = fraction. Of the 35s remain ing in the soil cores (Table 5 .8) , carbon-bonded S remained the dominant fract ion maki ng up 55% and 77% of the so i l 35s i n the SSP and S0 fert i l ized cores respectively. Ester-so4= made up a smaller percentage but was consistently higher in SSP fertilized core. Resu lts presented in both Tables 5 .6 and 5.8 indicate that i rrespective of ferti lizer form, a greater percentage of fertil izer S was immobilized into soil carbon-bonded S than the ester? so4= fract ion. Furthermore fert i l izer applied as S0 tended to promote this process. Data presented in Figure 5.68 shows that the majority of the carbon-bonded S formed rapidly (first 30 days) in the top 3 cm of the soil cores. The greatest accumulation of ester-so4= occurred after 60 days and was again in the top 3 cm of soi l . The percentage of 35s in this top-soil ester-S04 = fraction , however, declined gradually from 60-1 50 days, whereas the percentage in lower soi l depths remained u nchanged. The percentage of applied 35s remaining as carbon-bonded S in the topsoil did not decline from 60-1 50 days. If the data in Figure 5.6 is expressed as a percentage of 35s remaining in each soil depth then some interesting trends with depth become evident ( Figure 5.7). At all depths and for both fertilizers the percentage of 35s remaining as CaP-35s generally decreases with t ime (Table 5.78 and Figure 5 .4). With decreasing soil depth the proportion of the 35s present as ester? so4= increases but that present as carbon-bonded S decreases (F igure 5 .7) . The results probably reflect the relative input of carbon in each soil depth ; ester-so4= being formed in areas of low carbon input. David and Mitchell (1 987) found similar trends when they studied S transformations in a forest soi l in New York state, USA, by adding 35s sulphate to the forest floor. Recently Nguyen and Goh ( 1 990) also found simi lar trends i n New Zealand so i ls . Alternatively, the ester-so4= may remain more mobile than carbon-bonded S and is leached down the profile (Figure 5.7A and 5 .78) . /. Recovery 108 90 80 70 60 50 40 30 20 10 0 Figure 5.3 e 30 60 90 120 1s0 0 30 60 98 120 150 Superphosphate Days M icrof ine S Percentage of applied 35s labelled fertilizer recovered as total soil S in three soil layers and cumulative 35s uptake by pasture. fll] Top ? M idd le 0 Bottom EH S app l ied -?- S uptake -I. 1\) O'l 20 18 Percentage 16 of 14 app l i ed 12 35 10 s 8 6 4 2 0 Figure 5.4 EJ 30 60 90 120 150 EJ 30 60 98 120 150 Superphosphate Da!:}s M icrof ine S Percentage of applied 35s present as extractable S (CaP-35s) in three soil layers, cumulative plant 3Ss uptake and 35s taken up by pasture at each of five harvests. [[}] Top ? M idd le D Bottom ...o- Cumu I at i ve -?- S uptake _. N -...J /. Recovery Figure 5.5 68 se 48 38 28 18 8 I ' ' ? ' ' ? ' I l l I l l 1 1 I I I l l ' I I 1 1 I ' I I ' ? 38 68 98 120 15B Days Superphosphate 38 6B 9B 12B 158 M icrof ine S Percentage recovery of 3Ss applied as fertilizer in organic S in three soil layers. lill Top ? M idd le D Bottom _... 1\) ()) 46 T 36 ' 26 ' Percent _J_ 16 6 46 T m 36 . - .. 26 . . .. Percent .!m I ll 16 ..... - 6 38 ? it6 128 158 Superp osp ate Days 38 ( A ) m (B ) .. . . . . . . . Il l Il l Il l Il l 68 9? . 126f 158 S Icro 1ne Figure 5.6 Distributions of 35s ester sulphate (Ai and carbon bonded 35s (B) in three soil layers (value e;gJressed as percentage of 5s applied). [I]] Top &1 M idd le D Bottom [I]] Top ts1 M idd le D Bottom _.. 1\.) c.o Percent Percent Figure 5.7 100 (A ) ITII Top 50 0 1111'\l 11111\.J ltlll..1 I1IIIL.'1 t11!L'i 'I I'""' 1111m.? 'I""'' '1 "1'\. ' '1 ""' 1 'I tSJ M idd l e D Bottom 100 (B ) ill Top 50 0 pm'l 11 111'\1 t1 t iN ll lll\.1 IIIU\1 'I lllt.. '1 t1ua.'1 '!Ill "' '!"''-' 'I'">' 'I 30 60 98 120 150 Superphosphate Da?s 38 60 98 128 150 M icrof i ne S Distribution of 35s ester sulphate (A) and qrrbon-bonded 35s (B) in three soil layers (value expressed as percentage of total 3Ss remaining in soil core, 0-10 cm). ISl M idd l e D Bottom _.. (JJ 0 S uptake mg/core 35 S uptake MBq/core 80 60 40 y=0.84 + 0.45x r=D.99 Y=- 1 .8 + 0.48x r=0.99 (A) 20 " 0 \B) 4 3 2 1 0 y=3.2 - 3.8exp(-.02x) r=D.99 y=3.2 - 4. 1 exp(-0.02x) r=D.99 30 60 90 Superp hosphate 1 20 1 50 30 Days 60 90 1 20 1 50 Microfine S Figure 5.8 Cumulative S (A) and 35s (B) taken up by plants at five harvesting times; observed vs. predicted. 0 Observed * Predicted 0 Observed * Predicted ..... (..) ..... 1 32 Table 5.6 The percen?e of total 35s labelled fertilizers appearing in soil organic and inorganic 3 S fractions recovered from the 0-10 cm depth of undisturbed field soil cores at five harvesting times. Recovery percentage S forms Fertilizers % Days after application 30 60 90 120 1 50 ORGANIC S Total Superphosphate 25.4 22.4 22.5 21 .9 21 .5 Elemental S 1 48. 1 44.8 41 .9 40. 1 40.3 F-test * * * * * C.V. % 14.4 9.7 13.0 4.9 13.4 Ester sulphate Superphosphate 5.8 9.7 1 1 .3 8.8 8.8 Elemental S 1 8.9 18.3 10.9 7.3 7.8 F-test * * ns ns ns C.V. % 15.3 23.6 24.5 37.9 37.0 Carbon-bonded S Superphosphate 19.6 12.7 1 1. 1 13 .1 12.7 Elemental S 1 39. 1 26.4 30.9 32.8 32.5 F-test * * * * * C.V. % 19.2 16. 1 17.9 15.6 15.3 INORGANIC S (CaP-S) Superphosphate 18 .3 4.3 1 .6 1 .4 1 .4 Elemental S 1 9.7 7.7 3.2 2.3 1 .8 F-test * * * * * C.V. % 8.6 15.2 12.4 8.6 16.0 a Acetone extractable S0 355o 1 .8 0.8 0.4 0.07 0.06 so 3.7 1 .8 1 .0 0.80 0.60 ? = significant at 5% level; ns = not significant a Chatupote (1990); 1 particle size = < 1 0 J.U77 1 33 Table 5.7 Average extractable 35s (CaP-35s) expressed as a percentai?e of total 35s activity applied (PART A) and as a percentage of total S remaining (PART B)in three soil layers at five harvests. Days after fertilizer application Fertilizers Layers 30 60 90 120 150 PART A . .. . . % of applied .. . . . Superphosphate Top 10.6 2.0 0.9 0.6 0.8 Middle 3.7 1 . 1 0.3 0.3 0.3 Bottom 4.0 1 . 1 0.3 0.4 0.3 Elemental S 1 Top 6.4 4.8 1 .9 1 .3 1 .3 Middle 1 .9 1 .7 0.8 0.6 0.3 Bottom 1 .4 1.2 0.5 0.4 0.3 Lsd 5% Within typea 1 .8 1 .0 0.2 0.2 0.2 C.V. % 25.9 34.9 17.9 20.7 20.6 PARTE . . . . . % of remaining ..... Superphosphate Top 43. 1 13.4 6.9 4.7 6.7 Middle 39.4 18.8 5.4 6.2 5.7 Bottom 42.0 21.3 7.4 9.6 6.0 Elemental S 1 Top 13.7 12.0 6.4 4.4 4.8 Middle 26.8 19.7 9.9 6.7 3.4 Bottom 30.3 24.8 8.4 9.6 5.4 Lsd 5% Within typea 12.6 4.4 2.6 1 .9 1 .9 C.V. % 25.6 15.9 24. 1 18.5 23.8 ? = significant a t 5% level; a for comparison of means within each fertilizer type 1 particle size = < 1 OJ,tm 1 34 Table 5.8 Proportion of 35s labelled fertilizers appearing in soil organic and inorganic fractions; amount expressed as percent of total 35s recovered from the 0-10 cm depth of undisturbed soil cores at five harvesting times. Proportion % S forms Fertilizers of remaining Days 30 60 90 120 150 Organic S Superphosphate 57.5 83.8 93.4 94.0 93.6 Elemental S 1 80.4 83.9 91 .8 94.4 95.3 F-test * ns ns ns ns c.v. % 6.5 2.7 1 .5 0.5 1 . 1 Ester sulphate Superphosphate 13.0 36.5 47.5 37.9 38.5 Elemental S 1 15. 1 34.5 24.2 17.4 18.4 F-test ns ns * * * c.v. % 19.4 5.3 22.4 35.3 27.4 Carbon-bonded S Superphosphate 44.8 47.3 45.8 56.1 55. 1 Elemental S 1 65.3 49.5 65.7 77.0 76.9 F-test 1 1 .7 ns 18.3 1 1 .3 16.4 c.v. % 9.4 6.0 14.3 14.0 1 1 . 1 Extractable S (CaP-S) Superphosphate 42.0 16.2 6.7 5.9 6.3 Elemental S 1 16.4 14.5 7.2 5.4 4.6 F-test * ns ns ns ns c.v. % 15.1 15.0 18.7 9.0 18.8 35s0 (acetone extractable) 3.1 1.4 0.9 0.2 0.2 ? = significant at 5% level; ns = not significant 1 particle size = <10Jlm 1 35 5.4.2. 7 Prediction of the extent of S transformation The redistribution of 35s isotope with time can be used to estimate the rate of redistribution of soil S between the available S pool , plants and soil organic S forms. Over a period of t ime t, with a time step ?t = 1 harvest inte rval (i .e. harvest n to harvest n+ 1 ), 35s will be removed from the available S pool (E) by plant uptake (P) , microbial immobilization ( I ) and leaching (L) . After an initial period of immobilization some of the 35s may also be added to this pool as previously immobilized 35s is remineralized (M). The balance equation is as follows: = A similar equation can be written for the balance of 32s 32 SEtn+1 = (1) (2) Three pieces of evidence suggest that the 35sM is small. The amount of 35s in the inorganic pool conti nues to decrease with time (Figure 5 .4) not reaching equil ibrium. P lant uptake of 35s decreases rapidly with t ime ( Figure 5.4) as the majority of soil 35s appears confined to soi l o rgan ic S in relat ively constant proportions (F igure 5 .5) . Assuming then that 35sM is negligible, the model in equation 1 and 2 can be used to estimate the net amount of SI formed during any period, if it is assumed that the CaP extract samples the 35s specific activity of the available S pool, E. lt is also assumed that the change in CaP-35s specific activity between harvests is l i near, and the average specif ic activity of pool E can be calculated as the arithmetic mean of specific act ivities at harvest n and n+1 . The relationships derived to calcu late the net immobil ization and plant uptake of 32s are given below: ?35Sitn . . . tn+ 1 ?S1tn ... tn+1 = Similarly the amount of S taken up by plant can be estimated as ?SP tn . . . tn+ 1 = ?SSP tn ... tn+ 1 (3) (4) This relat ionships only hold if ?SI and ?SP are approx imately constant per unit t ime. For example , higher ?SP for the first haH of an uptake period, when the CaP-35s specific activity was higher, would lead to an underestimation of total plant uptake. The ?SSP between each 1 36 harvest was derived from the relationship between cumulative 35s uptake by plants as shown in Figure 5.8. The equations shown were used to smooth the experimental data. By using the relationships in equation (4) , the amount of S uptake in the superphosphate treated core during time periods of 30 days can be predicted are presented i n Table 5.9. Predicted and observed plant uptake are very s imilar ind icating that the specif ic activity of t he CaP-S (35SE/SE) is representative of the specific activity of the pool of S taken up by plants or immobi l ized into o rganic matter. Notable measure ?SP was essentially linear for the study period (30-1 50 days) . Table 5.9 Days 0-30 30-60 60-90 90-120 120-150 Total Observed plant S uptake and predicted values using relationships in equation 4 above. Observed 14.0 13.3 13.8 13.2 54.3 Plant S uptake (?SP) . . . . mg S core- ... . Predicted 14.9 10.8 14.5 8 .1 48.3 The calculat ion could not be completed for ?32SI (S immobil ized into soil organic matter) because apart from rapid immobilization of 35s into the organic fraction i n the first 30 days (Table 5 .6) for the remainder of the growth period the net amount of organic 35s changed little, decreasing slowly with time. Obviously a more detailed model including both mineralization and immobil ization of S is required to predict actual immobil ization rates. Such a model is d iscussed in Chapter 8. Similarly, ?SL ( leaching) occurs in intermittent events at varying i ntensit ies and requ ires a more complicated model wh ich is capable of describing daily drainage events which leach labelled sulphate from an SE pool of known daily specific activity. 1 37 5.4.3 EXPERIMENT 2, INVERSE ISOTOPIC DIL UTION 5.4.3. 1 Pasture dry matter yield and total S uptake Dry matters yields in both harvests and cumulative dry matter yields showed no significant d ifferences between treatments (Table 5 . 1 0) . Dry matter y ie lds in this experiment were comparable to those of the previous experiment . As mentioned before , there was no yield response to fertilizer application because there was probably sufficient available S in this soi l . Application of fertil izers , however, significantly increased S uptake at both harvests. During the first 30 days, the uptake from the SSP treated cores was larger than from the S0 treated core. By the end of the experiment (60 days) both fertilized cores had similar total S uptake and were significantly higher than the contro l plot. This indicated a slow release of available S from the S0 which was of a coarser size fraction than the 35so used in experiment 1 . Both fertilized cores produced larger S uptake than the control core and the results agreed with the small plot trial discussed in Section 5 .4 .2. 1 . I ncrease in S uptake indicated the luxury uptake of S by pasture plants which also occurred in the previous experiment. 5.4.3.2 Total recovery of 35 S in soil and plant On average approximately 1 00% of the isotope injected to label the native soil sulphate pool could be accounted for by plant and soil analysis at 30 days (Table 5.1 1 ) . The control core did have a lower recovery percentage than the fertilized cores, particularly the S0 core . This may result from experimental error .because no logical reason for th is exists. I n contrast, at 60 days, the highest recovery was in the control core and lowest in the SSP treated core. The recovery in the S0 treated cores and control cores were similar. The unaccounted isotope is presumed to have leached. Recovery of the 35s radioactivity in herbage from both fertil ized treatments were s imilar for both harvests, which suggested that fertilizer applications had no effect on the pasture uptake of the native SO 4 = soil pool in both harvests. In other words, S application caused no 'priming effect ' . This is contrary to the 'priming effect' observed In the d irect labelling experiment 1 38 (Section 5 .4.2. 1 ) but agrees with the results of field studies with 35s labelled gypsum reported by Gregg (1 976) who found that a priming effect only occurred when there were positive or negative y ie ld responses to fert i l izer applicat ion . In this experiment there was no yie ld response. lt is assumed that the injected carrier-free sulphate 35s solution and the soil sulphate pool at each depth were sampled by plant roots in a similar manner. The specific activity of 35s in pasture on the SSP treated core was lower than in pasture on the S0 and control cores at both harvests (as shown in Table 5 . 1 0) . lt is expected that the additional sulphate derived from SSP diluted the 35s either in the soil or in the plants. 5. 4.3.3 Recovery of injected 35 S in soil S fractions I n this experiment, it is assumed that the carrier-free 35s injected into the soil was rapidly incorporated into native soil S pools and equil ibrium was attained within less than one week. Any difference in the percentage of 35s in different soil S forms between fertilized and the control cores will be the result of the fertilizer treatment applied to the soil core. As anticipated, the injected 35s su lphate solution was well mixed with the soil S pools as indicated by the amount of 35s in the CaP-S fraction (Table 5 . 1 2) . The CaP-35s fraction was about 1 0% of the total soil 35s ( in the control core) which agreed with the proportion of native soil S present as CaP-S in the soil system before the trial. 5.4.3.4 Extractable 35 S (CaP-S) Results are shown in Table 5 . 12 . I n the first 30 days, there was a larger amount of CaP-35s activity remain ing in the SSP treated core in the 0-1 0 cm soil layer than the S0 and control cores. The amount of CaP-35s in the S0 treated core was slightly higher than that of the control core. By the end of 60 days, less than 7.2% of the applied 35s activity was present in the extractable fraction of al l cores. Larger amounts of CaP-35s activity were present in both fertilized cores relative to the control core but, whereas the 35s specific activities of the plant (discussed above) indicated isotope dilution by SSP addit ion, the CaP-35s specific activities did not reflect this dilution. 1 39 These results may indicate that application of S fertilizer (SSP and S0) resulted in an i ncrease of the soil CaP-S pool furthe r down the profile by leaching which transported 35s to a zone where dilution by organic S mineralization was slower. The effect was more pronounced in the SSP treated core than the S0 treated core and reflects an increase in the exchangeable sulphate, brought about by sulphate addition which may have led to more 35s sulphate initially exchanging with this larger pool than the smaller sulphate pool in the contro l soil . This confirmed the result of the previous experiment (see Section 5 .4.2) . Goh and Tsuji ( 1 979) also found an increase in native soil CaP-S after fertilizer S application. The specific activity of the CaP-35s was also higher in the ferti l izer treated cores. This may suggest that addition of fertilizer S has inhibited the mineralization of native soil S or that the rate of 35s di lution and immobilization was slower with the increased sulphate pool size. The total amount of CaP-S present in the soil cores (0-1 0 cm) at both harvests was h igher than in the first experiment ( labelled fertilizers experiment, Appendix 5.3) . This difference probably resulted from differences in soil drying methods (see Chapter 6) . I n the second experiment soil samples were dried in an oven at 65 ?C rather than freeze-dried which might cause an increase in the CaP-S in the soi l extracts. Many researchers have reported a large increase in CaP-S after the soil was dried at a high temperature (Barrow, 1 961 b; Peverill et al .. 1 975 ; David et al. , 1 982) which may be derived from microbial cell lysis (Sparling et a/. , 1 985). Since the particle size of S0 used in this experiment was a mixture of different particle sizes and larger (50%, <0.1 50 mm) than that of the previous experiment , oxidation was possibly slower. Thus, initially, the insignificant effects of the S0 on both CaP-S and 35s activity were not unexpected. By the second harvest the amounts of CaP-35s activity in the SSP treated core decreased by approximately 1 0% whereas the decreases in the S0 and control cores were about 4% and 3% respectively (Table 5 .1 2). The amount of decrease was considered to reflect the activity lost beyond 1 0 cm depth by leaching, rather than being taken up by plants or transformed to other soi l S fractions because the 35s activity taken up by herbage in the second harvest were similar in all treatments and concurrently there were larger decreases in total organ ic 35s in both fertilized cores. These overall greater decreases in 35s activity in fertilized cores imply that leaching losses were accelerated by both fertilizer applications. 1 40 5.4.3.5 Total organic 35 S The percentage of 35s recovered in organic form in the 0-1 0 cm soil layer, expressed as a percentage of the applied 35s are shown in Table 5 . 13 . As expected, the majority of carrier? free 35s was incorporated into the organic 35s fractions (ester-35so4= plus carbon-bonded 35s) . By 30 days after the injection, the amount ranged from 67% in the control core to 92% in the s0 treated core . The amount of organic 35s in the S0 treated core was higher than the other two cores. The control core had the smallest organic 35s fraction at 30 days. This indicated greater immobil ization which was caused by fertil izer S0 application. The higher organic 35s fraction in the S0 fertilized core at the first 30 days harvest agrees with the results of the previous experiment (see Section 5.4.2). By the end of the experiment (60 days) , however, smaller amounts of total organic 35s were detected in the fertilized cores (Table 5 . 13) . This suggests mineralization and leaching of the recently immobilized 35s. The amount of organic 35s in the control core did not change. As in experiment 1 the majority of organic 35s formed in the first 30 days was carbon-bonded 35s not ester-35so4= . The carbon-bonded 35s fraction also appears to mineralize faster decreasing in size between 30 and 60 days whereas the percent of added 35s in the ester? SO 4 = fraction changed l ittle. By the end of the experiment, approximately 90% of the 35s remaining in the soil was in organic form being approximately two-thirds carbon-bonded S and one-thi rd ester-so4=. This balance in the top 1 0 cm of soil was more in favour of carbon? bonded S than the natural soil S distribution given in Table 5.3 and may indicate that carbon? bonded 35s continues to mineralize. This supports the view of 35s being immobil ized by plant roots or growth of soil micro-organisms which senesced and slowly mineralized losing carbon and narrowing the C :S ratio which favoured ester-so4= formation (See Chapter 2, Sections 2.2 and 2.3) Table 5 . 10 Fertilizers Days after injection Superphosphate Elemental S 1 Control Lsd 5% C:V. % ns = not significant Dry matter 5. ield (DM), total S uptake, cumulative dry matter yield and S uptake and specific activity of 3 S in herbage from two harvests. DM Total Cumulative 3Ss s specific uptake activity DM s g core-1 mg core-1 _If? take KBq mg-1 S g m - 30 60 30 60 60 60 30 60 3.5 3.8 1 7.2 20.0 4 1 6 2. 1 137 70 2.9 4.7 12.5 20.8 436 1 .9 244 102 2.8 4.6 10.6 13.6 4 19 1 .4 236 129 ns ns 5.1 6.2 ns 0.4 35 .1 33.3 26.3 16.7 21 . 1 19 .3 1 7.9 19.8 12.5 14.1 1 particle size = 1 CXJ<'/o <.25{),lm and 50% < 15{),lm _. .j:>. _. 1 42 Table 5.1 1 Mean recovery o f 35s from herbage and soils (0- 10 cm depth) a t two harvests, 30 and 60 days after injection of 35so 4 = into soil. Fertilizers Days after injection Total recovery (herbage plus soil) Superphosphate Elemental S 1 Control Herbage Soil Lsd 5% fertilizers C.V. % Superphosphate Elemental S 1 Control Lsd 5% fertilizer C.V. % Superphosphate Elemental S 1 Control Lsd 5% Fertilizer C.V. % ns = not significant 1 particle size = 100% <250!J.m and 50% < 15o,.tm Recovery as % of radioactivity applied % % 30 60 96.9 57.7 1 16.1 69.3 85.2 77.5 17.4 1 1 .5 9.0 9 .1 9.2 6.0 12.6 8.7 10.4 7.4 ns ns 28.9 12.1 87.2 51 .7 103.5 60.6 74.8 70.2 13.9 7.6 9.8 7.8 Table 5. 12 Mean recovery of CaPJ5s and specific activity and total amount of CaP-S in 0- 10 cm soil depth. Recovery Recovery CaP-3)s Extractable % % specific s applied remaining activity (CaP-S) Fertilizers % % KBq mg-1 S mg S core- 1 Days after injection 30 60 30 60 30 60 30 60 Superphosphate 16.6 6.0 1 8.9 1 1 .8 6 1 .9 3 1 2 63.2 45.9 Elemental S 1 1 1 .2 7.2 10.9 12. 1 54.9 38.8 49.9 43 .9 Control 7.8 4.8 10.6 6.8 42.9 30.7 44..5 37.2 Lsd 10% 5.8 1 .5 6.5 3.8 5.5 4.6 13.0 4.5 C.V. % 35.9 19.3 35. 1 30.6 3 1. 1 32.2 18 .1 7.7 I particle size = 100% <250)..tm and 50% < 150j.lm ...... t; 1 44 Table 5.13 Mean recovery of soil 35s organic S fractions from the 0-10 cm soil depth. Recovery Proportion % % S forms Fertilizers applied remaining Days after injection 30 60 30 60 Total organic 35s Superphosphate 71 .2 45.7 8 1 .0 88.2 Elemental S 1 92.3 53.4 89. 1 87.9 Control 66.9 65.4 89.4 93.1 Lsd 5% 14.9 8.6 6.8 4.3 C.V. % 12.1 9.8 4.9 3.0 Ester sulphate 35s Superphosphate 12.3 16.1 13.9 3 1 .0 Elemental S 1 23.9 19.3 23.2 32.0 Control 18.1 24.2 24.3 34.5 Lsd 5% 5.1 3.5 6.1 ns C.V. % Carbon bonded 35s 20.5 13.1 21.8 9.6 Superphosphate 58.9 29.6 67.1 57.2 Elemental S 1 68.3 34. 1 65.9 55.9 Control 48.8 41 .2 65.1 58.6 Lsd 5% 12.2 6.9 ns ns C.V. % 15.2 14.4 8.6 7.3 ns = not significant 1 particle size = 100% <250j.Lm and 50% < 15??m 5.5 DISCUSSION 5.5.1 Pasture yield and S uptake 1 45 Pasture growth and S uptake in both the small plot and microplot were comparable with no d ry matter yield and S uptake responses to the applied fertilizer treatments. The applied fertilizer S was taken up at the expense of soil derived S. This was probably because the soil used in these studies had sufficient labile S for optimum pasture growth and because SSP has been applied annually to the paddock to maintain the optimum P fertility level. In this situation, S concentrations in mixed herbage throughout the trial period range from 0 .39-0.47%, well above the critical level of 0.28-0 .30% (Sinclair et al., 1 985) . lt has been suggested that for a paddock at this fertility level , applications of SSP can be withheld every two or three years without any depression in dry matter yield (Smith, 1 976; Ledgard et al., 1 99 1 ) . In this respect the majority of S taken up by plants was derived from the mineral ization of o rganic soil S bu ilt up from previous fertilizer applications. Maintenance S fertilization is needed only to replace losses by animal transfer and leaching (Sinclair and Saunders, 1 984) because pasture development has reached a stage where S immobi l izat ion i nto so i l o rganic matter equals organic matter mineralization. I n terms of S uptake and dry matte r y ie ld , the pe rformance of t he microf i ne S0 was comparable to SSP, but oxidation of this very f ine particle size S0 was very fast and probably complete within 25 days after application. 5.5.2 Short term fate of the fertilizer sulphur I n this experiment, interest was focused mainly on the immediate fate of fertilizers in the top 1 0 cm soi l layer since most plant root activities impact on this layer. A previous study us ing 35s in this soil revealed that about 90% S taken up by pasture plants was mostly derived from this layer (Horne et al., 1 992, unpublished data). Thus most fertilizer S moved beyond th is layer may be considered a loss from the immediate (short term) cycling pool. 5.5.2. 1 Uptake of 35 S labelled fertilizers 1 46 Labelled 35s SSP was taken up by pasture mainly during the first 60 days and amounted to 6 to 7% of applied 35s in each cut. Thereafter very small amounts were taken up. This pattern of isotope uptake is common for biologically active nutrients like S (Kennedy and Till , 1 981 b; Gregg and Goh, 1 982; Goh and Gregg, 1 982a) because microbially mediated immobilization and mineralization serve to remove 35s from the available pool and dilute it with unlabelled S as time progresses. This is discussed more fully below. Plant uptake of 35s during the earlier stages of the experiment (f irst 60 days) was greater in SSP fert i l ized cores but for the remainder of the experiment, 35s uptake was larger (60- 1 50 days) in the S0 cores. This probably reflects the slow release of 35s by oxidation in the first 30 days and the greater amount of 35s remaining in the s0 fertilized soil cores from 60-1 50 days. 5.5.2.2 Recovery of 35 S labelled fertilizer in the top 10 cm of soil Losses of 35s during the fi rst 30 days of the direct label l ing experiment were larger than expected and are unlikely to be completely explained by leaching loss because they represent S leaching losses from the SSP treatment exceeding 20 kg S ha-1 in the first 30 days even when the highest 35s specific activity of the CaP-S pool is considered ( i .e . 35s added/ fertilizer plus initial CaP-S at day 0) . Although the large loss of 35s remains largely unexplained by the measurement made in this experiment, the distribution of 35s in soil S forms and i n different soi l depths provides useful information. A larger amount of 35s from the labelled SSP could not be accounted for in the top 10 cm of soil (soil plus herbage) as compared to the labelled microfine S0 . By 30 days <3% of the S0 remaining in the soil was S0 . The greater retention of 35so probably resulted from its faster rate of incorporation into soil organic S rather than the non-susceptibility of S0 to leaching . With coarser S materials (0 .075-0 . 1 50 mm particle size) the non-susceptibility to leaching may be a more important mechanism in reducing S lost and is studied in Chapter 7. 5.5.2.3 Transformation of 35 S in top 10 cm of soil layer As discussed above as both experiments proceeded the activity of CaP-35s decreased in all soils and during the first 30 days there was a marked transformation of CaP-35s to organic S. Larger amounts of 35s labelled fertilizers remained as organic 35s in the top 0-3 cm soil layer and this may be attributed to and/or associated with larger root and microbial activity taking 1 47 place i n this surface horizon . Almost two-thirds of the 35s remaining in the top soil was transformed into carbon-bonded S (as shown i n Table 5 .6) . This trend in organic 35s part it ion i ng agrees with resu lts of other i nvest igators who observed t hat the rate of i ncorporation of radioactive 35s into organic S was h ighest in the soil surface layer which contains larger amounts of organic residues (Swank and Fitzgerald , 1 984; Schindler et al., 1 986; David and M itchel l , 1 987) where larger amounts of carbon-bonded S were formed (Strickland et al. , 1 987) . In these pasture soils, pasture roots decrease logarithmically down the profi le (Wil liams , 1 988) . Gregg (1 976) also fou nd that larger amounts of labelled 35s gypsum were incorporated into o rganic 35s in the improved pasture soi ls where larger amounts of organic matter had accumulated. Transformation of 35s from microfine S0 to organic forms was as fast as that of 35s labelled gypsum in SSP , but the amounts transformed were twice those of the SSP treatments. lt could not be explained why a larger amount of organic 35s occurred in the s0 treated cores. The transformation may be associated with or occur concurrently with the microbial oxidation processes. During these processes autotrophs derived energy from S0 oxidation to fix carbon (Alexander, 1 977) . More detailed studies are required to determine the impact of S oxidizing microbes on the form of organic S formed. However, it is of interest to note that the capacity of soi l to incorporate fert i l izer S into organic forms differs with different fert i l izer forms. In addition , it has been shown that the so i l capacity for incorporation of S into organic forms varies with different soils (Autry and Fitzgerald, 1 991 ). This was also indicated by the survey of pasture soil conducted by Jackman (1 964a, 1 964b) and relates to the climatically contro lled biological productivity of the site and the abil ity of the soi l to form organo-mineral complexes from plant and animal residues. As a nutrient conservation point of view, it may be considered that S0 is more preferable than SSP, since larger amount of S were retained in o rganic forms which appeared to be slowly mineralized as the experiment proceeded. A lthough larger amounts of o rgan ic 35s occurred i n it ia l ly in the s0 t reated cores, the proportions of 35s remaining in the soil which were subsequently transformed into the organic fraction were about the same for both fertilizers (Table 5.8). More than 90% of the remaining 35s was transformed into organic S. This i ndicates the importance of the transformation processes and suggests that it is possible that more 35s from the SSP may be transformed into organic 35s if losses by leaching can be reduced. 5.5.3 Comparison between labelled fertilizer and Inverse dilution techniques 148 Theoretically, the inverse dilution technique is employed to detect changes in the labile pool or exchangeable pool of nutrients in a system when a treatment is applied after the labile pool or the exchangeable pool has been labelled and a maximum equi l ibrium has been attained . Change in the labile pool can be measured through plant uptake. Only the changes of the native soil pools can be quantified. But a quantification of changes of the applied treatment (e.g. fertil izers) cannot be achieved if the treatment is concurrently transformed into an inactive form (organic forms). Shedley (1 982) employed this technique ( inverse dilution) in a study of ox idat ion of s0 and fou nd t hat a large amount of the oxid ized S (su lphate form) was transformed into organic forms. The author considered that using changes of soil su lphate levels to estimate S oxidation is inaccurate and underestimated the oxidation rates. I n general , resu lts of t he inverse d i lut ion techn ique were cons istent with the label l ing technique . Larger losses of 35s occurred in the SSP fert i l ized cores and more 35s was i ncorporated into organic S in the S0 fert i l ized cores. Addit ionally, the inverse d i lution technique also revealed changes in the soi l organic fractions, ester-so4= and carbon-bonded S. lt appeared that carbon-bonded S was likely to be a greater source of mineralized S. This trend in o rganic S mineralization has been observed by Freney et al. ( 1 975) , Mclaren and Swift ( 1 977) , McGill and Cole ( 1 981 ) , Mclaren et a!. (1 988) and Ghani et a!. ( 1 99 1 ) Isotope recovery and dilution data from both techniques remain difficult t o i nterpret without models which are more descriptive of the factors accounting for daily removals of S from the exchangeable pool, in particular leaching losses. 5.5.4 The mlcroplot technique ( undisturbed soil core) l t i s considered that the microp lot techn ique employed i n these studies gave results comparable to the small plot . The coeff ic ient of variation (%C.V.) among treatments was small . Some investigators employed th is technique to study the fate of fert i l ize r in soi l systems, e.g. Peveri l l et al. ( 1 977) ; Martin ( 1 985) ; Destain et al . ? ( 1 989) and Will iams et al. ( 1 990a, 1 990b) . lt was also considered that this method was less costly and presented few 1 49 practical difficu lties, especially concerning the u niformity of fertilizer application. All of the applied nutrients can be conserved with in the microplots and possible run-off of label led fertilizers is prevented (Martin, 1 985 ; Destain et al. , 1 989). 5.6 CONCLUSIONS 1 . Large unaccounted for losses of 35s from the soil cores during the first 30 days of the experiment are un l ikely to resu lt solely from so4= leaching and therefore reduce confidence in calculated amounts of soil and fertil izer S transformed into other S fractions. However, clear trends in the relative rates of transformation of 35s added as SSP or S0 are evident. 2. I ncorporation of 35s labelled ferti l izer S into the soi l o rganic fract ion was the process that conserved the largest amounts of 35s labelled fertil izers in the soil . 3. More of the microfine 35so was transformed into the organic 35s than 35s from SSP indicating that more soil organic S reserves may be formed through the use of e lemental S fert i l izer than SSP based ferti l izer. This may be a mechanism capable of reducing leaching loss in strongly leaching soil environments. Notably in all experiments the major organic S form labelled with 35s was carbon-bonded S particularly in upper soi l zones. 4. I f it i s assumed that u naccounted for loss of 35s was due to leaching then leaching losses of S beyond the 10 cm soi l layer were larger in the SSP treated cores and the majority of the loss occurred during the 30 days after application. Further studies, however, are required to confirm the nature of the loss because when expressed in terms of kg S leached per hactare the loss was unacceptably high. 5 . The inverse dilution technique indicated that carbon-bonded S was likely to be the source of mineralized S. Both of the results from the inverse dilution and standard techniques were consistent. 1 50 5 . In terms of plant S uptake, the microfine S0 was initially slightly less effective than SSP . Howeve r, over a l onger per iod both f e rt i l i ze rs s howed a s im i l a r performance, reflecting the larger conservation of 35s in the root zone of the S0 treated cores. The particle sizes of S0 used will have an important bearing on the rate of S0 oxidation and these studies should be repeated with a range of S0 particle sizes. CHAPTER 6 THE MEASUREMENT OF AVAILABLE SOIL SULPHUR 6.1 INTRODUCTION 1 51 A large number of methods have been proposed to estimate the size of the immediately plant available soil sulphate pool (see Chapter 2; Section 2.3) , among them extraction with 0.01 M Ca(H2P04)2. The amount of S extracted from air-dried soils (<2 mm) by 0.01 M Ca(H2P04)2 has been used by the New Zealand Min istry of Agricu lture and Fisheries to modify the maintainance S recommendation for pastures (Sinclair and Saunders, 1 984) into a short term fertilizer requirement. In this respect, the amount of S extracted from soil was considered to be an index of the soi ls abi l ity to provide S to plants duri ng the following growing season. A number of problems associated with this i ndex are recognized. D ifferent soil preparation techniques , in particu lar drying, storage, and grinding, influence the amount of S extracted (Peverill et al., 1 975 ; Searle and Sparling, 1 987) . Drainage events prior to sampling influence the soil test result (Ghani et al. , 1 990) and the test does not reflect the varying amounts of plant avai lable S that wil l be mineralized from the organic S present in the pastu re soils (Nguyen and Goh, 1 99 1 ). In general , the longer pasture has been influenced by fertil izer application and grazing, the greater the amount of o rganic S present in that soi l (Jackman, 1 964a, 1 964b; Saggar et al. , 1 990a , 1 990b) , the more S is m ineralized from that soi l (Sakadavan, 1 99 1 ) and the less responsive dry matter yields are to the application of fertilizer S at a g iven soi l S test (0 .01 M Ca(H2P04)2) value (Sinclair et al. , 1 985). Despite these problems most researchers adhere to the concept that 0 .0 1 M Ca(H2P04)2 o r 0 .04 M Ca(H2P04)2 (Sear le , 1 979) extracts from so i l are soluble and read i ly exchangeable (adsorbed) form of S, through which all S must pass before entering the plant root. Recently, Watkinson and Perrott ( 1 990) and Watkinson et al. ( 1 99 1 ) have demonstrated that up to 50% of total S in the extract can be present in the organic form. lt is assumed that the organic S extracted is a labile form and easily hydrolysed for uptake by plants. Currently, research is in progress to improve the usefulness of soi l S testing i n determining ferti l izer S requ irements (Watkinson et al. , 1 99 1 ) . The fie ld experiment d iscussed in Chapter 5 provided a series of soi l samples, taken at different t imes and from different soil depths under pasture fertil ized with radioactively labelled superphosphate and e lemental S (S0) . These samples provided the opportunity to evaluate soil preparation techniques and the ability of soil tests to determine the sources of the plant 152 available S pool in these soils by determining whether the specific activities of 35s of particular extract S fractions are simply related to the 35s specific activity in plants drawing their S supplies from these soils. Such information should be useful to those developing improved S soil tests. If it is assumed that the extractable S (soluble S and adsorbed S) is the precursor of plant S, then a simple relationship should exist between the specific activity (SA) of herbage S and the specific activity (SA) of extractable S. If this relationship is examined for extractable S from increasing soi l depths it may provide some indication of S uptake zones in soil and may indicate a soil depth which provides the best sample for analysis. For 600 days after fertil izing a paddock with label led 35s gypsum, Ti l l and May ( 1 971 ) found a significant relationship between the specific activity of extractable 35s (0.01 M Ca(H2P04)2) sulphate from the 7.5 cm soil depth and herbage 35s specific activity. Gregg (1 976) also found close relationships between specific activity of plant 35s and extractable 35s (0.01 M Ca(H2P04)2) sulphate but the relationship varied among sites and from harvest to harvest. At most sites the u ptake of 35s was confined to the upper 1 5 to 30 cm. With pasture soil, in New Zealand, it was reported that soil samples from the 0-7.5 cm depth gave satisfactory relationships between soil tests for available S and plant growth (Saunders and Cooper 1 975; Saunders et al., 1 988; Nguyen et al. , 1 989a, 1 989b). However, Probert and Jones (1 977) found that soil S extracted with 0 .05 M calcium phosphate solution could be used to predict S response in establ ishing legumes in native pastures provided the whole profile (0-1 00 cm) was considered. When extractable S was below 1 0 ppm in the surface soi l , us ing depth we ighted means for extractable S i n the profi le enabled them to de l ineate responsive from the non-responsive sites. Soils with more than 1 5 ppm extractable S in the surface horizon could be categorized as non-responsive. lt wou ld appear that relationships between extractable S and p lant yield may wel l remain dependent upon soil characteristics, plant rooting depth and climate. Thus it is appropriate to i nvestigate this relationship for permanent clover/ryegrass pasture on the Tokomaru si lt loam studied in Chapter 5. 6.2 OBJECTIVES The objective of the study reported i n this Chapter was to compare the inf luence of sol i sampling depth , soi l preparation method and extractant type o n t he abi l ity of soil tests to 1 53 identify the immediate source of plant available S i n soil samples taken from the field study described in Chapter 5. Two soil preparation techniques were evaluated, freeze-drying and grinding compared to field-moist soil. 6.3 MATERIALS AND METHODS 1 54 The field trial and soil and herbage sampling techniques were those described in Chapter 5, Section 5.3. A brief discription of soi l preparation and extraction techniques is g iven below. 6.3.1 Soli and herbage analyses 6.3. 1. 1 Extractable S The soil samples from the f ield were stored overnight at 5 ?C before being subdivided into three depths. Fresh moist soil was gently crumbled and then a known weight immediately extracted. A further sample of fresh soil was oven-dried at 1 05 ?C for 1 2 hours to calculate the moisture content. The extractions were performed on samples of both fresh, field-moist so il and freeze-dried, finely ground soil using 0.01 M calcium chloride (CaCI-S) and 0 .04 M calcium dihydrogen phosphate (CaP-S) , pH 4, as the extractants. Five grams of wet soil o r freeze? dried soi l were extracted with 40 ml of e ither extracting solution (1 :8, WIV) on an end-over-end shaker for 2 hours. The suspensions were centrifuged at 20,000 rpm, using a SS34 head on a Sorvall RC2B centrifuge, for 1 0 minutes and filtered with Whatman No. 42 paper. The filtrates were analysed for H I-reducible S as described in Chapter 3, Section 3 .3.6. Aliquots of 1 .0 ml of soi l extract are used to determine 35s activity as described in Section 3.3.8. The specific activity (SS) of the HI-reducible 35s in each extract was calculated as: ss = where = = KBq mg-1 S specific activity of 3Ss in the soil extracts amount of HI-reducible 35s activity in the soil extract; KBq kg-1 soil amount of HI -reducible S in the soil extract; mg S kg - 1 soil (6.3. 1) The 35s specific activity in the extract may have been s lightly overestimated because 35s associated with CaP-S soluble carbon bonded S would not be included in the H I-reducible S measurement. Comparisons of HI -reducible S and total S in the extracts from the Tokomaru soil suggest that carbon bonded S makes up less than 1 0% of the total S in the extract. 6.3. 1.2 Herbage analyses Total S and 35s activity in herbage were analysed as described in Chapter 5, Section 3.3.1 . 1 and Section 3.3.8 and the specific activity of 35s in herbage was calculated as: SH = where SH = Ah = sh = KBqmg-1 S specific activity of 3Ss in herbage amount of 35s activity in the herbage; KBq g- 1 herbage amount of total S in the herbage; mg S g-1 herbage 6.3.2 Statistical analyses 1 55 (6.3.2) Simple correlation and regression analyses were used to estimate the relationship between the specific activity of 35s in herbage and soil sulphate S. The average specific activity of 35s in the soi l extracts of the f irst soi l sampling and the second sampl ing were correlated and regressed against the specific activity of total 35s in the herbage at the second harvest and so on. The SAS (SAS Institute Inc .. , 1 985) and Minitab (Minitab I nc. , 1 989) computer programmes were employed. Specific activity data were paired as follows: Where z = soil depth 1, 2 or 3 (top, middle and bottom) H = harvest time, n or n-1 As, Ss, Ah and Sh = as described in Section 6.3.1 This treatment of data assumes that the rate of plant S uptake was constant over the period between the two harvests and that the decrease in the specific activity of the soil S extractable pool can be approximated by a l inear relationship. Analysis of variance was also performed on data where correlations and regressions were significant. 6.4 RESULTS AND D ISCUSSION 6.4.1 Effect of sol i preparation and extractant 156 As described in Section 6.4.2 below, unlike the s0 fertilized plots the specific activities of 35s derived from the extraction of freeze-dried soil samples from the superphosphate treated plots were not influenced by residual S0 slowly releasing sulphate into soil solution. Therefore, the effect of soil preparation and extractant type on the concentration of S and 35s activity in the soil extracts are presented and discussed using the results from superphosphate fertilized plots only. 6.4. 1. 1 Effect of soil preparations The statistical comparison between the extraction of S and 35s from fresh, field-moist soil and freeze-dried soil samples were analysed separately for each extractant, namely CaCI-S and CaP-S. Results are presented in Table 6.1 and Table 6.2. 6.4. 1 . 1. 1 Effect of soil preparations on extractable S The amounts of extractable soil S as affected by so i l preparations are presented in Tables 6. 1 A (CaCI-S) and 6.1 B (CaP-S) . Both extractants extracted more (approximately twice the amount) S from the freeze-dried soils than the moist soils. This increase, due to freeze-drying, may have resu lted from sulphate being released from plant and microbial cells destroyed during the freeze-drying and grinding processes. At any one time in soils, soil microbial biomass may contain about 2-5% of the total soil S (Saggar et al., 1 981 a ; Strick and Nakas, 1 984; Chapman, 1 987a, 1 987b; Ghani et al., 1 990) . Pasture stolen and root materials were not removed from the freeze-dried soil samples prior to analysis and additional sulphate may also be released from these materials. The soil sampled from the fresh field-moist soil had less roots and underground plant materials because they tended to remain as a root mat and not as part of the 'crumbled' soi l sample . After freeze-drying, however, the underground plant materials were very dry and brittle compared to those of the routine air drying method and they were easily ground and incorporated into the soil sample. The root , underground plant material and litter will contain some S in sulphate form. Horne et al. ( 1 992) (unpublished data, M .J. Hedley personal communication) showed that up to 30% of the S in pasture roots was in the H I-reducible form. Lee et al. ( 1 98 1 , 1 985) , Watkinson and 1 57 Perrott ( 1 990), Watkinson et al. ( 1 99 1 ) and Nguyen and Goh ( 1 99 1 ) have also shown that soil extracts using Ca(H2P04)2 as an extractant may contain about 30-50% of their S as soluble organic S (HI-reducible S) which may be of plant root and microbial o rigin. Throughout the experimental period, the concentration of extractable S from freeze-dried soil i n the top 3 cm was at least twice the concentration in the lower layers. Extractable S from moist soi l d id not show such marked d ifferences with soil depth after the effect of recently added fertilizers (first 30 days) diminished (60- 1 50 days) . This is probably associated with the larger activity of plant roots and soil micro-organisms in this layer. In general, the results from the freeze-dried soil samples were less variable with time than those of the fresh field-moist soi l . This is probably due to the more un iform soi l sample produced by freeze-drying and grinding . The concentration of extractable S i n the 30 day samples, in both moist and dry soils, were much higher than in later soil samples; this reflects the addition of superphosphate at day 0. I n general , after 30 days the S concentration in all layers did not change much. 6.4. 1. 1.2 Effect of soil preparation on 35 S activity The 35s activity in the soil extracts as affected by soil preparations within each extractant are presented in Table 6.2A (CaCI-35s) and PART B (CaP-35s) . In general, the average amount of extractable 35s from the freeze-dried soil was similar to that from the fresh, field-moist soils. However, more 35s was consistently extracted from the top layer of freeze-dried soil. The effect of so il preparation on CaP-S extractable 35s activity was similar to the effects on CaCI-S extractable 35s described above. The greater extraction of 35s afte r freeze-drying in the top layer probably reflects the release of 35s incorporated into microbial cel l and plant tissue. As discussed earl ier in Chapter 5 more organic 35s was synthesized in the top layer. Throughout the experiment , extraction of freeze-dried soil produced more consistent results and showed consistent d ifferences between the concentration of S and 35s activities in top and lower so il layers. This was probably because of the more un iform mixing of soil samples during grinding which caused less sampling e rror. I n summary, the concentration of extractable S derived from the freeze-dried soils was much higher than that of fresh f ie ld moist soi l whereas there were on ly sl ight differences in the concentration of the extractable 35s, ( i .e. the 35s specific activity in the extracts from freeze? dr ied so i l was less t han that from moist so i l , see d iscuss ion of Table 6 .4 later) . This i nconsistency between the concentration of extractable S and 35s activity suggests that 1 58 freeze-drying released H I-reducible S which had not exchanged with the isotope added. Such a source may be inactive soil microorganisms which are considered to make up 50% of the microbial population in soil (Sparling, 1 985) . 6.4. 1.2 Effect of extractants The effect of each extractant on the concentration of extractable S and 35s activity in the soil extracts are presented in Table 6.3. A statistical comparison between the effect of each extractant was made on data from freeze-dried soil samples of the superphosphate treated plots. 6.4. 1.2. 1 Effect of extractant on extractable S As shown in Table 6.3A, in general, the concentration of CaP-S was twice the concentration of CaCI-S (average CaCI-S : CaP-S ratios were 0.5-0.7). The larger amounts of S extracted by CaP-S are attributed to the ability of phosphate to displace sulphate from sorption sites ( Fox et al. , 1 964; Harward and Reisenauer, 1 966) . The chloride ion lacks this abil ity. As shown i n Appendix 7.1 4, S sorption capacity in this soil was about 24-31%. As mentioned before, the CaP-S extract may also contain some soluble organic S (Lee et al., 1 981 ; Nguyen and Goh, 1 99 1 ) . In some so i ls o rganic S may contribute to 50% of the S in the CaP-S extract (Watkinson and Perrot , 1 990; Watkinson et al., 1 99 1 ) . Therefore the greater extraction of soil S by CaP-S than CaCI-S was attributed to extraction of native soil organic S (soluble o rganic S) and desorption of adsorbed sulphate S. 6.4. 1.2.2 Effect of extractant on extractable 35 S activity The comparison of the concentrat ion of CaP-35s and CaCI-35s activit ies are presented in Table 6 .38. In general , amounts of 35s activities from the CaP-S extraction were only slightly higher than 35s activity from the CaCI-S the extraction (CaCI-35s : CaP-35s range from 0.7 to 1 .0) . lt appears that much of the 35s activity extracted is common to the CaCI-S and CaP-S extracts. Thus as discussed earlier the greater extraction of S by CaP-S than CaCI-S must be attributed to the extraction of a greater amount of non labelled (as described above) soil H I-S by CaP. This un labe l led S can o rig inate from H I -reducible S re leased from i nact ive microorganisms by the freeze-drying process and from strongly sorbed sulphate on sorption sites with in soi l aggregates p rotected from exchange with added 35s during the pot experiment in Chapter 5 . Such aggregates may break down during extraction but the sorbed H I-reducible S wou ld only be displaced by phosphate and not chloride . 1 59 Table 6.1 Concentrations of HI-reducible S in CaCl-S (Part A) and CaP-S (Part B) extracts from three soil layers using two soil preparations. Days after fertilizer application Methods Layers 30 60 90 120 150 PART A, EXTRACTABLE SULPHUR (CaC/-S) k - 1 "1 . .. . . mg g SOl ???? Field-moist soil Top 10.3 1 .2 0.9 2.3 0.6 Middle 5.6 0.6 0.3 0.5 1 .6 Bottom 5.2 0.9 0.2 0.6 0.4 Ave. 7.0 0.9 0.5 1 .4 0.8 Freeze-dried soil Top 20.7 5.9 5.9 4.5 6.0 Middle 14.9 4.3 1 .9 2.9 2.6 Bottom 17.6 2.8 2.1 2.7 2.4 Ave. 17.7 4.4 3.3 3.4 3.7 Lsd 5% method *** *** *** *** *** La ye? 3.66 1 .73 0.63 0.88 1 .53 Method* layer ns ns ns ns ns C.V. % 19.8 43.8 22.4 26.5 45.5 PART B, EXTRACTABLE SULPHUR (CaP-S) k - 1 "1 . . . . . mg g so1 ???? Field-moist soil Top 15.6 1 .2 4 .3 3.8 3 .3 Middle 7.9 2.7 2.5 2.7 2.1 Bottom 8.2 2.8 2.4 2.2 1 .4 Ave. 10.6 2.2 3.1 2.9 2.3 Freeze-dried soil Top 31 .9 10.6 8.6 8.4 8.3 Middle 24.2 4.2 5.4 6.2 4.8 Bottom 26.3 5. 1 5.8 6.4 5.7 Ave. 27.5 6.6 6.6 7.0 6.2 Lsd 5% Method *** *** *** *** *** La ye? 5.50 1 .70 2.3 1 1 .5 1 1 .62 Method*layer ns ns ns ns ns C.V. % 19.4 28.0 32.3 20.5 25.7 ... = significant at 0. 1% level; ns = not significant; 2 /sd for comparison between layers in each soil preparation Remark, data from labelled superphosphate treated plots 1 60 Table 6.2 Concentration of HI-reducible 3Ss in CaCl-S (Part A) and CaP-S (Part B) extracts from three soil layers using two soil preparation techniques. Methods Layers 30 PART A, EXTRACTABLE 3Ss (CaCl-S) Field-moist soil Top 3.09 Middle 0.94 Bottom 0.91 Ave. 1 .65 Freeze-dried soil Top 2.38 Middle 0.73 Bottom 0.69 Ave. 1 .27 Lsd S% Method ns La ye? 0.93 Method*Layer ns C.V. % 42.8 PART B, EXTRACTABLE 3Ss (CaP-S) Field-moist soil Top 2.39 Middle 1 .00 Bottom 1 .03 Ave. 1 .65 Freeze-dried soil Top 2.76 Middle 0.97 Bottom 1 .04 Ave. 1 .59 Lsd 5% Method ns La ye? 1 .22 Method*layer ns C.V. % 20.8 Days after fertilizer application 60 90 120 KB - 1 il ... . . q g so .... 0.26 0.18 0.08 0.19 0.13 0.06 0.39 0.09 0.07 0.27 0.13 0.07 0.38 0.26 0. 13 0.33 0.08 0.07 0.21 0.07 0.07 0.31 0. 13 0.09 ns ns ns 0.12 0.09 0.03 ns ns ns 27.6 46.0 23.8 KB - 1 i1 . . . . . q g so .... 0.39 0.12 0.09 0.53 0.51 0.07 0.51 0.35 0.07 0.47 0.32 0.08 0.53 0.25 0. 17 0.29 0.07 0.09 0.29 0.09 0. 1 1 0.37 0.14 0.12 ns ns ns 0.1 1 0.13 0.03 ns ns ns 17.4 38.9 24.1 ISO 0.04 0.02 0.02 0.02 0.21 0.06 0.05 0. 1 1 * 0.04 ns 38.7 0.06 0.04 0.04 0.04 0.22 0.08 0.07 0. 1 3 * 0.03 ns 23. 1 ? = significant at 5% level; ns = not significant; lsd for comparison between layers in each soil preparation Remark, data from labelled superphosphate treated plots 1 61 Table 6.3 Concentration of HI-reducible S (Part A) and 35s activity (Part B) in CaCl-S and CaP-S extracts from freeze-dried soils taken from three soil layers. Extractants Layers 30 PARTA, EXTRACTABLE SULPHUR CaCl-S Top 20.7 Middle 14.9 Bottom 17.7 Ave. 17.8 CaCl-S/CaP-S 0.7 CaP-S Top 31 .9 Middle 24.2 Bottom 26.3 Ave. 27.5 Lsd 5% Extractant * la ye? 4.1 Extr*layer ns C.V. % 12.3 PART B, EXTRACTABLE 35s CaCI-S Top 2.38 Middle 0.73 Bottom 0.68 Ave. 1 .27 CaCI-35sJcaPJ5s 0.8 CaP-S Top 2.77 Middle 0.97 Bottom 1.05 Ave. 1 .59 Lsd 5% Extractant * La ye? 0.46 Extr*layer ns C.V. % 22.0 Days after fertilizer application 60 90 120 k -1 '1 . . . . . mg g so1 .. . . 5.9 5.8 4.5 4.3 1 .9 2.9 2.8 2.1 2.7 4.3 3.3 3.3 0.7 0.5 0.5 10.5 8.5 8.4 4.2 5.4 6.2 5. 1 5.8 6.4 6.6 6.6 7.0 * ** ** 2.2 0.9 1 . 1 ns ns ns 28.0 12.3 14.3 KB -1 il . .. . . q g so .. . . 0.38 0.26 0. 13 0.33 0.08 0.07 0.21 0.07 0.07 0.31 0.14 0.08 0.9 1 .0 0.7 0.53 0.25 0. 17 0.29 0.07 0.09 0.29 0.08 0. 1 1 0.37 0.14 0.12 * ns * 0.10 0.10 0.03 ns ns ns 20. 1 29.6 21 .2 150 6.1 2.6 2.4 3.7 0.6 8.3 4.8 5.7 6.2 ** 1.9 ns 25.9 0.21 0.06 0.05 0. 1 1 0.9 0.22 0.08 0.07 0. 13 ns 0.04 ns 25.7 * and ** = significant at 5 and 1% level; ns = not significant, 2 lsd tor comparison between layers in each soil preparation Remark, data from labelled superphosphate treated plots 6.4.2 Specific activity of 35s In soli extracts and herbage 1 62 Average total S and 35s activity in herbage from superphosphate and microfine S0 treated cores are presented in Appendix 6 . 1 . (these results have already been discussed in Chapter 5) . The average specific activity of 35s in soil extracts in each soi l depth and herbage are shown in Table 6.4. In general , the specific activity of 35s from the moist soil samples was higher than that of freeze-dried soils. On average, the specific activity of 35s from the CaCI-S extraction was higher than that from the CaP-S extraction. Possible reasons for this are discussed above. 6.4.3 Relationships between the specific activity of 35s In herbage and In soli extractable 35s The specific activities of extractable 35s in soils harvested at the first 30 and 60 days were averaged (see methods in Section 6.3.2) and correlated with the specific activity of 35s in he rbage harvested at day 60 and so on for the subsequent growth period. The averaged specific activities of 35s in soil extracts and herbage are shown in Figure 6.1 . Decreasing 35s specif ic act ivi ty i n the so i l extracts and plants with t ime is cons istent w ith the slow mineralization of non-radioactively labelled soi l organic S. Simple linear relationships between specific activity of 35s in herbage and specific activity of 35s in soil extracts (Table 6.4) are presented in Table 6.5, Figures 6.2. 6.3, and 6.4. In general, better relationships were obtained from the correlation between the specific activity of 35s from the extraction of freeze-dried soil and the specific activity of herbage. Extractions of fresh field-moist soil yielded a poorer relationship and underestimated the specific activity of 35s taken up by the herbage, as shown in Table 6.5. I n addition the variabil ity in sampling moist soi ls caused apparent increases i n specific activity of 35s extracted from the top soi l layer with t ime. I n a moist so i l system it is impossible for the specific activity of the plant available S pool in the top soil layer to increase with time after sulphate fert i lizer application (S0 addition may cause this, as S0 is oxidized and also at lower soil depths this may occur as a single drainage event which may move labelled sulphate S into a lower depth). Thus the procedure of moist soil samplings would appear to produce non-uniform soil samples some containing 'hot spots' of applied 35s labelled fertilizer. Regardless of the number of soil depths considered, the specific activity of H I-reducible 35s from the extraction of freeze-dried soi l us ing e ither calc ium chloride (0 .01 M ) o r calc ium dihydrogen phosphate (0.04 M) most closely correlated with the herbage 35s specific activity for the superphosphate fertilizer plots. 163 The 1 :1 relationship was approximated better by considering the top and middle layers and using CaP-S as the extractant (F igure 6.3) . Other authors (Ti l l and May, 1 971 ; Gregg, 1 976) have suggested that soil extract : plant specific activity ratios of close to 1 indicate a form of extractable soil S available to plants. In the present study such an interpretation suggests that CaP-S is a bette r extractant of the plant avai lable poo l than CaCI-S. These authors suggestion, however, is flawed because no direct association can be made between CaP-S extractable S and S taken up by the plant. The resu lts with in the present study simply indicated that the pool of S taken up by the plant has a lower specific activity than CaCI-S extracts and is similar to the specific activity of CaP-S extracts. Notably by increasing the soil sampling depth the specific activity of CaCI-S extracts could be made to approach that of the plant (compare Figure 6.2A with Figure 6.4A) . The same improvement, but to a greater extent, was achieved with CaP-S extracts (compare Figure 6.28 with Figure 6.48) . The lower specific activity of plant S than CaCI-S (soluble S) results from the plants deriving a certain amount of their S from an unlabelled source, such as, the continued mineralization of native soil organic matter and/or unlabelled H I-reducible S from greater soil depths. The ability of CaP-S extracts to predict plant specif ic activity wi l l depend upon the p roport ional contribution to plant S from e ither source. If most of the un labelled S is derived from HI? reducible S from greater soil depths then the relationship in Figure 6.38 is a causal one; i .e . plants are deriving their 35s from a pool of soi l sulphate defined by CaP-S extractable soi l S. If mineralized soil organic S is the major plant source as indicated by the data d iscussed in Chapter 5 and the continued decline in CaP-S specific activity from the top soil depth (Table 6.4 and Figure 6.1 ) with time, then the pool size available for plant uptake will depend upon the mineral ization rate. lt is unlikely that this can be predicted from a single soil test taken at 1 point. Freeze-drying caused a decrease in specific activity probably due to the release of unlabelled, but labi le organic S or aggregate protected S. Thus, the better soil extract : plant specific activity relationships obtained using freeze-dried soils may suggest the importance of o rganic S in the soil extracts as a contribution to plant available S in a soil S test. H igher amounts of extractable S in freeze-dried soils, therefore resulted in lower specific activities of 35s in the soi l extracts and close relat ionships with the herbage 35s specific activities. This result further emphasizes the need to investigate the inclusion of the labile organic S fraction in soi l S testing procedures to improve the estimation of soil S status (Swift et al., 1 988; Nguyen and Goh, 1 99 1 ) . Of the three soil layers, the 35s specific activity of the CaP-S extract from the top 0-6 cm of soil had the best relationship with herbage specific activity. This also suggests that pasture S, in this superphosphate treated plot, was mostly taken up from this soil layer. 1 64 Table 6.4 The specific activity of 35s in CaCl-S and CaP-S extracts of field-moist and freeze-dried soils and plants. Days after fertilizer application Fertilizers Layers 30 60 90 120 150 FlEW-MOIST SOIL EXTRACTION . . . . . KBq mg-1 S .... . CaCl-S Superphosphate Top 312 203 246 41 77 Middle 176 390 552 108 16 Bottom 207 461 753 143 83 Elemental S Top 161 182 316 64 70 Middle 344 153 122 92 37 Bottom 509 289 77 104 72 CaP-S Superphosphate Top 174 472 37 25 20 Middle 130 201 206 29 1 8 Bottom 129 197 172 35 36 Elemental S Top 131 163 106 49 39 Middle 125 137 212 43 43 Bottom 102 140 213 38 27 FREEZE-DRIED SOIL EXTRACTION CaCl-S Superphosphate Top 1 16 69 43 29 36 Middle 50 80 43 25 27 Bottom 40 76 35 25 23 Elemental S Top 68 9 1 55 42 4 1 Middle 39 9 1 47 28 28 Bottom 26 84 40 3 1 23 CaP-S Superphosphate Top 86 50 29 20 28 Middle 40 75 13 14 1 8 Bottom 40 65 15 17 13 Elemental S Top 73 95 52 35 40 Middle 24 78 24 15 1 5 Bottom 18 61 20 19 12 PLANT35s SPECIFIC ACTIVITY Superphosphate 83 67 38 18 12 Elemental S 60 66 45 22 1 5 Table 6.5 165 Simple .linear relationships, y = bx between specific activity of total 35s in plants (y) and specific activity of 35s in extracts from different soil layers (x). Soil Preparation Field-moist soil Freeze-dried soil Relationships r Relationships r SOILLAYERS; TOP o.Bx*** o.63x*** 0.96*** CaCl-S Superphosphate ns Elemental S O. IOx*** ns 0.67x*** 0.68*** CaP-S Superphosphate o.o9x** ns o.86x*** 0.96*** Elemental S o.34x*** 0.66** o.6Ix*** 0.68*** SOIL LAYERS; TOP+MIDDLE2 CaCl-S Superphosphate o.12x*** ns o.nx*** 0.97*** Elemental S o.26x** o.55* 0.66x*** 0.57* CaP-S Superphosphate o.21x*** o.s5* LOix*** 0.99*** Elemental S o.29x*** ns o.7ox*** o.s5* SOIL LAYERS; TOP+MIDDLE+BOTTOM3 o.78x*** 0.95*** CaCl-S Superphosphate 0. 1 1x*** ns Elemental S o.28x*** 0.9 1*** o.68x*** ns CaP-S Superphosphate o.23x*** o.55* L09x*** 0.98*** Elemental S o.29x*** ns o.77x*** ns ?, ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; ns = not significant 1 Specific activity of extractable 35s (SS) was calculated as: SS = As1 / Ss1 where As1 = L 35sH(n-l)+Hn[top] = LSH(n-l)+Hn[top] 2 Specific activity of extractable 35s (SS) was calculated as: SS = As12 I Ss12 where As12 = = L35sH(n-1)+Hn[top + middle] LSH(n-l)+Hn[top + middle] 3 Specific activity of extractable 35s (SS) was calculated as: SS = As123 I Ss 123 where As123 Ss123 where = = L35sH(n-l)+Hn[top + middle + bottom] LSH(n-1)+Hn[top + middle + bottom] H = harvesting time, n; SS, As and Ss as described in section 6.3. 1 . 1 . KBq core-1 mg core-1 KBq core-1 mg core-1 KBq core-1 mg core-1 Figure 6.1 Speci f ic activity KBq/rng S 1 20 90 60 30 0 1 20 90 60 30 0 1 20 90 60 30 Top + m idd le Top + m idle + bottom 0+---?---+--?----+---? 0 30 60 90 1 20 1 50 Days 1 66 e CaCI-S * CaP-S ...,.. Plant e CaCI-S * CaP-S ...,.. P lant e CaCI-S * C a P-S -w. Plant Specific activity of 35s in plants and soil extracts (CaCl-S and CaP-S) from superphosphate treated cores at five harvests for three soil depths. Figure 6.2 1 20 S peci f ic CaCI-S act iv i ty 90 o f plantS 60 KBqirng S 30 0 1 20 Speci f ic CaP-S acti vi ty 90 o f p lants 60 KBqimg S 30 30 60 Specific activity of extracts 1 : 1 y=0 . 63x r=0.96 y .. Q .86x r=0.96 90 1 20 (A) 0 Observed * Predicted (B) 0 Observed * Predicted KBq/mg S Relationships between specific activity of 35s in plants and HI-reducible 35s in soil extracts from the !QQ. layer; (A) = CaCl-S and (B) = CaP-S. 1 67 1 68 90 (A) Speci f ic CaCI-S acti vi ty 60 0 Observed o f p lants y=0.72x * P re d i cted 30 KBqimg S r=0.97 0 90 / 1 : 1 ( B) CaP-S Speci f ic O/ act iv i ty 60 0 Observed o f p lants Y=i .Oi X * P redicted ? 30 r=0.99 KBqirng S o?------4-------?------? 0 30 60 90 Speci f ic act iv i ty o f extracts KBq/rng S Figure 6.3 Relationships between specific activity of 3Ss in plants and ID-reducible 35s in soil extracts from the too plus middle layers; (A) = CaCl-S and (B) = CaP-S. Figure 6.4 1 69 90 1 : 1 (A) Speci f ic CaCI-S Q/* act iv i ty /t?D o f 60 / 0 Observed p lants / 30 / y .. Q .78x * Predicted KBq/mg S / f-=0.95 0 Relationships between specific activity of 35s in plants and HI-reducible 35s in soil extracts from the top plus middle and bottom layers; (A) = CaCl-S and (B) = CaP-S. 6.5 CONCLUSIONS 170 1 . Of the two soil preparation techniques (extraction of moist-crumbled soil o r freeze? dried and ground so il) , freeze-drying and grinding causes significantly more S, but not more 35s , to be extracted from so i l samples. Therefore , freeze-drying exposes a form of soil S to extraction that was not freely exchangeable w ith added sulphate-35s even during 1 50 days of plant growth. 2. The average 35s specific activity in a CaP-S extract from a freeze-dried sample of the top 6 cm of pasture soil was the extract specific activity most closely related to that of S taken up by plants over several periods of plant growth. Calcium chloride (CaCI-S) extracts from freeze-dried soi l and CaCI-S and CaP-S extracts from moist soils had 35s specific activities that were generally higher than t hat of S taken up by plants. 3. Resu lts indicated that plant roots have access to forms of soil S that cou ld not be accessed by soil water and exchangeable sulphate ions ( i .e. extracted by CaCI2 from moist soil) during 1 50 days of plant growth. Such forms are likely to be : (a) o rganic S that is mineral ized during plant growth o r (b) aggregate-protected, adsorbed sulphate, (So4=) , which becomes accessible to the penetrating roots or root hairs during plant growth but not to CaCI2 extraction of moist so il. 4. The specific act ivities of CaP-S extractable S from samples of freeze-dried and ground soil (0-6 cm) were similar to that of S taken up by plants indicating that this extract from this soil depth bette r represents the plant available S pool than the other extraction methods, however, this result does not confirm that the S in the extract is the precursor of S taken up by plants. Confirmation that CaP-S extracts plant available S requires a quantitative assessment of plant S uptake relative to depletion of CaP-S. The CaP-S pool is dynamic and requires a consideration of all factors influencing its magnitude. This is attempted in Chapter 8 . CHAPTER 7 THE INFLUENCE OF FERTILIZER FORM ON THE FATE OF SULPHUR IN SOILS 7.1 INTRODUCTION 1 71 Extensive field trials in New Zealand (Ludecke, 1 965; Sinclair and Enright, 1 983 ; Sinclair et al. , 1 985; Boswell and Swanney, 1 986, 1 988, 1 991 ; Lee and Boswell, 1 988; Boswell et al. , 1 988a, 1 988b; Swanney et al. , 1 988) have demonstrated that finer particle sizes of elemental S (S0) oxidize more rapidly than coarser particles and create h igh i nit ial plant S uptake rates. Leaching losses of S were not measured in these experiments but it may be predicted that more rap id S oxidation may lead to greater S leaching as observed in Chapter 5 when microfine S0 was used. Thus for S conservation in a grazed pasture system, a compromise must be reached between oxidation rate, plant uptake and leaching loss . This compromise may best be achieved by varying the particle size distribution of the S0 ferti lizer for specific situation. Before such fertilizer products can be designed the effect of different S0 particle sizes on the fate of S in soils must be determined. Most s0 appl ied to pasture in New Zealand is combined with P fertilizer such as S0/RPR mixtures and sulphurized SSP in granu le forms. The presence of phosphate ferti l izer may influence S0 oxidation (Kittams and Attoe, 1 965; Attoe and Olsen, 1 966; Bloomfield, 1 967; Lee et al. , 1 987) and the fate of fertilizer S in the soi l/plant system (During, 1 984; Boswel l , 1 983; Bolan et al. , 1 986, 1 987) . In this series of experiments the effect of phosphate fertilizer on the fate of S from both sulphate-S and s0 was examined as was the effect of elemental S particle size. In Chapter 5, it was found that accumulation of o rganic S in soil was larger, and estimated leaching losses of S from field soils under pastures were smaller, when S0 was applied rather than single superphosphate (SSP) . Apparent sulphate leaching losses of applied fertilizer S beyond 1 0 cm depth in this field experiment were large, 61% for SSP and 44% for s0 treated soi l cores. These losses were uncontro llable due to the unpredictable nature of seasonal rainfall. The large leaching loss of S from S0 was probably due to the fact that the S0 used was microfine and was rapidly oxidized to sulphate . In the l ight of these resu lts it was necessary to repeat supplement this earlier experiment with particle size ranges of s0 which reflect the size ranges normally applied in fertilizer material. 7.2 OBJECTIVES 1 . To examine the influence of S0 particle size on the fate of S0 in pasture soils 1 72 2. To examine the effect of phosphate fert i l izer on the fate of S from S0 and sulphate-S fertil izers 3. To examine the effect of granulation on the fate of S from S0 and S0/phosphate rock fertilizers 7.3 MATERIALS AND METHODS A second experiment using undisturbed soi l cores, isolated in galvanized steel cylinders similar to those used in Chapter 5, was designed and conducted on two soil types; a Tokomaru silt loam (yellow-grey earth) , a pasture so il of lower ferti lity than that studied earlier and Ramiha s i lt loam, a yel low-brown earth/yel low-brown loam intergrade, sited in an area with higher rainfall and altitude. At each site only 1 0% of the soil cores prepared were left in situ for field studies, the remain ing undisturbed soi l cores were t ransferred to the glasshouse where leaching events could be controlled and the leachates collected for analysis 7.3.1 Soils The experiments were conducted both under f ield and glasshouse conditions at Massey Un ivers ity us ing und isturbed f ie ld so i l cores of Tokomaru and Ramiha s i lt loam soi ls contrasting in organic matter contents, S contents and P retention , vegetated by predominantly ryegrass/clover pasture. General soil properties of these two soils are shown in Table 7.1 and Table 7.2. These resu lts are derived form a series of small soi l cores ( 1 5 cores , 2 .5 cm diameter by 1 0 cm depth) sampled from each site prior to trial establishment. Soil cores were bu lked, air-dried and sieved through 2 mm prior to analysis. Tokomaru soil is genetically moderately l eached and moderately acidic, classified as a New Zealand yellow-grey-earth (Typic fragiaqualf). lt was formed on thick deposits of loess of fine sandy loam texture. Drainage is impeded during the wet season which results in a common occurrence of a pale-coloured horizon bearing iron/manganese concretions lying beneath the top and sub soils (Cowie, 1 978) . The field site was located at the Massey No. 4 Dairy farm 5 km from Palmerston North. The pasture composition on Tokomaru site were predominantly 1 73 clover (Trifolium repens) and ryegrass (Lolium perenne) and had not been fertilized for at least 1 3 years. The lack of recent fertilization is reflected in the low Olsen P and calcium phosphate extractable S (CaP-S) soil test values (Sinclair and Saunders, 1 984) . Ramiha soi l is derived from similar parent material to Tokomaru soi l , but formed under high rainfall and has been more strongly leached. The higher P retention is due to the weathering of volcanic ash present in the loess which produces amorphous type clay which are absent in Tokomaru s i lt loam soi l . Naturally, this soi l is low in exchangeable bases and is acid in react ion . The site used for the undisturbed so i l core study and collecting undistu rbed soil cores for glasshouse studies was a hil l soil ( intergrade between a New Zealand yellow-brown? earth/yellow-brown-loam) at Tuapaka farm at an altitude of 300 meters above the sea level, 1 0 km east of Palmerston North . The climate is cool-temperate with an evenly distributed rainfall throughout the year, normally peaking in June and December (Pollok and Mclaughlin, 1 986). On the Ramiha site there was a significant amount of browntop (Agrostis capillaris) , together with ryegrass and white clover. The Ramiha pasture had received superphosphate at the rate of 200 kg ha-1 ( 1 983-1 984) and 300 kg ha-1 ( 1 984-1 985) . I n 1 985-1 986, it received 200 kg ha-1 of Hyphos-S+Se (6% S) and no fertilizer was applied during the 1 986-1 987 trial period. Although the site had a low Olsen-P value , CaP-S levels were medium (Sinclair and Saunders, 1 984) . Both sites were under established pastures grazed by sheep. 7.3.2 Preparation of soli cores Galvanized stee l cylinders ( 1 mm wall th ickness, 1 5 cm diameter and 1 0 cm length) were driven into pasture of un iform sward content at both f ie ld s ites. The soil cores were then removed and the bottoms were sealed with nylon mesh ( 1 .00 mm opening). The cores were placed in a g lasshouse and maintained at a moisture content of approximate ly 90% of field capacity by weighing and watering every 3 days. This provided a simulated pasture situation, with soil microflora and fauna, top-soil profiles and plants virtually undisturbed. 7.3.3 Design of the experiments After surface application to the undisturbed soil cores the fate of S0 of three particle sizes, was studied for a pasture growth period of 1 80 days. An application rate of 30 kg S ha-1 was chosen for the study. Treatments consisted of three particle size ranges of 35s labelled S0 namely , <0 . 1 50 mm (SS) , 0 . 1 5 0-0 .250 mm (MM) , and 0 .250-0.500 mm ( LL) and two g ranu lated 35s0 labe l led , S0 mater ials made from the <0. 1 50 mm 35so labe l led so (g ranu lated s0 p lus f ine ground North Calorina phosphate rock and granu lated without 174 phosphate rock, SS/PR and SS/gr) . Two additional check soil cores, granulated phosphate rock (PR) and control (Ctrl) were also included in the study. These treatments were applied to the surface of pre-trimmed undisturbed soil cores on the 29 November, 1 987. Herbage and soil samples in both glasshouse and field trials were taken at various harvest intervals during the 1 80 days of growth. Treatments and replications are presented in Tables 7.3, 7 .4 , 7 .5 and 7.6. In the glasshouse trial, two additional treatments with four replicates each were also applied to both soils along with the above study. These treatments were 35s labelled gypsum (GP) and superphosphate (SSP) and were carried out for 90 days. Table 7.1 General properties of the soils used in the studies. Description GENERAL PROPERTIES Location Land form Rainfall (mm) Soil type NZ soil group Soil taxonomic classification Parent materials Clay minerals Bulk density (0-5 cm) (kg m-3) Soil pH, in water ( 1 :2) Organic carbon (%)c P retention (%)d Extractable pe (mg kg-1) Soils Rami ha 10 km east of Palmerston North Tararua range, rolling 1270-1520 Ramiha silt loam Yellow-brown earth/ yellow-brown loam Andic Dystrochrept Siliceous loess over graywackea Vermiculite, some illite, allophaneb 963 5.5 7.6 88 1 1 Tokomaru 5 km south of Palmerston North High river, flat 890- 1 140 Tokomaru silt loam Yellow-grey earth Typic Fragiaqualf Siliceous loessb Mica/illite some vermiculitea 1035 5.7 4.5 20 8 a Pollok (1975); b McLaughlin (1983); 0 Bremner and Tabatabai, 1971; d Saunders, 1965; e 0/sen et al. (1954). 175 Table 7.2 The forms and distribution of soil sa in three soil layers collected from field sites before the experiment. Layer Depth Total Organic Ester Carbon Extractable S b s s sulphate bonded cm s s CaP-S CaCI-S mg S kg- soil RAMI HA Top 0-3 540 524 201 323 15.2 10.0 Middle 3-6 434 425 198 327 8.2 4.2 Bottom 6- 10 445 437 21 1 226 7.7 3 .0 TOKOMARU Top 0-3 395 384 125 259 9.0 6.4 Middle 3-6 365 358 13 1 227 7.3 4.2 Bottom 6-10 320 3 13 138 175 7.1 3 . 1 a method of Landers e t al., (1983); b contains some soluble organic S (HI-reducible S) (Watkinson et al., 199 1) 7.3.4 Labelling fertilizer S 7.3.4. 1 Labelling SJ containing fertilizers Three different particle sizes, <0. 1 50 (SS) , 0 . 1 50-0.250 (MM) and 0 .250-0.500 mm (LL) , of S0 labelled with 35so were made for this study. The manufacturing methods were described in Section 3 . 1 .3. Radioactively labelled 35s0 was manufactured in two Lots : Lot I, low activity , destined for soil cores removed form the f ie ld and glasshouse during the early stages (0-60 days) of the experiment and Lot 1 1 , high activity, destined for soi l cores sampled late in the experiment (90- 1 80 days) Lot I . A 0.4 ml aliquot of 651 .2 MBq mr1 of carrier-free 35s i n toluene was added into 4 .509 g of finely ground e lemental S. This yielded 3.207 MBq per 53 g of S (standard deviation=1 .35) or a 35s specific activity of 60.51 MBq g-1 S. Soil cores that received this labelled material were destructively sampled at 1 5, 30, 45 and 60 days. Lot 11 1 76 A 2.1 ml aliquot of 651 .2 MBq mr 1 of carrier-free plus residual 35so from lot I were added into 7.306 g of finely ground e lemental S. This yielded elemental 35s with 94.295 MBq per 53 g S (standard deviation=2 .89) or a 35s specific activity of 1 779 MBq g-1 S. The soil cores that were destructively sampled at 90, 1 20 and 1 80 days received this labelled material. The particle size separates (<0 . 150 mm (SS) , 0 . 1 50-0 .250 mm (MM), and 0 .250-0.500 mm (LL)) of 35s labelled elemental S from both Lots, Lot I and Lot 1 1 , were prepared in the same manner by crushing and sieving through nylon cloth in a plastic bag in a fume hood. . The finer size (SS) had approximately 30% of the particles less than 0 .050 mm. During crushing and sieving approximately 2-4% of the s0 was not recovered from either mortar or sieve. The characteristics of the labelled fertilizers are shown in Table 7.3. 7.3.4.2 Labelling sulphate containing fertilizers Finely ground 3Ss labelled superphosphate (SSP) and gypsum (GP) were also prepared as described in Section 3.1 .4. For 35s labelled SSP 2.46 g of 65% H2so4 containing 92.5 MBq of 35so4= was added into 2.62 g of a 1 :1 mixture of Christmas Island A : Nauru phosphate rocks. This produced SSP containing 1 4.748 MBq g-1 (SD=0 . 1 04) or 35s specific activity of 1 47.48 MBq g-1 S ( 1 0.00% S sd=0.02) . The product was dried at room temperature and finely ground. About 1 . 65 g of calcium carbonate (99%) were mixed with 3.2 g of H2so4 (50% w/w) containing 92.5 MBq of 35so4= and dried in forced drought oven at 30 ?C. The GP produced had a specific activity of 1 48.02 MBq g-1 S (1 8 .5% S sd=.01 5) or 27.383 MBq g-1 of gypsum (sd=0.1 05) . The product was also finely ground after drying . Results of the 35s labelling of GP and SSP are shown in Table 7.4 7.3.4.3 Granulation of Ef> and Ef> /phosphate rock Granules of 35s0 with and without finely ground phosphate rock were prepared as described by Chatupote ( 1 990) . Fine 35s0 (<0.1 50 mm) and a mixture of fine 35so (<0.1 50 mm) and finely ground North Carolina reactive phosphate rock (NCRPR, 1 3 .2% P; 1 00% <0. 1 50 mm, 80% <0.075 mm) were granulated (0.5-1 mm granules size) with 1% agar and saturated KCL solution and yielded ferti lizer granules containing 9% S0, Table 7.3, (Chatupote, 1 990). Table 7.3 Characteristics of 35so labelled fertilizer used in this study. s s 35s 35s Lot Fertilizers content Applied a Activity Specific No. and applied activity sizes (mm) Abbrev. % mg S core-1 MBq core-1 MBq g-1 S Lot I <0.150 ss 100 53 3.207 60.51 0.150-0.250 LL 100 53 3.207 60.51 0.250-0.500 MM 100 53 3 .207 60.51 SS/PR* SS/PR 8.9 53 3.041 57.38 ss* SS/gr 8.6 53 3.207 60.51 Lot ll <0.150 ss 100 53 9.429 177.91 0.150-0.250 LL 100 53 9.429 177.91 0.250-0.500 MM 100 53 9.429 177.91 SS/PR* SS/PR 8.9 53 8.797 165.98 ss* SS/gr 8.7 53 9.429 177.91 ? Elemental S of particle size of <0. 150 mm was used and were granulated with/without finely ground phosphate rock and the total P content in the SS/PR granules was 10.6%. a equivalent to a rate of 30 kg S ha-1 Table 7.4 Characteristic of 35so4 = labelled superphosphate and gypsum used in this study. 35s 35s Amount s s Activity Specific applied applied content applied activitl Fertilizers mg core-1 mg core-1 % MBq core-1 MBq g- S GP 285 52.7 18.5 7.81 148.38 SSP 485 48.5 10.0 7.15 147.37 178 7.3.5 Experimental conduct 7. 3. 5. 1 General Before fertilizer application, herbage on soil cores from the glasshouse and field were cut to about 2.5 cm height and the herbage discarded. The fertilizers were surface applied on the undistu rbed soil cores at the rate of 53 mg S core?1 (30 kg S ha-1 ) . Treatments , repl ications, and harvesting and sampling schedules for the field trials are shown in Table 7.5 and those for the glasshouse trials are shown in Tables 7.6 and 7.7. A hand sprayer was used to s imulate small amounts of rainfall in the g lasshouse t rials. D isti l led water was sprayed onto the surface of the undisturbed soil cores twice a week. Additional watering was conducted by adding water to the plastic container in which the pot stood to maintain the core at 90% field capacity. Under the glasshouse conditions, a minus N, P , and S nutrient solution (M iddleton and Toxopeus, 1 973) was applied regularly to supply other nutrient to soil cores. At the field site the soil cores were fenced to prevent access by stock. 7.3.5.2 Leaching events I n addition to regular watering the undistu rbed soi l cores were leached on five occasions during the first few weeks with a total volume of water equivalent to 50 mm of rainfall ( 1 0 mm in each event) to s imulate early winter rain events. The leachates were collected in plastic bags, weighed and analysed for H I-reducible S and 35s activity. Three leachings (at days 7, 1 4 and 44) were made when the soil in the undisturbed soil core was close to maximum water holding capacity and two leachings (at day 2 1 and 56) were made when the cores were close to wilting point. 7.3.5.3 Soil and herbage sampling Field herbage harvesting and soil sampling schemes and the number of treatment replications sampled at each harvest are shown in Tables 7.5, 7.6 and 7.7. Herbages were cut at 2.5 cm above the soil surface and analysed as described in Table 7.8. Soil samples were sectioned into top, middle and bottom layers (0-33, 3-66, and 6-1 00 mm, respectively) . The soil samples from the radioactive treatments were frozen and freeze-dried to stop microbial activity before being ground to <1 mm particle size using a large hammer mil l . The non-radioactive treatment soil samples were air-dried at room temperature (20 ?C) 179 Table 7.5 Treatments, soil sampling and herbage harvesting schedule for treatments with different particle sizes of elemental S in the glasshouse trial. Days after fertilizer application Treatments Abbrev. Lot 15 30 45 60 90 120 150 1 80 Herbage Samplings Number of replications < 0. 150 mm ss I 1 3 1 3 < 0. 150 mm ss 11 10 1 10 1 10 7 7 7 0. 150-0.250 mm MM I 1 3 1 3 0.150-0.250 mm MM n 1 10 1 10 10 7 7 7 0.250-0.500 mm LL I 1 3 1 3 0.250-0.500 mm LL 11 1 10 1 10 10 7 7 7 SS + PR* SS/PR I 1 3 1 3 SS + PR* SS/PR 11 1 10 1 10 10 7 7 7 ss* SS/gr I 1 3 1 3 ss* SS/gr 11 1 10 1 10 10 7 7 7 PR1 PR nl2 1 13 1 13 10 7 7 7 Control Ctrl nl2 1 13 1 13 10 7 7 7 Soil Samplings Number of replications < 0. 150 mm ss 11 1 3 1 3 3 7 0.150-0.250 mm MM n 1 3 1 3 3 7 0.250-0.500 mm LL 11 1 3 1 3 3 7 SS + PR* SS/PR 11 1 3 1 3 3 7 ss* SS/gr 11 1 3 1 3 3 7 PR1 PR ni2 1 3 1 3 3 7 Control Ctrl nl2 1 3 1 3 3 7 ? fP of particle size of< 0. 150 mm was used and granulated with/without finely ground phosphate rock 1 Finely ground and granulated phosphate rock was used 2 nl = not labelled 1 80 Table 7.6 Treatments soil sampling and herbage harvesting schedule for treatments with different particle sizes, S0 in the field trials. Days after fertilizers application Treatments Abbrev. Lot 15 30 45 60 90 120 150 180 Herbage Samplings Number of replications < 0. 150 mm ss II 3 3 3 3 3 0.150-0.250 mm MM II 3 3 3 3 3 0.250-0.500 mm LL IT 3 3 3 3 3 SS + PR* SS/PR II 3 3 3 3 3 ss* SS/gr II 3 3 3 3 3 PR1 PR nl2 3 3 3 3 3 Control Ctrl nl2 3 3 3 3 3 Soil Samplings Number of replications < 0. 150 mm ss II - 0.150-0.250 mm MM 11 0.250-0.500 mm LL IT SS + PR* SS/PR II ss* SS/gr II PR1 PR nl2 Control Ctrl nl2 ? !fJ of particle size of <0. 150 mm was used and granulated with/without finely ground phosphate rock 1 Finely ground and granulated phosphate rock; 2 nl = not labelled 3 3 3 3 3 3 3 3 3 3 3 3 3 3 Table 7.7 Treatments and soil sampling and herbage harvesting schedule for treatments with 35s labelled gypsum and superphosphate fertilizers in the glasshouse trial. Days after fertilizer application Treatments Abbrev. 30 60 90 Number of replications Herbage sampling Gypsum GP 4 4 4 Superphosphate SSP 4 4 4 Control Ctrl 4 4 4 Soil sampling Gypsum GP 4 Superphosphate SSP 4 Control Ctrl 4 181 7.3.6 Chemical analyses 7.3.6. 1 Soil, plant and fertilizer samples Analyses for S and related 35s activit ies in soi l , plant and fert ilizer materials are listed in Table 7.8: Table 7.8 Analyses for S and related 35s activities in soil, plant and fertilizer materials. Described in Analyses Section Chapter Soil samples Total S 3.3.3 3 Total S0 (acetone extract) 3.3.4 3 Extractable S (CaP-S) 3.3.2 3 Organic S 3.3.5 3 Plant samples Total S 3.3.1 . 1 3 Fertilizer materials Sulphate S 3.3.7.1 3 so 3.3.7.2 3 HI -reducible S in soil extract and digested samples 3.3.6 3 7.3. 6. 2 Sulphate retention Sulphate retention capacities of Tokomaru and Ramiha soils from each soil depth (Appendix 7. 1 4) were determined as described by Gregg ( 1 976) . Five grams of soil were shaken for 1 6 hours i n an end-over-end shaker with 25 ml of 0.01 M CaCI2 containing different concentration of S (0, 50, 1 00, 1 50, 200, 250, 500, ppm S as Na2so4) . To suppress microbial activity, a few drops of chloroform were added before shaking. The soi l suspension were filtered through Whatman No 42 f i lter papers . Then the extracts were d i luted to a concentrat ion range between 5 to 1 0 mg S r1 and the S concentration measured using an automated, reduction? distillation technique for measuring H I-reducible S (as described Section 3.3.6). 1 82 The amount of su lphate retained by soil at each level of sulphate addition was calculated as followed: %S retention where A X y = = = = [(A + X) - Y]/A*100 amount of S contained in the 0.01 M CaC12 solution amount of soil sulphate released into 0.01 M CaC12 (no S added) solution in a separated extraction amount of sulphate left in solution after shaking the soil with 0.01 M CaC12 containing different concentrations of so4 = Results are presented in Appendix 7. 1 4 7.3.6.3 Recovery of added fertilizer 35 SJ activity from soil cores at day 0 A short term experiment was also designed to examine efficiency of the recovery of total 35s activity from soil cores fertil ized with 35so labelled ferti l izers at day 0 . Elemental S was labelled with small amount of 35so activity, about 3.3 MBq per g S0 (as described in Section 3 . 1 .3.2) and crushed into two particle sizes; < 0 . 1 5 0 mm and 0 . 250-0.500 mm. Each particle size was divided into three application rates, namely 1 2.5, 25.0 and 50.0 mg S0 core?1 ( 1 5 cm diameter) . After the labelled ferti l izers were evenly applied onto the so i l surface (0-3 cm layers) , the whole upper soil layer was freeze-dried for 7 to 1 0 days. The dried samples were then hammer mil led before total 35s determination ; total 35s (NaHC03 + Ag20; Landers et al., 1 983) , total S0 and 35so activity (Chatupote, 1 990) , as described in Section 3.3.7. 1 Resu lts are presented i n Appendix 7 . 1 5 . There was no significant d ifference between treatment means. Recoveries range from 70- 1 0 0% and grand means of the percentage recoveries of added fert i l izer 35s were used as correct ion factors to correct measured amounts of s0 to achieve amounts of 35s present in soil cores. 7.3.7 Statistical analyses Single replicated soil data collected at 1 5 days and 45 days used for monitoring rate of S0 oxidation, were not subjected to the statistical analysis and data were not shown. Other replicated data on herbage and soil according to Table 7.5 were subjected to analysis of variance to determine the significance of treatment effects . Data from both soils and herbage harvest were statistically analysed on a single harvest date basis. The Duncan Multiple Range 1 83 Test ( DMRT)and the Least Significant Diffe rence ( Lsd) were employed in comparisons of t reatment means where appropriate and resu lt o f the comparisons are presented in appropriate Tables and Appendices. The SAS (SAS Institute Inc. , 1 985) and Minitab (Minitab Inc. , 1 989) computer programmes were employed. Data on 35s activities were normalized to the day when the 35s0 labelled fertil izers were appl ied ( November 29 , 1 987) . Some related 35s calculat ions have been presented in previous Chapters (Section 4.3.4 and Section 5 .3 .5) . Addit ional relevant calculations are presented here : a. Percentage of applied 3Ss cumulatively taken up by pasture percentage of applied 35s cumulatively taken up by pasture at time t amount of 35s activity cumulatively taken up by pasture at time t total amount of 35so or 35so4 -2 activity in labelled fertilizer applied to soil core at time 0 b. Percentage of applied 35so in soil core = where amount of 35s0 activity recovered in acetone extract of soil cores at time t (corrected using recovery factors determined in Section 7.3.8.3 total amount of 3Sso activity in labelled fertilizer applied to soil core at time 0 c. Percentage of oxidized 3S S0 cumulatively taken up by pasture = where = = = percentage of oxidized 35so cumulatively taken up by pasture at time t percentage of applied 3Ss cumulatively taken up by pastures at time t percentage of applied 35so in soil core at time t 7.3.8 Calculation of specific rates (K) of s0 oxidation. The acetone extraction of S0 from soil samples in this Chapter was conducted by W.Chatupote as a part of h is Ph.D. thesis (Chatupote, 1 990) . An iterative computer programme developed in the thesis was used to calculate the K values (specific oxidation rate Jl9 S cm-2 day-1 ) . The 1 84 programme assumes that S0 particles are spheres and that specific oxidation rate is constant per un it surface area of S0 exposed. A description of these relat ionships was given by (Watkinson, 1 988, 1 989) 7.4 RESULTS AND DISCUSSION Weekly amounts of rainfal l , and drainage water, and average weekly maximum and minimum temperatures which represented the climate conditions in the Palmerston North district and of the field sites during these experiments were presented in Appendix 7. 1 6 (Taken from DSIR Grasslands, P .N . Meteorological records). Drainage water percolating beyond 10 cm, was calculated as described by Scotter et al. ( 1979). Drainage occurred in the field soils at both sites before the trials were laid down and thereafter no drainage occurred. Rainfall during the trial period was approximate ly 450 mm. However , there were some dry periods which produced soil water deficits that probably limited pasture growth during weeks 14-20 and 26. On average , maximum and min imum temperatu re were about 20 and 1 0 ?C which was considered to be adequate for S0 oxidation in soils (Janzen and Bettany, 1 987b). In general soil pH and available P (Table 7. 1 ) for both soils were sl ightly below opt imum for pasture growth (Cornforth, 1 981 ; Cornforth and Sinclair, 1 984) . Extractable S (CaP-S, Table 7.2) , in the Ramiha soil was optimum, but was considered to be low for the Tokomaru soil (critical level = 1 0 ppm S, according to Sinclair et al. , 1 985). The majority (>95%) of S in both soils was present as organic S (Table 7.2). The S retention (adsorption) of both soils was considered to be low (see so4= adsorption isotherm results in Appendix 7. 1 4) despite the fact that the Ramiha soil had a high phosphate retention capacity. The S retention capacity in the top 0-3 cm soil layer of Ramiha soil was lower than that of Tokomaru soi l , probably due to larger amounts of organic S and adsorbed P (as indicated by Olsen-P in the Ramiha soi l relative to the Tokomaru soil ) . Searle ( 1 982) , Johnson et al. ( 1 979) , Johnson and Todd ( 1 983) have shown the negative influence of organic matter on sulphate adsorption. In general, fertility status of the Rami ha soil was considered to be higher than that of Tokomaru soil partly as a result of its better fertil izer history. 7.4.1 The percentage recovery of 35so and %SDFF In pasture as Influenced by fertil izer specific activity Data on percentage recovery of applied 35s in pasture and the percent of accumulated plant S derived from fert i l izer (%SDFF) at 30 and 60 days after fert i l izer application were used to compare results obtained using 35s0 of different specific activity (Lot 1 and Lot 11, Table 7.3). 1 85 Only data from the SS treatments ( <0 . 150 mm) could be used for this comparison during early stages of the experiment because the amount of 35so released from the coarser particle sizes MM and LL was low and subject to variation. Resu lts are presented in Table 7.9. Sample count rate had no s ignificant effect on the apparent recovery of fertil izer S in the plants and therefore such resu lts obtained from the two d ifferent sources of labelled 35so could be compared directly when expressed as a percentage of the original isotope added. Table 7.9 Effect of fertilizer specific activities on 35s recovery in pastures and percent of plant S derived from fertilizers (%SDFF). 30 days 60 days Soils Lot No. Recovery %SDFF Recovery %SDFF % % % % Ramiha Lot I 2.5 24.3 4.8 67.8 Lot ll 3 . 1 23.5 4.4 58.5 F-test ns ns ns ns C.V. % 25.4 17.9 23.5 23.8 Tokomaru Lot I 1 .3 17.4 2.3 36.6 Lot H 1.5 13 .7 1 .5 35.2 F-test ns ns ns ns C.V. % 23.3 19.4 35.7 22.7 ns = not significant; 1 average of seven replications; 2 average of three replications 7.4.2 THE INFLUENCE OF PARTICLE SIZE AND FERTILIZER FORM ON THE FATE OF SJ 186 I n this Section, discussion will be focused on the fate of 35s0 as influenced by 35s0 particle size (SS, MM and LL) , the effect of granulation of fine S0 (SS/gr) and granulation of fine S0 with phosphate rock (SS/PR) . 7.4.2. 1 Recovery of radioisotope in soils and plants The cumulative uptake of 35s radioactivity by plants and that remaining in the top 1 0 cm of the glasshouse and field soil are shown in Table 7.1 0 for the Ramiha soil and Table 7. 1 1 for the Tokomaru soil. By 1 80 days cumulative plant uptake under glasshouse conditions accounted for less than 20% of the isotope and the majority (75-90%) remained in the soil. The amount of isotope leached beyond the 1 0 cm soi l in glasshouse cores was negligible (see later, Table 7 . 1 5) . I n many cases the total (soil+plant) recovery of added 35s was apparently greater than 1 00%. This error is attributable to the difficulty of obtaining a uniform distribution of the particu late S0 i n each soi l sample taken for analysis for total 35so (acetone extraction) and total 35s (NaC03+Ag2o digestion) , and was a particular problem with the coarse S 0 particle size (LL) . This problem has also been noted by Watkinson et al. , ( 1 987) and Barrow (1 968) . These errors could be reduced by increasing the number of undisturbed cores in each replicate but this would have produced an unmanageable workload. Notably, there is a closer spread of recoveries at 1 80 days when seven replicate cores were harvested. The recovery of 35s applied to Ramiha field cores was lower than g lasshouse cores. There may have been greater losses of sulphate S due to leaching at the higher rainfall Ramiha site. Recoveries of 35s activity lower than glasshouse cores were not recorded for the Tokomaru field cores. Table 7.10 Treatments 187 Recovery of labelled 35s fertilizers in pasture, soil and total recovery in Ramiha soil cores in glasshouse and field trials (average of three replications from individual microplots). 30 Days after fertilizer aplication 60 90 . ... . % recovered .... . Field 180 CUMULATIVE PASTURE UPTAKE ss 2.5a 7.5a 12.5a 20.5a 9.8a LL 0.4c 1 .4c 1 .9c 4.2c 1 .7b SS/PR 1 .5b 4.0b 7.7b 14.7b 8 .1a SS/gr 1 .7b 6.1b 6.2b 13.7b 8.3a F-test ** ** ** ** *** C.V. % 39.9 24.7 36.4 24.8 34.2 SOIL ss 1 14.9 91 .6 9 1 .3 75.4 74.5 LL 13 1 .9 1 19.8 109.9 97.8 87.7 SS/PR 104.4 101.9 98.4 99.2 81 .3 SS/gr 98.6 88.3 87.0 80.3 78.2 F-test ns ns ns ns ns C.V. % 14.3 15.8 4.9 1 1 .2 20.7 TOTAL ss 1 17.4 99. 1 102.8 95.9 84.3 LL 132.4 121.2 1 1 1 .6 101 .9 89.4 SS/PR 106.9 105.9 105.3 1 13 .8 89.4 SS/gr 99.8 94.4 90.8 93.4 86.5 F-test ns ns ns ns ns C.V. % 13.9 15.4 4.2 8.8 13.4 . ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; ns = not significant; mean separation by DMRT ' at 5% level denoted by letters 1 average of seven replications 1 88 Table 7. 1 1 Recovery of labelled 35s fertilizers in pasture, soil and total recovery in Tokomaru soil cores in glasshouse and field trials (average of three replications from individual microplots). Days after application application Field Treatments 30 60 90 1801 180 .... . % recovered . .. . . CUMULATIVE PASTURE UPTAKE ss 1 .2a 3.5a 6.3a 13.ga 4.2a MM 0.4b 1.9b 1 .4c 3.5c ub LL 0.2c 0.9c 0.8c 1.9c 0.6b SS/PR o.sb 2.1b 4.7b 12.6b 5.8a SS/gr 0.4c 1 .3b 4.ob 10.4b 3.9a F-test ** ** ** ** *** C.V. % 35.5 55.9 34.7 28.7 34.2 SOIL ss 1 15.1 100.9 88.4 92.9 94.9 MM 136.9 132.1 1 13 .1 101 .8 100.4 LL 126.7 122.6 141.3 104.4 105.5 SS/PR 1 16.3 1 1 1 . 1 109.4 99.1 101.4 SS/gr 120.6 107.5 86.2 87.3 91 .5 F-test ns ns ns ns ns C.V. % 13.7 15.2 12.7 13 .7 1 1 .4 TOTAL ss 1 16.3 104.5 92.8 106.9 99.2 MM 137.2 133.9 1 15.3 105.4 101.5 LL 125.8 123.5 142.3 106.3 106.1 SS/PR 1 16.7 1 12.6 1 13.9 122.5 107.1 SS/gr 120.9 108.8 90.9 97.7 95.4 F-test ns ns ns ns ns C.V. % 13 .7 14.9 12.5 12.6 20.0 ?, ?? and ... = significant at 5, 1 and 0. 1% level, respectively; ns = not significant; mean separation by DMRT at 5% level denoted by letters 1 , average of seven replications 7. 4.2.2 Plant uptake of S and 35 S and percentage of plant S derived from fertilizer (%SDFF) 1 89 Cumulative S uptake (soil S plus fertilizer S) by the herbage in glasshouse and field trial are shown in Figures 7.1 A and 7.2A and Appendices 7.1A and 7.28, respectively. By the end of the tr ial ( 1 80 days ) , on ly the smalle r part ic le size (SS) p roduced increased S uptake significantly higher than unfertil ized control (Ctrl) in both soils in the glasshouse trial which is partly due to the higher oxidation rate of finer S0 (Watkinson et al., 1 987). Throughout the trial period the <0.150 mm S0 (SS) treatment on Tokomaru soil which in itially had a lower sulphate status than the Ramiha soil (Table 7.2) , had significantly higher S uptake than the other two treatments (MM and LL) which produced more plant S than the unfert i l ized control (Ctrl). These significant differences did not occur on the higher S status Ramiha soil. In general, the S uptake from soil plus fertilizer was slightly higher on the Ramiha soil in both glasshouse and field trials. This soil had h igh initial sulphate status (Table 7.2) . Fertilizer application caused no increase in plant S uptake in the field trials which suggested that other factors (e .g. P, N avai lability, soil moisture) might have l imited S uptake which was 3-5 times lower than that occurring under glasshouse conditions where nutrients solution (K , Mg, Ca and some micro? nutrients) was also provided. I n general , both SS/PR and SS/gr treatments yie lded less cumu lative S uptake (soil plus ferti l izer S uptake at 1 80 days) than that of SS treatments i n g lasshouse trials ( Figure 7.2, Appendix 7. 1 A) , particularly on the Tokomaru soil . I n the field trials there was no effect of granulation (SS/gr) and phosphate rock addition (SS/PR) on S uptake (Appendix 7.2A) . Lower cumulative plant S uptake on the Tokomaru so i l than on Ramiha soi l cores in the glasshouse trials may be attributed to lower phosphate extractable S and total S in the intact soil cores of Tokomaru soil (Table 7.2) . Accumulated dry matter yields in the glasshouse and field trials are shown in Figures 7.3 , 7.4 and Appendices 7.1 B, 7.28, respectively. Appl ications of S0 and/or phosphate rock d id not significantly increase plant dry matter yields i n both glasshouse and field trials. Dry matter yield of herbage grown on the glasshouse undisturbed soil cores were consistently higher than on the f ield cores. This is due to the more favou rable growth condit ions (temperatu re and moisture) in the g lasshouse than in the f ie ld. The dry matter yields on both soi ls in the glasshouse trials were similar but were slightly higher for the Rami ha soil in the field trials. The smaller particle size (SS) consistently, but not always significantly, Increased herbage S concentrations at the last five harvests under glasshouse conditions and at the last f ield 1 90 harvest (Apppendix 7.3A and 7.38). At the first sampling, under the glasshouse conditions S concentrations of pastures were above the critical level of 0.30% (McNaught and Christoffe ls, 1 961 ; Cornforth et al. , 1 983 ; Sinclair et al. , 1 985). Thereafter, the S concentration of pastures fel l be low the critical leve l , except at the last sampling time. The increases in the plant S concentration at the last sampling might be attributed to increased mineralization of o rganic S as ind icated by increase in phosphate extractable S (CaP-S) in the soi l cores at the last sampling (Table 7.1 38). Although, there were some significant differences in the S concentration of pasture at the last harvest of the field trials (Appendix 7.38) , the S concentrations were below the optimum level (0.30%). The low plant S concentrations in the field trial may be partly attributed to periods of low soi l moisture content du ring the trials which would l imit mineralizat ion. A lso h igher evapotranspiration under glasshouse conditions would contribute to more plant S uptake than in the field, i .e . greater water use and therefore more S uptake. Results also showed no effect of granulation and phosphate rock on pasture S concentration. This indicated that surface applied granules (SS/PR and SS/gr) did breakdown and released S i n the same manner as the powder form (non-granulated, SS) . However, the increases in plant S uptake did not generate increased dry matter yield in e ither the glasshouse or field trials. There was larger cumulative recovery of 35s labelled fertil izer in pastures under glasshouse t han f ie ld condit ions ( F igure 7 .5 and Appendix 7.4) . The g lasshouse grown swards accumulated approximately twice the activity of 35s accumulated by the fie ld pasture. The smallest particle size (SS) produced the highest recovery percentage throughout the whole trial period i rrespective of whether it was granulated in either the presence or absence of phosphate rock. The greater amount of pasture grown on the Ramiha soil took up more 35s from labelled fertilizer than that grown on the Tokomaru soil . I n general , i n both soils pasture fertilized with the small particle size (SS) took up 4-5 times more 35s than the larger particle sizes (MM and LL) and this can be attributed to faster S oxidation rates of the finer materials (see Tables 7. 1 2 and 7 . 1 3) . On Tokomaru soi l , in both g lasshouse and field trials, the recovery of 35s by pastures grown using MM S0 was greater (but not significantly) than the recovery of 35s from LL fertilized soil. The trend in cumulative 35s uptake (Figure 7.6) revealed the effect of granulat ing SS with (SS/PR) and without (SS/gr) phosphate rock better than cumulative S uptake (Figure 7.2) . In the glasshouse trial, throughout the period of pasture growth, granulated s0 alone (SS/gr) and 191 s0 granulated with phosphate rock (SS/PR) produced significantly less cumulative 35s uptake than the non-granulated S0 (SS) (Figure 7.6 and Appendix 7.4A) . But the negative effect of granulation with or without phosphate rock was less pronounced in the field trials and addition of phosphate rock in the granulation mixture sl ightly increased the cumulat ive 35s uptake at the final samplings in the field trial on Tokomaru soil (Figure 7.8B, Appendix 7.4B). Percentages of plant S cumulatively derived from fertilizer (%SDFF) are presented in Figure 7.7 and Appendix 7.5 and show simi lar trends to data on recovery of 35s in pasture , as mentioned above. These results demonstrated that S0 of finer particle size (SS) provided substantially more S to plants than the medium (MM) and coarser S0 fractions (LL) . lt was interesting to note that both glasshouse and field trials produced similar %SDFF values and this was in contrast to the results of cumulative 35s recovery and cumulative S uptake of pastures. The same %SDFF in both glasshouse and field trials might indicate that the volume of soil occupied by the fertilizer 35s were the same in both situations. The %SDFF tended to increase during the trial period, this indicates that the 35s labelled S0 continued to oxidize and release plant avai lable 35s into the plant available pool during the experiment. This was confirmed by the recovery of residual S0 from soi l (see later discussion , Tables 7. 1 2A and 7. 1 3B). Similar %SDFF's in the glasshouse and field trial indicate similar efficiencies of S0 use. But as discussed above, there were large d iffe rences in the amount of S and 35s uptake in both systems as a result of differences in the plant growth environments. This suggests that as climatic conditions suitable for the uptake of soil S improved (moving from field to glasshouse) so the rates of S0 oxidation must have also increased such that the amount of 35s made available to the plant per unit of soil S remained essentially constant. Notably granulat ion of the SS part icle s ize a lone generally decreased the %SDFF, but not always significantly (Figure 7.8 and Appendix 7.5) . In itially, up to 90 to 1 20 days significantly smaller %SDFF's occurred in treatments with SS/gr and SS/PR in both glasshouse and field trials. In the g lasshouse t rial on Ramiha soil and in the f ie ld trial on Tokomaru so i l , the negative effect of g ranu lat ion (SS/gr) on %SDFF prevai led throughout the t ria l period . However, by the end of the trial the differences between the %SDFF for SS and SS/gr were less pronounced. Addition of ground phosphate rock (SS/PR) signif icantly al leviated the negat ive effect of granu lation on the %SDFF of pasture in f ie ld trials on Tokomaru soi l . Overall , i n both glasshouse and f ie ld tr ia ls addit ion of phosphate rock (SS/PR) sl ightly m it igated the negative effect of g ranu lat ion on %SDFF. This could have resu lted from improved granule dispersion or through the positive influence of phosphate rock (PR) on so oxidation ( Chatupote, 1 990) . In all soils fertilized with <0.1 50 mm S0 (SS) , 27-43% of plant S (%SDFF) was derived from 1 92 the fertilizer, however, several of these treatments did not generate increased dry matter yield or increased plant S (soil plus fertilizer S) uptake. This indicated that fertilize r S was taken up at the expense of native soil S (Rennenberg, 1 984) . When the results were expressed as the percentage of oxidized 35so cumulatively taken up by pasture, it was evident that at most harvests in the glasshouse experiment and a t the final harvest in the f ie ld trial, plants recovered oxidized 35s more efficiently from S of smaller particle size (Figure 7.9 and Appendix 7.6) . The higher percentage of oxidized 35so taken up by pasture fertilized with finer S0 particles (SS) was probably because at the same rate of S0 application the finer particles (SS) fertilized a greater soil volume and thereby created a greater chance for root interception of the su lphate released from S0. At the final glasshouse harvest, however, plants grown on Ramiha soil fertilized with fine (SS) and coarse (LL) S0 recovered similar percentages of oxidized S0. lt is unclear why this occurred because in the Tokomaru soil there were consistent reductions in the percent of oxidized 35so taken up by plant as the S0 became coarser. Notably, the percentage recoveries of oxidized S0 by plants were lower in fie ld cores than glasshouse cores again probably reflecting the greater water use by and S demand of the glasshouse plants. Granulation of the SS particle size with (SS/PR) and without (SS/gr) phosphate rock in itially reduced amounts of oxidized 35so taken up by plants (Figure 7.1 0, Appendix 7.6) but as the experiment continued inclusion of phosphate rock on average produced higher percentage recoveries of oxidized 35so. 2 Cumulative S 1 uptake 1 .5 -2 1 t g m _/! (A) ?? o.5 T? 0 Cumulative S 2 J uptake 1 . 5 -2 1 t g m ? (B) ?? ? o.5 T ? _.,_-0 Figure 7.1 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glasshouse Days Field The effect of S0 particle size on the cumulative S taken up by pasture grown on Ramiha (A) and Tokomaru (B) soils. e ss * LL + Ctrl e ss + MM * LL ....._ Ctrf ..... (!) (..) Cumulative S 2.5 uptake 1 .? t (A) - ? g rri 2 1 o .5 1?- -L? I@ ? 0 Cumulative S 2.5 1 uptake 2 1 .5 g rri 2 1 0.5 1 0 (B) A ? ' , 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 G lasshouse Days Fie ld Figure 7.2 The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative S taken up by pasture on Ramiha (A) and Tokomaru (B) soils. e ss * SS/PR * SS/gr e ss * SS/PR * SS/gr .... (0 +:> Cumulative dry matter yield -2 g m Cumulative dry matter yield -2 g m Figure 7.3 800 600 l ? (A) e ss 400 t ? - ? * LL 2oo T ? ? -*- Ctrl 0 800 1 (B) e ss 600 -':) + MM 400 t ? ,, ? * LL 2oo T? ? - -*- Ctrl 0 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glashouse Days Field The effect of S0 particle size on the cumulative dry matter yield of pasture on Ramiha (A) and Tokomaru (B) soils. .... I.D 01 Cumulative dry matter yie ld -2 g m Cumulative dry matter yield -2 g m Figure 7.4 800 600 t (A) e ss ? * SS/PR 400 200 t?' ? SS/gr 0 BOO T (B) e ss 6oo I ? * SS/PR ?'- 400 200 I ? ? -? ? SS/gr 0 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glasshouse Days Field The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative dry matter yield of pasture on Raimiha (A) and Tokomaru (B) soils. .... <.D Ol 35 Cumulative S 25 uptake 20 l o/o 1 5 1 0 s l 0 35 25 Cumulative S 20 I uptake 1 s T o/o 1 0 s i 0 ? (A) / ,, ?- (B) ? ? 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glasshouse Days Field e ss * LL e ss + MM * LL Figure 7.5 The effect of S0 particle size on the cumulative percentage 35s uptake by pasture on Ramiha (A) and Tokomaru (B) soils. ..... (0 ..., 35 Cumlative S uptake o/o 35 Cumulative S uptake % Figure 7.6 25 20 t ? 1 5 1 0 5 f ? 0 25 l 20 1 5 1 0 I ? 5 f ? 0 30 60 90 1 20 1 50 1 80 Glasshouse ""' - (A) ? - - (B) ? 30 60 90 1 20 1 50 1 80 e ss * SS/PR - SS/gr e ss * SS/PR - SS/gr Days Field The effect of granulation of S0 with or without phosphate rock (SS/PR and SS/gr compared with SS) on the cumulative 35s taken up by pasture on Ramiha (A) and Tokomaru (B) soils. .... (J:) 0) %SDFF %SDFF Figure 7.7 SO T (A) 40 30 + / ? ? n-/ 20 1 ? t ? . X . X . X . )( . . )E . * ---1( * X X 50 (B) 40 30 20 1 0 0 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glasshouse Days Field The effect of S0 particle size on the percentage of cwnulative plant S derived from fertilizer (%SDFF) on Ramiha (A) and Tokomaru (B) soils. e ss * LL e ss + MM * LL ..... <0 <0 %SDFF %SDFF Figure 7.8 so ? (A) 40 t B 8 e A-_ - ? e ss 30 t 0 - . ? * SS/PR 20 1 0 -*- SS/gr 0 so I 40 (B) e ss 30 t ? G--EY? * SS/PR 20 1 0 1? ? -*- SS/gr 0 30 60 90 1 20 1 50 1 80 30 60 90 1 20 1 50 1 80 Glasshouse Days Field The effect of granulation of S0 with or without phosphate rock on the percentage of cumulative plant S derived from fertilizer (%SDFF) on Ramiha (A) and Tokomaru (B) soils. I\) 0 0 Ox id ized taken up /. Ox idized taken up /. 35 38 s 25 t V7'] B88'dll 28 15 t ? - ? ?? ? I 18 5 - 8 35 38 s 25 t 28 15 ? ? 18 I V7.A VA&}BSSBI 5 8 38 68 98 188 G lasshouse Da!:;Js Figure 7.9 The effec t of 35s0 particle size on the percentage of oxidized 35so cumulatively taken up by pasture on Ramiha (A) and Tokomaru (B) soils. (A ) ? ( B ) 188 F ie ld FLi ss BBI LL r2] ss ? MM IBB LL "' 0 ..... 35 3B I Ox id ized S IJClOIJI CA ) taken up ZB ! .. . 1B ? 111::::?::1 ' :::::: : : 1111 11 /. lllN B 35 3B Ox id ized s I ( B ) taken up ZB ' ?? /. 1B f ? ? Mlk/1 I ? ?m ?a:::::::1 lf$lor::::::l ?Ufm B 3B 6B 9B 18B 18B G lasshouse Da!:Js F i e ld Figure 7.10 The effect of granulation of S0 with or without phosphate rock on the percentage of oxidized 35so cumulatively taken up by pasture (SS/PR and SS/gr compared with SS). ma ss ? SS/PH [] SS/gr ma ss w.3 SS/PH D SS/gr "' 0 "' 7.4.2.3 Residual 35 SJ activity and SJ (acetone extracts), extractable S and 35 S (CaP-S) activity in soil cores 203 Negligible amounts of 35so activity were detected in acetone extracts of soil samples taken below 3 cm depths from some selected glasshouse and field soil cores. The percentage of applied 35so recovered in the top layer (0-3 cm) soil cores are presented in Table 7.1 2A. At sampli ng dates up to 60 days, the residual activity of 35so was not significantly different in cores fertilized with S0 different particle size. The reason for this lack of significance was large variability in measurement of 35so by acetone extraction and total 35s determination in soil samples. I n Tokomaru so i ls , at l east 60-90 days of incu bat ion were requ i red before differences i n the activity of 35so remain ing in soi l fert i l ized wit h d ifferent particle s ize treatments were evident . For both soils the rate of _oxidation of S0 , as indicated by residual 35so activity in soils from g lasshouse and field trials, appeared to be approximately similar, ranging from 1 0 to 1 4 llg S0 cm-2 (specific oxidation rate per un it surface area) for the SS particle s ize in Tokomaru and Ramiha soils, respectively (Chatupote, 1 990) . In general, by 1 80 days, the amount of S0 oxidized increased significantly with a decrease in S0 particle size. In general, the percentage recoveries of 35so in the Ramiha soil were lower than those in the Tokomaru soil . This was attributed to a slightly faster oxidation rate of S0 in the Ramiha soil as suggested by Chatupote ( 1 990) . The faster oxidat ion rate of S0 in the Ramiha soil may have resulted from the higher available moisture content in this soil (Moser and Olsen, 1 953; Wainwright, 1 984) and higher organic matter content (Wainwright et al., 1 986) . After 1 80 days the extent of S0 oxidation as determined from residual 35s0 (Table 7.1 2A) , was similar in both glasshouse and field trials. Oxidation rates as measured by the amount of S0 remaining in soil (Table 7.1 3A) showed similar trends to results obtained from the amount of 35so act ivity remaining (Table 7 . 1 2A) . At early samplings, however, the 35so provided a more sensitive ( less variability in repl icate data) measure of S0 oxidation than measuring the amount of S0 residues (Table 7.1 2A) . In general , data on the amount of 35so activity (Table 7 .1 2A) and S0 (Table 7. 1 3A) remaining in the soil cores fertilized with granulated materials SS/gr and SS/PR did not show significantly different S0 oxidation rates in both glasshouse and field trials. Percentages of 35s0 bio logically oxidized into phosphate extractable 35s (CaP-35s , HI? reducible 35S) forms in the 0-1 0 cm soi l cores are shown in Figures 7.1 1 , 7. 1 2, Table 7 .1 28 and Appe ndix 7. 7. Throughout the 1 80 days, the f iner part ic le s ize (SS granulated or ungranulated) produced larger amounts of extractable 35s and by 1 80 days, in the glasshouse trial, about 1 2 and 1 8% of applied elemental 35s0 appeared in extractable 35s in Tokomaru and Ramiha so i l s , respect ively. The percentage of 35so t ransformed Into phosphate extractable 35s in field trials was approximately 50% of that in the g lasshouse trial. Given that 204 s0 oxidation rates in glasshouse and field soils were similar, these results suggested that the net immobi l ization of 35s label led ferti l izer was greater in f ield soi ls . This supported the greater recovery of added 35s as organic S in field soils at 1 80 days (Table 7 . 1 48). lt may be that the warmer g lasshouse temperature i nduced g reater re-mineral ization of previously immobilized 35s which would maintain higher 35s activity in the CaP-S extracts. Very low amounts of 35s activity remained i n the CaP-extract of the MM and LL treatments suggesting that the rate of S0 oxidation in these treatments did not exceed the rate at which solution SO 4 = was immobilized by synthesis of plant and soil organic S. As shown in Appendix 7.7, the majority of 35s activity recovered by CaP-extracts was in the 0- 3 cm soil depth. Only in this depth was the effect of particle S0 size noticeable. The data i ndicated that there was little movement of 35s down the profile. In field conditions more extractable 35s was measured at lower soi l depths of Ramiha soi l as compared to the Tokomaru soi l as evident by significant difference between the amounts of extractable 35s in the SS and LL treatment in the lower layer of the Ramiha field soil. This probably reflects a greater leaching of sulphate S and a greater amount of adsorbed sulphate (CaP extractable) in the lower depths of the Ramiha soil (Appendix 7.8 compared to the Tokomaru soil , Appendix 7.9) . As finer particle sizes of S0, whether granulated or not, increased the 35s activity in the extractable S pool so did they maintain higher extractable S levels (Appendix 7.8 and 7.9) in the upper part (top 0-3 cm) of the glasshouse and field Ramiha and Tokomaru soils. This was particular noticeable in the earl ier harvests. Extractable S values in the 3-6 (middle) and 6-1 0 (bottom) cm soil depths showed no effect of S ferti lizer form although i n the bottom depths of the glasshouse cores extractable S concentrations tended to increase at final harvest. At no t ime during the experiment did the top soi l extractable S values decrease to a level which wou ld be expected to l imit plant growth ( i .e . below 1 0 mg S kg?1 , S inclair and Saunders, 1 984). In general, the Ramiha soi l had larger amounts of extactable 35s than the Tokomaru soil in both glasshouse and field trials. This was consistent with the cumulative 35s uptake, %SDFF and the percentage of oxidized 35s0 taken up by pasture , as discussed Section 7.4.2.2. above. These data reflected greater S oxidation rate in the Ramiha soils which resulted from greater soil moisture and soil o rganic matter as mentioned above. The Ramiha soil fertil ity status (Table 7 . 1 and 7 .2) , i n genera l , was h igher than the Tokomaru soil ; i .e . h igher extractable P and S contents. This indicated that S oxidation was favoured by higher soil fert i l ity status which provide more essential nutrients for the S oxidizing micro-organisms allowing their population to increase. The data on extractable 35s activity (CaP-S) in the soi l cores revealed some effects of granulat ion and phosphate rock (Figure 7. 1 2 and Appendix 7.7) . In the g lasshouse trial , in itially, up to 60 days, granulation of S0 (SS/gr, SS/PR) produced sl ightly less extractable 35s but thereafter had less effect on the activity of 35s In CaP extracts particularly In field soi ls (F igure 7.1 2, Table 7.1 28 and Appendix 7.7) . 205 Addition of ground phosphate rock at granulation did increase the release of extractable 35s from granules (SS/PR cf . SS/gr) in the soil cores (0-1 0 cm) , throughout the trial period in both the glasshouse and field trials (Figure 7 . 1 2 Table 7 . 1 28 and Appendix 7.7). Data on the amounts of 35so (acetone extract) remaining in the top layer of soil cores were presented in Table 7 . 13A. These data did not show that phosphate rock increased S0 oxidation as d id the extractable 35s data as mentioned above (see Table 7. 1 28) . An explanation for i ncreased Ca-35s in the presence of PR when extractable S (Table 7. 1 38 and Appendix 7.8 and 7.9) did not increase with time is that addition of PR stimulated remineralizat ion of soi l organic 35s. The accumu lation o r reductio n of organic 35s (as d iscussed below) , however, does not indicate this trend. 7. 4.2.4 Immobilization into organic matter There was no significant effect of S0 particle size on the immobi l ization of 35so into soi l organic S (Table 7.1 48) and by 1 80 days 1 9-55% of the added 35s0 had been immobilized in the glasshouse Rami ha soi l and 30-5 1% in the glasshouse Tokomaru soi l . I n field soils, the extent of immob i l izat ion was greater be ing 5 1 -55% in Ramiha soi l and 70-76% in the Tokomaru soil . The majority of the immobilization appeared to occur during the first 30 days after ferti l ization. There are, however, a number of discrepancies between the activity of 35s in organic S and that remaining as acetone extractable S0 ? For example, it is impossible to incorporate 35% and 38% of the 35s0 label (LL treatment) into organic s by 30 days in Tokomaru and Ramiha so i l , respect ively when only 1 0% and 1 7% of the 35so cou ld be accounted for as acetone extractable 35so (Tab le 7 . 1 2A) . Obvious ly , sampl ing and analysis e rror involved i n determini ng activit ies of 35s in acetone extractable s0 and total 35s (NaHco3 + Ag20 digestion) compound to produce large e rror in estimating the activity of 35s in o rganic S. Bearing these errors in mind, little can be said about the rate of 35s incorporation into organic S or the effect of S particle size of granulation on the rate of incorporation. There was no significant effect of particle size, granulation with and without phosphate rock on the immobi l izat ion of 35s0 into o rgan ic S (Table 7 . 1 48) and this may also be due to inaccuracies in the determination of 35s0 in acetone extraction and total 35s (NaHC03+Ag2o digestion) in soi l samples (Table 7 . 14A) . The greatest single fate of 35s0 in this experiment was immobilization into soi l organic S. 206 7. 4.2.5 Movement ot35s down the profile and leaching losses As d iscussed earlier in Section 7.4.2. 1 , 75-1 00% of the 35s remained in the soil cores and was not taken up by plants or leached. The vertical d istribution of this total 35s in the Ramiha (Figure 7 . 13) and Tokomaru (Figure 7 . 14) glasshouse soils after 1 80 days showed that little of the 35s moved beyond the upper 0-3 cm of the undisturbed soil columns. These results and others for other harvest dates and fert i l izer treatments are presented in Appendix 7 . 1 0 . Notably, in the field cores there was greater movement of S isotope t o depth. The main mechanisms that would move isotope to depth are e ither leaching of 35so4=, translocation of 35s down roots or through the action of earthworms. There was evidence of earthworms casting activity in all soil cores. A lack of S0 oxidation would prevent movement by these mechanisms unless earthworms ingested small particle of unoxidized 35so. I n the field trials leaching losses during the 1 80 days of the experiments could be est imated from the amount of total 35s not recovered (soi l plus cumulative pasture uptake) from the undisturbed soi l cores (0-1 0 cm) and data are presented in Tables 7.1 0 and 7. 1 1 . Results i ndicated that in spite of larger pastu re uptake, s l ightly larger amounts of 35so labelled fe rt i l izers were u naccounted for on the Ramiha so i l in re lat ion to Tokomaru so i l and approximately 1 0-1 6% of 35s appeared to move beyond the 1 0 cm depth. Despite the greater overall loss of isotope from the Ramiha soil there was more movement of 35s to the middle and bottom depths of field cores in the Tokomaru soil (Figures 7. 1 4, 7. 1 6 and Appendix 7 . 1 0) . A smal l f ract ion o f 35s lost from t h e Ramiha soi l cores could have been lost lateral ly, particularly if some of the heavier rainfall had caused particle run-off. By 1 80 days , in the g lasshouse tria l , about 5 and 1 % and 1 5 and 5% (mean of seven replications) of the total 35s applied were found in the middle and bottom layer of the Ramiha and Tokomaru soils, respect ively. No significant difference was found among the total 35s within the lower soil layers of the soil cores treated with three particle sizes of the 35s labelled so. Smaller amounts of total 35s in the top layer of the SS (<0 . 150 mm particle size) treated soil cores during the 90 and 1 80 sampling times resulted from greater pasture uptake rathe r than movement down the profiles. In the field soils, at 1 80 days, a similar trend to the glasshouse trials was observed. The total 35s recovered from the <0.1 50 mm S0 (SS) treated top layers In both soils were significantly less than those treated with the 0 .250-0.500 mm S0 (LL) . No significant differences between 207 the amounts of total soi l 3Ss were found in the middle and the bottom layers treated with different s0 particle sizes. At 1 80 days larger amounts of total soil 35s appeared in the lower depths of the field cores in relation to glasshouse data. In the field trials more than 25% of labelled fertilizers moved beyond the top layer (0-3 cm) where it was in itially placed. Only data from the Ramiha soi l showed any effect of particle size in the lower layers (see Appendix 7. 1 0) . Granulation and addition of phosphate rock did not show any effect on the movement of 35so labelled fertilizer as evident by similar amounts of 35s activity in the lower layers in both soils in the glasshouse and field trials (Figures 7. 15 , 7.1 6 and Appendix 7 . 1 0) . Analysis of the leachates from the glasshouse cores showed that no sign ificant differences were detected in the amounts of HI -reducible 35s in leachates from soi l cores treated with different particle sizes and granules of 35s labelled S0 (Table 7.1 5A). Total amounts of 35so labelled fertilizer in leachates in each event were considered negligible amounting to less than one percent of the appl ied 35s , even though during the leachi ng period up to 56 days, approximately 65% and 29% of the SS particle size had oxidized in the Ramiha and Tokomaru soils respectively (Table 7 . 12) . The small amount of H I-reducible 35s present in the leachates is mostly attributable to rapid immobil ization of 35s into non-leachable organic S and the sulphate adsorption capacity of both soils (see Appendix 7 . 14) since little 35s was taken up by plants. The retention of soil sulphate against leaching loss due to rapid incorporation of ferti l izer 35s into soil organic S was also reported by Gregg (1 976) , Gregg and Goh ( 1 978 and 1 979) and Goh and Gregg ( 1 982a) . The amounts of H I-reducible S in leachates were also measured and the results are presented in Table 7 . 1 58. No significant effect of S0 particle size was observed due to large variation in the observed data. The concentration of H I-reducible S in leachates from each leaching event was less than one J.l.g mr1 (data not shown). Similar concentrations of S were also observed in l eachates collected from a mole-drained field plots of Tokomaru soi l (Smith , 1 979; Heng, 1 99 1 ) . Total amount of H I-reducible S in leachates from five leaching event were about 0 .3- 0 .8 mg S core-1 which is equ ivalent to 0 .1 7-0.45 kg S ha-1 . 35 25 CaP- s 28 ill ill CA ) 15 :?:?:?: ? ? : :: : : :: /. 18 I /C 11 5 j?f[ ? - ?fi? 8 ? - 35 25 CB ) CaP- s 28 15 m /. ? Hl I 5 ? 1 )j1j?l 1;?;?;?;1 I ''Nol 1:::::::1 r:::::::l ::::: : : : :;::: : 8 . . . 38 68 98 188 188 G lasshouse Days F ie ld Figure 7.1 1 The effect of S0 particle size on the recovery of extractable 35s (CaP-35s) in soil cores (0-10 cm) in Ramiha (A) and Tokomaru (B) soils. D ss ? LL lli ss lj MM ? LL N 0 OJ 35 CaP- S :1. 35 CaP- S :1. Figure 7.12 3B (A ) 2B 1B B I wml/3 ? ? ? '110000 I 3B (B ) 2B 1B B I m.JIN)I .. . SJWIIS I ? 3B 6B 9B G lasshouse 18B Da?s 18B F ie l d The effect of granulation of S0 with or without phosphate rock on the recovery of extractable 35s (CaPJ5s) in soil cores (0- 10 cm) (SS/PR and SS/gr compared with SS) in Ramiha (A) and Tokomaru (B) soils. fm ss [] SS/PR 1mJ SS/gr ? ss E8 SS/PR ? SS/gr 1'.) 0 tO 210 Table 7.12 Percentage of applied 35s recovered as residual 35so (acetone extracts), in 0- 3 cm soil depth (PART A) and extractable 35s (CaP-S) (PART B) in soil cores (0- 10 cm) in glasshouse trials and after 1 80 days in the field trials (average of three replications). Days after fertilizer application Treatments 30 60 90 1801 PART A, RESIDUAL ELEMENTAL 35SO Ramiha ss 58.7 % 35so recovered bacetone extract) 45.4b 42.3 21 .4b LL 82.7 94.5a 84.6a 75.6a SS/PR 63.4 52.5b 52. 1b 25.5b SS/gr 66.9 53.8b 63.4b 21 .2b F-test ns ** * ** C.V. % 18.5 10.4 8.8 30.8 Tokomaru ss 64.8 61 .8 50.6b 30. 1b MM 86.5 70.0 61 .9a 69.8a LL 90.4 75.9 79.6a 69.9a SS/PR 69.5 66.3 65.5b 35.6b SS/gr 68. 1 75.6 43.9b 34.3b F-test ns ns ** ** C.V. % 20.1 16.8 19.7 19.1 PART B. CaP-S EXIRACTABLE 35s 35 .... % S recovered as CaP-S .... Ramiha ss 21 .4a 12.oa 13.9a 18.5b LL uc 1.6b 1 .9c 3 .0c SS/PR 12.4b 12.8a 15.3a 25.5a SS/gr 8.4ab 1 1 .6a 5.9b 15.sb F-test ** ** *** *** C.V. % 44.8 24.9 24.6 27.7 Tokomaru ss 10.3a 9.oa 13.6a 12. 1a MM 1 .2b 3 .2b 2.4d 2.4c LL 1.2b ub I . 3d 0.8c SS/PR 14.5a 1 l .la 10.2b 12.7a SS/gr 10.7a 2.8b 6.? 8.7b F-test ** * ** *** C.V. % 50.2 46.5 27.2 34.3 Field 180 21 .7b 63.2a 18.7b 22.3b ** 17.7 31 .4c 49.9b 64.3a 29.4c 28.3c * 13.3 10.9a 2.3b 12.9a 12.7a *** 27.8 6.4b 1.3c 0.9c 9.4a 6.4b *** 27.2 ?, ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; ns = not significant; mean separation by DMRT at 5% level denoted by letters 1 average of seven replications 21 1 Table 7.13 Percentage of applied S0 recovered as residual S0 in 0-3 cm layers (PART A) and total amounts of extractable S, CaP-S, (PART B) in soil cores (0-10 cm) in glasshouse trials and after 180 days in the field trials (average of three replications). Treatments 30 PART A, ELEMENTAL S RECOVERED Ramiha ss 61 .7 LL 93.2 SS/PR 70.6 SS/gr 80.0 F-test ns C.V. % 15.1 Tokomaru ss 79. 1 MM 100.0 LL 102.3 SS/PR 87.7 SS/gr 72.6 F-test ns C.V. % 13.5 PART B, EXTRACTABLE S (CaP-S) Ramiha ss 48.1 LL 37.2 SS/PR 50.8 SS/gr 46.9 PR 39.7 Ctrl 46.9 F-test ns C.V. % 14.8 Tokomaru ss 32.8 MM 29.2 LL 27.6 SS/PR 40.3 SS/gr 48.9 PR 30.0 Ctrl 23.9 F-test ns C.V. % 30.2 Days after fertrilizer application 60 90 1801 .... . %S0 recovery .. . . . 57.5b 54.2c 22.5b 1 15.1a 100.8a 48.7a 66.4b 64.2b 18. 1c 81 .3b 79.1b 22.5b * ** ** 16.9 9.3 3 1 .7 68.9 55.5 35.5b 76.6 68.3 58.4a 80.6 8 1 . 1 77.7a 68.7 63.4 40.4b 81 .3 48.7 41 .7b ns ns ** 15.2 19.3 18.0 s - 1 ... . . mg core .. . . . 58.3bc 52.8a 65.5ab 52.0c 36.8ab 53.5c 68.5a 47.3ab 75.2a 57.7bc 443ab 68. 1ab 61 .6ab 35.ob 57.7c 54.2c 46.sab 36.sd * ** ** 7.0 18.8 16.8 32.8 43.4a 39.8ab 36.2 25. 1bc 33.? 25.8 22.sbc 32.9c 37.8 25.3bc 43.3a 32.3 3 1 .4b 41 .2a 29.6 23.5bc 34.sc 25.8 18.9c 35.1ab ns ** ** 17.2 18.7 12.5 Field 180 20.7b 41 .5a 15.8b I5.ob ** 26.3 38.5b 54.8b 73.2a 32.7b 34.3b ** 23.8 50.3ab 41 .6c 54. la 45.sbc 44.2bc 40.8c ** 7.3 32.oab 24.9bc 26.sbc 35.1a 33.5a 30.9ab 33.3a ** 9.9 ?, ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; ns ? not significant; mean separation by DMRT at 5% level denoted by letters; 1 average of seven replications Table 7.14 Treatments 21 2 Percentage recovery of total 35s (PART A) and organic 35s (PART B) in soil cores (0-10 cm) at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications). Days after fertilizer application Field 30 60 90 1801 180 PART A, TOTAL 35s .. . . . % recovered . .. . . Ramiha ss 1 14.9 91 .6 91 .3 75.5 74.5 LL 13 1 .9 1 19.8 109.9 97.9 87.7 SS/PR 104.3 101 .9 98.4 99.2 8 1 .3 SS/gr 98.6 88.3 87.6 80.2 72.1 F-test ns ns ns ns ns C.V. % 14.4 15.2 12.3 1 1 .2 1 1 .4 Tokomaru ss 1 15.1 100.9 88.4 92.9 95.0 MM 136.9 132. 1 1 13.7 101.8 100.4 LL 126.7 122.6 141.3 104.4 105.5 SS/PR 1 16.3 1 1 1 .4 109.4 99.9 101.4 SS/gr 120.6 107.5 86.3 87.3 91 .0 F-test ns ns ns ns ns C.V. % 13.7 15.2 12.7 13.7 20.7 PART B, ORGANIC 35sa .. . . . % recovered .. . . . Ramiha ss 34.8 34.2 35.1 37.4 51 .9 LL 48.2 23.7 23.9 18.9 50.6 SS/PR 28.6 36.5 3 1 .1 52.5 56. 1 SS/gr 23.2 23.3 17.8 34.3 55.0 F-test ns ns ns ns ns C.V. % 40. 1 68.8 43.8 56.4 13.5 Tokomaru ss 40.1 30.2 24.2 50.8 70.5 MM 49.2 58.9 48.6 29.5 72.4 LL 35.1 45.6 60. 1 33.7 68.9 SS/PR 32.3 33.8 37.5 50.5 76.3 SS/gr 41 .2 29.3 32. 1 43.5 69.9 F-test ns ns ns ns ns C.V. % 52.1 39.4 50.0 21 . 1 14.8 ns = not significant; 1 average of seven replications; a calculated as: Soil organic 35S=(totai35S) - (residual35sD) - (CaP-35s), as described in Section 3.3.5 Figure 7. 1 3 G l asshouse top Layers m i dd le Bottom F ie ld Layers m idd l e bottom 0 ss /. recovery 50 100 l i l l i l l l i /. reco very 0 0 213 LL /. recovery 50 100 /. recovery 0 50 100 I , Total recoveries of 35s in three layers of Ramiha soil 180 days after fert ilization with two particle sizes of 35so. Figure 7 . 14 214 ss MM LL G lasshouse /. recovery /. recovery 40 80 40 80 1 1 ! 1 1 1 11 1 1 1 1 1 1 1 1 1 1 I I Dot tom F i e ld /. recovery /. recovery /. recovery bottom Total recoveries o f 3Ss in three so i l layers of Tokomaru soil s 1 80 days after fertil ization with three particle sizes of 3 5so. Figure 7.15 G l asshouse SS/PR /. recovery 0 50 100 Layers top 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 m idd le Dot?tom F i e l d top Layers m i dd l e bottom /. recovery 0 50 100 215 SS/gr % recovery 0 50 100 1 1 1 ! 1 1 1 1 1 ! 1 1 1 ! 1 /. recovery 0 50 100 Total recoveries of 35s in three soil layers of Ramiha soil 180 days afler fertilization with fine 35s0 (SS) granulated with and without phosphate rock. Figure 7 . 1 6 G l asshouse top Layers m idd l e llottom F i e l d top Layers m idd l e bottom SS/PR :1. recovery 8 58 1?8 : : % recovery 8 58 188 216 SS/GR :1. recovery 8 58 188 1 1 1 1 1 1 1 ! 1 1 1 1 1 1 % recovery 8 58 Total recoveries of 3 5s in three so i l layers of Tokoqwru so i l 1 80 days afler fertilization with fine 3Sso (SS) granulated with and without phosphate rock. 217 Table 7.15 Percentage recovery of 35 S in leachates (PART A) and total amounts of sulphur and HI-reducible S (PART B) in leachates at five leaching events in glasshouse trials. Treatments 7 14 PART A, 35s RECOVERY Rami ha ss 0.0014 0.026 LL 0.0003 0.001 SS/PR 0.0047 0.019 SS/gr 0.0003 0.001 F-test ns ns C.V. % 160.6 228.6 Tokomaru ss 0.0120 0.006 MM 0.0076 0.007 LL 0.0032 0.002 SS/PR 0.0094 0.031 SS/gr 0.0120 0.014 F-test ns ns C.V. % 102.6 123.6 PART B, HI-reducible S Ramiha ss 94.0 53.0 LL 62.9 45.0 SS/PR 95.6 86.6 SS/gr 67.7 64.3 PR 60.6 63.2 Ctrl 60.2 5 1 .2 F-test ns ns C.V. % 34. 1 30.6 Tokomaru ss 90.5 87.9 MM 91.4 90.5 LL 64.4 82.8 SS/PR 67.7 91 .0 SS/gr 69.3 62.2 PR 1 17.8 142.9 Ctrl 76.4 162.3 F-test ns ns C.V. % 120.1 209.0 ns = not significant at 5% level Days after fertilizer application 21 44 ... . . . % recovery ...... 0.017 0.064 0.001 0.001 0.007 0. 1 12 0.003 0.003 ns ns 247.7 203.6 0.009 0.074 0.008 0.038 0.006 0.001 0.026 0.012 0.013 0.073 ns ns 87.6 409.9 s w-3 - 1 . . . mg event . . . . 49.8 181 .5 73.6 100.5 75.9 238.5 52.0 123.3 77.9 49.3 77.5 88.3 ns ns 32.5 120.9 92.9 204.0 127.6 216.9 108.5 80.0 108.6 147.0 51 .2 190.3 93.8 58.9 77.9 66.0 ns ns 45. 1 54.9 56 Total 0.163 0.272 0.015 0.018 0.1 10 0.254 0.059 0.058 ns ns 101 .8 82.1 0.333 0.434 0.016 0.206 0.009 0.019 0.749 0.155 0.1 19 0.23 1 ns ns 140.3 3 10.1 Total 90.7 469.4 32.5 3 14.6 102. 1 598.8 74.9 380.9 72.2 324.3 65.8 342.9 ns ns 138.5 120.1 409.9 885.3 52.4 578.4 53.6 389.4 79. 1 502.2 121.7 494.7 34.4 447.9 19.0 403.4 ns ns 138.6 40.1 218 7.4.2. 6 Summary Despite the fact that for much of the experimental period plant S concentrations in field and glasshouse experiments were well below optimum level , only glasshouse soils ferti l ized with the finer S0 (SS) particle size consistently had higher S uptake , but this was not translated into increased dry matter production. However , N and P deficiency may have limited pasture growth. There was a lways a large influence of available soil-S or organic S mineralization on the amounts of plant S taken up and the amount of CaP extractable soil S, therefore the non radioactive data show little or no effect of added S0 fertilizers on plant S uptake and soil S status as measured by CaP. Measurement of the fate of 35s derived from ferti l izer showed clearly that the extent of S oxidation, plant uptake of S and the percent of oxidized S0 taken up by plants were always higher with the finest particle size of S0. Except for the study of Lee et al., ( 1 987) previous experiments in New Zealand soils have only inferred these fertilizer effects from differences in plant S uptake or levels of extractable SO 4 = in S fertilized vs. non-fertilized soils (Boswell and Swanney, 1 988 ; Smith and McDougal l , 1 988; Swanney et al. , 1 988; Boswell and Swanney, 1 99 1 ) . Recently published work by Nguyen and Goh (1 991 ) has also shown similar effects. Notably the percent of oxidized S taken up by plants was greater in g lasshouse soil than field soils where evapotranspiration rate were higher. This result suggests that available S is used more efficiently when water used by plant is high. Such evidence agrees with the concept that sulphate uptake by plants occur mostly by mass flow of soil water to roots (Barber, 1 984) . The finer particle size of s0 produced greater amounts of soil extractable 35s (CaP-S) and the majority of this fraction remained in the top layer (0-3 cm) . The amounts of 35s leached were negligible and the majority of fertilizer S remained in the top soi l layer (0-3 cm) , more than 50- 90%, mostly present in organic form. No significant effects of particle size or granulation of the fine S0 with and without phosphate rock on the amounts of 35s leached or immobilized into organic S were observed. I nitially, 0- 60 days , the amounts of fert i l ize r S0 and 35s leached beyond 1 0 cm soi l depth were negligible . The failure to detect differences in the rate of transformation of 35s into o rganic S caused by ferti l izer form were as much due to the inaccuracy of total 35s determination (NaHC03 + Ag20 digestion) and total 35s0 (acetone extraction) when particulate s0 was present in small soil samples. 219 I n general , granulation of fine s0 with and without phosphate rock slightly reduced the efficient use of the oxidized s0 by plants. But this effect was not consistent in both soils and in the Tokomaru soil, addition of phosphate rock (PR) appeared to give improved uptake of oxidized S compared to s0 granulated alone. Data on the percent of plant S derived from fertilizer (%SDFF) and the percentage of oxidized 35so cumulatively taken up by plants, indicated that the efficiency of S0 in providing S for pasture plant was h igher in the Ramiha soil which had a h igher soil ferti l ity status than the Tokomaru soil . Data on acetone extraction for residual S0 and 35s0 activity revealed similar trends of S0 oxidation rate. However, accuracy in determination of the residual S0 and total S in the S0 treated soil samples needs further improvement. 7.4.3 FATE OF SULPHATE-BASED FERTILIZERS This sub-experiment was carried out for the first 90 days of the glasshouse exper iment described above. Previously, in Chapter 5, the fate of sulphate-based (SSP) and microfine S0 fert i l izers has been investigated under field conditions in the undisturbed soil cores i n the Tokomaru so i l . In t his exper iment two su lphate-based fert i l i zers were invest igated in glasshouse conditions. The main objective was to compare the fate of 35s labelled gypsum and superphosphate in soils contrasting in organic matter content, S content, and P retention with that of S0. 7.4.2. 1 Recovery of 35 S in soils and plants The cumulative uptake of 35s radioactivity by plants and that remaining in the top 1 0 cm soil in the glasshouse after 90 days are shown in Table 7.1 6 for both Ramiha and Tokomaru soils. Cumulative plant uptake of 35s under glasshouse conditions accounted for approximately 20% of the isotope which was greater than that of S0 fertilized soils, 6-1 2%, during the same period (see Appendix 7 .4) . The majority (67-80%) remained in the soi l . The amount of isotope leached beyond the 1 0 cm soil cores was about 4% for both soils (Table 7. 1 9) being more than double the percentage of 35s0 that was leached (Table 7 .1 5). 7.4.3.2 Plant uptake of S and 35s and percentage of plant sulphur derived from fertilizers (%SDFF) 220 There were no significant d ifferences in S concentration and cumulative S uptake (so il plus ferti l izer S) of pastu res between pastures grown on soil treated with GP and SSP (Figure 7. 1 7B, Appendix 7.1 1 ) . I n terms of S concentration and cumulative S uptake, pasture grown on GP and SSP responded ve ry wel l to the appl ication of these two ferti l izers . Both S concentration and cumulative S uptake were significantly h igher than the control soil cores. The increased S uptake only s lightly increased dry matter yield of pasture grown on the Ramiha soil but significantly increased the dry matter yield in the Tokomaru soil (Figure 7. 1 7 A and Appendix 7.1 1 ) . The lack of dry matter yield response in the Ramiha soils might be due to larger phosphate extractable S in soil in relation to the Tokomaru soil (Table 7.2 and Appendix 7. 1 3) Pasture treated with 35s labelled GP and SSP cumulatively took up the same amounts of 35s on both soils and also showed a s imilar trend i n the %SDFF ( Figu res 7 . 1 8A, 7 . 1 8B and Appendix 7 .1 2) . However, cumulative 35s uptake from soils treated with 35s labelled GP and SSP were much higher than those treated with the <0. 1 50 mm elemental 35so (SS) . In Ramiha soil, initially both 35s labelled GP and SSP treated cores had higher %SD FF (34- 50%) than the <0.1 50 mm (SS) treated soil cores (24-28%) , as presented in Figure 7. 1 8B and Appendix 7.1 2 . But thereafter comparable %SDFF values were observed. Notably, the %SDFF was not always a good indicator for a comparison on the relative efficiencies of use between the two sou rces, sulphate and S0 ( i .e. GP or SSP vs. finer particle of S0, SS). This contrasts to the discussion in section 7 .4.2.2, since the GP and SSP fertilized cores yielded larger amounts of 35s uptake. The greater S uptake in the GP and SSP probably i ndicated that environments in the GP and SSP soil cores were more favourable for S uptake than that of SS soil cores. The SS treatment stil l provided less S for plant demand than the GP and SSP treatments. Therefore both %SDFF and the amount of 35s uptake should be considered together in comparison of two sources of fertilizer S (see also the discussion in section 7.4.5 and 7.4.2.2). 7.4.3.3 Immobilization into organic matter The percentage of applied 35so4= converted into soil organic 35s in the soil cores (0- 1 0 cm) treated with 35s labelled gypsum and superphosphate were similar (Tables 7 . 1 7 and 7. 1 8) . However, larger amounts o f organic 35s occurred i n the Tokomaru soil (49-53%) than I n the 221 Raimha soil (39-40%) . In comparing these resu lts to those of previous experiment in Chapter 5 (see Tables 5.6 and 5.8), extents of applied 35s incorporation into organic S were similar and occurred at similar rates. The proportion of soil organic 35s (value expressed as percentage of total S remaining in the soi l cores) in this experiment were less than those reported in Chapter 5 (Table 5.8). I n Chapte r 5 , d u ring t he same per iod (90 days) about 90% of 35s from 35s labe l led superphosphate remained as organic S whi le in this experiment only 52 and 60% were observed. The reason may be attributed to the lower leaching losses in the g lasshouse experiment. Even after five leaching events, which were approximately equivalent to 0.7 pore volume , about 35 to 44% of the 35s was accounted for in the phosphate extractable 35s fractions mostly in the top soi l indicating that much of the drainage water must bypass the soil pores containing the majority of the 35so 4 = or that significant quantities of surface adsorbed organic 35s are extracted by calcium phosphate (Watkinson and Perrott, 1 990; Watkinson et al., 1 99 1 ) . As shown in Table 7.1 8 when the amounts of organic 35s were expressed as the percentage of the total 35s remaining in the soil cores, the majority of total 35s in the middle and bottom layer were in organic form (about 70 to 80%). Despite the large experimental errors involved in the comparison , it appeared that at 90 days more 35s had been converted into organic S in the soil fertilized with gypsum and SSP (Table 7. 1 7) than in the same soil fertilized with 35s0 (Table 7.1 4). Over this period (0-90 days) it is expected that the rate of 35s immobilization wou ld be limited by the extent of S oxidation, however, because of the large errors involved in measuring total 35s and residual 35so in the presence of particu late 35s this relationship is not always clearly shown. For example, at 30 days in LL treated Tokomaru soi l the percentage of 35s in acetone extractable 35s (Table 7.1 2A) was 90.4% and organic 35s (Table 7.1 48) was 35.1% g iving a total of 1 40% whereas for all SS treatments there appeared to be a better mass balance of isotope. 7.4.3.4 Movement ot35s down the profile and leaching losses Vertical d istribut ion of soi l 35s fract ions i n each soi l layer , presented i n Tab le 7 . 1 8 , demonstrate the movement of 35s released from labelled gypsum and superphosphate i n the intact soil cores. Data presented were the results of soil samples taken after 90 days which was about 40 days after the last leaching events (Table 7. 1 9) were applied. Results Indicated that both treatments exhibit similar 35s movement in soil cores. About 50, 1 0 and 5% of the 222 35s re leased form labelled ferti l izer were present in the top , m iddle and bottom layers, respectively and the majority were in organic forms. Five leaching events were also applied to both soil cores as described in Section 7.3.6. No significant difference occurred between the amounts of H I-reducible 35s in the leachates from the two soils (Table 7. 1 9) . About 0.4 to 1 .6% of 35s labelled fertilizer was recorded in each event during the leaching. On average, the total amounts of the H I-reducible 35s leached beyond the 1 0 cm depth was about 4% of total applied 35s labelled fertilizers. This represents a larger leaching loss of fertilizer S than in the S0 treatments discussed in Section 7.4.2.5. As mentioned in Section 7.4.3.3 above , the majority of 35s in the middle and bottom layer were i ncorporated i nto organic forms by microbial processes (F reney et al. , 1 97 1 , 1 975; Gregg, 1 976) . Again, this process along with the sulphate retention capacity of the soils (see Appendix 7.1 4) will play a role in reducing leaching losses of the 35s0 labelled fertilizers in the leaching events. The amounts of H I-reducible S in the leachates were also measured (Table 7 .1 98). There was no significant difference in the amount leached from GP or SSP treated soil but there were large variations in the data for each event. Much larger amounts of H I-reducible S, equivalent to 0.9-1 .0 kg ha- 1 , were leached from the GP and SSP treated cores when compared to control and the S0 treated cores (particularly the <0.1 50 mm, SS cores, Section 7 .4 .5.2) . 7.4.3.5 Summary Both GP and SSP increased extractable soil S (CaP-S) and raised pasture S concentration above the 'critical level' and in turn generated dry matter yield i ncreases for pasture grown on Tokomaru soi l with a low level of extractable soil S (Tokomaru soil) The efficiency of plant use of S from GP and SSP were similar. In these two soil materials there was no significant effect of the phosphate in SSP on the amount of S taken up by plants, leached down soil profile or immobilized into organic S. The majority of radioisotope (total S , extractable S and organic S) remained in the top 3 cm soil layer with the majority being converted to organic S in 90 days experiment. Further experiments on the fate and residual value of fertilizer S should concentrate on this top layer rat he r than i nc lus ion of so i l at deepe r depths , because the observed rates of 35s immobilization indicated high biological turnover of nutrients in this zone. There may, however, 223 be strong interaction between the amount of fertilizer S leached and soil status because in contrast to the resu lt in the field trial on higher S status, Tokomaru soil (Chapter 5) , less radioactivity was apparently leached. This agrees with the findings of Saggar et al. ( 1 990a, 1 990b) who reported greater S leaching loss from soil with h igh S status . 7.4.4 Comparison of sulphate-based fertil izers and elemental S To facil itate comparisons of the efficiency of use of sulphate and S0 fertilizers ( Figu re 7. 1 9) , simple relationships between S and 35s taken up by pastu res and time were constructed using the combined means of the GP and SSP treatments and the individual means of the SS and control treatments during the first 90 days of the glasshouse experiment. Plant uptake of S and 35s from su lphate-based ferti l izer (GP and SSP) was more readily avai lable to pastu re plants than the S in the finer S0 particles (SS) . Plants utilized greater amounts of ferti l izer S in the GP and SSP ferti l ized cores than in the SS fert i lized cores. However, the utilization of oxidized 35so was similar to the soluble 35so4= from GP and SSP in the Ramiha soil, but on average lower in the Tokomaru soil . The resu lts in this Chapter indicated a superiority of GP and SSP over S0 in providing ferti lizer S for pasture plants in the short term. To be as effective as sulphate S fertilizer in providing short term available S for pastures, the particle size of S0 must be as small as microfine S0 , discussed in Chapter 5. In Chapter 5, the microfine S0 was as effective as SSP. One problem with the microfine S0 is that apparent losses from the plant-soil system were approximately 45% (Chapter 5 ) . I n this Chapter where losses of 35s were small there was no evidence to support the finding reported in Chapter 5, that there was greater immobilization of 35so into soi l organic matter than 35s from SSP. Such results may vary with soil S status and climate and S particle size. Daily uptake of S by pasture (as indicated by slope of regression lines in Figure 7 . 1 9) in this experiment was much lower when compared with the results discussed in Chapter 5 ( Figure 5.2 and Table 5 .5 compared with Figure 7 . 1 9 and Appendix 7 .1 1 ) . The mean daily S uptake for pasture fertilized with suphate-based (GP and SSP) , S0 (SS) fertilizers and Ctrl were about 0 . 1 8 , 0 . 1 and 0 .09 kg ha- 1 day-1 , respectively, (compared with f ie ld resu lts of SSP and microfine S0 t reated cores in Chapter 5 which were about 0 .25 and 0 .28 kg ha- 1 day- 1 , respectively) . Pasture S concentrations on GP and SSP treated cores were similar i n both of these experiments ( in Chapter 5 and 7) . Dry matter production and the absolute amounts of S taken up were influenced by the general fertility of each site. The second s ite, although the same soil type, had a poor fertilizer history reflected in the total soi l S content (395 mg S kg?1 soil, Table 7.2) compared to the fertilized site in Chapter 5 (540 mg S kg-1 soil , Table 5 .3) . I n 224 these legume-based pastures regular fertilizer history leads to increase N fixation. Nitrogen availability is a primary factor limiting pasture yield and probably l imited pasture growth. For example, the percent of plant S derived from S0 in the higher fertil ity site (Chapter 5, Appendix 5.2) was at least 50% lower than observed at the lower fertility site ( Figure 7.8). These results will reflect the faster oxidation rate of microfine S0 and greater leaching loss of S observed at the first site and the lower availability of so il S at the second s ite. Similar d ifferences were observed for SSP and GP but the comparison between field results in Chapter 5 (Appendix 5.2) and the glasshouse results in Chapter 7 ( Figure 7. 1 88) are less valid because of the much higher water use of the glasshouse grown plants. lt is important therefore to consider the fertility status of soil and climate when attempting to extrapolate results pertaining to the efficiency of fertilizer use to other sites and situations even on the same soil type as shown by Gregg (1 976) . 7.4.5 Comparison of g lasshouse and field trials Cumulative S and 35s uptake, dry matter yield, S concentration of pasture plants, percentage of oxidized S0 taken up by plants and CaP-35s were higher in the glasshouse trial. Only the %SDFF's from S0 fertil izers were similar. The difference in plant growth conditions in the g lasshouse and field produced some large differences in some measurements of plant growth and S uptake (Chapter 7 only) but not the %SDFF. Such data suggest that the climate d ifferences (temperature and soi l moisture content) between g lasshouse and f ie ld so i l cores had paral le l effects on S0 oxidation, mineralization of soil organic S. and plant S uptake as the ratio at which soil S and 35so were made available stayed constant. This effect needs investigating further because they indicate that glasshouse studies using undisturbed soil cores may be relevant to field conditions when a fertilizer with biologically controlled release rate (S0 oxidation) is being evaluated. Cumulative dry matter yield -2 g m S uptake -2 g m Figure 7.17 Rami ha Tokomaru 400 T :X (A1. * GP 300 t .../?- ?? + SSP 200 T /? ?? e ss 1 00 I V ? -*-- Ctrl 0 (B) * GP 2 1 .6 l ? __-:i: + SSP 1 .2 T _/ - /? e ss 0.8 0 .4 I ? ?---* ? Ctrl 0 30 60 90 Days 30 60 90 Cumulative dry matter yield (A) and S uptake (B) of pasture on Ramiha and Tokomaru soils; GP and SSP vs. SS. 1\) 1\) ttl 35 50 S uptake 40 30 o/o 20 1 0 0 50 40 %SDFF 30 20 1 0 0 Figure 7.18 Rami ha Tokomaru (A) + GP * SSP ? ? e ss (B) ? ? + GP * SSP ? e ss 30 60 90 30 60 90 Days Cumulative 35s u:f:take (value expressed as percentage of 3 5s applied) by pasture (A) and %SDFF (B); Ss labelled GP and SSP vs. 35so labelled SS. I\) I\) (J) 35 S uptake % S uptake -2 g m 1 y = 0.260X R2 = .96 2 y = 0.224X R2 = .91 3 y = 0. 130X R2 = .98 4 y = 0.01 8X R2 = .98 5 y = 0.01 0X R2 = .95 6 y = 0.009X R2 = .97 25 t 20 1 5 1 ? f 3 0 2 1 .6 4 1 .2 5 0.8 6 0.4 0 0 30 60 90 7 y 8 y 9 y 1 0 y 1 1 y 1 2 y 0 = 0.230x R2 = .94 = 0.100x R2 = .96 = 0.060x R2 = .98 = 0.0 1 9x R2 = .92 = 0.009x R2 = .94 = 0.006x R2 = .99 (A) / 7 * GP-SSP /'' 8 + Oxidized ?9 e ss (B) 1 0 / * GP-SSP C\ 1 1 e ss ? - -1 2 -*- Ctrl 30 60 90 Rami ha Days Tokomaru Figure 7.19 Simple relationsh?s between cumulative 3Ss uptake (A) (value expressed as percentage of 3 S applied) and S uptake (B) by plants against time (a comparison between sulphate-S and elemental S fertilizers). "' "' --I Table 7.16 228 Total recovery of 35s in soil and plant in two soils resulting from applications of 35s labelled gypsum and superphosphate after 90 days . Treatments PLANT SOIL F-test C.V. % F-test C.V. % TOTAL F-test C.V. % ns = not significant GP SSP GP SSP GP SSP Ramiha % 22.6 19.6 ns 10.8 69.7 67.7 ns 14.9 92.3 87.4 ns 1 1 .36 . . . . . . . Soils . . . . . . . Tokomaru % 19.4 22.4 ns 16.7 8 1 .4 80.2 ns 8.2 100.8 102.7 ns 4.3 Table 7.17 Soils 229 Fractions of 35s in soils (% recovery) 90 days after applying 35s labelled gypsum (GP) and superphosphate (SSP). Total Or:!:;anic Extractable 35s s 35s Treatments % % % Rami ha GP 69.7 39.1 (56.1) 30.6 (43 .9) SSP 67.4 40.3 (59.6) 27.4 (40. 1) F-test ns ns ns C .V . % 14.9 19.1 19.5 Tokomaru GP 8 1.4 52.7 (64.7) 28.4 (34.8) SSP 80.2 49.4 (61 .6) 30.8 (38.3) F-test ns ns ns C .V . % 8.2 7.4 8.4 numbers in brackets were proportion expressed as the percentage of total S remained in the soil cores; ns = not significant Table 7.18 Distribution (% recovered) of total 35s (TT), organic 35s (Org) and phosphate extractable 35s (CaP-S) in three soil depths of two soils 90 days after application of 35s labelled gypsum (GP) and superphosphate (SSP). Ramiha Tokomaru Fertilizers Layers TT Org CaP-S TT Org CaP-S . . . % ... . .. % . . . GP Top 55.5 28.4 27.1 62.1 37.9 24.2 Middle 10.3 7.9 2.5 12.8 10. 1 2.7 Botom 4.1 2.7 1 .4 6.5 5.0 1 .4 SSP Top 53.3 30.6 22.6 67.5 38.7 28.9 Middle 9 .1 6.3 2.8 8.0 6.8 1 .9 Bottom 5.3 3.4 1 .9 4.7 3 .9 0.7 F-test Fertilizers ns ns ns ns ns ns Within fertilizer1 10.7 7.2 6.8 5.8 3.2 3.7 C.V. % 3 1 .5 37. 1 47.8 14.5 14.3 25.7 ns = not significant at 5% level; 1 Lsd at 5% level 230 Table 7.19 Percentage recovery of 35s labelled fertilizer sulphur in leachates (PART A) and HI-reducible S in leachates (PART B) at five leaching events after 35s labelled gypsum (GP) and superphosphate (SSP) applications. treatments 7 PART A, 35s RECOVERY Ramiha GP 1.4 SSP 0.6 F-test ns C.V. % 127.5 Tokomaru GP 1 .2 SSP 2.3 F-test ns C.V. % 101.5 PART B, HI-REDUCIBLE S Ramiha GP 0.33 SSP 0.10 Ctrl 0.06 F-test ns C.V. % 143.1 Tokomaru GP 0.33 SSP 0.54 Ctrl 0.08 F-test ns C.V. % 95.9 Days after fertilizer application 14 21 44 35 .... % S recovered ... 1.5 0.8 ns 168.6 0.4 0.6 ns 84.5 0.52 0.30 0.05 ns 129.1 0. 15 0.21 0.02 ns 44.9 0.4 0.8 1 .6 1 .2 ns ns 38.9 88.5 1 .0 0.9 0.6 1 .2 ns ns 127.5 67.2 s -1 . . . mg event . . . 0.14 0.40 0.43 0.60 0.08 0.09 ns ns 165.8 83. 1 0.34 0.50 0.20 0.52 0.08 0.06 ns ns 1 1 1 .8 68.7 56 0.2 0.1 ns 74.8 0.2 0.2 ns 75.9 0.20 0. 15 0.07 ns 75.7 0.27 0. 10 0.02 ns 94. 1 ? = significant at 5% level; ns = not significant at 5% level; mean separation by DMRT at 5% level Total 4.2 4.3 ns 57. 1 3 .6 4.9 ns 45.9 Total 1 .60'l 1 .55a o.3ob * 44.5 1 .60'l 1 .55a 0.40b * 35.5 7.5 CONCLUSIONS 231 The implications of the findings in this Chapter are that the efficiency of plant use of S0 can be improved by decreasing s0 particle size. Unlike the findings in Chapter 5 using higher fertility soi l , S derived from S0 was not conserved to a greater extent in soil organic matter than S derived from SSP. This suggests that the particle size of S0 , climate effects and soil fertil ity status have a large impact on transformations of S in soil . Granulation with or without ground phosphate rock, slightly decreased the oxidation rate or the efficiency of finely ground S0 in supplying S to plants in the short term. Granule breakdown and s0 dispersion wi l l be an important factor to consider when evaluating granular S0 fertilizers. The use of 35s tracer has confirmed that S0 (coarser particle SS, MM , LL) is more suitable as a maintenance ferti l izer for maintaining the S status in soils and plants rather than rapidly increasing pasture yield. Superphosphate and gypsum are more suitable for this. lt is recommended that experiments are carried out to examine closely the effects of existing soi l fertility status on the transformations and loses of fertilizer S in pasture soils. Some such studies have been undertaken by Sakadevan ( 1 99 1 ) . In such studies, techniques need to be developed to overcome the varability inherent in analysing mixtures of particu late fertilizers and soils, particularily, measurements of total S and residual S0. The inablil ity to remove this variability in the experiments discussed above has l imited the interpretation of the results. CHAPTER 8 MODELLING THE FATE OF SULPHUR IN THE PASTURE SYSTEM 8.1 INTRODUCTION 232 Recognized methods for calculating the fluxes of isotopically labelled compounds o r nutrients from one 'pool' to another in 'open' systems ( Shipley and Clarke, 1 972) cannot be applied to aquic soil environments when the added labelled nutrient (in this case 35so4=) or a product of the added label led ( i . e . 35s0 oxidized to 35so4=) is suscept ibl e to randomly occurring drainage events that vary in their intensity and specific activity of S leached is not easily quantified. For example after 35so4 =-based fertilizer addition to soil the specific activity of the so i l CaP-S pool w i l l i nc rease as the fe rt i l i zer ente rs the so i l and then decrease as mineralization transforms un labelled soil organic S into so4=, which is then removed by plant uptake, immobilization and leaching processes. Leaching, being an intermittent process, will transfer S of varying specific activity to lower soil depths depending on when it occurs after isotope addit ion in soi l . I n order to use i sotopic label l ing techniques to calculate actual mineralization and immobilization rate of S in field soils, measurement of soil solution 35so4= specific activity would need to be made at very frequent time intervals, particularly after rain, at a number of soil depths. Alternately, if there were long time intervals between measurements then a model of daily events that influence S movement and transformation (e.g. rainfall and plant uptake) is required to interpret the significance of measurements of S pool size and specific activity made at each time interval. The objective of th is Chapter is two fo ld , ( 1 ) to develop a model , wh ich if it accurately describes the observed movement and transformation of S and 35s in Chapter 5, will be used to calculate the actual rates of mineralization and (2) to summarize information gained in this thes is and from other studies on the soil-plant part of S cycle in a form which is usefu l for indicating future research directions. A simple computer simulation model recently constructed by Sakadevan ( 1 991 ) was modified to describe the movement, transformation and losses of S from the soil-plant system studied in Chapter 5 . The conceptual S pools are outlined in Figure 8 . 1 . These concepts are explained in more detail in Section 8 .3 and 8 .4 . The model output is compared to field observations made in Chapter 5 . Fig. 8 . 1 . A Simple Conceptual Dynrunic Model Showing the S - Cycle Product Loss and Excrcla Transfer f(S1 ,L ) t f(S 1,U ) -----.... - \ Animal Planl '---------' Ferti l izer f(A) ro; .R,w,K? Immobilization Soil Organic Sulphur "' Rainfall .. Soil Sulphate Atlsorption Coefficient ? Temperature - Transpiration .. Stocking Rate ? Plant Organk Maller, Quality .. Animal Organic Matter, Quality N u f L D 0 .. Particle Size plw; Oxidation or Solubility Constant ? Extractable Sulphate .. Drninnge .. Allimal Excretion Pallcms, iHcludi..llg S m1d L = Soil Volwuetric Water Content r = GrowU1 Limiting Nutrient .. Pasture Utilisation .. function of a variable lisle ? Landslope ? Diffusion Coefficient of Sulphate 233 8.2 MATERIALS AND METHODS Methods employed in making the field observations were discussed in Chapter 5 . &3 MODEL DEVELOPMENT 234 The simulation model develop?d in Section 8.3. 1 . 1 to 8 .3 .3.3 builds on a model developed by Sakadevan ( 1 991 ) and for clarification much of his discuss ion of the model development is repeated and modified where appropriate. 8.3.1 Water balance 8. 3. 1 . 1 Calculating drainage volumes. The vector for the movement of plant nutrients in the so il and loss through leaching is the soil water. A drainage model developed by Sakadevan (1 99 1 ) was used to predict the movement of water through the soil. Daily drainage water volumes for a soil column of known depth were predicted using the following simple balance equations: wf = Wi + R - EP 8.1 where W? 1 = the initial depth of water (mm) in a soil column of specified depth. wf = the final depth of water (mm) in a soil column of same depth. R = daily rainfall (mm day-1). EP = daily potential evapotranspiration (mm day-1) calculated from daily air temperatures and sun light hours by the method ofPrestley and Taylor (1972). To keep the model simple, it was assumed that rainfall intensity did not exceed the infiltration rate of water into the soil and that run-off did not occur. The amount of drainage (D, mm day? 1 ) leaving a certain soil depth (Z) is given by the equation D = 8.2 where = depth of soil water (mm) to soil depth Z (mm) = depth of soil water at field capacity (mm) in soil depth Z (mm) 235 In a field trial where actual drainage volumes were measured (Sakadevan, 1 991 ) , this equation explained 98% of the variation (1 :1 relationship) in the observed drainage volumes 8.3. 1 .2 Estimating actual daily evapotranspiration. The actual evapotranspiration (AEP) from these pasture soils will be used to estimate the flux of water from the soil through the plant. With complete cover of the soil by pasture plants 1 00% of evapotranspiration (EP) can be considered to occur from the leaves of pasture plants (Coulter, 1 973 ; Jensen, 1 973; McNaughton et al. , 1 979; Payne, 1 988) . This will be the vector which carries soil solution su lphate into the plant (Figure 8 . 1 ) to predict plant S uptake (Section 8 .3.3). The effect of decreasing soil water content is to decrease the rate of evapotranspiration (Payne, 1 988; Gregory, 1 988) . This effect can be approximated by the relationship g iven by Scotter et al. ( 1 979) and Rickard et al. ( 1 986). Rickard et al. (1 986) considered that for the Lismore stony silt loam plant growth and actual evapotranspiration (AEP!EP = 0 at permanent wilting point, PWP) completely stops when the soil water deficit was approximately 70% of the maximum soil water deficit and the potential evapotranspiration (EP, AEPIEP = 1 ) could be achieved when the soil water deficit was less than 50% of maximum. In the absence of data on the relationship between pasture growth and soil water deficit, for the Tokomaru field site, the relationship derived by Rickard et al. ( 1 986) was modified and applied. At the Tokomaru site Wfc was measured at approximately 45 mm for the top 1 00 mm of soi l and the maximum observed deficit was approximately 0 .3 Wfc = PWP (see below) . Actual evapotranspiration (AEP) was allowed to equal the potential evapotranspiration (EP) at so i l water contents between Wfc and 0 .65(Wfc) ? The ratio of AEP!EPwas allowed to decrease linearly, however, when the soi l water content decl ines from 0 .65( Wfc) to 0 .3 (Wfc) which was considered appropriate for the Tokomaru silt loam (Heng, 1 991 . personal communication) . The soil water depth at which the soil water deficit is zero ( Wfc) were estimated as 1 5 mm, 1 5 mm, 1 5 mm and 9 1 mm for 0-33 mm, 33-66 mm, 66-1 00 mm and 1 00 -300 mm (Heng , 1 991 , personal communication) soil depths, respectively. Using this relationship the AEPwas calculated from a knowledge of daily EP and the soil water content using a computer program written in OuickBASIC (Microsoft Corporation, 1 987) by Heng (1 991 ) . A detailed description of the movement of soil water through the plant-soil system is given by Gregory ( 1 988) . I n the simple model presented in this Chapter, however, it is assumed that the amount of water removed by the plant from the soi l depends only upon the distribution of root weight in each soil depth. Evidence that this method provides a reasonable description of 236 water removal from different so i l depths i s presented by Payne ( 1 988) . Wil l iams ( 1 988) showed that, the root mass distribution of a ryegrass-clover dominant sward on Tokomaru silt loam was 55%, 20%, 1 5% and 1 0% for 0-33 mm, 33-66 mm and 66-1 00 mm and 1 00-300 mm soi l depths, respectively. The total EPwas partitioned to draw water from each depth relative to the percentage root distribution in each soil layer. EP from Jh soil layer was calcu lated as follows: EP ? ,1 where = = total root mass for iW the soil layers ( I = 1 to 4) root mass for the I layer. 8.3 EP from each depth was then converted to AEP based on the relationship between AEPI EP and soil water deficit in each depth. 8.3.2 A single sulphate pool for plant uptake, Immobilization and leached sulphate. The simple configuration in Figure 8 . 1 assumes that there is one available su lphate pool from which S is taken up by plant and lost by leaching (Syers and Curtin, 1 987; McCaskill and Blair, 1 988) . In the experiment described in Chapter-6, evidence was provided that the specific activity of plant 35s formed during a period tn to tn+ 1 was similar to that of CaP extractable 35s over the same period. Sakadevan ( 1 99 1 ) has also shown that the specific activity of 35s leached from mini-lysimeters for time tn to tn+ 1 were similar to that taken up by plants (plant 358/S = 0.73 leached 35s/S) over the same period. The reasons given by Sakadevan ( 1 99 1 ) for sl ightly h igher specific activities of leached so4= were: That early leaching o f 35so4= immediately after labelled SSP application, moved 35so4= to soil depths where root activity and so4 = turnover were slower than in upper more densely rooted soil horizons. Subsequent rainfall events were suspected of leaching some of the less d i luted 35s from the lower soil depths. Also in the plant rhizosphere , which includes micropores exploited by fine roots and root hairs (Barber, 1 984) , the rate of sulphate turnover may be more rapid than in the bulk soil or in macropores. More leached S may be derived from the larger pores (mobile phase) rather than micropores (immobile phase) . Plant root hairs normally derive S from the immobile phase by diffusion ( Barber, 1 984) . The sulphate in the immobile phase may be d i luted rapidly because of increased mineralization associated with root activity. Thus, in general, the specific activity of su lphate in larger pores may be diluted at a slower rate than that in areas of i ntense root activity. 237 Evidence indicates that both plant S and leached S are derived from a pool with similar specific act iv i ty , but t he res u lts of Sakadevan ( 1 99 1 ) i nd icated that a robust leach i ng and mineralization/immobilization (SO 4 = turnover) model is required before the specific activities of measurable outputs of the plant-soil system can be adequately predicted. Sakadevan ( 1 99 1 ) concluded that for a computer model to simulate combined plant uptake, organic turnover and SO 4 = leaching it would be wise to partition the soil into defined zones of differing root activity. In this respect a vertically layered model with root activity and mineralization decreasing down the soil profile would seem acceptable if root activity is known. 8.3.3 Modelling the fluxes of S between various pools The following discussion represents the S mass balance central to mode lling changes in the size of the extractable S pool in the soil . The amount of sulphate present in the soil solution for a given layer (/) , which is potentially available for plant uptake and loss by leaching at any time ' f (M normally = 1 day) is given by the balance equation Es,t where Es,t Es,tn-1 p s Ls MS Is F s = = = = = = = = sulphate (kg S ha-1) present in the plant available pool solution at any time i for ith larr sulphate (kg S ha- ) present in the plant available solution at time t-1 for i tfi layer sulphate taken up by giant (kg S ha- 1 day- 1) during the period from t-1 to t from the it layer sulphate lost by leaching (kg S ha-1 day-1) during the period from the ith layer amount of sulphate (kg S ha-1 day-1) mineralized to the sulphate pool from soil organic matter in one day in ith layer. amount of S (kg S ha-1 day-1) entering the sulphate pool in the ;th layer from rainfall or other solution ii1futs amount of sulphate (kg S ha-1 day- ) entering the sulphate pool from fertilizer. 8.4 At the field site , estimated S input through rainfall is not high(<2 kg S ha-1 y-1 , Smith, 1 979 ; Heng, 1 99 1 ; Ledgard and Upsdell, 1 99 1 ) The majority of the S input to the sulphate pool is from the mineralization of soil organic matter, which includes the decomposition of animal excreta, plant litter, root and older soil organic matter, and the amount of fertilizer S, if applied. In the model plant S , immobil ized S and leached S are derived from solution sulphate and the order in which the processes occur on a daily basis is firstly leaching fol lowed by, plant uptake and mineralization/immobilization (Figure 8 .2) . The following sections construct submodels to describe the dynamic features of the components of equation 8.4 Fig. 8 .2. 1-licrarchical Order of the Processes in the S Cycle Daily End ?I + l Mineralization Submodcl i = 1 j 0 3 Sorpt ion Submodcl Daily S tart Fertilizer Submodel Tn i .. 1 j "' 1 Plant Uptake Submodel (AE1w . } 11 I 3 i = 2 1---:JIIII..... I: P . 1------1 j . 1 S,1 j "' 3 Leaching Submodel i "' 1 i = 2 i = 3 i 0 1 - 3 f(R ,w. ,wr ) 11 I c Ss = Soil Solution Sulphate E s = Extractable Soil Sulphate Rate = Rate of Fertilizer Application i = Soil Layer 1 , 2, or 3 For definition of symbols sec Figure 8 . 1 . 238 239 8.3.3. 1 Predicting the soil solution sulphate concentration. The method used here was similar to that used by Heng ( 1 99 1 ) and Sakadevan ( 1 99 1 ) using the relat ionships between laboratory determined CaP extractable S and f ield-soil solution so4 = concentrations. For a model to predict the amounts of S taken up by plants or leached (Figure 8 . 1 ) the soil solution sulphate must be estimated on a regular basis. Ideally, functions are required to describe the relationship (sorption isotherm) between adsorbed and solution sulphate in field soils. lt is difficult to measure the soil solution sulphate concentrations in field soils, particularly when the soil moisture content is lower than field capacity. In dry periods the soi l solution cannot be collected using porous, porcelain cups and suction sampling. lt is however, relatively easy to core sample soi ls and carry out laboratory measurements . The common soil test to measure solution plus adsorbed sulphate in the soil is to extract the soil with 0 .01 M or 0.04 M Ca(H2P04)2 (Searle, 1 979 ; Saunders et al., 1 98 1 ) . Some of the S extracted from pasture top soils is in organic form (Watkinson and Perrott, 1 990; Watkinson et al., 1 991 ) . The slopes of the Freund lich isotherm describing the relationships between CaP-S and the solution sulphate concentration for so il samples taken from the Tokomaru silt loam pasture soil are presented in Table 8 . 1 . Heng ( 1 99 1 ) also used a Freundlich 'isotherm' to described the relationship between the SO 4 = concentration in soil water sampled from 200 mm soil depth in the Tokomaru soil and the CaP-extractable S concentration in cores taken from that depth (Table 8. 1 ) . Notably the 'field' isotherms (Heng , 1 99 1 ) differ markedly from the 'laboratory' i sotherms for approximately the same so i l depth . The laboratory isotherms show an increasing S04= sorption power as the soil depth increase. Estimated field isotherms for the 0-33 mm and 33-66 mm soil depths were produced by changing slope of field isotherms i n the same ratio that the laboratory isotherms changed with soil depth. As discussed by Sakadevan (1 991 ) the re lationships were not strictly adsorption isotherms, because the extractable sulphate measured using 0 .04 M Ca(H2P04)2 contains solution, adsorbed and some organic S (Watkinson and Perrott, 1 990; Watkinson et al., 1 99 1 ) . For this reason true adsorption isotherms were not generated using plots of adsorbed sulphate vs. solut ion su lphat e . The Fre undl ich isotherms represented the re lat ionships between extractable S and soil solution sulphate for all the three soil layers (Table 8 . 1 ) as described above. 1 240 Table 8.1 Freundlich type equation fitted to laboratory and field mesured isotherms. Isotherm Soil Fitted1 method layers equations (mm) cs = acl b Laboratory batch method (Heng, 1991) 2.87q0.379 0-50 50-300 2.87q0.379 300-450 3.78Cl0.382 Field method (Heng, 1991) 3.03q0.627 0-200 C8 = extractable soil S cl = solution so4 = concentration a and b = empirical constants Model Equations1 depths used in model (mm) cs = acl b 0-33 2.3ci 0.627 33-66 2.3ct627 66-100 3.03CI 0.627 100-200 3.03q0.627 kg S ha-1 mm-1 soil depth kg S ha-1 mm-1 soil depth 241 8.3.3.2 Accounting for leaching of sulphate 8.3.3.2. 1 Layered drainage model with mobile and immobile water phases. Only a few mechanistic models have attempted to describe sulphate leaching from pastures (Heng, 1 991 ; Sakadevan, 1 99 1 ) . The approach used in this study was that of Sakadevan ( 1 99 1 ) . As sulphate adsorption capacities for each soi l depth (measured by Heng, 1 99 1 ) increasing with depth and because, i n this study, forms of soil S had been measured i n three soil layers (0-33 mm, 33-66 mm and 66-1 00 mm, Chapter 5), a three layered leaching model was considered an appropriate way of dealing with the S sorption differences and the soil S data available. The main assumption in this three layer model is that an input of water e ither rainfal l or drainage from the layer above (/-1 ) (R, mm) moves into a layer I of the soi l profile, once the soil water depth ( ?. mm) is equal to soil water depth at field capacity ( "Ytc? mm) the incoming water displaces an unknown proportion of native soil water in this layer in relation to the i nput volume (the condition when R exceeds "Ytc is discussed later) . This assumption simulates the condition where some of the input water, Ra mm (a fraction a) wil l move through larger pores interacting litt le with the native soi l water, while in smaller capillaries native soil water will be displaced ahead of the incoming water, R(1 -a) mm. A value for a of 0 .3 was determined for Mahoenui silt loam (Sakadevan, 1 99 1 ) by iteratively fitting a drainage model , similar to that discussed below, to the amounts of so4= leached from an undisturbed soil core subjected to simulated rainfall. In the absence of a fitted value of a for the Tokomaru silt loam, a value for a of 0.3 was used in this study. The soil solution sulphate concentration (C1 i kg S mm-1 soi l depth ha-1 ) for a given amount of extractable sulphate (Es, i) for Jh layer befo , re input water passes through the layer is given by the equation Cl ? ,1 = where = = (E ? I a?)( l!bi) S,l 1 the amount of extractable sulphate (kg S ha -I) present in ith layer coefficients of the Freundlich isotherms (empirical constant) explaining the relationship between extractable sulphate and the soil solution sulphate for the ;th layer. 8.5 242 The functions for all the three layers were given in Table 8 . 1 . Prior to reaching field capacity ( Wfc mm) for the layer I = 1 ( i .e , Ri + H'i ;;::: Wfc) then the new soil solution concentration for layer I = 1 is given by the equation Cl ? ,1 = where = = ((E ? + I ? * R ?) I :?.plbi) S,1 S,1 1 I. the concentration of sulphate (kg S mm -l) in the incoming water. the amount of input water to the i = I layer day-1 . 8.6 The remaining parameters have been described earl ier. In the case of the surface soil layer (I = 1 ) the i ncoming water is rainfall and the concentration of S is negl igib le , except when fertilizer particles are present on the soil surface (see later discussion) . Under such conditions the concentration of su lphate in the soi l solution is the one given in equation 8 .5 ( i .e . the change in concentration of su lphate in the soil solution of layer I = 1 due to incoming rainwater is assumed to be negligible unless leaching occurs) . When the soil water depth of the f.h layer ( Wi , mm) has reached Wfc? then the drainage volume (Di , mm day-1) is equal to the input water volume (Ri , mm day-1 ) and the total amount of S lost by leaching (Ls , i ? kg S ha- 1 day-1 ) from the /-h layer is sum of two components. 1 . The amount of native soi l solution sulphate displaced (Ld , i ? kg S ha- 1 day-1 ) by the slow moving input volume water (Di( 1 -a)) entering the layer which is given by the equation. Ld ? ,1 = [(E ? I a,p!bi)] * D?(l-a) S,I I. 1 8.7 2 . The amount of S i n the mobile fraction (Lm , i ? kg S ha- 1 day- 1 ) of the incoming water (Dia) which passes directly through the soil layer. When the input water passes rapidly through the surface layer without interacting with the native soil wate r i ts drainage path to lower soil depths is presumed to be un-impeded. Therefore this water wi l l pass through the rest of the layers without any interaction with the native soi l water. For the purposes of the model H'i? for the layers I = 2 and 3, is only increased by the water volume D(1 -a) d isplaced from the /-1 layer. Also this means that drainage Di=1 a occurs from the whole soil profi le as soon as Ri=1 + Wi=1 > Wfc,i=1 , irrespective of the H'i=2 and H'i=3. The input water sulphate concentration Us} has have two distinct components : (a), the fast moving drainage water volume Dia with S concentration /5,a, and (b) , slow moving drainage 243 water Di(1 -a) with S concentration ls(1 -a) . The amounts of S in the mobile leached volume would be given by the equation = Therefore = I ? * a * D? S,l 1 Ld i + Lm i ' ' 8.8 8.9 In the case of rain water at the Tokomaru site which contains negligible sulphate, ls, i is zero then Lm, i is equal to zero . At other sites closer to coast ls , i would require a positive value (Ledgard and Upsdel l , 1 991 ). If the displaced drainage water volume (D(1 -a)i_1 + "'i - "'fc) is negative then 0(1 -a)i = 0 and no leaching occurs. i .e. Ls, i is zero in equation 8.9. After leaching, the equilibrium between adsorbed and solution sulphate was allowed to establish prior to plant uptake. The amount of extractable sulphate (kg S ha-1 ) remaining in the Jh soil layer after an input event or a drainage event is given by the size of initial extractable sulphate pool , Es, i plus the amount of sulphate leached from the layer above (Ls, i-1 ) minus the sulphate leached from that layer (Ls}- Es,i(t+l) = where Es,i(t+l) = Es,i,t + Ls,(i-1) - Ls,i the amount of extractable sulphate (kg S ha- 1) present after leaching. The new concentration of sulphate in the solution is cl,i(t+1) = (Es,i(t=1) I a/llbi) where t = a time (normally = 1 day) or volume step (see discussion later). 8.10 8. 1 1 This i s the solution concentration that i s used to calculate plant uptake on a day when leaching occurs (Section 8 .3.3.3) . 244 8.3.3.3 Plant uptake of sulphur The amount of S removed from each of the three layers depends upon plant uptake , immobil ization and drainage. Drainage was considered to be immediate ( instantaneous) relative to plant u ptake (during periods of transpiration) and mineralization/immobil ization (cont inuous processes) . Therefore i n th is mode l l each ing of S fo l lowed by t he re? establ ishment of the equi librium between adsorbed and solution S prior to plant u ptake , immobilization and mineralization takes place. The second step in the model process is the plant uptake of S from soil solution. The amount of any nutrient taken up by plant for a particular period of time is a function of the amount of water removed by the plant (Scotter et al., 1 979 ; Hayman and Stocker, 1 982; McAneney and Judd, 1 983 ; Rickard et al. , 1 986; Martin , 1 990) and the soils ability to supply the nutrient to the soi l solution. The amount of water removed by plant can be considered to be the actual evapotranspiration (AEP) from pasture soils (see discussion 8 .3.1 .2) . So , assuming that most sulphate moves to plant roots by mass flow (Barber, 1 984) the amount of S removed by plant (Ps, kg S ha- 1 ) is proportional to the AEP (mm) mu ltipl ied by the soi l solution su lphate concentration (C1, kg S mm?1 ha-1 ) . This assumes that c1 in the Tokomaru soil never reaches concentrations higher than that which can be adsorbed by pasture roots. Even in campsites of ve ry high S status , Sakadevan ( 1 991 ) fou nd that AEP * c1 , i gave a good prediction of observed plant uptake. The amount of water removed from each layer was calculated from daily AEP multiplied by the soil solution su lphate concentration (calculated from equations 8 .5 or 8 .6 or 8 .1 1 in each layer) to give the amount of S removed from each layer. In itially it has been assumed in this model that all of the S removed by the plant was moved to the root zone by mass flow. Thus when drainage occurs from the soi l , the amount of S removed from the particular layer by plant uptake will be lower than that removed without drainage . Mathematically, the amount of S removed by plant from Jh layer is given by the equation Where AEPt,i Cl ? ,1 = = = = c1 . * AEP ? ,I t,I the amount of S (kg S ha -1 day-1) removed by plant from the ith layer, the volume of AEP removed from the ith layer (mm day?1) discussed earlier or if drainage occurs c1 i is recalculated using equation 8.10 and 8. 1 1 . ' 8 .12 As mentioned in Section 8 .3 .3.2 ; there exists an equilibrium between the soil solution sulphate concentration and adsorbed su lphate. As the soi l solut ion sulphate is decreased by plant uptake, su lphate will be released into soil solution from the amount of sulphate on the surface 245 of the soil. This equilibrium was allowed to re-establish on a daily basis after mineralization and immobilization fluxes of S to and from the extractable S pool had also been accounted for. Uptake of S by roots was accounted for as part of below ground immobil ization. 8.3. 3. 4 Accounting for immobilization and mineralization of soil organic sulphur Immobilization of s into soil organic matter and its subsequent mineralization were considered to be continuous root and microbial processes which extract S from and input S i nto the extractable S (Es) pool. As reported in the l iterature review (Sect ion 2 .3) the net d ifference between these two processes is h ighly dependent on the C : S ratio of organic substrates available for soil micro? organism growth and the soi l so lution su lphate concentration . Other than experiments examin ing net so4= re lease o r i mmob i l izat ion i n laboratory stud ies , except that from experiment carried out by May and eo-workers (May e t al., 1 972; May e t at, . 1 973; Til l , 1 979) on dry land Austral ian pastu res , l itt le i nformat ion is avai lable on individual rates of immobi lization and mineralization in fie ld soils. Immobilization requires a source of carbon. The dominant sources of carbon in subsurface layers of field soils are plant roots. Thus the extent of immobil ization of so4= in any soil layer was considered to be proportional to the sulphate uptake activity of roots in that layer which is al ready dependent upon soi l moisture content , max imum/min imum temperatu res, sunsh ine hours and root mass (wh ich are embodied in the calculation of AEPi) . This can be conveniently expressed as: where Mmmobilization l M ilimmobilization Kimm Ps,i = = = = Kimm * Ps,i 8.13 amount of S immobilized during time ?t an immobilization constant, unitless, proportionally constant plant uptake from soil layer I per day as discussed earlier Other authors, however, Gregg ( 1 976) and May et al. , ( 1 972 , 1 973) have noted that after addition of so4= to field soils , there was initial rapid immobilization of sulphate , indicating that above certain soil sulphate l evels ( Cak) there is additional immobilization stimulated by h igh soi l solution su lphate concentration. Thus equation 8 . 13 was expanded with a pre-condition this: I F c, . , I > where = (Cak) * K THEN Kimm * (( Cl , i - Cak) 246 8 . 1 4 8 . 1 5 threshold Es i above which immobilization increases in proportion to the difference between the threshold Cak and current Es Mineralization of soil organic S is the major process supplying S to pasture plants in aquic environments of the central north island, New Zealand. The development of a pasture index (PD I) to account for pasture S supply from organic S (Sinclair et al. , 1 985) in fertilized pastures reflects this. The importance of mineralization in supplying both plant S and leached S in New Zealand, No rth I s land h i l l pastu re has been conf i rmed recently by Sakadevan ( 1 99 1 ) . Therefore it can be deduced that in a long established permanent pasture in this environment the extent of mineralization in each soil layer must be closely related to the root activity in each layer. Unlike immobilization, however, mineralization may not always be positively related to root S uptake because mineralization may be stimulated by lower soil so4= concentrations and inhibited by higher soil solution so4= levels increased through fertilizer addition. As discussed in Chapter 2, Section 2.3, Tabatabai and eo-workers have shown that sulphatase enzyme levels are inhibited in the presence of high so4= concentrations. For the purpose of developing a prel iminary model it was considered that mineralization rate was proportional to the soil organic S content (Dj, kg S soil layer ha-1 ) , the root mass, temperature and soil water availabil ity ( i .e . AEPi) and inversely proportional to the soil sulphate concentration in each soil layer. This was expressed as: where ?mineralization I M = K ? * AEP? * 0? I E mm 1 1 s = = a mineralization constant kg S ha-1 mm-1 day-1 a length of time 8.16 At the end of each day ( t = 1 day) the new amount of S (kg S soil layer ha-1 ) in the extractable pool was calculated by subtracting Ps and ?immobilization 8.3.3.5 Executing the model The whole model was executed by writing a program in QuickBASIC (Microsoft Corporation, 1 987) and executing the program on an IBM compatible personal computer. The i nput data required to run the model d iscussed above were those appropriate to the experimental conditions in Chapter 5, they include: ? 247 climatic data consisting of Julian day, daily rainfall (mm) , daily maximum and minimum temperatu re and sunshine hours ; i nitial soi l water content data consisting of soi l volumetric water contents (mm) at f ie ld capacity and i nit ial soi l volumetric water contents (mm) for each soil layer of depth z mm; the percentage distribution of plant roots in each soil layer and the total organic S (Table 5.3) and the initial Es, i value, i .e . calcium phosphate extractable S (freeze-dried preparation technique, Table 5.3 and 6.3A) in each soil layer (kg S ha-1 layer- 1 ) ; The Freundlich isotherm constant for the 'isotherms' describing the distribution of so4= between soil solution and soil surfaces (Table 8 .1 ) . The mobile water volume a remained at 0.3 , that determined by Sakadevan ( 1 99 1 ) for Mahoenui silt loam, because there was insufficient time to conduct such experiments on undisturbed cores of Tokomaru silt loam soil. 8.4 RESULTS AND DISCUSSION 8.4.1 Prediction of plant S uptake on unfertll lzed soli To simu late 1 50 days of S transformation, the model takes 15 seconds to run on an IBM compatible personal computer. Several iterations were conducted adjusting only Kmin and Kimm values until the model accurately predicted the plant uptake on the unfertilized control plots described in Chapter 5 (Table 5.5) and gave what were considered acceptable readout for the changes in soil CaP-S levels (equivalent to Es in the previous discussion) in all depths. The model output and observed plant S uptake values are compared in Figure 8 .3A and 8.38. The model gave a remarkably accurate prediction of changes in plant S uptake with time (Figure 8.3A) . The CaP-S values (not presented) however could not be compared directly with those measured at 30 and 60 days (Table 5. 1 2) which , u nl ike the i nitial soi l test CaP-S (Appendix 5.3) , had been determined on oven dried, not freeze-dried, soil. The model output was not very sensitive to small changes in a (?1 0%) so that the value a = 0.3 was retained. Plant S uptake and S leaching loss (data not shown) were dependent upon the relative size of Kmin and Kimm? Changes in the difference between them rapidly increased or decreased the s ize of the CaP-S-pool , which i nfluenced plant uptake . The final Kmin and Kimm values selected to g ive the best fit of plant uptake indicated that actual rate of mineralization and immobilization were 1 .5 to 2 fold greater than the rate of plant uptake from any particular soil depth. 8.4.2 Prediction of plant S uptake on superphosphate fertil ized soli 248 To predict plant S uptake on fertilized soils required that a submodel be added which could predict the release of sulphate from superphosphate into soil. The non-mechanistic approach of McCaskill and Blair ( 1 989) was considered but it was decided that a mechanistic approach may be more portable between soils. 8.4.2.1 Movement of sulphate from superphosphate Into soli Superphosphate (SSP) can be considered to contain gypsum, monocalcium phosphate , less soluble phosphates and minor impurities. When a g ranule of SSP is applied to moist soil , moisture absorbed by the granule forms a ferti l izer solution which is similar in Ca(1 .3-1 .5 M) and P(3 .4-3 .8 M) concentrat ion to the metastable triple po int so lut ion (MTPS) when an excess of monocalcium phosphate is dissolved in water (Wil liams, 1 971 a) . This solution has a pH between 1 .6-1 .8 and initial S concentration of 0 .02-0.07 M . This concentration will remain higher than that of saturated solution of gypsum (0.01 4 M) while the pH of the fertilizer solution remains low. Sulphate will d iffuse from th is fert i l ize r solut ion into the soil and Wil l iams ( 1 971 b) has observed the hemispherical nature of the sulphate concentration gradient below a SSP particle. In order to model the rate and extent of 35so4 = movement into soil cores fertilized with SSP in Chapter 5 this diffusive flux of so4= was simulated in the following manner: Consider a SSP particle ( 1 2% S, density of p = 1 900 rng cm-3 and radius rn) sitting on a soil surface. The soil has a volumetric moisture content of e at f ield capacity, a tortuosity (or impedance) factor (Nye and Tinker, 1 977 ; Barbe r, 1 984) of f and a so il su lphate buffer capacity oCs/oC1 (where oCs/oC1 = a * b * c1 (b-1 ) , where a and b are the coefficients of a Freundl ich sorption isotherm representing the relationship between adsorbed soil sulphate (Cs) and the soil solution sulphate concentration (C1) . The equation describing the diffusive flux (F rng cm2 s-1 of sulphate from the fertilizer solution at the particle surface across distance X cm to the soil solution can be written as: F where OCjiCox D = = = n * e * f * oc1 1 ocx 1 [a * b * c1(b- 1)J the concentration gradient of so4 = between the fertilizer solution and the soil solution (mg cm-3 cm-1) diffusion coefficient of sulphate in water 1 * 10-5 cm2 s?l (Barber, 1984) 8.17 249 For the purposes of simulation this re lationship can be solved by using a finite difference method similar to that used by Kirk and Nye ( 1 986) to model the dissolution and diffusion of P away from particles of d icalcium phosphate . In the case of the superphosphate particle (described above) it is considered to have its lower hemispherical surface (21tr2) in contact with the so il water film. D iffusion of so4= away from the particle can be considered to occur in a number (n = 2 to 6) of d istance steps through a series of hemispherical shells (volume 2/31t? ( r3n+1 - r 3 n) where the width of each shel l (r(n+1 ) - rn) = dx. Each t ime step ( t) a new concentrat ion of su lphate per un it soi l vo lume ( Cs + e * c1 , mg cm-3) and soi l so lution concentration ( c1) are calculated: = 8. 18 The f i n i te differe nce solut ion of equat ions 8 . 1 7 and 8 . 1 8 were written in QuickBAS IC programming language (M icrosoft Corporat ion . , 1 987) and executed using IBM compatible personal computer. Time steps of 1 day and hemispherical shell widths of 0 . 1 cm were used with initial soi l solution so4= concentrations of 7.5 * 1 0- 4 mg S cm-3 and the fertilizer solution at the surface of a granule was 0.64 mg S cm-3 in the Tokomaru si lt loam soil while e = 0 .45, f = 0.3 and the Freundlich su lphate sorption isotherm is that of top soil layer calculated i n Table 8 . 1 . Each step, the flux of S into the soil was subtracted from the particle mass and a new radius calculated. Three simulations were conducted with particle radius ranging from 0 . 1 5 to 0.05 cm at an application rate of 30 kg S ha-1 . The simulation model indicated that particle radius wou ld have a marked effect on the rate of movement of fertilizer i nto the soil by diffusion (Table 8 .2) . The most distant radi i of the hemispherical shell of S penetration into soi l are less than those observed by Wil l iams ( 1 971 b) possibly because the sorption isotherm for the Tokomaru soi l is steep in the region of soil solution concentration, 7.5 * 1 0-4 rng cm-3. Field observations conducted during the normal addit ion of fertilizer in field trials on winter wet Tokomaru soi ls indicated that 3 mm SSP granules disappeared in less than the 1 28 days indicated i n Table 8 . 2 ( M .J . Hedley, personal communication ) . Addit ional mechanisms explaining the disappearance of granules are coverage of granules by earthworm casts (Syers and Springett, 1 984) and the impact of rainfall (McCaskill and Blair, 1 989) on the granule itself. The effect of rainfall was simulated as follows: Table 8.2 Initial particle radius cm 0. 15 0. 10 0.05 250 The simulated time taken for 30 kg S ha - 1 to diffuse into Tokomaru silt loam and depth of sulphate movement (radius of outer hemispherical shell from center of granule). Diffusion only Rainfall plus diffusion ????????????????????????????????????? ????????????????? Days taken for Radius of Days taken for 30 kg S ha-1 final 30 kg S ha- 1 to enter soil hemisphericala to enter soilb shell days cm days 128 1 .6 73 68 0.9 46 26 0.3 22 a = i.e. maximun soil depth penetrated. b = distance travelled > 100 mm due to leaching. A term was added to the diffusion model to estimate the volume of rain (cm3) impacting o n the remaining granule surface area (7tr n 2) . The volume of rain was allowed to form a saturated solution of superphosphate S, concentration = 0.64 mg S cm3, before passing into the soil by mass flow. The combined diffusive and mass flow flux of S leaving the granule accelerated the decrease in granu le size (Table 8 .2) . If the soil was already at f ield capacity, a fraction (a, mobile volume discussed in Section 8 .3.3.2 . 1 ) of the SSP solution derived from the rain was allowed to move freely through the soil as macropore flow to be included as leached sulphur. The addition of rainfall rapidly increased the rate of SSP dissolution and depth of entry i nto the soil (see model output Figure 8.4 to 8 .7) . The combined rainfall d iffusion model was used to simulate the input of S into the surface layer (I = 1 ) of the SSP fertilized soil . 8.4.2.2 Prediction of plant uptake on SSP fertilized plots The rainfall diffusion submodel was added to the simulation model discussed in Section 8 .4 .2 .1 and the simulation model was executed using the same parameters described previously. The simulated entry of fertilizer S into the soil ( Figure 8 .4) indicates how the onset of rain rapidly accelerates fertilizer d issolution and movement. The comparison of plant S uptake and CaP-S in each soil layer on the SSP fertilized cores with the observed values are presented in Figures 8.3B and 8.5. The prediction of actual plant S uptake was very good (1 : 1 r2 = 99.7) but prediction of CaP-S concentrations during the period of the trial were less accurate (F igure 251 8.5) , in particu lar, observed CaP-S concentration during the first 30 days were double those that could be predicted by the model. Conversely after 60 days while there was a reasonable prediction of CaP-S level in the top 0-33 mm of soil there was overestimation at lower depths. The model-predicted CaP-S levels were sensitive to major drainage events (cf. Figure 8.4 and 8.5) , reflecting leaching of S, particularly severe just after 1 20 days from fertilizer application. Field observations by Ghani et al., 1 990 also showed that CaP-S decreased markedly after recent rainfall . The total accumulated leaching loss 21 kg S ha"1 ( Figure 8 .4) compared favorably with leaching losses of S (8-70% of that applied annually) that occur from hil l country pastures in the North Is land of New Zealand (Saggar et al. , 1 990a, 1 990b) . This was equivalent to an average drainage water su lphate concentration of approximately 1 2 ppm S which is within the range (6-1 4 ppm S) measured by Heng et al., ( 1 991 ) for SSP fertilized paddocks on a Tokomaru soil 8.4.3 Prediction of 35s movement and transformation In undisturbed soli cores One main objective of construct ing this model was to overcome the difficulties that are normally encountered in predicting the fate of isotopic tracers in open systems where the rate and intensity of isotope loss is haphazard and l imits the interpretat ion of 35s isotope to determine actual rates of soil organic S mineralization. The movement of isotope into soil was simulated in the model by using the rainfall/diffusion simulation of SSP input into layer I = 1 to add 35s at a rate proportional to the amount of S released (i .e. S release rate multiplied by SSP 35s specific activity) . The specific activity of fertilizer 35s released into the soil was immediately diluted by the amount of CaP-S present in layer I = 1 . This is to simulate plant uptake of non-labelled and labelled S from the soil layers and to accommodate the fact that soils were sampled in 33 mm layers and thoroughly mixed prior to analysis. lt is not expected that 35s wou ld normally mix with all the CaP-S in field soil without the soil being artificially mixed. Leaching, plant uptake and immobilization removed 35s from this pool in proportion to the amount of S removed multiplied by the CaP-S 35s specific activity. Leached 35s entered the next layer I = 2 (3 o r 4) where the process of m ixing and removal were repeated. Immobil ized S and 35s in each layer entered the o rganic S pool i n each soi l l ayer. The specific activity of mineralized organic S was simulated by calculating a running total of the S and 35s in the organic poo l . The predicted and observed percentages of added isotope present in soil CaP-S fractions, plant S and soil organic S are presented in Figures 8 .6-8.8. Whereas the model simulates well the percentages of added 35s remaining in the CaP-S fractions from all soil depths. The model seriously over predicts plant 35s uptake and the percentage of added 35s remaining in the top soil as organic S. The main reason for this is 252 that the current parameters of the model fail to predict the large observed loss of isotope from the soil cores during the f i rst 30 days (Appendix 5 .4) . Although the model predicted that drainage occurred at 1 1 , 1 2 and 23 days, leaching was insufficient to remove large amounts of 35s from the core . I rrespective of increas ing a to accommodate larger leaching loss of fertilizer solution on days when rainfall caused leaching, it was not possible to simulate such large losses of isotope by 30 days. The relative distribution of 35s between soil and plant S forms predicted by the model were ranked in the same order as the observed values for variation between forms and soil depths ( i .e . observed and predicted 35s recovery was the greatest in soil o rganic matter > plant > CaP-S and the end of the first 30 days and throughout the remainder of the experiment and also organic S in the top soil layers was greater than the middle and lower soil depths) . This comparative similarity between observed and predicted 35s distribution gives some support to us ing root activity as a modif ier of mineral ization and immobil ization rates to describe the relative extent of these processes in different soil depths. The large amounts of unaccounted for 35s activity at 30 days (Appendix 5 .4) presumed leached remain a problem in evaluating the model. Uncertainty concerning the actual fate or cause of isotope loss ( leaching or lateral movement from the core) means that confidence cannot be placed in adjusting ferti l izer release rate (or rainfall solution rate of SSP) and leaching parameters to account for this loss . The data in Chapter 7 concerning the fate of 35s0 applied to the Tokomaru soil provide a set of data i n which most added 35so was accounted for and could be used to further test the simulation model . A sub model predicting 35so4= release rates from oxidizing S0 developed by Chatupote ( 1 99 1 ) could be used to simulate 35s input into soil layer /. Unfortunately t ime has not allowed this to be completed prior to submitting this thesis. S uptake (kg/ha) S uptake (kg/ha) 50 40 I 30 20 1 0 0 50 40 30 20 1 0 Contro l cores (A) (B) SSP cores 0 ?. ?--?--?----?--?--? 0 30 60 90 1 20 1 50 Days 2 X PV = 0.67; R = 99.6% 2 PV = 0.70 ; R = 99.7% 2 + PV = 0.73 ; R = 99.7% 0 Observed 2 Predicted ; R = 96.3% 0 Observed Figure 8.3 Comparison of predicted and observed plant S uptake on (A) control (unfertilized) and (B) SSP fertilized soil cores (experimental details Chapter 5). The coefficient of determination (R2) represents the variation in observed data accounted for by the model prediction (PV stands for the less mobile fraction of soil water; 1-a). 1\) 01 UJ S disappearance (kg/ha) Rainfal l (mm) S leached (kg/ha) Drainage (mm) Figure 8.4 50 40 30 20 1 0 0 -1 0 -20 -30 -40 -50 0 30 60 90 Days 1 20 1 50 The measured rainfall and predicted drainage occurring from the top 100 mm of Tokomaru soil during the experiment conducted in Chapter 5. The dotted line shows the disappearance of surface applied superphosphate (30 kg S ha -I) as it dissolves and moves into soil. The solid line shows the predicted accumulated leaching loss of S from top 100 mm. 0 Rainfall - S disappearance ? Drainage - S leached I\) 01 .p.. Figure 8.5 255 1 5 0-33 mm 0 Predicted CaP-S 1 0 ( kg/ha) 0 Observed 5 0 1 5 66- 1 00 mm P redicted CaP-S 1 0 0 (kg/ha) 0 Observed 5 0 0 0 0 0 0 30 60 90 1 20 1 50 Days Comparison of predicted (output every 5 days) and observed amounts of CaP? S in each soil depth in SSP fertilized cores. Figure 8.6 % 3? i n CaP-S % 3? i n CaP-S 3S in CaP-s 20 1 5 1 0 5 0 20 1 5 1 0 5 0 20 1 5 1 0 0-33 mm 33-66 mm 66-1 00 mm 5 0???????==? 0 30 60 90 1 20 1 50 Days 256 Predicted 0 Observed Predicted 0 Observed Predicted 0 Observed Comparison of predicted and observed percentages of added 3S S recovered in CaP-S fractions from different soil depths. 257 30 25 o/o 20 35 S uptake 1 5 0 0 Figure 8.7 0 0 Predicted 1 0 5 0 0 Observed 0 0 30 60 90 1 20 1 50 Days Comparison of predicLed (oulput every 5 days) and observed percentage of added 35s recovered in pasture plants. Figure 8.8 60 (a % 40 3? in so i l o rgan ic 20 0 60 % 40 3=t in soi l o rganic 20 0 60 % 40 3es i n soi l 20 o rganic 0 0 30 0-33 mm -- 0 0 0 0 33-66 mm 66- 1 00 mm 60 90 1 20 1 50 Days 258 pred icted 0 Obse rved Predicted 0 Observed Pred icted 0 Obse rved Comparison of predicted and observed percentage of added 35s recovered as organic S in different soil depths. 8.5 CONCLUSIONS 259 The simulation model provided a very accurate method of predicting plant uptake of S from both S fertilized and unfertilized soil cores over 1 50 days of plant growth. Due to the lack of quantitative information on factors influencing plant uptake, immobilization and mineralization rates of S in f ield soi ls many unsubstantiated assumpt io ns were made to construct the simulation model developed i n this Chapter. If al l the assumptions made to construct the model are true the model indicates that, on an average, actual rates of mineralization and immobilization exceed root uptake of S by 1 .5 to 2 fold (mg S turned over per unit of S taken up by plants) . The model was less accurate at predicting the changes in the CaP-S extractable pool size and was relatively inaccurate at predicting the transformations of 35s tracer added to the soil as SSP. The observed relative distribution of 35s between organic soil S and CaP-S at different soil depths and plant S forms and the relative distribution predicted by the model were similar, which gave support to the concept of using root activity as a modifier of plant uptake, and organic S turnover at different soil depths. This concept may be usefu l in other studies on the fate of nutrients in pasture systems. There was great uncertainty, however, concern ing the quantity of 35s tracer data to test the model because of the large unaccounted for losses of 35s which had been presumed to result f rom leaching. The model fail to predicted large leaching losses in the first 30 days. lt was not possible, therefore, to test with any confidence many of the assumptions made in constructing the model. Output from the current model suggested that, because of the small initial distance that S will move into soils from SSP or ferti lizer in general, emphasis on the study of S transformation processes particu larly immobilization and mineralization should concentrate on studying narrow horizontal soi l slices close to the soil surface (a point also raised by Til l , 1 979) . Various outputs from this model are sufficiently close to observed values that further research should be undertaken to test its various assumptions. In the current model the role of litter return in immobilization of S at the soil surface was not considered. To complete the cycle observations on litter and the return of grazing animal dung and u rine detailed by other authors (Gillingham, 1 980; Gill ingham et al., 1 980; Boswel l , 1 983 ; Saggar et al., 1 990a, 1 990b) could be added in order to test the ability of the model to predict plant S uptake over longer periods of time and under grazing conditions. CHAPTER 9 9.1 REVIEW OF LITERATURE SUMMARY Sulphur is an important major nutrient, particu larly important in the nutrition of legume-based pastures. During the S cycle, highly mobile sulphate is formed which is susceptible to leaching losses. Because of this, the abundance of S in grazed-pasture systems in aquic environments decreases with t ime u nt i l pasture production becomes l imited by S deficiency. Su lphur fertilizers are, therefore, required to maintain pasture productivity. The persistence of plant available soil S in pasture grazed by sheep is influenced by several factors. These include : the rate and form of ferti l izer input , pasture growth rate and its S status, the amount of pasture consumed by animals, uneven returns of S in dung and urine, losses from the removal of an imal products , rainfal l , su rface run-off, sulphate retention characte rist ics of so i l and leach ing , and the extent to which the applied fert i l izer S is immobilized into soil organic S reserves. New Zealand's traditional pastu re ferti l izer for S and P has been superphosphate (SSP). Recent studies have shown large leaching losses of S from SSP fertilized pastures. Sulphur fertilizer applied as elemental S (S0) is non-water soluble and is a slow so4= release form that has been shown to be more efficiently used by plants than so4 =-based ferti l izers in soil environments where leaching is severe. Whereas this slow-release mechanism may reduce S leaching loss from S0 fertilizer in the short-term, it is unclear whether reduced S leaching can be achieved in the long term because S leaching is a function of the rate of S cycling which is a function of pasture productivity ( i .e . stocking rate) . Knowledge of the comparative fate of sulphate and S0 based fertilizers in soils is required to formulate future fertil izer application strategies. This study was undertaken with the main objective of tracing the short-term fate of both S0 and su lphate-based ferti l izers applied to undistu rbed grazed pasture soil cores. A method for radioactively labelling fertilizer sulphur was developed. Different 35s0 particle size diameters, ranging from microf ine partic les of l ess than 0 .0 1 0 mm to part ic les of 0 .500 mm, were employed i n a study i n conjunct ion wit h t he su lphate-based fert i l izers , gypsum and superphosphate . Transformations of these products were studied i n a series of field and 261 glasshouse trials using undisturbed cores of pasture so i l . The effect of sheep dung on the short term immobi l izat ion of so i l and fert i l izer S was also measured. Final ly , a s imple computer simulation model was constructed to explain the fate of fertilizer 35so4= in pasture soil. 9.2 LABELLING TECHNIQUES Methods for manufacturing radioactively labelled (35s) sulphur fert i l izers (S0 of d iffe rent particle size, SSP and gypsum) were developed. The level of enrichment of fertilizers with 35s activity was based on concepts regarding the fate of S in the plant and soil system being studied. These included dilution of the label by soil and herbage S, the sensitivity of detection of 35s, sample s ize and the amount of fertil izer needed. These techniques, which do not i nvolve using any specialized equipment other than a small Geiger counter monitoring for safety, gave ferti l izer materials which were un iformly labelled with more than 95% of the radioactive 35s recovered in the fertilizers. 9.3 THE EFFECT OF SHEEP DUNG ON THE SHORT? TERM IMMOBILIZATION OF SOIL AND FERTILIZER S A field study showed that in a short term (from June 27 to October 30, 1 985) the pasture dry matter yield i n the area directly adjacent to the dung was insignificantly i nfluenced by the nutrients P and S contained in the dung. This result was influenced by the high soil fertil ity of the experimental sites which historically, had been well maintained with fertilization to ensure an optimum pasture production. Very small amounts of plant P and S (about 2-5%) , within 1 0 cm of the dung, were derived from the dung. The majority of plant P and S were derived from soi l sou rces, a large part of which wil l be derived from s lowly mineralizing dung that has accumulated with time. Furthermore, the presence of the dung did not significantly influence the immobilization of soil or fertilizer derived phosphate and sulphate. This study confirms findings of others that the short-term effect of dung on nutrient availability was smal l . Therefore , the ro l e of dung is one of a s low re lease component as it is decomposed by soil organisms and is converted into soil humus which in turn continues to release S over a longer period of time. lt was concluded that in short-term (up to one year) field trials examining the fate of fertilizer S in well developed permanent pastures, it is not necessary to apply sheep dung to experimental 262 plots in order to simulate the effect of dung return because probably this will have little effect on the size of the 'bank' of decomposing dung and soil humus in pasture soils. For long term (several years) experiments on permanent or less well developed pasture where soil fertility is very low, animal dung should be deposited to maintain soil organic S reserves. 9.4 THE FATE OF S FROM S0 AND SSP IN SOIL For a period of 1 50 days after applying 35s labelled m icrofine S0 and SSP, to separate undisturbed cores of pasture soil, the transformation and movement of 35s to plant and soil S forms was monitored. Duri ng the f i rst 30 days of t he exper iment , there we re la rge unaccounted for losses of 35s from al l soil cores. At the t ime this was presumed to result from leaching. A drainage model developed in Chapte r 8, however, suggested that there was i nsuff ic ient drainage to generate such a large leaching loss. If it is assumed t hat the unaccounted for losses of 35s were due to leaching then leaching losses of S beyond the 1 0 cm soi l layer were larger i n the SSP treated cores. The large unaccounted for loss of 35s, however, reduced the confidence of using the measured transformation of 35s to calculate amounts of soil and ferti lizer S transformed into plant and soil S fractions. Despite th is large loss of isotope, clear trends in the relative rates of transformation of 35s added as SSP or S0 were evident. Within 30 days there was rapid oxidation of the microfine S0 to calcium phosphate extractable sulphur and movement of 35s from both S0 and SSP to plants and deeper soil depths. During 1 50 days, plant uptake accounted for 1 3 .6% of the S0 and 1 6.3% of the SSP. Incorporation of 35s labelled fertilizer S into the soil o rganic fraction was the process that conserved the largest amounts ( 22-40%) of 35s labelled fertilizer in the soil. Notably in all experiments the major organic S form labelled with 35s was carbon-bonded S, particularly in upper soil zones. More of the microfine 35s0 was transformed into the o rganic S than 35s from SSP, indicating that more soil organic S reserves may be formed through the use of S0 fertilizer rather than SSP. Such an indication has not been reported previously. An inverse dilution technique where undisturbed soil cores were un iformly labelled (rather than the ferti l izer) indicated that carbon-bonded S was likely to be the source of mineralized organic S. Both results from the inverse dilution and labelled fertilizer techniques were consistent in this respect. I n terms of plant S uptake, the microfine S0 was initially slightly less effective than SSP. Over a longe r period (90-1 50 days) , however, both fert i l izers showed a simi lar performance, 263 reflecting the larger conservation of 35s in the root zone of the S0 treated plot. One problem with the rapidly oxidizing microfine S0 was that apparent losses of S from the plant-soil system were only slightly lower than that from SSP. 9.5 MEASUREMENT OF PLANT AVAILABLE SOIL SULPHUR Using soi l samples, labelled with different forms of 35s (derived from the experiment described in Chapter 5) , a series of studies were undertaken to evaluate soil preparation and extraction techniques. The objective was to find a technique where the specific activity of S in the soil extract was the same as S taken up by pasture. The initial concept applied was that if the extract : plant 35s specific activity ratio was close to unity then the extract and the plant must draw S from a similar biologically active pool . Simi lar 35s specif ic activity does not confi rm , however, that the S i n the extract is the precursor of S taken up by plants. Confirmation that CaP quantitatively extracts plant available S requires a quantitative assessment of plant S uptake re lative to depletion of CaP-S pools. The CaP-S pool is dynamic and requires a cons iderat ion of a l l factors influencing its magnitude. Of the two soil preparation techniques (extraction of moist-crumbled soil or freeze-dried and ground soil) the average 35s specific activity in a CaP-S extract from a freeze-dried sample of the top 6 cm of pasture soil was most closely related to that of S taken up by plants over several periods of plant growth. Calcium chloride (CaCI-S) extracts from freeze-dried soil and CaCI-S and CaP-S extracts from moist soils had 35s specific activities that were generally higher than that of S taken up by plants. Freeze-drying and grinding caused significantly more S, but not 35s, to be extracted from soil samples. Therefore, freeze-drying exposes a form of soil S to extraction that was not freely exchangeable with added sulphate-35s even duri ng 1 50 days of plant growth. Results indicated that plant roots had access to forms of soil S that could not be accessed by soil water and exchangeable sulphate ions ( i .e . extracted by CaCI2 from moist soil) during 1 50 days of plant growth. Such forms are likely to be: (a) organic S that is mineralized during plant growth or (b) aggregate-protected, adsorbed sulphate (So4=) , which becomes accessible to the penetrating roots or root hairs during plant growth but not to CaCI2 extraction of moist soil . 9.6 INFLUENCE OF FERTILIZER FORM ON THE FATE OF S IN SOIL 264 The influence of s0 particle size and granulation of S0 with and without phosphate rock on the transformations of S in soil were studied using undisturbed soil cores and 35s labelled fertilizer. In addition, the relative fate of S from SSP and gypsum was also studied. Some of the soil cores were removed to a glasshouse where controlled simulated rainfall events were applied. Both glasshouse and field studies continued for 1 80 days. As other authors have observed, there was a marked decreased in the rate of S0 oxidation as s0 particle diameter increased from <0. 1 50 mm to 0 .1 50-0.250 to 250-500 mm. These findings also indicated that the efficiency of plant use of S0 can be improved by decreasing S0 particle size. Most of the 35so4= oxidized from S 0 remained in the top 33 mm of soil and was transformed into soil organic S . Negl igible 35so 4 = was leached from glasshouse o r field cores which contrasted with the observations in Chapter 5 . Granulation of the finest S0 particle size (<0. 1 50 mm) with or without phosphate rock had little practical longer-term effect on its rate of oxidation or fate in field or glasshouse soils. This i ndicated that provided granules can d isperse into their component particles, granular S0 products should be acceptable for agronomic use. Under the simulated environmental conditions of the glasshouse study, plants took up more fertilizer S from gypsum and SSP than from fine S0 (<0.1 50 mm particle size) . The fate in soil of the S derived from e ither gypsum, SSP or fine S0 were essentially simi lar , with most fertilizer S being converted to soil organic S. There was marginally more leaching of S to the 6-1 0 cm soil depths with the sulphate-based fertilizers. The main indication from this study and that conducted in Chapter 5 is that unless severe leaching occurs immediately after the application of sulphate-based fertilizers then the fate in soil of the S applied either as sulphate or S0 will be similar. The major fate pathway wil l be i ncorporation into so i l organic matter. In the short term, plant uptake of fertilizer S will be greater, the more quickly the fertilizer becomes soluble , providing of course that conditions are suitable for plant growth. 9.7 MODELLING THE SHORT-TERM FATE OF FERTILIZER S IN SOIL A pre lim inary mechanist ic compu te r s imu lat ion mode l was modif ied to d escribed S transformations in the soil-plant system of the pasture studied. lt provided a very accurate 265 method of predicting plant uptake of S from both so4=-s fertilized and unfertilized soil cores . If al l the assumptions made to construct the model are true the model i nd icated that on average actual rates of mineralization and immobil ization exceed root uptake of S by 1 .5 to 2 fold (mg S turned over per unit of S taken up by plants). The assumptio ns made i n the model were to be tested by attempt ing to pred ict t he redistribution of fertilizer 35s between plant and soil S forms using the measured data from Chapter 5. There was only a relative similarity between observed and predicted proportional 35s distribution between soil and p lant S forms. The model was relatively inaccurate at predicting the actual amounts of isotope present in each form and could not simulate the large loss of isotope that had previously (Chapter 5) been attributed to leaching. lt was not possible, the refore , to ve rify any of the assumptions made in the model or the indicated rate of minera l izat ion. The relative sim ilarity between observed and predicted 35s proportional distribution between soi l S forms g ives some support to using root activity as a modifier of mineralization and immobil ization rates to describe the extent of this process in different soil depths. 9? SUGGESTIONS FOR FURTHER RESEARCH As S is a much more mobile nutrient in soil than P , understanding S transformations in soil remains the key to understanding the sustainabil ity of New Zealand legume-based pasture systems. Techniques need to be developed to overcome the variability inherent in analyzing mixtures of particulate ferti l izers and soils, particularly, measurements of total S and residual S0. The inab i l ity to remove th is variab i l ity in the experiments d iscussed above has l im ited the i nterpretation of the resu lts . A sequential extraction p rocedure may be appropriate for determining the S0 , CaP-S, total S and organic S on the same soil sample. In this study, CaP? extracts of freeze-dried soil reflected the same S transformations (same time dependent-35s specific activity) as S taken up by plants , however, more studies of undisturbed soils are required to determine the actual size of the labile S pool which CaP and plants sample. Results suggest that the climate d ifferences (temperature and soil moisture content) between g lasshouse and f ie ld soil cores had parallel effects on S0 ox idat ion , mineralization of soil organic S and plant S uptake as the ratio at which soil S and 35so were made available stayed constant . This effect needs i nvest igat i ng further because they Indicate that g lasshouse 266 studies, us ing undisturbed soil cores, may be d irectly relevant to field conditions when a fertilizer with biologically contro lled release rate (S0 oxidation) is being evaluated. Further research on field transformations ( incorporation into soil o rganic S and l eaching) of applied fertilizer S should focus on changes in these processes during the first 30 days after fertilizer application because much t ransformation of the applied S into soil organic S mainly occurred during this period. Studies should be undertaken in soils of different fertility status, so i l textu res and ra infa l l . Furthermore , un l ess leaching is severe , stud ies s hou ld be concentrated separately on the top 0-30 mm soil layer, rather than inclusion of soil at deeper depths, because the observed rates of 35s immobilization indicated high biological turnover of nutrients in this zone. This will result in a better estimat ion of the rate of immobilization of applied S fertilizers. Many assumpt ions made in the short term fate model need fu rther c larificat ion . The relationships between soil solution S and extractable S need further investigation to test and refine the model , however, modell ing plant uptake of S by evapotranspiration appears to accurately describe temporal changes in actual plant uptake both in this study and that of Sakadevan ( 1 991 ) . Care should be taken in soils of low S status where so4= diffusion to root may be more important than in soils of high S status. The importance of litter and dung return needs investigation and could be used to modify the model in order to test its ability to predict plant S uptake over longer periods of time. 267 BIBLIOGRAPHY AI-Khafaji, A.A. and Tabatabai, M.A. 1979. Effect of trace elements on arylsulfatase activity in soil. Soil Science 127: 129-133. Alexander, M. 1977. Introduction to Soil Microbiology. John Wiley & Sons Inc., London. Andrew, C.S. 1975. Evaluation of plant and soil sulphur tests in Australia. In Sulphur in Australasian Agriculture (Ed. by K.D. McLachlan). Sydney University Press, pp 1%-200. A.O.A.C., 1984. Sulfur in fertilizers. Association of Analytical Chemists, 14th edition, AOAC Inc., Arlington, Verginia, pp 32-34. Asher, C., Blarney, F. and Mamaril, C.P. 1983. Sulfur nutrition of tropical annual crops. In Sulphur in South East Asian and South Pacific Agriculture (Ed. G.J. Blair), pp 54-64. Attoe, O.J. and Olson, R.A. 1966. Factors affecting rate of oxidation in soils of elemental sulfur and that added in rock phosphate-sulfur fusions. Soil Science 101: 3 17-325. Autry, A.R. and Fitzgerald, J.W. 1991 . Potential for organic sulfur accumulation in a variety of forest soils in saturating sulfate concentration. Biology and Fertility of Soils 10: 281 -284. Aylmore, L.A.G. and Karim, M. 1968. Leaching of fertilizer ions in soil columns. Transactions of 9th International Congress of Soil Science. 1 : 143-155. Barber, S .A. 1962. A diffusion and mass-flow concept of soil nutrient availability. Soil Science 93: 39-49. Barber, S.A. 1984. Soil Nutrient Bioavailability, A mechanistic approach. John Wiley and Sons. 398 p. Barber, S .A., Walker, J.M. and Vassey, E.H. 1%3. Mechanisms for the movement of plant nutrients from soil and fertilizer to plant root. Journal of Agricultural and Food Chemistry 1 1 : 204- 207. Bardsley, C.D. and Kilmer, V.J. 1963. Sulfur supply of soils and crop yields in Southeastern United States. Soil Science Society of America Proceedings 27: 197-199. Barley, K.P. 1964. Earthworms and decay of plant litter and dung: A review. Proceedings of the Australian Society of Animal Production 5: 236-240. Barrow, N.J. 1960a. The effect of varying the nitrogen, sulphur, and phosphorus content of organic matter on its decomposition. Australian Journal of Agricultural Research 1 1 : 3 17-330. Barrow, N.J. 1960b. S timulated decomposition of soil organic matter during the decomposition of added organic material. Australian Journal Agricultural Research 1 1: 331-338. Barrow, N.J. 1960c. A comparison of the mineralization of nitrogen and of sulphur from decomposing organic materials. Australian Journal of Agricultural Research 1 1 : 960-969. Barrow, N.J. 1961a. Studies on mineralization of sulphur from soil organic matter. Australian Journal of Agricultural Science 12: 306-319. 268 Barrow, NJ. 1961b. Mineralization of nitrogen and sulphur from sheep faeces. Australian Journal of Agricultural Research 12: 644-650. Barrow, NJ. 1967a. Studies on extraction and availability to plants of adsorbed plus soluble sulfate. Soil Science 104: 242-249. Barrow, N.J. 1967b. Some investigation of the effect of grazing on the nutrition of pasture. The Journal of Australian Institute of Agricultural Science 33: 254-262. Barrow, N.J. 1968. Determination of elemental sulphur in soils. Journal of Science , Food and Agriculture. 19: 454-456. Barrow, NJ. 1969a. Effects of adsorption of sulfate by soils on the amount of sulfate present and its availability to plants. Soil Science 108: 193-201 . Barrow, N.J. 1969b. The accumulation of soil organic matter under pasture and its effect on soil properties. Australian Journal of Experimental Agriculture and Animal Husbandry 9: 437- 444. Barrow, NJ. 1970. Sulphate adsorption by soil as a determinant of sulphur status and methods of supplying sulphur on non-adsorbing soils. Proceedings of 1 1 th International Grassland Congress. 370-373. Barrow, N J. 1975. Reaction of fertilizer sulphate in soils. In Sulphur in Australasian Agriculture. (Ed. K.D. McLachlan) Sydney University press, pp 50-54. Barrow, NJ. 1985. Reaction of anions and cations with variable-charge soils. Advance in Agronomy 38: 183-230. Barrow, N.J. and Lambourne, LJ. 1962. Partitioning of excreted nitrogen, sulphur and phosphorus between the faeces and urine of sheep being fed pasture. Australian Journal of Agricultural Research 12: 461-471. Barrow, N.J., Spencer, K. and McArthur, W.M. 1969. Effect of rainfall and parent material on the ability of soils to adsorb sulfate. Soil Science 108: 120-126. Bates, CJ. 1971 . Factor affecting critical nutrient concentrations in plants and their evaluation - A review. Soil Science 122: 1 16-130. Beaton, J.D. 1987. Desirable characteristics of sulphur containing fertilizers suitable for Asia and the Pacific region. Proceeding of the Symposium on Fertilizer Sulphur Requirements and Sources in Developing Countries of Asia and the Pacific, Bangkok, pp 100-1 1 1. Beaton, J.D., Burns, G.R. and Platou, J. 1966. Determination of sulphur in soil and plant material. The Sulphur Institute. Technical Bulletin No. 14. 56 p. Beaufils, E.R. 1973. Diagnosis and recommendation integrated system (DRIS). Soil Science, Bulletin No. 1: 1- 132, University of Natal, Pietermaritzburg, South Africa. Beckman Liquid Scintillation systems. Model LS 1801/3801/5801 , Operating Manual. Bettany, J.R., Saggar, S . and Stewart, J.W.B. 1980. Comparison of the amounts and forms of sulphur in soil organic matter fractions after 65 years of cultivation. Soil Science Soc iety of America Journal 44: 70-75. 269 Bettany, J.R., Stewart, J.W.B. and Halstead, E .H. 1973. Sulfur fractions and carbon, nitrogen and sulfur relationships in grassland, forest and associated transitional soils. Soil Science Society of America Proceedings 37: 9 15-918. Bettany, J.R., Stewart, J.W.B. and Halstead, E.H. 1974. Assessment of Available soil sulphate in an 35s growth chamber experiment. Canadian Journal of Soil Science 54: 309-315. Bettany, J.R., Stewart, J.W.B. and Saggar, S. 1979. The nature and forms of sulfur in organic matter fractions of soils selected along an environmental gradient. Soil Science Society of America Journal 43: 981-985. Biederbeck, V.O. 1978. Soil organic sulfur and fertility. In Soil Organic Matter. (Eds. M. Schnitzer and S .U. Khan) Elsevier Scientific Publising Co., New York, pp 273-3 10. Bird, P.R. 1971 . Sulphur metabolism and excretion studies in ruminants, 2. organic and inorganic sulphur excretion by sheep after intraruminalor intraduodenal infusion of sodium sulfate. Australian Journal of Biological Science 24: 1329-1339. Blair, G.J. 1979. Sulfur in the tropics. The Sulphur Institute. 69 p. Blair, G.J., Chinoim, N., Lefroy, R.B.D., Anderson, G.C. and Cracker, G.J. 1991 . A soil sulphur test for pastures and crops. Australian Journal of Soil Research 29: 619-626. Blakemore, L.C., Collie, T.W. and Metson, A.J. 1968. Sulphur. New Zealand Soil Bureau Bulletin No 26(2), pp 103-109. Blakemore, L.C., Ludecke, T.E., Leamy, M.L and Metson, A.J. 1969. A study of the pattern of sulphur response shown by dry-land lucerne on some brown-grey earths. New Zealand Journal of Agricultural Research 12: 333-351. Blanchar, R.W. 1986. Measurement of sulfur in soils and plants. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 455-490. Bloomfield, M. 1%7. Effect of some phosphate fertilizers on the oxidation of elemental sulphur in soil. Soil Science 103: 219-223. Bohn, H.L., McNeal, N.L. and O'Conner, G.A. 1979. Soil Chemistry. John Wiley & Sons, New York. 329 p. Bohn, H.L., Barrow, N.J., Rajan, S .S .S. , and Parfitt, R.L. 1986. Reaction of inorganic sulfur in soils. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 233-250. Bolan, N.S., Scatter, D.R., Syers, J.K. and Tillman, R.W. 1986. The effect of adsorption on sulphate leaching. Soil Science Society of America Journal 50: 1419-1424. Bolan, N.S., Syers, J .K. and Sumner, M.E. 1991 . Calcium induce sulfate adsorption by soil. in press. Bolan, N.S., Hedley, M.J., Syers, J.K. and Tillman, R.W. 1987. Single superphosphate-reactive phosphate rock mixtures 1 . Factors affecting chemical composition. Fertilizer Research 13: 223-239. Bolan, N.S . , Syers, J .K., Tillman, R.W. and Scatter, D.R. 1988. Effect of liming and phosphate application on sulphate leaching. Journal of Soil Science 39: 493-504. 270 Bornemissza, G.F. and Williams, C.H. 1 970. An effect of dung beetle activity on plant yield. Pedobiologia 10: 1 -7. Boswell, C.C. 1983. An Assessment of Phosphorus and Sulphur Interactions on Aspects of the Sulphur Cycle in Pasture. Ph.D. thesis, University of New England, Australia. Boswell, C.C. 1985. Sulphur fertilizer strategies for the post-superphosphate era. Proceedings of 20th Technical Conference, New Zealand Fertilizer Manufacturer Research Association, pp 554- 603. Boswell, C.C. 1988. Sulphur fertilizer requirements. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 253-256. Boswell, C.C. and Swanney, B. 1986. Alternative sulphur fertilizers in New Zealand. Proceedings of the New Zealand Grassland Association 47: 233-242. Boswell, C.C. and Swanney, B. 1988. The effect of particle size, rate and frequency of application of elemental sulphur on the release of sulphate to lucerne in the upper Waitaki basin. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 159-168. Boswell, C.C. and Swanney, B .S . 199 1 . Comparison of the long-term effects of different sulphur fertilizers on legume pasture production on a topo-sequence of North Otago soils, New Zealand. New Zealand Journal of Agricultural Research 34: 95-104. Boswell, C.C., Owers, W.R. and Swanney, B. 1988a. Sulphur/sodium bentonite mixtures as fertilizers 1 . the effect of S/Na-betonite ratio on the rate of dispersion and particle size distribution of elemental sulphur dispersed from laboratory-produced priUs. Fertilizer Research 15 : 13-3 1 . Boswell, C.C. , Swanney, B . and Owers, W.R. 1988b. Sulphur/sodium betonite prills a s sulfur fertilizers. 2. Effect of sulfur-sodium bentonite ratio on the availability of sulfur to pasture plants in the field. Fertilizer Research 15: 33-46. Bouma, D. 1967. Nutrient uptake and distribution in subterranean clover during recovery from nutritional stress 2. experiment with sulphur. Australian Journal of Biological Science 20: 613-621 . Bouma, D. , Spencer, K. and Dowling, EJ. 1969. Assessment of the phosphorus and sulphur status of subterranean clover pastures 3. plant tests. Australian Journal of Experimental Agriculture and Animal Husbandry 9: 329-340. Bouma, J. 1980. Field measurement of soil hydraulic properties characterizing water movement through swelling clay soils. Journal of Hydrology 45: 149-158. Bremner, J.M. and Tabatabai, M.A. 1971 . Use of automated combustion techniques for total carbon, total nitrogen and total sulfur analyses of soils. In Instrumental Methods for Analysis of Soil and Plant Tissue. Soil Science Society of America, Inc., Madison. Bromfield, S .M. 1 96 1 . Sheep faeces in relation to the phosphorus cycle under pasture. Australian Journal of Agricultural Research 12: 1 1 1 -123 . Bromfield, S .M. and Jone, O.L. 1970. The effect of sheep on recycling of phosphorous in hayed-off pasture. Australian Journal of Agricultural Research 21 : 699-71 1 . 271 Brown, K.A. 1982. Sulphur in the environment: a review. Environment Pollution. Series B. 3: 47-80. Brown, K.A. 1985. Sulfur distribution and metabolism in waterlogged peat. Soil Biology and Biochemistry 17: 39-43. Burns, R.G. 1978. Soil Enzymes. Academic Press, London, 386 p. Camberato, J.J. and Kamprath, E.J. 1986. Solubility of adsorbed sulfate in coastal plain soils. Soil Science 142: 21 1-213. Castellano, S .D. and Dick, R.P. 1988. Distribution of sulfur fractions in soil as influenced by management of organic residues. Soil Science Society of America Journal 25: 1403-1407. Chaiwanakupt, S . , Parkpain, P., Pongsakul, P. and Phetchawee, S. 1 987. Sulphur status and requirements in Thailand. Proceeding of the Symposium on Fertilizer Sulphur Requirements and Sources in Developing Countries of Asia and the Pacific, Bangkok, pp 89-94. Chao, T.T., Harward, M.E. and Chang, S.C. 1962a. Movevement of 35s tagged sulfate through soil columns. Soil Science Society of America Journal 26: 27-32. Chao, T.T., Harward, M.E. and Fang, S.C. 1962b. Soil constituents and properties in the adsorption of sulfate ions. Soil Science 94: 276-283. Chapman, S.J. 1987a. Microbial sulphur in some Scottish soils. Soil Biology and Biochemistry 19: 301-305. Chapman, S .J. 1987b. Partitioning of ryegrass residue sulphur between the soil microbial biomass, other soil sulphur pools and ryegrass (Lolium perenne L). Biology and Fertility of Soils. 5; 253-257. Chapman, S.J. 1989. Oxidation of micronized elemental sulphur in soil. Plant and Soil 1 16: 69-76. Chatupote, W. 1990. An Investigation on Some Factors Influencing the Rate of Oxidation of Elemental Sulphur Fertilizer. Ph.D. thesis, Massey University, New Zealand. Chaudhry, LA. and Cornfield, A.H. 1967a. Effect of moisture content during incubation of soil treated with organic material on changes in sulphate and sulphide levels. Journal of the Science of Food and Agriculture 18: 38-4 1 . Chaudhry, LA. and Cornfield, A.H. 1967b. Effect of temperature on sulphate levels in aerobic and sulphide levels in anaerobic soil. Journal of the Science of Food and Agriculture 18: 82-84. Clark, P.E. and Paul E.A. 1970. The microflora of grassland. Advance in Agronomy 22: 375-435. Cooper, M. 1968. A comparison of five methods for determining the sulphur status of New Zealand Soils. 9th International Congress of Soil Science, Transaction. II: 263-271 . Cooper, M . and Hogg, D.E. 1966. Leaching losses o f sulphate from superphosphate and gypsum applied on soil from Rhyolitic pumice. New Zealand Journal of Agricultural Research 9: 345-349. Cornforth, I.S. 1981 . Soil testing and fertilizer requirements. Proceedings of Technical Conference of New Zealand Fertilizer Manufacturers Research Association 18: 46-53. 272 Comforth, I.S. and Sinclair, A.G. 1984. Fertilizer and Lime Recommendations for Pasture and Crops in New Zealand. 2nd revised edition, Ministry of Agriculture and Fisheries Bulletin New Zealand. Comforth, I.S . and Sinclair, A. G. 1986. The MAF fertilizer advisory service. Proceedings of the New Zealand Grassland Association 47: 197-202. Comforth, I.S., Smith, G.S. and Wheeler, D.M. 1983. Soil testing and plant analysis for sulphur. In Sulphur in New Zealand Agriculture. (Eds. P.E.H. Gregg and J.K.Syers) Occasional report No, 5, Massey University, Palmerston North, pp 49-60. Coughenour, M.B. 1978. Grassland Sulphur Cycle and Ecosystem Responses to Level of Sulphur Dioxide. Ph.D. thesis, Colorado State University, Fort Collins, Colorado. Coulter, J.D. 1973. Prediction of evapotranspiration from climatological data. Proceedings of Soil and Plant Water Symposium, Inf. Ser. No. 96, pp 38-45. Cowie, J .D. 1978. Soil and Agriculture of Kairanga, North Island, New Zealand. New Zealand Department of Scientific and Industrial Research. Soil Bulletin No. 33. Crouchley, G. and Sinclair, A.G. 1982. Response to sulphur and phosphorus fertilizers on improved pastures on some soils of W airarapa. New Zealand Journal of Experimental Agriculture 10: 185-189. Curll, M.L. 1982. The grass and clover content of pastures grazed by sheep. Herbage Abstracts 52: 403-41 1 . Curtin, D. and Syers, J.K. 1990. Mechanism of sulphate adsorption by two tropical soils. Journal of Soil Science 41 : 295-304. David, M.B . and Mitchell, M.J. 1.987. Transformations of inorganic and organic sulfur: Importance of sulfate flux in Adirondack forest soil. JAPCA, the International Journal of Air Pollution and control and Harzardous Waste Management. 37; 39-44. David, M.B., Mitchell, MJ. and Nakas, J.P. 1982. Organic and inorganic constituents of a forest soil and their relationship to microbial activity. Soil Science Society of America Journal 46: 847- 852. David, M.B., Mitchell, M.J. and Schnindler, S .C. 1984. Dynamics of organic and inorganic sulfur constituents in hardwood forest soils and their relationships to nutrient cycling. In Forest Soils and Treatment Impacts. (Ed. E .I. Stone) Proceedings North America Forest Soil Conference 6th Knoxville, Tennessee, pp 221-245. David, M.B., Mitchell, MJ. and Scott, T.J. 1987. Importance of biological processes in the sulphur budget of northern hardwood ecosystem. Biology and Fertility of Soils 5: 258-264. David, M.B., Schindler, S .C., Mitchell, M.J. and Strick, J.E. 1983. Importance of organic and inorganic sulfur to mineralization processes in forest soil. Soil Biology and Biochemistry 15: 671-677. Davidson, J.L. and Milthorpe FL. 1966a. Leaf growth in Dactylis glomerata following defoliation. Annals of Botany 30: 174- 185. Davidson, J.L. and Milthorpe, F.L. 1966b. The effect of defoliation on the carbon balance in Dactylis glomerata. Annals of Botany 30: 185-189. 273 Day, F.H. 1963. The Chemical Elements in Nature. George G. Harrolo & Co Ltd., London. De Long, W.A. and Lowe, L.E. 1962. Carbon-bonded sulphur in soil. Canadian Journal of Soil Science 42: 223. Dean, G.A. 1966. A simple colorimetric finish for the Johnson-Nishita micro-distillation of sulphur. Analyst 91 : 530-532. Destain, J.P., Guiot, J., Francios, E and Riga, A. 1989. Fertilizer nitrogen budgets for two dose of NaN03 dressings split-applied to winter wheat in microplot on a loam soil. Plant and Soil 1 17: 177-183. Devaud, E. and McFarlance, J.D. 1980. The fate of radioactive sulphur applied in irrigated lucerne. Australian Journal of Agricultural Research 31 : 887-897. Dick, W.A. and Tabatabai, M.A. 1977. An alkaline oxidation method for determination of total Phosphorus in soils. Soil Science Society of America Journal 41 : 5 1 1-514. Dijkshoorn, W. and Van Wijk, A.I. 1967. The sulphur requirement of plants as evident by the sulphur? nitrogen ratio in the organic matter: a review of published data. Plant and Soil 26: 129-157. Dijkshoom, W. Lampe, J.E.M. and Van Burg, D.F.G. 1960. A method for diagnosing the sulphur nutrition status of herbage. Plant and Soil 13: 227-241 . Doak. B.W. 1952. Some chemical changes in the nitrogenous constituents of urine voided on pasture. Journal of Agricultural Science, Cambridge 42: 162-17 1 . Donald, A.D. and Leslie, R.T. 1969. Population studies on the effective stage of some nematode parasites of sheep. Parasitology 59: 141-157. Doyle, P.T. and Moir, R.J. 1979. Sulfur and methionine metabolism in sheep 1 . first approximation of sulfur pools in and sulfur flows from the recticulo-rumen. Australian Journal Biological Science 32: 51-63. Duke, S .H. and Reisenauer, H.M. 1986. Roles and requirements of sulfur in plant nutrition. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 124-168. During, C. 1984. Fertilizers and Soils in New Zealand Farming. 3rd edition. Goverment printer, Wellington New Zealand. 355 p. During, C. and Cooper, M. 1974. Sulphate nutrition and movement in a soil with high sulphate sorption characteristics. New Zealand Journal of Experimental Agriculture 2: 45-51 . During, C. and Martin, D.J. 1968. Sulphate nutrition, movement and sorption, with special reference to a gley podzol, West coast, South island. New Zealand Journal of Agricultural Science 1 1 : 665-676. During, C. and McNaught, K.J. 1961 . The effect of cow urine in growth of pasture and uptake of nutrients. New Zealand Journal of Agricultural Research 4: 591-595. During, C. and Weeda, W.C. 1973. Some effect of cattle dung on soil properties, pasture production and nutrient uptake, 1 . dung as a sources of phosphorus. New Zealand Journal of Agricultural Research 16: 423-430. 274 Eaton, F.M. 1966. Sulfur In Diagnostic Criteria for Plant and Soils. (Ed. H.D. Chapman), University of California Press, pp 444-475. Ensminger, I.E. 1954. Some factor affecting the adsorption of sulfate by Alabama soils. Soil Science Society of America Proceedings 19: 259-264. Ensminger, L.E. 1958. Sulfur in relation to soil fertility. Alabama Agricultural experimental Station, Bulletin 312. 19 p. Evan, P.S. 1971 . Root growth of Lolium perrenne L. 2. effect of defoliation and chloroplast pigments. Botanical Gazette 1 15: 225-234. Faires, R., and Boswell, G.GJ. 1981. Radioisotope Laboratory Techniques. 4th edition, Butterworths. Fitzgerald, J.W. 1976. Sulfate ester formation and hydrolysis: a potentially important yet often ignored aspect of the sulfur cycle of aerobic soils. Bacteriological Reviews 40: 698-721 . Fitzgerald, J.W 1978. Naturally occurring organo sulfur compounds in soils. In Sulfur in the Environment, part If Ecological Impacts. (Ed. J.O. Nriagu), Wiley, New York. pp 391-443. Fitzgerald, J.W., Strickland, T.C. and Swank, T.W. 1982. Metabolic fate of inorganic sulfate in soil samples from undisturbed and managed forest ecosystems. Soil Biology and Biochemistry 14: 529-536. Fitzgerald J.W., Ash, J.T. and Strickland, T.C. 1983. Formation of organic sulfur in forest soils: a biologically mediated process. Canadian Journal of Forest Research 13 : 1077-1082. Fox, R.L. 1976. Sulphur requirements of sugarcane. Agronomy Journal 68: 891-896. Fox, R.L. 1980. Responses to sulphur by crops growing in highly weathered soils. Sulphur in Agriculture 4: 16-22. Fox, R.L., Asghar, M. and Cable, W. 1983 . Sulfate accretions in soils of tropics. In Sulphur in South East Asian and South Pacific Agriculture. (Ed. G.J. Blair), pp 39-53. Fox, R.L., Kang, B.T. and Nangju, D. 1977. Sulfur requirements of cowpeas and implications for production in the tropics. Agronomy Journal 69: 201-205. Fox, R.L., Olsen, R.A. and Rhoades, H.F. 1964. Evaluation of sulfur status of soils by plant and soil tests. Soil Science Society of America Proceedings 28: 243-246. Franklin, R.E. 197 1 . Cation effects on chloride, sulfate and phosphate uptake by excised roots. Soil Science 1 12: 343-347. Freney, J.R. 1958. Determination of water-soluble sulfate in soils. Soil Science 86: 241-244. Freney, J.R. 1961. Some observations on the nature of organic sulphur compounds in soils. Australian Journal of Agricultural Research 12: 424-432. Freney, J.R. 1967. Sulphur Containing Organics. In Soil Biochemistry. (Eds. A.D. McLaren and G.H. Peterson) Marcel Dekker Inc., New York. Freney, J.R. 1986. Form and reactions of organic sulfur compounds in soils. In Sulphur in Agriculture. (Ed. M.A. Tabatabai) , Agronomy series number 27, American Society of Agronomy, pp 207-232. 275 Freney, J.R. and Spencer, K. 1960. Soil sulphate changes in the presence and absence of growing plants. Australian Journal of Agricultural Research 1 1 : 339-345. Freney, J.R. and Stevenson, F.J. 1966. Organic sulfur transformations in soils. Soil Science 101 : 307- 3 16. Freney, J.R. and Swaby, R.J. 1975. Sulphur transformation in soils. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 22-30. Freney, J.R. and Williams C.H. 1983. The Sulphur Cycle in Soil. In Global Biogeochemical Sulphur Cycle. (Eds. M.V. Ivanov and J.R. Freney) John Wiley & Sons, pp 129-201 . Freney, J.R., Melville, G.E. and Williams, C.H. 1969. Extraction, chemical nature and properties of soil organic sulphur. Journal of Science of Food and Agriculture 20: 440-445. Freney, J.R., Melville, G.E. and Williams C.H. 1970. The determination of carbon-bonded sulphur in soil. Soil Science 109: 3 10-3 18. Freney, J.R.5 Melville, G.E. and Williams, C.H. 1971 . Organic sulphur fractions labelled by addition of 3 S-sulphate to soil. Soil Biology and Biochemistry 3 : 133-14 1 . Freney, J.R., Melville, G.E. and Williams, C.H. 1975. Soil organic matter fractions as sources of plant? available sulphur. Soil Biology and Biochemistry 4: 227-221 . Freney, J .R ., S tevenson, F.J. and Beaver, A.H. 1972. Sulphur containing amino-acids in soil hydrolysates. Soil Science 1 14: 486-476. Fuller, R.D., David, M.B. and Driscoll, C.T. 1985. Sulfate adsorption relationships in forest Spodosols of Northeastern USA. Soil Science Society of America Journal 49: 1034-1040. Garwood, E.A. 1967. Seasonal variation in appearance and growth of grass roots. Journal of the British Grassland Society 23: 1 17-128. Ghani, A., McLaren, R.G. and Swift, R.S. 1988. The incorporation and re-mineralization of 3Ss in soil organic sulphur fractions. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 32-46. Ghani, A., McLaren, R.G. and Swift, R.S. 1990. Seasonal fluctuations of sulphate and soil microbial biomass-S in the surface of a Wakanui soil. New Zealand Journal of Agricultural Research 33: 467-472. Ghani, A., McLaren, R.G. and Swift, R.S. 1991 . Sulphur mineralization in some New Zealand soils. Biology andFertility ofSoi/s 1 1 : 58-74. Gillingham, A.G. 1978. Phosphorus Cycling in Grazed, Steep Hill Country. Ph.D. thesis, Massey University, New Zealand. Gillingham, A.G. 1980. Phosphorus uptake and return in grazed, steep-hill country 1 . Pasture production dung and litter accumulation. New Zealand Journal of Agricultural Research 23: 3 13 -321 . Gillingham, A.G. and During, C . 1973. Pasture production and transfer o f fertility within long? established hill Pasture. New Zealand Journal of Experimental Agriculture 1: 227-232. 276 Gillingham, A.G., Syers, J.K., and Gregg, P.E.H. 1980. Phosphorus uptake and return in grazed, steep? hill country 1. above ground components of the phosphorus cycle. New Zealand Journal of Agricultural Research 23: 323-330. Gillman, G.P. 1973. Studies on some deep sandy soils in Cape York Peninsula, North Queensland 3. losses of applied phosphorus and sulphur. Australian Journal of Experimental Agriculture and Animal Husbandry 13: 418-422. Goedewaagen, M.A.J. and Schuurman, J.S. 1950. Root production by agricultural crops on arable land and grassland as source of organic matter in soil. Transaction 4th international Congress of Soil Science 2: 28-3 1. Goh, K.M. 1988. Sulphur transformation in field soils - a review summary. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 78-94. Goh, K.M. and Gregg, P.E.H. 1980. Re-utilization by perennial ryegrass (Lolium perenn L.) of labelled fertilizer sulphur incorporated in field-growth tops and roots of pasture plants added to soils. Fertilizer Research 1: 73-85. Goh, K.M. and Gregg, P.E.H. 1982a. Field study on the fate of radioactive sulphur fertilizer applied to pastures. Fertilizer Research 3: 337-351 . Goh, K.M. and Gregg, P.E.H. 1982b. Field studies with radioactive sulphur-labelled gypsum fertilizer Ill. Sulphur pools and their significance. New Zealand Journal of Agricultural Research 25: 135-139. Goh, K.M. and Nguyen, M.L. 1990. Effect of grazing animals on the plant availability of sulphur fertilizers in grazed pastures. Proceeding of New Zealand Grassland Association. 52: 181- 185. Goh, K.M. and Tsuji, T. 1979. Changes in soil sulphur during incubation with and without added sulphur. New Zealand Journal of Agricultural Research 22: 585-594. Goodrich, R.D. and Garrett, J .E. 1986. Sulfur in livestock nutrition. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 617-634. Grayston, S .J. and Germida, J.J. 1990. Influence of crop rhizospheres on populations and activity of heterotropic sulphur-oxidizing microorganisms. Soil Biology and Biochemistry 22: 457-463. Gregg, P.E.H. 1976. Field Investigations into the Fate of Fertilizer Sulphur Added to Pasture-Soil Systems. Ph.D. thesis, University of Canterbury, New Zealand. Gregg, P.E.H. and Goh, K.M. 1978. Field studies with radioactive sulphur-labelled gypsum fertilizer I. Soil factors affecting the movement of fertilizer sulphur. New Zealand Journal of Agricultural Research 21 : 593-601 . Gregg, P.E.H. and Goh, K.M. 1979. Field studies with radioactive sulphur-labelled gypsum fertilizer II. Climatic and management factors affecting the movement of fertilizer sulphur in soils. New Zealand Journal of Agricultural Research 22: 425-429. Gregg, P.E.H. and Goh, K.M. 1982. Field studies with radioactive sulphur-labelled gypsum fertilizer IV. recovery by pasture and soil component. New Zealand Journal of Agricultural Research 25: 345-353. 277 Gregg, P.E.H., Goh, K.M. and Bash, D.W. 1977. Isotope studies on the uptake of sulphur by plants 2. Uptake from various soil depths at several field sites. New Zealand Journal of Agricultural Research 20: 299-233. Halstead, R.L. and Rennie, P.J. 1977. The effects of sulphur on soils in Canada. In Sulphur and Its Organic Derivatives in Canadian Environment. Environmental Secretariat, National Research Council, Bubl. NRCC, 15015, Ottawa Gregory, P .J. 1988. Growth and function of plant roots. In Russell's Soil Condition and Plant Growth (Ed. A. Wild), John Wiley & Sons, New York. Harada, T. and Spencer B. 1960. Choline sulphate in fungi. Journal of General Microbiology 22: 520- 527. Harward, M.E. and Reisenauer, H.M. 1966. Reactions and movement of inorganic soil sulfur. Soil Science 101 : 326-335. Harward, M.E., Chao, T.T. and Fang, S.C. 1962. The sulfur status and sulfur supplying power of Oregon soils. Agronomy Journal 54: 101-106. Hasan, S.M., Fox, R.L. and Boyd, C.C. 1970. Solubility and availability of sorbed sulfate in Hawaiian soils. Soil Science Society of America Proceedings 34: 897-901 . Hayes, M.H.B. and Swift, R.S. 1978. The chemistry of soil organic matter colloids. In Chemistry of Soil Constituents. (Eds. D.J. Greenland and M.H.B. Hayes) John Wiley & Sons Ltd., New York, pp 179-320. Hayman, J.M. and Stacker, R.V. 1982. Soil Water extraction patterns under pasture and lucerne on two soil types in Mid Canterbury. Proceedings of Agronomy Society of New Zealand 12: 61- 63. Haynes, R.J. and Williams, P.H. 1991. Effect of grazing animals on the nutrient status of pasture soils. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities (Eds. R.E. White and L.D. Currie) Occasional Report No. 5, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 224-236. Hedley, MJ. and Stewart, J.W.B. 1982. Method to measure microbial phosphate in soils. Soil Biology and Biochemistry 14: 377-385. Hedley, MJ., Tillman, R.W. and Roger, P. 1990. The environment: nutrient losses from New Zealand agriculture. In Proceedings of the New Zealand Fertilizer Manufacturers' Research Association Conference, The Challenge for Fertilizer Research in 1990, pp 62-91 . Heng, L.K. 199 1 . Measurement and Modelling of Chloride and Sulphate Leaching from a Mole? Drained Soil. Ph.D. thesis, Massey University, New Zealand. Heng, L.K., White, R.E., Bolan, N.S. and Scatter, D.R. 1991 . Leaching losses of major nutrients from a mole-drained soil under pasture. New Zealand Journal of Agricultural Research 34: 325- 334. Herriott, J .B .O. and Well D.A. 1 963 . The grazing animal and sward productivity. Journal of Agricultural Science, Cambridge 61 : 89-99. Hesse, P.R. 1957. The effect of colloidal organic matter on the precipitation of Barium sulphate and modified method for determining soluble sulphate in soils. Analyst 82: 710-712. 278 Higgins, DJ. 1987. Some recent developments in sulphur fertilizer technology. Proceeding of the Symposium on Fertilizer Sulphur Requirements and Sources in Developing Countries of Asia and the Pacific, Bangkok, pp 1 12-122. Hilder, E.J. 1964. The distribution of plant nutrients by sheep at pasture. Proceeding of Australian Society of Animal Production 5: 241-248. Hilder, E.J. 1966. Distribution of excreta by sheep at pasture. Proceedings of the 10th International Grassland Congress, Helsinki. pp 977-981 . Hingston, FJ. 1959. The loss of applied phosphorus and sulphur from soils under pasture in W.A. The Journal of the Australian Institute of Agricultural Science 25: 209-213. Hingston, F.J . , Atkinson, R.J. , Posner, A.M. and Quirk, J.P. 1967. Specific adsorption of anions. Nature 215: 1459- 1461 . Hoeft, R.G. , Walsh, L.M. and Keeney, D.R. 1 973. Evaluation of various extractants for available sulfur. Soil Science Society of America Proceedings 37: 401-404. Hogg, D.E. 1965. Studies on leaching losses of sulphur from pumice soils, yellow brown loams and sands. New Zealand Grassland Association 27: 123-128. Hogg, D.E. and Toxopeus, M.RJ. 1966. Studies on the retention of sulphur by yellow-brown pumice soils. New Zealand Journal of Agricultural Research 9: 93-97. Hoque, S ., Heath, S .B . and Killham, K. 1987. Evaluation of method to asses adequacy of potential soil S supply to crops. Plant and Soil 101 : 3-8. Hue, N.V. , Adam, F. and Evan, C. V. 1 984. Plant available sulphur as measured by soil solution sulphate and phosphate extractable sulphate in Ultisol. Agronomy Journal 76: 726-730. I.A.E.A. 1976. Tracer manual on crops and soils. Technical report No. 171 . International Atomic Energy Agency, Vienna. Jackman, R.H. 1964a. Accumulation of organic matter in some New Zealand soils under permanent pasture II. rate of mineralization of organic matter and the supply of available nutrients. New Zealand Journal of Agricultural Research 7: 472-479. Jackman, R.H. 1964b. Accumulation of organic matter in some New Zealand soils under permanent pasture I. pattern of change of organic carbon, nitrogen, sulphur and phosphorus. New Zealand Journal of Agricultural Research 7: 445-471 . Janzen, H.H. and Bettany, J.R. 1987a. Measurement of sulfur oxidation in soil. Soil Science 143: 444-452. Janzen, H.H. and Bettany, J.R. 1987b. Effect of temperature and water potential on sulfur oxidation in soil. Soil Science 144: 8 1-89. Janzen, H.H. and Bettany, J.R. 1987c. Oxidation of elemental sulfur under field conditions in central Saskatchewan. Canadian Journal of Soil Science 67: 609-618. Jenkinson, D.S . 1976. The effect of biocidal treatments on metabolism in soils 4 . the decomposition of fumigated organisms in soil. Soil Biology and Biochemistry 8: 167-177. 279 Jenkinson, D.S., Powlson, D.S. and Weddweburn, R.W.W. 1976. The effect of biocidal on metabolism in soil 3. the relationship between soil bio-volume measured by optical microscopy, the flush of decomposition caused by fumigation. Soil Biology and Biochemistry 8: 189-202. Jensen, M.E. 1973. Consumptive Use of Water and Irrigation Water Requirements. American Society of Civil Engineers, New York. Johnson, C.M. and Nishita, H. 1952. Micro-estimation of sulphur in plant materials, soils and irrigation waters. Analytical Chemistry 24: 736-742. Johnson, D.W. and Cole, D.W. 1977. Sulfate mobility in an out-wash soil in western Washington. Water, Air and Soil Pollution 7: 489-495. Johnson, D.W. and Henderson, G.S . 1979. Sulfate adsorption and sulfur fractions in a highly weathered soil under a mixed deciduous forest. Soil Science 128: 34-40. Johnson, D.W. and Todd, D.E. 1983. Relationships among iron, aluminium, carbon and sulfate in variety of forest soils. Soil Science Society of America Journal 47: 792-800. Johnson, D.W., Hornbeck, J.W., Kelly, J.M., Swank, W.T. and Todd, D.E. 1979. Regional pattern of soil sulfate accumulation: relevance to ecosystem sulfur budgets. In Proceedings; Atmospheric Sulphur Deposition Environmental Impact and Health Effects. (Ed. D.S . Shriner, C.R. Lichmond and S .E. Lindberg), Ann Arbor Science Publishing Inc., the Butterworth Group, pp 507-520. Jones, J .B. and Isaac, R.A. 1972. Determination of sulfur in plant material using a Leco sulfur analyzer. Agricultural and Food Chemistry 20: 1292-1294. Jones, L.H.P., Cowling, D.W. and Lockyer, D.R. 1972a. Plant available and extractable sulphur in some soils of England and Wales. Soil Science 1 14: 104-1 14. Jones, M.B. 1964. Effect of Applied sulphur I. uptake of various California dryland pasture species. Agronomy Journal 56: 235-237. Jones, M.B. 1975. Plant assays for the recognition of sulphur deficiency. In Sulphur in Australasian Agriculture (Ed. K.D. McLachlan). Sydney University Press, pp 175-182. Jones, M.B. 1986. Sulfur availability indexes. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 549-566. Jones, M.B., Martin, W.E. and Ruckman, J.E. 1970. Effectiveness of various sulfur sources applied to annual type grasslands of California. Proceedings of 9th International Grassland Congress, pp 373-376. Jones, M.B., Martin, W.E. and Williams, W.A. 1968. Behavior of sulfate sulfur and elemental sulfur in three California soils in lysimeters. Soil Science Society of America Proceedings 32: 535- 540. Jones, M.B., Oh J.H. and Ruckman, J.E. 1970. Effect of phosphorus and sulphur fertilization on the nutritive value of subterranean clover. In Proceedings of 32rd Conference on New Zealand Grassland Association, pp 69-76. Jones, M.B . , Rickman, J.E. and Lawler, P.W. 1972b. Critical levels of sulphur in bur clover. Agronomy Journal 64: 55-58. 280 Jones, S.L. and Smith, I.K. 198 1 . Sulfate transport in cultured tobacco cells- effect of calcium and sulfate concentration. Plant Physiology 67: 445-448. Kamprath, E.J . , Nelson, W.L. and Fitts, J.W. 1956. The effect of pH, sulfate and phosphate concentrations on the adsorption of sulfate by soils. Soil Science Society of America Proceedings 21 : 463-466. Kane, A. 1983. How I handle my fertilizer program. Proceedings of the 1983 hill and high country. Centre for Resources Management, Lincoln, New Zealand, pp 45-48. Kenendy, A.P. and Till, A.R. 1981a. The distribution in soil and plant of 35s from sheep excreta. Australian Journal of Agricultural Research 32: 339-351 . Kennedy, A.P., and Till, A.R. 198 1b. Effect of rate of application of gypsum and rainfall on the movement of 35s in soil and plant. The Journal of Australian Institute of Agricultural Science. 47: 1 10-1 13. Kennedy, P.M. 1974. Utilization and excretion of sulphur in cattle fed on tropical roughages. Australian Journal of Agricultural Research 25: 1018-1022. King, K.L. and Hutchison, K.J. 1 976. The effects of sheep stocking intensity on the abundance and distribution of meso-fauna in pasture. Journal of Applied Ecology 13 : 41-45. Kirk, G.J.D. and Nye, P.H. 1986. A simple model for predicting the rate of dissolution of sparlingly soluble calcium phosphates in soil, 2. appliction of the model. Journal of Soil Sceince 37: 541-554. Kittams, H.A. , and Attoe, O.J. 1965. Availability of phosphorus in rock phosphate-sulphur fusions. Agronomy Journal 57: 331-334. Kononova, M.M. 1966. Soil Organic Matter. Pergamon Press, Oxford. Konopka, A.E., Miller, R.H., Sommers, L.E. 1986. Microbiology of the sulfur cycle. In Sulphur in Agriculture. (Ed. M.A. Tabatabai), Agronomy series number 27, American Society of Agronomy, pp 23-56. Korentajer, L., Byrnes, B.H. and Hellums, D.T. 1984. Leaching losses and plant recovery from various sulfur fertilizers. Soil Science Society of America Journal 48: 671-676. Kowalenko, C.G. and Low. L.E. 1975. Evaluation of several extraction methods and of a closed incubation method for soil sulfur mineralization. Canadian Journal of Soil Science 55: 1-8. Kowlenko, C.G. 1978. Organic nitrogen, phosphorus and sulphur. In Soil Organic Matter. (Eds. M. Schnitzer and S.U. Khan) Elsevier Scientific Publishing Co., New York, pp 95-136. Kurmarohita, K. 1973. Distribution, Nature and Availability of Sulphur in Soil of Thailand. Ph.D. thesis, University of Illinois. Lambert, M.G. , Clark, D.A., Costall, D.A. and Budding, P.J. 1988. Phosphorus and sulphur accumulation after long-term superphosphate application to grazed hill country. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2 , Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 95- 100. Landers, D.H., David, M.B . and Mitchell, M.J. 1 983. Analysis of organic and inorganic sulfur constituents in sediments, soils and water. International Journal of Environment, Analytical Chemistry 14: 245-256. 281 Langlands, J.P., Sutherland, H.A.M. and Playne, M.J. 1973. Sulfur as nutrient for merino sheep 2. utilization of sulfur in forage diets. British Journal of Nutrition 30: 537-543. Lantinga, E.A., Keuming, J.A., Groenwold, J. and Deeman, P J.A.G. 1987. Distribution of extreta nitrogen by grazing cattle and its effect on sward quality, herbage production and utilization. In Animal Mannure on Grassland and Fodder Crops. (Eds. H.G. Meer), Martinus Nijhoff Publishers. Laskin, A.I. and Lechevalier, H.A. 1973. Handbook of Microbiology vol 2. Microbial composition CRC press, Cleveland. 1060 p. Ledgard, S .F. and Upsdell, M.P. 1991. Sulphur inputs from rainfall throughout New Zealand. New Zealand Journal of Agricultural Research 34: 105- 1 1 1 . Ledgard, S .F., Sporsen, M.S., Brier G.J. and Saunders, W.M.H. 1991 . Temporal changes in various soil and plant sulphur analyses in relation to the development of pasture response to sulphur fertilizer. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities (Eds. R.E. White and L.D. Currie) occasional Report No. 5. Fertilizer and Lime Research centre, Massey University, Palmerston North, pp 1 15-121. Lee, A. 1983. Sulphur oxidation in soils. In Sulphur in New Zealand Agriculture. (Eds. P.E.H. Gregg and J.K.Syers) Occasional report No, 5, Massey University, Palmerston North, pp 29-38. Lee, A. and Boswell, C.C. 1988. Laboratory and field measurements of elemental sulphur oxidation in some New Zealand soils. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 101-108. Lee, A., Boswell, C.C. and Watkinson, J.H. 1988. Effect of particle size on the oxidation of elemental sulphur, Thiobacilli numbers, soil sulphate, and its availability to pasture. New Zealand Journal of Agricultural Research 31 : 179-186. Lee, A., Watkinson, J.H. and Lauren, D.R. 1988. Factors affecting oxidation rate of elemental sulphur in soil under a ryegrass dominant sward. Soil Biology and Biochemistry 20: 809-816. Lee, A., Watkinson, J.H. and Lauren, D.R. 1988. Factor affecting oxidation rates of elemental sulphur in a soil under ryegrass dominant sward. Soil Biology and Biochemistry 20: 809-8 16. Lee, A. , Watkinson, J.H., Orbell, G. and Bagyaraj, J. and Lauren, D.R. 1987. Factors influencing dissolution of phosphate rock and oxidation of elemental sulphur in some New Zealand soils. New Zealand Journal of Agricultural Research 30: 373-385. Lee, R. and Speir, T.W. 1979. Sulphur uptake by ryegrass and its relationships to inorganic and organic sulphur levels and sulphatase activity in soil. Plant and Soil 53: 407-425. Lee, R., Blakemore, E.J., Gibson, EJ. and Daly, B.K. 198 1. Effect of extraction time and charcoal treatment on the adsorbed sulphate values of several New Zealand topsails. Communication in Soil Science and Plant Analysis 12: 1 195-1206. Lee, R., B lakemore, L.C., Daly, B .K., Gibson, EJ., Speir, T.W. and Orchard, V.A. 1985. Sulphur supply to ryegrass during a pot trial and correlations with soil biological activity: the influence of two difference methods of determining the adsorbed sulphate status of soils. Communication in Soil Science and Plant Analysis 16: 97-1 17. 282 Lehninger, AL. 1982. Principles of Biochemistry. Worth Publishers Inc., New York. 101 1 p. Lewis, D.C. and Quirk, J.P. 1967. Phosphate diffusion in soil and uptake by plants; 3. 3 1p movement and uptake by plants as indicated by 32p autoradiography. Plant and Soil 26: 445-453. Lewis, D.C., Clarke, A.L. and Hall, W.B. 1987. Accumulation of nutrients and changes in soil properties of sandy soil under fertilized pasture in South-eastern South Australia 2. Total sulfur and nitrogen, organic carbon and pH. Australia Journal of Agricultural Research 25: 203-210. Li, P and Caldwell, A.C. 1966. The oxidation of elemental sulfur in soil. Soil Science Society of America Proceedings 30: 370-372. Lotero, J., Woodhouse, W.W. jr. and Petersen, R.G. 1966. Local Effect on fertility of urine voided by grazing cattle. Agronomy Journal 58: 262-265. Lowe, L.E. 1964. An approach to the study of the sulphur status of soils and its application to selected Quebec soils. Canadian Journal of Soil Science 44: 176-179. Lowe, L.E. 1965. Sulphur fractions of selected Alberta soil profiles of the Chernozem and Podzolic order. Canadian Journal of Soil Science 45: 297-303. Lowe, L.E. 1969a. Forms of sulfur in soils. Report of Meeting of Western Section, National Soil Fertility Committee, Edmonton, Alberta, pp 72-80. Lowe, L.E. 1969b. Distribution and properties of organic fraction in selected Alberta soils. Canadian Journal of Soil Science 49: 129-141 . Lowe, L.E. and De long, W.A. 1963. Carbon bonded sulphur i n selected Quebec soils. Canadian Journal of Soil Science 43: 151-155. Ludecke, T.E. 1965. Further aspects of sulphur nutrition of legumes. Proceedings of New Zealand Grassland Association 27: 129-138. MacDiarmid, B.N. and Watkin, B.R. 1971 . The cattle dung patch 1. Effect of dung patch on yield and botanical composition of surrounding and underlining pasture. Journal British Grassland Society 26: 239-245. MacDiarmid, B .N. and Watkin, B.R. 1972a. The cattle dung patch 2. The effect of dung patch on the chemical status of soil, and ammonia nitrogen losses from the patch. Journal of British Grassland Society 27: 43-48. MacDiarmid, B.N. and Watkin, B.R. 1972b. The cattle dung patch 3. Distribution and rate of decay of dung patches and their influence on grazing behavior. Journal of British Grassland Society 27: 48-54. MacFadyen, A. 1978. The role of fauna in decomposition processes in grasslands In Proceedings of the Symposium of Grassland Fauna. Scientific Proceedings of the Royal Dublin Society, Series A. 6(1 1) : 197-206. MacLusky, D.S. 1960. Some estimation of the area of pasture foule by excretion of dairy cows. Journal ofThe British Grassland Society 15: 181-188. Manzel, R.G. and Smith, SJ. 1984. Soil fertility and plant nutrition. In Isotope and Radiation in Agricultural Sciences. (Eds. M.F. L'annunziata and J.O. Legg), Academic Press. 292 p. 283 Martin, J.K. 1985. A single technique for field studies of distribution of 32P-labelled phosphate between plant and soil pools. Plant and soil 86: 4 15-423. Martin, R.J. 1990. Measurement of water use and pasture growth on Templeton silt loam. New Zealand Journal of Agricultural Research 33: 343-349. Martin-pervel, P. and Gagnard, J. 1986. Plant analysis and a guide to nutrient requirements for temperate and tropical crops. Lavoiser, Paris. May, P.F., Till, A.R. and Downes, A.M. 1968. Nutrient cycling in grazed pastures 1 . a preliminary investigation of the used of 35s gypsum. Australian Journal of Agricultural Research. 19: 531-543. May, P.F., Till, A.R. and Cumming, M.J. 1972. System analysis of 35s sulpur kinetics in pastures grazed by sheep. Journal of Applied Ecology 9: 25-49. May, P.F., Till, A.R. and Cumming, M.J. 1973. System analysis of the application methods on the entry of sulphur into pastures grazed by pasture. Journal of Applied Ecology. 10: 607-626. Maynard, D.G., Stewart, J.W.B. and Bettany, J.R 1983a. Sulphur and nitrogen mineralization in soil compared using two incubation techniques. Soil Biology and Biochemistry 15: 251-256. Maynard, D.G., Stewart, J.W.B. and Bettany, J.R. 1983b. Use of plant analysis to predict sulfur deficiency in rapeseed (Brassica napas and B. campestris). Canadian Journal of Soil science 63: 387-391. Maynard, D.G. , S tewart, J .W.B. and Bettany, J.R. 1985. The effects of plants on soil sulfur transformations. Soil Biology and Biochemistry 127: 127-134. McAneney, K.J. and Judd, H.J. 1983. Pasture production and water use measurements in Waikato. New Zealand Journal of Agricultural Research 26: 7-13 . McAuliffe, C . and Peech, M. 1949. Utilization of plant of phosphorus in farm manure, ! .Labelling of phosphorus in sheep with 32P. Soil Science 68: 179- 184. McAuliffe, C. and Peech, M. 1949. Utilization of plant of phosphorus in farm manure, 1. Availability to plants of organic and inorganic forms of phosphorus in sheep manure. Soil Science 68: 1 85-195. McCaskill, M.R. and Blair, G.J. 1 987. Particle size and soil texture effects on elemental sulfur oxidation. AgronomyJournal 79: 1079-1083. McCaskill, M.R. and Blair, G.J. 1988. Development of a simulation model of sulphur cycling in grazed pastures. Biogeochemistry 5: 165-181 . McCaskill, M.R. and Blair, G .J. 1 989. A model for the release of sulphur from elemental S and superphosphate. Fertilizer Research 19: 77-84. McCaskill, M.R. and Blair, G.J. 1990. A model of S , P and N uptake by a perennial pature, calibration and predictions. Fertilizer Research. 22: 173-179. McGill, W.B. and Cole, C. V. 1981. Comparative aspects of cycling of organic C, N, S and P through soil organic matter. Geoderma 26: 267-286. 284 McLachlan, K.D. and Marco, D.G. 1971 . Current and residual effects of calcium sulphate applications on sulphur deficient basaltic soil. Australia Journal of Agriculture and Animal Husbandry 1 1: 64-70. McLachlan, K.D. and De Marco, D.G. 1975. Changes in soil sulphur fractions with fertilizer additions and cropping treatment. Australian Journal of Soil Research 13: 169- 176. McLaren, R.G. and Swift, R.S. 1977. Changes in soil organic matter fractions due to the long term cultivation of soil. Journal of Soil Science 28: 445-453. McLaren, R.G., Ghani, A. and Swift, R.S. 1988. Studies on sulphur mineralization from soils using open incubation. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 66-75. McLaren, R.G., Keer, J.I. and Swift, R.S. 1985. Sulphur transformations in soils using sulphur-35 labelling. Soil Biology and Biochemistry 17: 73-79. McLay, C.A.D., Cameron, K.C. and McLaren, R.G. 1988. Sulphate leaching in undisturbed soil monoliths and repacked soil columns. In Towards the More Efficient Use of Soil and Fertilizer Sulphur (Eds. RE. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 129-136. McLaughlin, B. 1983. Taupaka Soil Survey-an exercise in classification and detail mapping in New Zealand hill country. M. Phil. thesis, Massey University, New Zealand. McNaught, KJ. 1970. Diagosis of mineral deficiencies in grass-legumes pasture by plant analysis. Proceedings of 11 International Grassland Congress, pp 334-338. McNaught, KJ. and Christoffels, PJ.E. 1961 . Effect of sulphur deficiency on sulphur and nitrogen levels in pastures and lucrene. New Zealand Journal Agricultural Research 4: 177- 196. McNaughton, K.J., Clothier, B.E. and Kerr, J.P. 1979. Evaporation from land surfaces. In Physical Hydrology- A New Zealand Experience. (Eds., D.L. Murray and L.D Ackroyd) New Zealand Hydrological Society, Wellington, pp 97-1 19. Mengel, K. and Kirkby, E.A. 1978. Principles of Plant Nutrition. International Potash Institute, Berne, Switzerland. Metson, AJ. 1969. Sulphur status of New Zealand soils. Proceedings ofTechnical Conference of New Zealand Fertilizer Manufacturers Association 12: 73-84. Metson, AJ. 1973. Plant Analysis as a Guide to Sulphur Status of Forage Grasses and Legumes. The Sulphur Institute. 24 p. Metson, A.J. 1979a. Sulphur in New Zealand soils 1 . A review of sulphur in soils with particular reference to adsorbed sulphate-sulphur. New Zealand Journal of Agricultural Research 22: 95- 1 14. Metson, A.J. 1979b. Sulphur in New Zealand Soils 3. Sulphur distribution in soils derived from parent materials of predominantly sedimentary origin. New Zealand Soil Bureau Scientific Report No. 41 : 42 p. Metson, AJ. 1979c. Sulphur in New Zealand Soils 4. Sulphur distribution in soils derived from parent materials of predominantly volcanic origin. New Zealand Soil Bureau Scientific Report No. 42: 37 p. 285 Metson, AJ. and Blakemore, L.C. 1978. Sulphate retention by New Zealand soils in relation to the competitive effect of phosphate. New Zealand Journal of Agricultural Research 21 : 243- 253. Meyer, B. 1977. Sulfur, Energy and Environment. Elsevier Scientific Publishing Company. Microsoft Inc. 1987. Microsoft QuicBASIC 4.0. Microsoft Inc. Middleton, K.R. and Toxopeus, M.KJ. 1973. Diagnosis and measurement of multiple soil deficiencies by subtractive technique. Plant and Soil 38: 219-226. Mini tab Inc. 1989. Minitab Reference Manual. Minitab Inc. Morton, J.D. and Baird, D.B. 1990. Spatial distribution of dung patches under sheep grazing. New Zealand Journal of Agricultural Research 33: 285-294. Moser, U.S. and Olson, R.V. 1953. Sulphur oxidation in four soils influenced by soil moisture tension and sulphur bacteria. Soil Science 76: 251-257. Muller, F.B . and McSweeney, G. 1974. Effect of granulation of superphosphates on movement of phosphorus and sulphur in two soils. New Zealand Journal of Agricultural Research 2: 53- 62. Murphy, J. and Riley, J.P. 1962. Modified single solution method for the determination of phosphate in natural waters. Analytica Chemica Acta 27: 3 1 -36. Murphy, M.D. 1990. Fifteen year of sulphur research in Ireland. Sulphur in Agriculture 14: 10-15. Murphy, M.D. and Borgan, J.M. 1988. Sulphur deficiency in herbage in Ireland 1 . cause and extent. Irish Journal of Agricultural Research 27: 83-90. Nearpass, D.C., Fried, M. and Kilmer, V J. 1961. Greenhouse measurement of available sulfur using radioactive sulfur. Soil Science Society of America Proceedings 25: 287-289. Neptune, A.M. , Tabatabai, M.A. and Hanway, J.J. 1975. Sulphur fractions and carbon-nitrogen? phosphorus-sulphur relationships in some Brazilian and Iowa soils. Soil Science Society of America Proceedings 39: 51-55. Nguyen, M.L. 1982. An Evaluation of S Topdressing Strategies in Eastland Pasture. M. Agr. Thesis, Massey University. New Zealand, New Zealand. Nguyen, M.L. 1988. Direct determination of elemental sulphur in soils and the application of this method in measuring oxidation potentials of elemental sulphur fertilizers. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R .E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 9-25. Nguyen, M.L. and Goh, K.M. 1990. Accumulation of soil sulphur fractions in grazed pasture receiving long-term superphosphate application. New Zealand Journal of Agricultural Research 33: 1 1 1-128. Nguyen, M.L. and Goh, K.M. 1991. Applicability of the MAP soil sulphate quick test and the MAF sulphur model in predicting short-term sulphur requirement of grazed pastures in Canterbury. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities. (Eds. R.E. White and L.D. Currie) Occasional Report No. 5, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 280-294. 286 Nguyen, M.L. and Rickard, D.S. 1988. Soil organic sulphur and its contribution to pasture growth in the absence of superphosphate application. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 292-297. Nguyen, M.L., Tillman, R.W. and Gregg, P.E.H. 1989a. Evaluation of sulphur status of East Coast pasture of North Island, New Zealand 1. soil and herbage survey. New Zealand Journal of Agricultural Research 32: 263-271 . Nguyen, ML., Tillman, R.W. and Gregg, P.E.H. 1989b. Evaluation of sulphur status of East Coast pasture of the North Island, New Zealand 2. field trials. New Zealand Journal of Agricultural Research 32: 273-280. Nguyen, M.L., Rickard, D.S. and McBride, S .D. 1989. Pasture production and changes in phosphorus and sulphur status in irrigated pastures receiving long-term application of supherphosphate. New Zealand Journal of Agricultural Research 32: 245-262. Nunn, RJ. and Dee, T.P. 1954. Superphosphate production: the influence of various factors on speed of reaction and the composition of the product. Journal of Science and Food and Agriculture 5: 257-265. Nye, P.H. and Tinker, P.B.H. 1977. Solute movement in the soil-root system. In Studies in Ecology No. 4. Scientific Publications. 342 p. Oertli, J.J. 1990. Limitation to the diagnostic information obtained from soil analysis. Fertilizer Research 26: 1 89-196. Olsen, S .R., Cole, C. V. Watanabe, F.S. and Dean, L.A. 1954. Estimation of available phosphorus in soil by extraction with sodium bicarbonate. United States Department of Agriculture, Circular 937. Parfitt, R.L. and Smart, R.C. 1978. The mechanism of sulfate adsorption on iron oxides. Soil Science Society of America Journal 42: 48-50. Payne, D. 1988. The behaviour of water in soil. In Russel/' s Soil Conditions and Plant Growth (Ed. A. Wild), John Wiley & sons, New York. Perrott, K.W. and Surathchandra, S .U. 1987. Nutrient and organic matter levels in a range of New Zealand soils under established pasture. New Zealand Journal of Agricultural Research 30: 249-259. Petersen, R.G., Lucal, H.L. and Woodhouse, W.W. 1956. Distribution of excreta by freely grazing cattle ant its effect on pasture fertility. Agronomy Journal 48: 440-444. Peverill, K.I. and Douglas, L.A. 1976. Use of undisturbed cores of surface soil for investigating leaching losses of sulphur and phosphorus. Geoderma. 193-199. Peverill, K.I., Briner, G.P. and Douglas, L.A. 1975. Changes in extractable sulphur and potassium levels in soil due to oven drying and storage. Australian Journal of Soil Research 13: 67-75. Peverill, K.I., Douglus, L.A. and Greenhill, N.B. 1977. Leaching losses of applied P and S from undisturbed cores of some Australian surface soils. Geoderma 19: 91-96. Pirela, HJ. and Tabatabai, M.A. 1988. Sulfur mineralization rates and potential of soils. Biology and Fertility of Soils 6: 26-32. 287 Pollok, J .A. 1975. A Compartive Study of Certain New Zealand and German Soil Formed From Loess. Ph.D. thesis, Friedrich-Wihelms Universitat, Bonn. Pollok, J.A and McLaughlin, B. 1986. The Soils of Taupaka Farm. Massey University, Taupaka Farm Series. Publication No. 3. Postgate, J.R. 1968. Inorganic Sulphur Chemistry. In The Sulphur Cycle. (Ed. G. Nickless) Elsevire Scientific Publishing Co., New York, pp 259-272. Priestly, C.H.B. and Taylor, R.J. 1972. On assessment of surface heat flux and evaporation using a large-scale parameters. Monthly Weather Review 100: 81-92. Probert, M.E. 1976. Studies of 'available' and isotopically exchangeable sulphur in some North Queensland soils. Plant and Soil 45: 461-475. Probert, M.E. and Jones, K.R . 1977. The use of soil analysis for predicting the response to sulphur of pasture legumes in the Australian tropics. Australian Journal of Agricultural Research 15: 137-146. Rajan, S .S .S. 1978. Sulfate adsorbed on hydrous alumina, ligannd displaced ? and changes in surface charge. Soil Science Society of America Journal 42: 39-44. Randall, P. and Sakai, H. 1983. Sulfur analytical methods for soils. plants and animal samples. In Sulphur in South East Asian and South Pacific Agriculture. (Ed. G.J. Blair), pp 255-269. Raymant, G.E. 1983. Prediction of response to sulphur by established sirato/grass pasture in Southeast Queensland. Australian Journal of Experimental Agriculture and Animal Husbandry 23: 280-287. Reisenauer, H.M. 1975. Soil assays for the recognition of sulphur deficiency. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 182-187. Reisenauer, H.M., Walsh, L.M. and Hoeft, R.G. 1973. Testing soil for sulfur, boron, molybdenum and chlorine. In Soil Testing and Plant Analysis. (Eds. L.M. Walsh and J.D. Beaton). Soil Science Society of America Inc.. pp 173-200. Rennenberg, H. 1984. The fate of excess sulpuur in higher plants. Annual Review of Plant Physiology 35: 121-153 . Reuter, DJ. and Robinson, J .B . 1986. Plant Analysis: an interpretation and manual. Itaka Press, Melbourne, Australia. Rhue, R.D. and Kamprath, E.J. 1973. Leaching losses of sulfur during winter months when applied as gypsum, elemental S or prilled S. Agronomy Journal 65: 603-605. Richard, I.R. and Wolton, K.M. 1976. The spatial distribution of excreta under intensive cattle grazing. Journal of British Grassland Society 3 1:89-92. Rickard, D.S., McBride, S .D. and Fitzgerald, P.D. 1986. The effect of soil moisture on pasture yield. New Zealand Journal of Agricultural Science 20: 7-12. Risk, W.H. and Boswell, C.C. 1988. Sulphate sulphur in the subsoil of Southland and Otago soils. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 1 18- 128. 288 Rogers, D.A. 1985. The importance of the form of sulphur and the sulphur to phosphorus ratio in fertilizers. Proceedings 20th Technical Conference of New Zea land Ferti lizer Manufacturers' Research Association, pp 573-583. Rorison, I.H. 1973. The effect of soil acidity on nutrient uptake and physiology of plants. In Acid Sulphate Soils. (Ed. H. Dost) Wageningen, International Institute of Land Reclamation and Improvement 1 : 223-254". Rothbaum, H.P., Sewell, J.R. and Evason, W. 1980. Water-degradable sulphur/anhydrite pellets for fertilizer use. New Zealand Journal of Science 23: 371-382. Rowarth, J.S. 1987. Phosphorus Cycling in Grazed Hill-Country Pasture. Ph.D. thesis, Massey University, New Zealand. Rowarth, J.S., Gillingham, A.G. and Syers, J.K. 1988. Effects of season and fertilizer rate on phosphate concentrations in pasture and sheep faeces. New Zealand Journal of Agricultural Research 31 : 187-193. Rowarth, J.S. , Gillingham, A.G., Tillman, R.W. and Syers, J.K. 1985. Release of phosphorus from sheep faeces on graze hill country pasture. New Zealand Journal of Agricultural Research 28: 497-504. Rowarth, J.S., Tillman, R.W., and Gillingham, A.G. 1990. comparison of release of phosphorus from monocalcium phosphate and sheep faeces. New Zealand Journal of Agricultural Research 33: 473-477. Roy, A.B. and Trudinger, P.A. 1970. The Biochemistry of Inorganic Compounds of Sulphur. Cambridge University Press, Cambridge 400 p. Sachdev, M.S. and Chhabra, P. 1974. Transformation of 35s labelled sulphate in aerobic and flooded soil conditions. Plant and Soil 41 : 335-341 . Saggar, S . 1980. Studies of Sulfur in Relationships to Carbon and Nitrogen in Soil Organic Matter. Ph.D. thesis, University of Saskatchewan. Saggar, S and Hedley, MJ. 1989. Some recent development and areas for future research into soil and fertilizer in New Zealand. SCOPE Workshop on Sulphur Cycle in the Wet/and and Terrestrial Ecosystems. Peterborough, Canada. Saggar, S . , Bettany, J.R. and Stewart, J.W.B. 1981a. Measurement of microbial sulfur in soil. Soil Biology and Biochemistry 13: 493-498. Saggar, S . , Bettany, J.R. and Stewart, J.W.B. 1981b. Sulfur transformation in relation to carbon and nitrogen in incubated soils. Soil Biology and Biochemistry 13: 499-51 1 . Saggar, S . , Mackay, A.D., Hedley, MJ., Lambert, M.G. and Clark, D.A. 1990a. A nutrient transfer model to explain the fate of phosphorus and sulphur in grazed hill-country pasture. Agriculture, Ecosystem and Environment 30: 295-3 15. Saggar, S . , Hedley, M.J. , Gillingham, A.G., Rowarth, J.S . , Richardson, S . , Bolan, N.S. and Gregg, P.E.H. 1990b. Predicting the fate of fertilizer sulphur in grazed hill country pasture by modelling the transfer and accumulation of soil phosphorus. New Zealand Journal of Agricultural Research 33: 129-138. 289 Saggar, S . , Hedley, M.J ., White, R.E., Gregg, P.E.H., Perrott, K.W. and Comforth, I.S. 1991 . A glasshouse assessment of the ability of Olsen and resin P extractions to predict the relative yield of ryegrass in soils fertilized with phosphate rock. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities. (Eds. R.E. White and L.D. Currie) Occasional Report No. 5, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 149- 150. Sakadevan, K. 1991 . Factors Influencing the Transformation and Fate of Sulphur and Nitrogen in Grazed Hill Country Pasture. Ph.D. thesis, Massey University, New Zealand. SAS. 1985. SAS User's Guide: Statistics. Version 5, SAS Institute Inc. 956 p. Saunders, W.M.H. 1965. Phosphate retention by New Zealand soils and its relationships to free sesquioxides, organic matter, and other soil properties. New Zealand Journal of Agricultural Research 8: 30-57. Saunders, W.M.H. and Cooper, D.M. 1975. Evaluation of soil and plant analysis to determine sulphur status of New Zealand soils. In Sulphur in Australasian Agriculture (Ed . by K.D. McLachlan). Sydney University Press, pp 188-195. Saunders, W.M.H. and Hogg, D.E. 1971 . Significance of phosphate and sulphate retention. Technical report 1971/1 Ruakura Soil Research Station. pp 1-9. Saunders, W.M.H., Cooper, D.M. and Sinclair, A.G. 198 1 . Soil and fertilizer- Sulphur: use and interpretation of sulphur in soil tests for pastures. Aglink, FPP 649, Ministry of Agriculture and Fisheries, Wellington, New Zealand, pp 1-2. Saunders, W.M.H., Cooper, D.M. and Gravett, I.M. 1988. Response to sulphur and comparison of soil tests for predicting sulphur status of soils for grazed clover-grass pastures in New Zealand. New Zealand Journal of Agricultural Research 31 : 195-203. Scharpenseel, W.H. and Krausse, R. 1963. Radiochromatographic investigations on the cycle of sulfate and the S-amino acid cysteine and methionine. Z. Pflanzenemahr, Dung Bodenk 101 : 1 1-23. Schindler, S.C. and Mitchell, M.J. 1987. Dynamics of 3Ss in horizons and leachates from a Hardwood forest Spodosol. Soil Biology and Biochemistry 19: 531- 538. Schindler, S.C., Mitchell, MJ., Scott T.J., Fuller, R.d. and Driscoll, C.T. 1986. Incorporation of 35s sulfate into inorganic and inorganic constituents of two forest soils. Soil Science Society of American Journal 50: 457-462. Schnug, E. 1990. Sulphur nutrition and quality of vegetables. Sulphur in Agriculture 14: 3-7. Scott, N.M. 1981. Evaluation of sulphur status of soil by plant and soil tests. Journal of the Science of Food and Agriculture 32: 173-191 . Scott, N.M. 1985. Sulphur i n soils and plant. I n Soil Organic Matter and Biological Activity. Development in plant and soil sciences 16: 379-402, (Eds. D Vaughan and R.E. Malcolm) Martinus Nijhoff/Dr W Junk Publishers, pp 379-402. Scatter, D.R., Clother, B.E. and Turner, M.A. 1979. The soil water balance in Fragiaqualf and its effect on pasture growth in central New Zealand. Australian Journal of Soil Research 17: 455-465. Searle, P .L. 1979. Measurement of adsorbed sulphate in soils : effects of varying soil:extractant ratios and method of measurement New Zealand Journal of Agricultural Research 22: 287-290. 290 Sear le, P.L. 1982. Sulphate retention studies of three Yellow-brown loams. Soil of New Zealand, Part 6, Department of Scientific and Industrial Research, pp 68-75. Searle, P.l . and Sparling, G.P. 1987. The determination of air-drying and storage conditions on the amounts of sulphate and phosphate extracted from a range of new Zealand topsails. Communication in Soil Science and Plant Analysis 18: 725-734. Searle, P.L., and Speir, T.W. 1988. An anion exchange membrane method for extracting soil sulphate and a research application. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 47-53. Sears. P.D. 1950. Soil fertility and pasture growth. Journal of British Grassland Society. 5: 267-280. Sears, P.D. and Newbold, R.P. 1942. The effect of sheep droppings on yield and botanical composition, and chemical composition of pasture 1. establishment of the trial, technique of measurement and result for 1940-41 season. New Zealand Journal of Agricultural Science and Technology A:24 : 36-61 . Sears, P.D. and Goodall, V.C. 1948. The effect of sheep droppings on yield, botanical composition and chemical composition of pasture 2. results for the years 1942-1944 and final summary of the trial. The New Zealand Journal of Science and Technology 30: 23 1 -250. Shamoot, S., McDonald, L. and Batholomew, W.W. 1968. Rhizo-deposition of organic debris in soil. Soil Science Society of America Proceedings 32: 8 17-820. Sheard, R.W., Jackman3 R.H. and Butler, G.W. 1978. Utilization by white clover and ryegrass so sulphur from 5s labelled gypsum. Communication in Soil Science and Plant Analysis 9: 935-954. Shedley, C .D. 1982. An Evaluation of Elemental Sulphur as a Pasture Fertilizer. Ph.D. thesis, University of New England, Australia. Shedley, C.D. and Till, A.R. and Blair, GJ. 1979. A radiotracer technique for studying of nutrient release from different fertilizer materials and its uptake by plants. Communication in Soil Science and Plant Analysis 10: 737-745. Shipley, R.A. and Clark, R.E. 1972. Tracer Method for in vivo Kinetics: theory and applications. Academic Press Inc. 330 p. Shock, C.C., Jones, M.B., Williams, W.A. and Center, C.M. 1983. Effect of sulfur fertilization on three annual range species 1 . Laughlin soil experiment. Agronomy Journal 75: 515-520. Siems, H.B . 1953. Chemistry and manufacture of superphosphates and phosphoric acid. In Fertilizer Technology and Resource. (Ed. K.D. Jacob), pp 167-204. Sinclair, A.G. 1 982. Soil and fertilizer sulphur: sulphate leaching index. AgLink, Ministry of Agriculture and Fishery, New Zealand. Sinclair, A.G. 1983. Problems in modelling sulphur requirements in New Zealand. In Sulphur in New Zealand Agriculture. (Eds. P.E.H. Gregg and J.K.Syers) Occasional report No, 5, Massey University, Palmerston North, pp 106-1 17. Sinclair, A. G. and Boswell, C. C. 1983. The agronomic properties of sulphur fertilizers. Proceedings of the 19th Technical Conference of the New Zealand Fertilizer Manufacturer' s Research Association, pp 477-389. 291 Sinclair, A.G. and Cornforth, I .S. 1981 . Superchoice: a model of calculating superphosphate requirement in grazed pastures. Proceedings Lincoln College Farmer Conference, pp 57-65. Sinclair, A.G. and Enright, D.P. 1983. Effectiveness of elemental sulphur fertilizers. In Sulphur in New Zealand Agriculture. (Eds. P.E.H. Gregg and J.K.Syers) Occasional report No, 5, Massey University, Palmerston North, pp 164-175. Sinclair, A.G. and Jones, M.B. 1991 . Use of the diagnosis and recommendation integrated system (DRIS) on New Zealand clover-based pastures. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities (Eds. R.E. White and L.D. Currie) Occasional Report No. 5, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 304-3 17. Sinclair, A.G. and Saunders, W.H.M. 1982. Sulphur fertilizer in New Zealand. Sulphur in Agriculture 6: 15-22. Sinclair, A. G. and Saunders, W.H.M. 1984. Sulphur. In Fertilizer and Lime Recommendations for Pastures and Crops in New Zealand (Eds. I.S. Cornforth and A.G. Sinclair) New Zealand Ministry of Agriculture and Fisheries, Wellington, pp 15-17. Sinclair, A.G., Boswell, C.C., Comforth, I.S ., Lee, A., Morgan, C., Morton, J., Nguyen, M.L., Saunders, W.H.M., Shannon, P., Smith R.G. and Whelan, G. 1985. Agronomic requirements of sulphur in New Zealand Pastures. Proceeding of the 20th Technical Conference of the New Zealand Fertilizer Manufacturer's Research Association, pp 538-572. Skjelbreid, E. and Nissen, P. 1980. Effect of Calcium on sulphate uptake by barley roots. Physiology and Plant 49: 383-387. Skrijka, P. 1987. Investigation of the fertilizer value of sheep excrements left of pasture. In Animal Manure on Grassland and Fodder Crops. (Ed. H.G. v.d. Meer), Martinus NijhoffPublishers, Netherlands, pp 325-327. Smith, C.M. 1976. Occurrence of Sulphur Deficiency in some Manawatu Soils. Diss. Dip. of Agri Sci. Massey University, New Zealand. Smith, C.M. 1979. Leaching and Runoff Losses of Sulphur and Potassium from a Tokomaru Soil. M. Agr. Thesis, Massey University, New Zealand. Smith, C.M., Gregg, P.E.H. and Tillman, R.W. 1983. Drainage losses of sulphur from a yellow-grey earth soil. New Zealand Journal of Agricultural Research 26: 363-371 . Smith, F. and Siregar, M.E. 1983. Sulfur requirements of tropical forages. In Sulphur in South East Asian and South Pacific Agriculture. (Ed.3 GJ. Blair), pp 76-86. Smith, R .G . and McDougall, D.B. 1988. Evaluation of sulphur/sodium bentonite prill as sulphur fertilizers in the Manawatu. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North pp 169-178. Sorn-srivichai, P. 1 980. Mineralization of Soil Organic Sulphur in the Tokomaru Silt Loam. M.Agr.Sci. thesis, Massey University, New Zealand. Som-srivichai, P 1985. An Evaluation of Water Extraction as Soil Testing Procedure for Phosphate. Ph.D. thesis, Massey University, New Zealand. 292 Sparling, G.P. 1985. The soil biomass. In Soil Organic Matter and Biological Activity. Development in plant and soil sciences (Eds. D Vaughan and R.E. Malcolm), Martinus Nijhoff/Dr W Junk Publishers 16: 223-263 Spading, G.P., Whale, K.N. and Ramsay, A.J. 1985. Quantifying the contribution from soil microbial biomass to extractable P level from fresh and air-dried soils. Australian Journal of Soil Research 23: 613-621 . Speir, G.A. and McGill, W.B . 1979. Effect of phosphorus addition and energy supply on acid phosphatase production and activity in soils. Soil Biology and Biochemistry 1 1 : 3-8. Speir, T.W. 1977. Studies on climo-sequence of soils in Tussock grasslands 2. urease, phosphatase, and sulphatase activities with other properties including plant available sulphur. New Zealand Journal of Science 20: 159-166. Spencer, K. 1975. Sulphur requirements of plants. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 98- 108. Spencer, K. and Freney, J.R. 1960. Comparison of several procedures for estimating sulphur status of soils. Australian Journal of Agricultural Research 2: 948-959. Spencer, K. and Glendinning, J.S. 1980. Critical soil test values for predicting the phosphorus and sulfur status of sub-humid temperate pastures. Australian Journal of Soil Research 18: 435- 445. Spencer, B. and Harada, T. 1960. The role of choline sulphate in the sulphur metabolism of fungi. Biochemistry Journal 77: 305-315 . Spencer, K. , Jones, M.B. and Freney, J.R. 1977a. Diagnostic indices for sulphur status of subterranean clover. Australian Journal of Agricultural Research 28: 401-4 12. Spencer, K., Freney, J.R. and Jones, M.B . 1977b. Diagnosis of sulphur deficiency in subterranean clover. Sulphur in Agriculture 1 : 12-17. Stevenson, FJ. 1986. Cycle of Soil. John Wiley & Sons, New York. Stewart, B .A. , Porter, L.K. and Viets, F.G. Jr. 1966a. Sulfur requirements for decomposition of cellulose and glucose in soil. Soil Science Society of America Proceedings 30: 355-358. Stewart, B .A., Porter, L.K. and Viets, F.G. Jr. 1966b. Effect of sulfur containing content of straw on rate of decomposition and plant growth. Soil Science Society of America Proceedings 30: 453-456. Strick, J.E. and Nakas, J.P. 1984. Calibration of microbial sulfur technique for use inn forest soils. Soil Biology and Biochemistry 16: 289-29 1 . Strickland, T.C. and Fitzgerald, J.W. 1 985. Incorporation o f sulphate-sulphur into organic matter extracts of litter and soil: involvement of ATP sulphurylase. Soil Biology and Biochemistry 17: 779-784. Strickland, T.C. and Fitzgerald, J.W. 1987. Bacterial production of organic sulphur in a forest litter extract Soil Biology and Biochemistry 19: 771-774. Strickland, T.C., Fitzgerald, J.W., Ash, J.T. and Swank, W.T. 1987. Organic sulfur transformations and sulfur pool sizes in soil and litter from a southern Aplachian hardwood forest. Soil Science 143: 453-458. 293 Stuanes, A.O. 1985. Yearly and seasonal variation in extractable soil sulphate at two locations in Norways. Acta Agriculture Scandinavian 35: 33-40. Stumm, W. and Morgan, JJ. 1970. Aquatic Chemistry. John Wiley & Sons Inc., New York. Swank, W.T. and Fitzgerald, J.W. 1984. Microbial transformation of sulfate in forest soils. Science 223: 182-184. Swanney, B., Boswell, C.C., Enwright, P.D. and Sinclair, A.G. 1988. A comparative field evaluation of some potential sulfur fertilizer materials. New Zealand Journal of Agricultural Research 16: 183-192,. Swift, R.S. 1977. Mineralization of nitrogen and sulphur from soil organic matter. In Soil Organic Studies. Proceedings; symposium IAEA Viena, pp 275-281 . Swift, R.S. 1983. Mineralization and immobilization of sulphur in soils. In Sulphur in New Zealand Agriculture. (Eds. P.E.H. Gregg and J.K.Syers) Occasional report No, 5, Massey University, Palmerston North, pp 18-28. Swift, R.S., McLaren, R.G. and Keer, J.I. 1988. Some method for the extraction and fractionation of soil organic sulphur. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 55-65. Syers, J.K. and Springett, J.A. 1984. Earthworms and Soil fertility. Plant and Soil 76: 93-104. Syers, J.K., Curtin, D. 1987. Soil and Fertilizer Fulphur in UK Agriculture. Paper read before The Fertilizer Society of London on 10 December 1987. Tabatabai, M.A. 1982. Sulfur. In Methods of Soil Analysis, Part 2, Chemical and Microbiological Properties. (Ed. A.L. Page), Soil Science Society of America Inc. Tabatabai, M.A. 1987. Physicochemical fate of sulfate in soils. JAPA the International Journal of Air Pollution and control and Harzardous Waste Management 37: 34-39. Tabatabai, M.A. and Al-khafaji, A.A. 1980. Comparison of nitrogen and sulfur mineralization in soils. Soil Science Society of America Journal 44: 1000-1006. Tabatabai, M.A. and Bremner, J.M. 1970a. An alkaline method for the determination of total sulfur in soils. Soil Science Society of America Proceedings 34: 62-65. Tabatabai, M.A. and Bremner, J.M. 1970b. Factors affecting soil arysulfatase activities. Soil Science Society of America Proceedings 34: 427-429. Tabatabai, M.A. and Bremner, J.M. 1972a. Forms of su1fur, and carbon, nitrogen and sulfur relationships, in Iowa soils. Soil Science 1 14: 380-386. Tabatabai, M.A. and Bremner, J.M. 1972b. Distribution of total and available sulfur in selected soils profiles. Agronomy Journal 64: 40-44. Takebe, I. 1960. Choline sulphate as major soluble component of condiospores of Aspergillus niger. Journal of General Applied Microbiology 6: 83-89. Tallis, G.H. and Donald, A.D. 1964. Model for the distribution on pasture of infective larvae of the gastrointestinal nematode parasites of sheep. Australian Journal of Biological Science 17: 504-513. 294 Taylor, J.A. 1980. Marino Sheep and the Interpaddock Pattern of Herbaceous Species on the Northern Tablelands of New South Wales, Australia. Ph.D. thesis, University of New England. Thorrold, B.S., O'conner, K.F. and White, J.G.H. 1985. Management influences on sheep behaviour, dung distribution and soil phosphate. Proceedings of the New Zealand Grassland Association 46: 127-134. Till, A.R. 1975. Sulphur cycle in grazed pasture. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 68-76. Till, A.R. 1979. Nutrient cycling. In Grassland Ecosystems of the World: Analysis of Grassland and Their Uses. (Ed., R.T. Coupland), Cambridge University Press, London, pp 277-285. Till, A.R. and Blair, GJ. 1978. The utilization by grass of sulphur and phosphorus from clover litter. Australian Journal of Agricultural Research 29: 235-242. Till, A.R. and May, P F. 1970a. Nutrients cycling in grazed pastures 2. Further observation with (35-S) gypsum. Australian Journal of Agricultural Research 21 : 253-260. Till, A.R., and May, P.F. 1970b. Nutrient cycling in grazed pastures 3. Studies on labelling of the grazed pasture system. Australian Journal of Agricultural Research 21 : 455-463. Till, A.R. and McCabe, T.P. 1 976. Sulphur leaching and lysimeter characterization. Soil Science 121 : 44-47. Till, A.R., and May, P.F. 1971 . Nutrient cycling in grazed pastures 4. the fate of sulphur-35 following its application to small area in grazed pasture. Australian Journal of Agricultural Research 22: 39 1 -400. Till, A.R., Blair, G .J. and Dalal, R.C. 1 982. Isotopic studies of the recycling of carbon, nitrogen, sulphur and phosphorus from plant material. In The Cycling of Carbon, Nitrogen Sulphur and Phosphorus in Terrestrial and Aquatic Ecosystems. (Ed. I.E. Galbally and J.R. Freney) Australian Academic of Science, pp 51-59. Till, A.R., Saunders, W.H.M., S inclair, A.G. 1 987. Plant nutrient requirements; Oceania. In World Regional Plant Requirements Plant Nutrient Sulphur, Trade, Promotion and Marketing. Sulphur 87. Proceedings, Houston, Texas pp 102-1 15. Tisdale, S.L. 1977. Sulphur in Forage Quality and Ruminant Nutrition. Technilcal Bulletin No. 22 The Sulphur Institute. Tisdale, S .L., Reneau Jr., R.B. and Platou, J .S . 1 986. Atlas of sulfur deficiencies. In Sulphur in Agriculture. (Ed. M.A. Tabatabai) , Agronomy series number 27, American Society of Agronomy, pp 295-322. Toxopeus, M.R.J. 1 970. Sulphur deficiency in grassland on young soils from volcanic ash and its possible prediction by laboratory tests. Proceeding of nth International Grassland Congress, pp 346-346. Toxopeus, M.R.J. and Gordon, R.B. 1971 . Rational fertilizer practice on pumice soils. Proceedings of New Zealand Grassland Association. 33: 35-41 . Troughton, A . 1957. The underground organ of herbage grasses. Bulletin 44. 163 p. 295 Trudinger, P.A. 1979. The biological of sulphur cycle. In Biogeochemical Cyling of Mineral Forming Elements. (Eds. P.A. Trudinger and DJ. Swaine) Elsevier, Amsterdam, pp 293-3 13 . Tsuji, T. and Goh, K.M. 1979. Evaluation of soil sulphur fractions as sources of plant-available sulphur using radioactive sulphur. New Zealand Journal of Agricultural Research 22: 595-602. Van Praag, H.J. 1973. Cycle interne de I' azote et du soufre dans horizons humiferes forestiers des sols bruns acides III cycle interne du soufre. Plant and Soil 39: 61-69. Vaughn, C.E. 1987. Sulphur soil test on Northern California subclover-annual grass pastures surface soils. Soil Science 143: 184-191 . Vitolin, M.I. and Swaby, R.J. 1969. Activity of sulphur-oxidizing microorganism in some Australian soils. Australian Journal of Soil Research 7: 171-183. Vose, P.B. 1980. Introduction to Nuclear Techniques in Agronomy and Plant Bilogy. Pergamon Press. 391 p. Wainwright, M. 1978. Microbial Sulphur Oxidation in Soil. Science Progress, Oxford, 65: 459-475. Wainwright, M. 1984. Sulphur Oxidation in soils. Advance in Agronomy. 37: 350-392. Wainwright, M., Nevel, W. and Grayston, SJ. 1986. Effect of organic matter on sulphur oxidation in soil and influence of sulphur oxidation in soil nitrification. Plant and Soil 96: 369-376. Walker, T.W. 1957. The sulphur cycle in grassland soils. Journal of British Grassland Society 12: 10- 18. Walker, T.W. 1968. Changes induced in the soil by pastoral farming. In Soil of New Zealand. part 1 , Soil Bureau Bulletin 26(1), New Zealand Department of Scientific and Industrial Research, pp 104- 109. Walker, T.W. and Adams, A.F.R. 1958a. Studies on soil organic matter I. influence of phosphorus content of parent materials on accumulations of carbon, nitrogen, sulphur and organic phosphorus in grassland soils. Soil Science 85: 307-318. Walker, T.W. and Adams, A.F.R. 1 959. Studies on soil organic matter 2. influence of increased leaching at various stages of weathering on level of carbon, nitrogen, sulphur, and organic and total phosphorus. Soil Science 87: 3-10. Walker, T.W. and Adams, A.F.R. 1958b. Competition for sulphur in a grass-clover association. Plant and Soil 9: 353-366. Walker, T.W. and Gregg, P.E.H. 1975. The occurrence of sulphur deficiency in New Zealand. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 145- 154. Walker, T.W., Thupa, B.K. and Adams, A.F.R. 1959. Studies on soil organic matter 3 . accumulation of carbon, nitrogen and sulphur organic and total phosphorus in improved grassland soil. Soil Science 87: 135-139. Watkin, B.R. 1954. The animal factor and level of nitrogen. Journal of British Grassland Society 30: 35-40. Watkinson, J.H. 1988. Rate equation for the oxidation of elemental sulphur in soil. In Towards the More Efficient Use of Soil and Fertilizer Sulphur. (Eds. R.E. White and L.D. Currie) Occasional Report No. 2 , Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 26-3 1 . 296 Watkinson, J.H. 1989. Measurement of the oxidation rate of elemental sulphur in soil. Australian Journal of Soil Research 27: 365-375. Watkinson, J.H. and Perrott, K.W. 1990. A new soil test for sulphate and mineralized organic sulphur. Proceeding of the new Zealand Fertlizer Manufacturer'S Research Association Conference, 1990 'the Challenges for Fertilizer research in 1990' , pp 188-201 . Watkinson, J.H., Lee, A. and Lauren, D.R. 1987. Measurement o f elemental sulfur in soil and sediments: Field sampling, sample storage, pretreatment, extraction and analysis by High? performance liquid chromatography. Australian Journal of Soil Research 25: 167-178. Watkinson, J.H., Perrott, K.W. and Thorrold, B.S. 1991 . Relationship between the MAF pasture development index of soil and extractable organic sulphur. In Soil and Plant Testing for Nutrient Deficiencies and Toxicities (Eds. R.E. White and L.D. Currie) Occasional Report No. 5, Fertilizer and Lime Research Centre, Massey University, Palmerston North, pp 66-71 . Watson, E.R. and Lapins, P. 1969. Loss of nitrogen from urine on soils from South-western Australia. Australian Journal of Experimental Agriculture and Animal Husbandry 9: 85-91 . Weeda, W.C. 1967. The effect of cattle dung patches on pasture growth, botanical composition and pasture utilization. New Zealand Journal of Agricultural Research 10: 150-159. Weir, R.G. 1 975. The oxidation of elemental sulphur and sulphides in soils. In Sulphur in Australasian Agriculture. (Ed. by K.D. McLachlan). Sydney University Press, pp 40-49. Westerman, D.T. 1974. Indexes of sulfur deficiency in alfalfa. 1 . extractable soil S04 =-s. Agronomy Journal 66: 578-581. Whitaker, R.H. 1970. Communities and ecosystems. In Current Concepts in Biology Series. The Macmillan Co., Collier-Macmillan Ltd., London. White, J.G. 1959. Mineralization of nitrogen and sulfur in sulfur deficient soil. New Zealand Journal of Agricultural Research 2: 255-258. Widdowson, J.P. 1970. Available of Sulphur in Iowa Soils. Ph.D. thesis, Iowa State University. Wilkinson, S.R. and Lowrey, R.W. 1973 . Cycling of mineral nutrient in pasture ecosystem. In Chemistry and Biochemistry of Herbage. (Eds. G.W. Buttler and R.W. Bailey) 2: 247-3 15. Williams, C.H. 1967. Some factors affecting the mineralization of organic sulphur in soils. Plant and Soil 26: 205-223. Williams, C.H. 1968. Seasonal fluctuations in mineral sulphur under subterranean clover pasture in Southern New South Wales. Australian Journal of Soil Research 6: 13 1-1 39. Williams, C.H. 1971a. Reaction of surface-applied superphosphate with soil, 1 . the fertilizer solution and its initial reaction with soil. Australian Journal of Soil Research 9: 83-94. Williams, C.H. 1971b. Reaction of surface-applied superphosphate with soil, 2. movement of the phosphate and sulphur into the soil. Australian Journal of Soil Research 9: 95-106. Williams, C.H. 1974. The chemical nature of sulphur in some New South Wales soils. In Handbook on Sulphur in Australian Agriculture. (K.D. McLachlan), CSIRO, Melbourne. 297 Williams, C.H. 1975. The chemical nature of sulphur compounds in soils. In Sulphur in Australasian Agriculture. (Ed. K.D. McLachlan). Sydney University Press, Sydney, pp 21-30. Williams, C.H. and Steinberges, A. 1958. Sulphur and phosphorus in some Eastern Australian soils. Australian Journal of Agricultural Research 9: 483-491 . Williams, C.H. and Steinbergs, A. 1959. Soil sulphur fractions as chemical indices of available sulphur in some Australian soils. Australian Journal of Agricultural Research 10: 340-352. Williams, C.H. and Steinbergs, A. 1962. The evaluation of plant-available sulphur in soils 1. The chemical nature of sulphate in some Australian soils. Plant and Soil 17: 279-294. Williams, C.H. and Steinbergs, A. 1 964. The evaluation of plant-available sulphur in soil II. the availability of adsorbed and soluble sulphates. Plant and Soil 21 : 50-62. Williams, P.H. 1988. The Fate of Potassium in Grazed Dairy Pastures. Ph.D. thesis, Massey University, New Zealand. Williams, P.H. and Morton, J.D. 1985. Pasture responses to superphosphate and elemental sulphur on soils of West Coast, South Island. New Zealand Journal of Experimental Agriculture 13 : 369-374. Williams, P.H., Gregg, P.E.H. and Hedley, M.J. 1990a. Fate of potassium in diary cow urine applied to intact cores. New Zealand Journal of Agricultural Research 33: 15 1 -158. Williams, P.H., Hedley, M.J.H. and Horne, D.J. 1988. The effect of draina? on some aspects of nutrient loss from silt loam under dairying . Proceedings of the 4 National Drainage Seminar, New Zealand. Williams, P.H., Hedley, M.J. and Gregg, P.E.H. 1989. Uptake of potassium and nitrogen by pasture from urine-affected soil. New Zealand Journal of Agricultural Research 32: 4 15-421 . Williams, P.H., Hedley, M.J. and Gregg, P.E.H. 1990b. The effect of preferential flow of dairy cow urine and simulate rainfall on movement of potassium through undisturbed top soil cores. Australian Journal of Soil Research 28: 857-867. Young, R.D., Westfall, D.G. and Colliver, G.W. 1985. Production, marketing and use of phosphorus fertilizer. In Fertilizer Technology and Use. (Ed. O.P. Engelstad), Soil Science Society of America Inc., Madison, USA. Zucker, A. and Zech, W. 1985. S ulfur status of four uncultivated soil profiles in Northern Bavaria. Geoderma 36: 229-240. 298 Appendix 3 . 1 Results from NaOBr digestions of plant material compared with median values published by the International Plant-Analytical Exchange, Wageningen Agricultural University, Netherlands. Batch Number Materials Total-S Median MAD21 analysed concentration (Y)2 (X) March-April, 1988 (33)3 .... . . . mmole kg- .... . . . . 99(1 .98)4 821 Bouvardia plant 98 7.5 868 Vine leaf 42(2.28) 42 5.5 757 Spinach 136(4.75) 127 6.5 757 Spinach 128(2.25) 126 6.0 869 Lettuce 74(2.53) 7 1 3 .0 565 Barley (grain) 43(0.8 1) 44 4.0 May-June, 1988 (40) 757 Spinach 1 18(1 .34) 1 3 1 6.0 854 Potato tuber 47(2.87) 44 6.0 841 Endive 1 1 1(3.45) l lO 5.0 870 Leek 150(4.56) 140 1 0.0 556 Letuce 49(0.78) 47 3 .0 September-October, 1988 (31) 551 Pea (grain) 81 (3.40) 75 3 .0 872 Gladiolus 36(1 .46) 32 3 .0 662 Oil plam (leaf) 60(4.66) 56 3 .0 677 Maiz (plant) 38(1 .84) 37 3.5 722 Amaryllis (shoot) l l0(2.54) 95 5 .0 757 Spinach 141(4.87) 128 7.0 January-February, 1989 (33) 890 Cauliflower 214(1 .59) 207 25.0 891 Courgettes 80(2. 18) 75 6.0 875 Cabbage 222(4.12) 217 16.0 875 Cabbage 229(4.90) 220 10.0 873 Gladiolus leaf 38(1 .04) 34 5.0 9 1 1 Tabac leaf mixture 142(3.65) 138 1 1 .0 1 MAD2 Median of Absolute Deviations (median of the absolutes of the observations minus their median) calculated after omitting unusual values 2 Regression of Y on X: Y = 1 .7 + 1.02X; ? = 0.99 3 = Number of laboratories from other countries which had reponed the analysis 4 = Numbers in brackets are standard deviations 299 Appendix 3.2 Calculation for 35s enrichment of fertilizer materials (a) Divide the desired liquid scintillation counting rate (B, Bequerels), [suppose the desired counting is 16.67 Bq or 1000 cpm] by counting efficiency (CE),[90%] , of the cocktail mixture, this yields the amount of radioactivity (AJ required per sample vial. = B/CE = 16.67* 100/90 = the desired count rate in V ml of extractable S (CaP-S) from the top 0-2.5 cm soil at the end of the experiment cpm or Bq per ml Bq Bq (b) Calculate the initial activity of (Ao> of 35s (half life, t 112 = 87.4 days) when the duration of the experiment is T days [ 150]. A0 = AtEXP(T*0.693/t112) = 18.52*EXP(150*0.693/87.4) Bq Bq (c) From the aliquot of extractant counted (V ml), [ 1 .00 ml], and the soil to extractant ratio [S g soil : E ml extr. soln. = 5:40], calculate radioactivity per g of soil (AS a> at time 0 (tJ. = A0*E/(V*S) 60.83*40/(1 *5) Bq g?1 soil Bq g?1 soil (d) Calculate the total amount of radioactivity to be added per gram of soil (AF 0) if, at steady, state extractable S makes up X percent [2%] of total S . AF0 = = AS0*100/X 486.66*100/2 Bq g?1 soil Bq g?1 soil (e) Calculate the total amount of radioactivity (AP J required from the desired agronomic specification. where AP0 = AF0*W*l000 = 24333. 15*0.442*1000 W = soil dry weight to 2.5 cm depth (of a soil core Bq Bq 15 cm in diameter) Kg This is a total amount needed for applying into a soil core of 15 cm in diameter. 300 Therefore specific activity of fertilizer (SAf) is SAf = AP 0/(R*D/10 ) Bq g- 1 S = 10.75* 106/(0.0177*30/10) Bq g-l S where D = soil area to be fertilized m2 [15 cm diameter] i.e. 0.0177 m2 kg ha-1 R = rate of S fertilization (30 kg S ha-1) 301 Appendix 3.3 The manufacturing of 35s labelled superphosphate Phosphate rock used in these experiments was a mixture of two unreactive phosphate rocks, Nauru Island and Christmas Island grade A (15% P) at the ratio of 1 : 1 . The stoichiometric equation for the reaction between apatite rock and sulphuric acid is as follows: According to the equation, the ratio of phosphate rock (molecular weight=1008 g) to superphosphate (1608 g) is about 1 : 1 .6, then: Amount of superphosphate required = s g Weight of 1 : 1 ; Christmas:Nauru RP required = S/1 .6 g = A g Weight of sulphuric acid (100%) required for 100% acidulationa Weight of 98% sulphuric acid (w/w) required = S/1 .6*0.62 g = D g = D/0.98 g or = W g = W/1 .84 ml Where 1 .84 is specific gravity of sulphuric acid. The acid is normally diluted to 65% (w/w) by addition of water. Therefore, the amount of water (B) needed to be added to make 65% sulphuric acid: B or = 98*W/65 = 98*W/65 g ml Therefore the superphosphate is prepared by mixing 'B' g of water and 'W' g of sulphuric acid (98% w/w) to 'A' g of phosphate rock. Before mixing, the dilute acid is heated to about 65-70 ?C. After mixing, the mixture is dried at room temperature (20 ?C). a Based on 0.62 acid rock ratio; Bolan et al., (1987) Appendix 4.1 302 Activities of 35s and 32p in pasture per unit area (square centimetre) as influenced by different application rates of 35s and 32p labelled dung and distances of pasture from dung. Distances PART A (FIRST HARVEST) RI R2 R3 R4 Means D1 Dung D2 . . . . . 35s KBq cm-2 ... .. D3 0. 1 1 0.32 0.45 0.01 0.02 0.06 0.00 0.00 0.01 0.00 0.00 0.00 0.03 0.09 O. I3 Lsd 5% Dung = 0.03; distance = 0.04; within dung = 0.06 c.v. % = 59.74 PART B (SECOND HARVEST) RI R2 R3 R4 Means . . . . . 35s KBq cm-2 .... . 0.09 0.25 0.36 0.02 0.03 0.08 O.OI O.OI 0.02 0.01 0.01 0.01 0.03 0.08 O. I2 Lsd 5% Dung = 0.03; distance = 0.04; within dung = 0.24 c.v. % = 45.I PART C (FIRST HARVEST) . . . . . 32p K.Bq cm-2 .. . . . RI 0.02 R2 O.OI R3 0.00 R4 0.00 Means 0.01 Lsd 5% Dung = 0.01 ; distance = O.OI; within dung = 0.06 c.v. % = 45.1 0.06 0.03 0.0 1 0.00 0.02 O.I3 0.05 0.01 O.OI 0.05 Means 0.29 0.03 0.00 0.00 0.23 0.04 O.OI 0.01 0.07 0.03 0.01 0.00 Appendix 4.2 303 Influence of dung application rates and distances of P.asture from 35s and 32P labelled dung on the percent recovery of 35s and 32p by pasture. Dung Distances DI D2 PART A (FIRST HARVEST) RI 35 .. ... % s .... . R2 R3 R4 Means 1 .95 0. 16 O.I7 0.07 0.59 Lsd 5% Dung = ns; distance = 0.25; within dung = 0.87 C.V. % = 5 1 .38 2.8I 0.24 O. IO 0.02 0.79 PART B (SECOND HARVEST) . . . . . % 35s ..... R I 1.54 R2 0.37 R3 0.55 R4 0.30 Means 0.69 Lsd 5% Dung = ns; distance = 0.25; within dung = 0.44 c.v. % = 49.I 2.23 0.36 0.46 O. I7 0.8 1 PART C (FIRST HARVEST) 32 .... . % P .. . . . R I 0.47 R2 0.24 R3 0.24 R4 0.22 Means 0.29 Lsd 5% Dung = ns; distance = O. I3; within dung = 0.23 c.v. % = 50.4 0.73 0.44 0.30 0. 13 0.40 D3 1 .99 0.34 O. I3 0.03 0.62 1 .59 0.44 0.33 O. IO 0.62 0.78 0.37 0.35 0. 1 1 0.40 Means 2.25 0.24 O. I3 0.04 1 .79 0.39 0.49 O. I9 0.66 0.35 0.29 O. I 5 Treatments Dl RI R2 R3 R4 D2 RI R2 R3 R4 D3 R1 R2 R3 R4 Appendix 4.3 Specific activity of 3Ss or 32P in pastures (S .A.) and the percentage of plant S and P derived from dung (%SDFD and %PDFD) calculated using the S.A. of 35s and 32p in the total pool (TP) 'and extractable pool (EP), CaP-S and Olsen-P, of dung samples. Harvest 1 Harvest 2 Harvest 1 35s 35s 32p S.A. %SDFD SA %SDFD SA %PDFD TP EP TP EP TP EP KBq mg-1 S % % KBq mg- l S % % KBq mg- 1 P % % 21 .8 1 .83 0.18 8 .4 0.70 0.07 2.8 1 .4 1 1 .1 3 1.8 0.15 0.02 1 .7 0 . 14 0.01 1 . 3 0.63 0.50 0.4 0.03 0.00 0.5 0.05 0.01 0.3 0.14 0.1 1 0.3 0.03 0.00 0.4 0.03 0.00 0.4 0. 18 0.14 59.7 5.02 0.5 1 19.7 1 .65 0.17 7.7 3.77 3.00 4.8 0.40 0.04 2.4 0.20 0.02 3.6 1 .76 1 .4 1 0.9 0.07 0.01 0.9 0.07 0.01 1 . 1 0.52 0.4 1 0.2 0.01 0.00 0.4 0.03 0.00 0.4 0.21 0. 17 63 . 1 5 .30 0.54 27.6 2.23 024 14.4 7.07 5.64 1 0.3 0.87 0.09 5.5 0 .45 0.05 5.7 2.8 1 2.24 1 .5 0 .13 0.01 1 .4 0 . 12 0.01 2.3 1 . 14 0.9 1 0.4 0.03 0.00 0.4 0.03 0.00 0.8 0.41 0.32 w s: Appendix 4.4 305 Percentage of plant S (PART A and B) and P (PART C) derived from dung applied at different rates (%SDFD and %PDFD)a in pasture at different radial distances away from dung. Dung weights (g) Distances 0.5 1 .0 PART A (FIRST HARVEST) . . . . . %SDFD ... . . R1 0.188 0.5 14 R2 0.016 0.041 R3 0.003 0.008 R4 0.003 0.001 Means 0.052 0.141 Lsd 5% Dung = 0.045; distance = 0.015; within dung = 0.08 c.v. % = 59.74 PART B (SECOND HARVEST) . . . . . %SDFD ... . . R l 0.071 0. 167 R2 0.014 0.021 R3 0.005 0.007 R4 0.003 0.003 Means 0.023 0.050 Lsd 5% Dung = 0.014; distance = 0.016; within dung = 0.028 c.v. % = 45.1 a a PART C (FIRST HARVEST) . . . . . %PDFD ... . . a RI 1 . 13 3.00 R2 0.49 1 .41 R3 0. 1 1 0.41 R4 0. 14 0. 16 Means 0.47 1 .24 Lsd 5% Dung = 0.014; distance =0.016; within dung = 0.96 C.V. % = 57.2 2.0 0.543 0.089 0.013 0.004 0.162 0.238 0.047 0.01 1 0.004 0.075 5.65 2.25 0.90 0.32 2.28 a calculated using specific activities ot35s anCJ32P in the exchangeable pool of dung S and P Means 0.4 15 0.049 0.008 0.003 0. 159 0.027 0.008 0.003 3.26 1.38 0.47 0.21 306 Appendix 4.5 Diffusion of sulphate and phosphate in soil. The initial flux of either phosphate or sulphate from under a moist dung pellet out into the soil can be described by the following equation (Nye and Tinker, 1977; Barber, 1984): Fp where ? p Assumption: = = = = = = = flux ofH2ro4=or S04= diffusion coefficient of H2P04 = (or S04 =) in water volumetric soil water content impedance factor reciprocal of buffer power; unit change, C, (flg mm-3 solution) in solution H2P04= or so4 = concentration with a unit change in the amount of soil surface, S, (llg mm-3 soil) H2P04 = (or so4=) concentration gradient between soil solution affected by the dung to the bulk soil solution. For the purpose of calculation, X, is considered to be 1 mm It is assumed that, for a 2 g dung pellet which contains 275 and 5454 mg S and P/kg, respectively, inititally covers an area of 10 cm2 and a volume of 10 cm3 and when wet contains 40% moisture. Then; soluble sulphate was measured at 0.55 mg/10 cm3, i.e. 0.055 mg/cm3 given a dung solution containing l38 flg S mm-3 and similarly; suluble phosphate was measured at 1 1.0 mg/10 cm3 i.e. 1 .1 mg/cm3 given a dung solution containing 2750 flg P mm-3. The following table presents parameters used for theoretical comparison of sulphate and phosphate flux away from dung. Ion D e f OCIOS OCIOX F mm2 s-1 flg mm-3 mm-1 flg mm-1 s- 1 H2P04= 1*10 -3 0.3 0.3 0.015a 2750c 3.7* 10-3 so4= 1*10-3 0.3 0.3 0.830b 138d 10.3* 10-3 a Sorn-srivichai, 1985; b Appendix 7.14; c and dTable 4.1 and see assumption above; This simple calculation suggests that 2-3 times more sulphate should initially diffuse away form the dung ?llet than ?chosphate. The different rates of diffusion through soil do not explain why relatively more 2P than 3 S was recovered at greater radial distances from dung. Appendix 4.6 Calculation for area of a paddock influenced by sheep dung. A. Stock Unit (SU) su = where DM = PU = RQ = Therefore su = = DM*PUIRQ dry matter production3 pasture utilization dry matter requirement/SU/year 12000*70/100/550 15.27 = 15 B. Quantity of dung (PY) 1 . number of patches PY = ND*SU*365 where ND = number of excretions per dayb PY = 6*15*365 = 32850 2. dung weight (M) M = PY*E where E = mass excreted per patch b M = 32850*0. 1 = 3285 C. Fraction of paddock influenced by dung (PI) PI = 1t*(Rp+Rd)2*PY/10000 where Rp = radius of dung patchc Rd = extended radius of area of influence PI = 22n*(0.064+0.I o)2*82125/1 oooo = 0.2775 or = 27.75 percent of paddock a Home, D. 1985; b and0 C/eland, A. Personal communication kg ha-1 year- 1 percent kg patch year - 1 patch year - l kg day-1 kg ha- 1 year- 1 metre metre 307 58 Temperature 48 38 Ra infa l l 28 (mm ) 18 8 Dra inage (mm ) 18 28 38 48 e Appendix 4.7 ? .. ___ ,., .. 4 8 12 16 28 Weeks Weekly rainfall (Rain) , drainage water (Drainage), average maximum (Max) and minimum (Min) temperature during the field trial period (June 27, 1985 - August 29, 1985). 0 Ra inf a l l 0 ??? Max C ?? M in ?C ID Dra inage (J.l 0 00 Appendix 5 . 1 s Applied mg S core- 1 Gypsum Mean Elemental S 1 Mean 12.5 25.0 50.0 F-test C.V. % 12.5 25.0 50.0 F-test C.V. % ns = not significant 1 particle size = < 1??m 309 Recovery of 35 S from 35s labelled gypsum S0 fertil izers after application onto the surface of soil cores. Recovery percentage of totat 35s % 69. 1 65.2 72. 1 68.8 ns 14.3 69.4 77.5 72.4 73.1 ns 16.2 Appendix 5.2 3 10 Cumulative S uptake by pasture and the percentage and amount of plant S derived from fertilizers (SDFF) with time. Days after fertilizer application Fertilizers 30 60 90 120 150 Cumulative plant S uptake .... . mg S core- .... . Superphosphate 14.2 28.3 41 .6 51 .3 69.5 Elemental S 1 12.6 28. 1 41 .4 55.8 71 .8 F-test ns ns ns ns ns c.v. % 21 .7 13 .1 17.6 1 1 .7 19.2 %SDFF . . . . . . % ... . . . . . Superphosphate 28.9 23.4 18.5 17.4 13.5 Elemental S 1 19.3 1 5.7 16.9 13.4 10.6 F-test *** *** ns ns ns c.v. % 1 1 . 1 8 .2 8.1 15.2 14.2 Cumulative S derived from fertilizers - 1 . . . . . mg core .... . Superphosphate 4.1 6.6 7.7 8.9 9.3 Elemental S 1 2.4 4.4 6.9 7.5 7.6 F-test *** *** ns ns ns c.v. % 29.5 13.0 16.9 16.6 10.6 Number of replications 20 16 12 8 4 ??? = significant at 0. 1% level; ns = not significant 1 particle size = <10j.l.m Appendix 5.3 3 1 1 Total CaP-S in three soil layers; CaP-S as percentage of S derived from, PART A, fertilizers (number in parentheses) and PART B , native extractable S and PART C, total extractable S in 0-10 cm depth of soil cores. Days after fertilizer application Fertilizers & Layers 0 30 60 90 PART A, Total CaP-S (% of CaP-S derived from fertilizers)a ..... mg S layer-1 ..... Superphosphate Top 22.4(27) 7.4(15) 6.0(8) Middle 16.9(12) 2.9(21) 3.8(5) Bottom 18.4(13) 3 .5(17) 4.0(5) Elemental S 1 Top 18.9(20) 10.7(26) 7.5(13) Middle 16.2(6) 4.7(19) 5 .1(10) Bottom 16.2(13) 4.2(17) 5.0(6) Lsd 5% Within typec 2.9 1.7 0.9 C.V. % 10.7 20.6 1 1 .8 PART B, Native CaP-Sb ..... mg S layer-1 ..... Superphosphate Top 7.4 16.3 6.3 5.5 Middle 3.5 14.8 2.3 3.6 Bottom 3.5 16.1 2.9 3.8 Elemental S 1 Top 7.4 15.2 7.9 6.5 Middle 3.5 15.2 3.8 4.6 Bottom 3.5 15.6 3 .6 4.7 Lsd 5% Within typec ns 1 .4 0.8 C. V. % 1 1 .0 2 1 .8 1 1.8 PART C, Total CaP-S (0-10 cm) s - 1 ..... mg core ..... Superphosphate 14.4 57.8 1 3.9 13.9 Elemental S 1 14.4 51 .4 19.7 17.6 F-test ns * * C.V. % 7.8 10.5 9.0 ? = significant at 5% level; ns = not significant; 1 particle size = <10 J.UTI a calculated as 1 00 * (CaP? 35s 1 SAF) I CaP-S b. calculated as CaP-S - (CaP-35s 1 SAF) mg core-1 120 5.9(7) 4.4(5) 4.5(7) 7.7(10) 5 . 1 (6) 4.9(5) 1 .0 12.3 5.5 4.2 4.2 6.9 4.8 4.7 0.9 12.0 14.8 17.8 * 14. 1 where SAF CaP-35s Specific activity of 35s in labelled fertilizers (see Table 5 .1 ) Amounts of 35s activity in CaP-S extraction CaP-S Amounts of extractable S, 0, for comparison of means within each fertilizer type 1 50 5.8(9) 3.3(3) 3.9(3) 6.9( 1 1) 4.0(3) 4.5(4) 1 .6 23 . 1 5.3 3 .2 3 .8 6 . 1 3 .9 4.3 1 .3 12.4 13 . 1 1 5.4 * 1 3 . 1 Appendix 5.4 3 12 Cumulative 35s uptake in herbage and total amount of 35s in 0-10 cm soil depth at five harvesting times. Recovery percentage Fertilizers % Days 30 60 90 120 150 HERBAGE Superphosphate 6.4 12.9 13.3 15.7 16.3 Elemental S 1 3.6 8.1 1 1 . 1 1 3.2 13 .6 F-test * * * ns ns C.V. % 28.2 13.0 17.5 16.6 10.6 SOIL Superphosphate 43.7 26.7 24. 1 23.3 22.9 Elemental S 1 59.7 53.3 45.6 42.5 42.3 F-test * * * * * C.V. % 7.5 1 3.2 12.6 4.4 16.9 TOTAL (soil+herbage) Superphosphate 50.9 39.6 37.4 39.0 39.2 Microfine-S0 64.4 61 .3 56.6 55.8 55.8 F-test * * * * * C.V. % 7.5 1 1 . 1 12.6 5.3 9.7 UNACCOUNTED FOR (losses beyond 10 cm) Superphosphate 49.1 60.4 62.6 6 1 .0 60.8 Elemental S 1 35.6 39.7 44.4 44.2 44.2 ? = significant at 5% level; ns = not significant 1 panicle size = <1f>t.lm Appendix 5.5 Recovery of total 35s in three layers of soils at five harvest times. Fertilizers Layers Days after application 30 Superphosphate Top 24.6 Middle 9.5 Bottom 9.6 Elemental S 1 Top 48.8 Middle 6.9 Bottom 4.6 Lsd 5% within typea 3.5 C.V. % 13.7 ? = significant at 5% level; ns = not significant a for comparison among means within each fertilizer type 1 particle size = <10J.m Percent recovery % 60 90 120 150 15.5 14.2 13.5 13.9 5.9 5. 1 5.4 5.4 5.3 4.7 4.4 4.6 39.6 3 1 .3 29. 1 28.5 8.8 8.5 8.8 8.9 4.8 5 .8 4.6 4.8 3.2 3 .9 1 .6 3 .3 16.3 22.3 9.7 20.7 3 13 Appendix 5.6 3 14 Organic 35s expressed as a percentage of total 35s activity applied (PART A) and as a percentage of total 35s remaining (PART B) in three soil layers at five harvests. Days after fertilizer application Fertilizer Layers 30 60 90 120 150 PART A . . . . . % of applied ..... Superphosphate Top 13.9 13 .4 13.2 12.8 12.1 Middle 5.8 4.8 4.8 5.1 5 .1 Bottom 5.6 4.2 4.5 3.9 4.3 Elemental S 1 Top 39.8 33.9 29.9 27.7 27. 1 Middle 5.1 7 .1 7.7 8.2 8 .6 Bottom 3.2 3.7 5.3 4.2 4.6 Lsd 5% Within typea 14.5 7.6 13.4 2. 1 1 1.2 C.V. % 21 .1 16.9 17.9 9.8 22.3 PARTE . . . . . % of remaining .... . Superphosphate Top 56.9 86.6 93.5 95.3 93.3 Middle 60.6 8 1 .2 94.6 93.8 94.3 Bottom 57.9 78.7 92.6 90.4 93.9 Elemental S 1 Top 82.4 85.9 92.9 95.3 95.0 Middle 73.2 80.3 90.0 93.3 96.9 Bottom 69.7 75.2 91 .6 90.4 94.6 Lsd 5% Within typea 12.6 4.3 2.7 1 .9 1 .9 C.V. % 12.5 15.9 2.0 1.3 1 .3 ? = significant at 5% level; ns = not significant a for comparison of means within each fertilizer type 1 particle size = <10jlm Appendix 5. 7 3 15 Carbon-bonded 35s expressed as a percentage of total 35s activity applied (PART A) and as a percentage of total 35s remaining (PART B) in three soil layers at five harvests. Days after fertilizer application Fertilizers Layers 30 60 90 120 150 PART A . .. . . % of applied .. . . . Superphosphate Top 1 1 .2 8.8 8.6 9.8 9.2 Middle 4.7 2.5 1.4 2.4 2.3 Bottom 3 .7 1 .4 1 . 1 0.9 1.2 Elemental S 1 Top 34.8 21 .3 23. 1 25.6 25.0 Middle 3.3 4.3 5.8 5.6 5.7 Bottom 0.9 0.8 2.0 1 .6 1 .8 Lsd 5% Within typea 18.4 12.2 7.9 3.7 6.0 C.V. % 30.0 36.6 27.5 17.3 22.2 PARTE . . . . . % of remaining ..... Superphosphate Top 46.0 57.1 60.7 72.6 70.8 Middle 48.8 43.1 27.4 43.6 43.5 Bottom 36.5 25.6 18.2 19.8 24.6 Elemental S 1 Top 72.0 53.9 74.2 88.3 88.4 Middle 47.8 48.4 67.6 62.3 63.3 Bottom 20.6 16.1 3 1 .9 3 1 .8 33.7 Lsd 5% Within typea 23.9 16.1 23. 1 19.4 15.8 C.V. % 35.4 26. 1 32.9 24.3 19.4 ? = significant at 5% level; ns = not significant a for comparison of means within each fertilizer type 1 particle size = < 1 O)lm Appendix 5.8 316 35s Ester sulphate expressed as a percentage of total 3 5s activity applied (PART A) and as a percentage of total 35s remaining (PART B) in three soil layers at 5 harvests. Day after fertilizer application Fertilizers Layers 30 60 90 120 1 50 PART A . . . . . % of applied ..... Superphosphate Top 2.7 4.6 4.6 3 .1 2.8 Middle 1 . 1 2.3 3.4 2.7 2.8 Bottom 1 .9 2.8 3.4 3.0 3 .4 Elemental S 1 Top 4.9 12.7 5.8 2. 1 2 .1 Middle 1 .8 2.8 1 .9 2.6 2.9 Bottom 2.3 2.9 3.3 2.6 2.8 Lsd 5% Within typea 2.4 13 .5 ns ns ns C.V. % 44.4 53.4 33.3 45.5 43.3 PARTE . . . . . % of remaining ..... Superphosphate Top 10.8 29.4 32.4 22.6 22.4 Middle 1 1 .8 38.8 67.3 50.3 50.7 Bottom 21 .5 53.1 74.3 70.6 69.4 Elemental S 1 Top 10.4 3 1 .9 17.9 7. 1 6.6 Middle 25.4 3 1 .9 22.4 30.9 33.2 Bottom 49. 1 59. 1 59.7 58.5 60.9 Lsd 5% Within typea 16.7 15.0 22.6 19.5 15 .8 C.V. % 48.6 24.4 32.9 32.3 25.8 ? = significant at 5% level; ns = not significant a for comparison of means within each fertilizer type 1 particle size = <10Jlm Temperature Ra i nfa l l (rnrn ) Dra inage (rnm ) 100 80 60 40 20 0 I t1"'11h-rlt1 t 11?11 1thitb-11b:lt lh-i!th?Mrithit t tW11 11 llfr?ttrJI1 11 I 11 11 1?1 20 40 60 0 Appendix 5.9 i start i end 4 8 12 16 20 24 Weeks Weekly amount of rainfall (Rain) and drainage water (Drainage), average maximum (Max) and minimum (Min) temperature during the field trial period from October 1985-March 1986. 28 D Ra inf a l l ?+- M in ? C -?- Max ?C []] Dra inage lJ.J ...... -.J 3 18 Appendix 6. 1 Total S and 35s activity taken up by herbage from 35s labelled superphosphate and microfine S0 treated plots in five harvests. Fertilizers TOTALS Superphosphate microfine S0 F-test C.V. % 35s ACTIVITY Superphosphate Microfine S0 F-test C.V. % Number of replicates ? = significant at 5% level; ns = not significant 30 14.2 12.6 ns 22.9 1 174 757 * 29.4 20 Days after fertilizer application 60 90 120 s - 1 . . . . . mg core .. . . 14.4 13.3 13.8 15.5 13 .3 14.4 ns ns ns 18.8 19.5 35.3 ..... KBq core- 1 .... 967 5 1 1 255 1028 577 315 * * ns 13.0 17.5 16.6 16 12 8 150 13.2 15.9 ns 36.4 155 245 ns 10.6 4 319 Appendix 7. 1 Cumulative sulphur taken up by pastures (PART A) and cumulative dry matter yield, DM, (PART B) in glasshouse trials (average of seven replications). Days after fertilizer application Treatments 30 60 90 120 150 180 PART A, CUMULATNE S UPTAKE -2 . ... . g m ..... Ramiha ss 0.40 0.66 0.93 1 . 15 1 .30 1 .74a LL 0.33 0.48 0.69 0.89 1 .02 1 .34ab SS/PR 0.40 0.57 0.78 0.99 1 . 12 1 .47ab SS/gr 0.38 0.53 0.73 0.93 1.09 1 .50b PR 0.40 0.57 0.77 0.98 1 . 13 1 .59b Ctrl 0.40 0.56 0.72 0.89 1 .02 1 .35b F-test ns ns ns ns ns ** C.V. % 22.4 22.2 18.1 16.9 15.9 15.7 Tokomaru ss 0.39a 0.58a 0.8oa 1 .01a 1 .16a 1.53a MM 0.27b 0.42b 0.63b 0.77b 0.88b 1 . 15b LL 0.25b 0.41b 0.61b 0.75b 0.87b 1 . 19b SS/PR 0.27b 0.44b 0.66b 0.84b 0.97b 1 .29b SS/gr 0.27b 0.39b 0.62b o.8ob 0.93b 1 .25b PR 0.29b 0.43b 0.63b 0.79b 0.89b 1 .21b Ctrl 0.24b 0.32c 0.51c 0.59c 0.68c 0.97c F-test * ** ** ** ** ** C.V. % 3 1 .9 27.4 20.1 17.0 16.5 15.4 PARTB, CUMULATNE DM -2 .. . . . g m ..... Rami ha ss 88 193 333 421 480 593 LL 83 180 313 393 453 544 SS/PR 103 192 3 15 403 462 562 SS!gr 101 188 298 377 439 545 PR 100 194 315 409 471 608 Ctrl 1 13 214 325 414 472 593 F-test ns ns ns ns ns ns C.V. % 18.2 13.1 15.4 16.2 14. 1 13.8 Tokomaru ss 106 210 319 410 494 610 MM 79 173 283 362 435 532 LL 81 185 298 384 467 590 SS/PR 77 170 292 379 455 564 SS/gr 89 174 294 376 450 555 PR 94 197 3 19 409 478 604 Ctrl 85 167 271 338 402 506 F-test ns ns ns ns ns ns C.V. % 14.5 22.1 16.8 14.8 14.6 14.3 ?, ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; ns = not significant; mean separation by DMRT at 5% level denoted by letters 320 Appendix 7.2 Cumulative sulphur taken up (PART A) and cumulative dry matter yield, DM, (PART B) of pastures in Field trials (average of three replications). Days after fertilizer application Treatments 30 60 90 120 150 180 PART A, CUMULATIVE S UPTAKE -2 .... . g m ????? Rami ha ss 0.15 0.20 0.35 0.50 0.58 0.68 LL 0. 15 0.20 0.37 0.48 0.55 0.63 SS/PR 0.17 0.22 0.39 0.49 0.55 0 .62 SS/gr 0.23 0.29 0.47 0.56 0.61 0.70 PR 0. 1 1 0.16 0.27 0.35 0.38 0.42 Ctrl 0.12 0.16 0.29 0.38 0.43 0.55 F-test ns ns ns ns ns ns C.V. % 33.0 28. 1 22.6 21 .7 23. 1 2 1 .9 Tokomaru ss 0.05 0.06 0.20 0.30 0.34 0.38 MM 0.08 0.1 1 0.27 0.36 0.39 0.44 LL 0.07 0.1 1 0.27 0.37 0.40 0.45 SS/PR 0.10 0.10 0.3 1 0.42 0.47 0.53 SS/gr 0.08 0.10 0.29 0.39 0.44 0.50 PR 0. 12 0.13 0.3 1 0.38 0.42 0.46 Ctrl 0.08 0. 10 0.23 0.31 0.34 0.37 F-test ns ns ns ns ns ns C.V. % 43.7 46. 1 28.9 26.1 27.3 26.5 PART B, CUMULATIVE DM -2 .... . g m . ... . Rami ha ss 102 132 236 301 350 408 LL 106 146 244 303 353 416 SS/PR 106 140 244 300 334 378 SS/gr 160 202 3 1 1 361 390 447 PR 76 1 12 195 240 261 2% Ctrl 98 127 196 266 304 404 F-test ns ns ns ns ns ns C.V. % 33.3 28. 1 22.6 21 .7 23. 1 2 1 .9 Tokomaru ss 35 45 130 194 202 238 MM 67 82 198 267 295 322 LL 68 89 201 275 305 341 S/PR 72 87 210 279 3 1 1 353 SS/gr 67 84 208 280 305 350 PR 60 108 193 249 327 359 Ctrl 64 83 182 238 264 293 F-test ns ns ns ns ns ns C.V. % 47. 1 48.5 34. 1 29.9 29.7 28.8 ns = not significant 321 Appendix 7.3 Sulphur concentration of pastures in glasshouse (PART A) and field (PART B) trials as influenced by different particle size of elemental S . Days after fertilizer application Treatments 30 60 90 120 150 1 80 PART A, GLASSHOUSE1 . . . . . % .. . . . Rami ha ss 0.47 0.23a o.2oa 0.25a 0.26a 0.39a LL 0.38 0.14c 0.17b 0.25b 0.21c 0.34b SS/PR 0.41 0.19b 0.17b 0.23b 0.23b 0.36a SS/gr 0.38 0. 17b 0. 18b 0.25a 0.25a oAoa PR 0.40 0. 18b 0.17b 0.23b 0.24b 0.33b Ctrl 0.34 0.17c 0. 15c 0.19c 0.21c 0.27c F-test ns ** * * ** ** C.V. % 13.6 28.5 1 1 .4 13 .2 13.6 13 .4 Tokomaru ss 0.36 0.18a o.2oa 0.24a 0.17a 0.32a MM 0.33 0.17a 0.19b 0. 18b 0. 15b 0.26b LL 0.31 0. 15b 0.19b 0.16c 0.14b 0.26b SS/PR 0.35 0.17a 0. 19b 0.21a 0.18a 0.3 1a SS/gr 0.3 1 0. 14c 0.19b 0.22a 0. 18a 0.3 1a PR 0.32 0. 14c 0. 17c 0.17c 0. 15b 0.26b Ctrl 0.29 0. 14c 0.14c 0.16c 0.14b 0.29b F-test ns * * ** ** *** C.V. % 12.5 16.5 13.6 16.3 13.8 13 .6 PART B, FIELD TRIAL2 . . . . . % .... . Ramiha ss 0. 15 0.15 0.15 0.22 0.17 0.17a LL 0.14 0. 13 0.16 0. 18 0. 14 0.1 3b SS/PR 0.16 0.15 0.17 0. 18 0.14 0.16a SS/gr 0.16 0.14 0.16 0. 19 0. 14 0.16a PR 0.14 0. 14 0. 13 0.17 0. 14 0 . 13b Ctrl 0.13 0. 13 0.15 0. 17 0. 13 0. 12c F-test ns ns ns ns ns ** C.V. % 9.7 1 1 .4 9.7 12.7 16.0 1 1 .8 Tokomaru ss 0.14 0. 1 1 0. 17 0.17 0. 17 0.16a MM 0.12 0. 13 0.14 0.14 0.14 0.13b LL 0.12 0. 13 0. 15 0.14 0. 12 0 . 13b SS/PR 0. 15 0.1 1 0.16 0.15 0.16 o. 15b SS/gr 0.1 1 0. 12 0.16 0.16 0. 15 0. 14b PR 0.13 0. 13 0. 13 0.13 0.13 0 . 13b Ctrl 0. 12 0. 1 1 0. 13 0.14 0. 12 0. 12c F-test ns ns ns ns ns ** C.V. % 15.3 12. 1 15.3 17.7 15.6 9.5 ?? ?? and ??? = significant at 5, 1 and 0. 1% level respectively; ns = not significant; mean separation by DMRT at 5% level denote by letters; 1 and2 average of seven and three replications, respectively 322 Appendix 7.4 Cumulative percentage of 3Ss taken up by pastures in glasshouse (PART A) and field (PART B) trials at six sampling times. Days after fertililzer application Treatments 30 60 90 120 150 180 PART A, GLASSHOUSE TRIALS1 .... . %recovery .. . . . Ramiha ss 3. 1a 8.6a 12.5a 14.9a I6.6a 20.5a LL o.sc uc 1 .9c 2.5c 2.9c 4.1c SS/PR 1 .9b 4.8b 7.7b 10.3b 1 1 .7b 14.7b SS/gr 1 .6b 3.9b 6.2b 8.4b 10.3b 13 .7b F-test * *** *** *** *** *** C.V. % 42.5 44.4 36.5 32.9 29.4 24.7 Tokomaru ss 1.8a 4.oa 6.3a 8.6a 10.6a 13 .9a MM 0.4c 0.9c 1 .4c 1.9c 2.5c 3 .5c LL 0.2c o.sc 0.8c l .Oc uc 1.9c SS/PR 0.8b 2.6b 4.7b 6.?b 8.9ab 12.6ab SS/gr o.8b 2. lb 4.ob 5.8b 7.7b 10.4b F-test ** ** *** *** *** *** C.V. % 28.5 41 .2 34.6 34.9 32.2 28.7 PART B, FlEW TRIAL 2 . .. . . % recovery .. . . . Rami ha ss 1 .4a 1 .8a 4.6a 6.6a s.oa 9.8a LL 0.26b 0.3b o.8b ub 1.4b 1 .7b SS/PR 1 .4a 1 .8a 4.2a 5.7a 6.6a 8. la SS/gr 1 .5a 1 .9a 4.4a 5.8a 6.6a 8.3a F-test ** ** *** *** *** *** C.V. % 27.5 28.2 3 1 . 1 3 1 .8 35.3 32.5 Tokomaru ss 0.5a 0.6a 2.2a 3.3a 3.7a 4.2a MM 0.2b 0.2b 0.6b 0.9b LOb l . lb LL O.lb O. lb 0.3b o.sb o.sb 0.6b SS/PR 0.6a 0.7a 2.7a 3 .9a 4 .6a 5.7a SS/gr 0.4ab 0.4ab 1 .9a 2.7a 3.2a 3 .? F-test * *** ** *** *** *** C.V. % 54.0 45.5 32.1 29.2 35. 1 30.4 ?, ?? and ??? = significant at 5, 1 and 0. 1% level, respectively; mean separation by DMRT at 5% level denoted by letters; 1 and2 average of seven and three replications, respectively 323 Appendix 7.5 Percentage of plant S cumulatively derived from fertilizers (%SDFF) in glasshouse (PART A) and field (PART B) trials. Days after fertilizer application Treatments 30 60 90 120 150 180 PART A, GLASSHOUSE TRIAL1 .. . . . % SDFF ... . . Rami ha ss 23.8a 39.0a 40.2a 38.8a 38. la 35.2a LL 4.2c 7.3c 8.3c 8.4c 8.6c 9.3c SS/PR 14.9b 26.2b 29.3b 3 1 .5b 3 1 .7b 30.5b SS/gr 12.1b 21 .5b 24.8b 26.4b 27.9b 26.8b F-test ** *** *** *** *** *** C.V. % 35.2 33.5 26.7 23.0 21 .2 18.3 Tokomaru ss 14. 1a 20.6a 23.7a 23.4a 27.3a 27.3a MM 5.3c 6.5c 6.9c 7.6b 8.4b 9.3b LL 3 .0c 4.0c 4.0c 4.2b 4.4b 4.6c SS/PR 9.4b 17.5ab 21 .5ab 24.5a 27. 1a 28.7a SS/gr 8.5b 15.2b 19.1b 21 .6a 24.6a 24.6a F-test * ** *** *** *** *** C.V. % 28.7 34.5 32.4 3 1 . 1 28. 1 22.2 PART B, FlEW TRIAL 2 .. . . . %SDFF ..... Ramiha ss 27.7a 26.8a 39.0a 39.5a 4I .(il 43.la LL 4.sC 4.3c 6.2b 7.0b 7.3b s.ob SS/PR 25.7a 24.8a 32.la 34.7a 35.7a 38.9a SS/gr 19.6b 19.3b 27.6a 29.9a 3 1 .4a 34.7a F-test ** *** *** *** *** *** C.V. % 13.1 14.7 23.1 19.1 18.7 15.2 Tokomaru ss 29.5a 28.7a 33.2a 32.0a 32.4a 33.5a MM 7.0c 6.4c 7.lc 7.3c 7.3c 7.SC LL 3.4c 4.0c 4.0c 4. 1c 4.2c 4.4c SS/PR 15.4b 14.7b 24.7b 27.3a 28.8a 32.la SS/gr 15.4b 12.9b 19.4b 20.3b 21 .4b 23.4b F-test *** ** *** *** *** *** C.V. % 28.1 26.2 17.8 14.7 12.6 10.1 ?, ?? and ... = Significant at 5, 1, and 0. 1% level, respectively; mean separation by DMRT at 5% level denoted by letters; 1 and 2 average of seven and three replications, respectively 324 Appendix 7.6 Percentage of oxidized elemental 35s0 cumulatively taken up by pastures 1 in glasshouse and field trials (average of three replications). Days after fertilizer application Treatments 30 60 90 1802 Field 180 . . . % ... Rami ha ss 6.2a 13.8 19.8a 26.2a 1o.oa LL 1 .9b 8.6 10.6b 25.7a 1 .7b SS/PR 4.0a 8.5 14.5a 21 .0b 8.3a SS/gr 3.7a 12.9 10.4b 19.4b 8 .5a F-test * ns * ** *** C.V. % 35.8 25.6 22.8 16.7 32.8 Tokomaru ss 3.4 6.4 8.8a 20.8a 4.3a MM 2.8 5.5 5.9b 16.8b ub LL 1 .6 3.8 5.3b 9.4c 0.6b SS/PR 1 .4 6.6 13.oa 2l . la 5.9a SS/gr 1 . 1 5 .2 8.lb 17.1b 4.oa F-test ns ns * *** ** C.V. % 50. 1 49.6 28.4 20.1 33.5 ?, .. and ... = significant at 5, 1 and 0. 1%, level respectively; ns = not significant; mean separation by DMRT at 5% level denoted by letters 1 calculated as described in Section 7.3.9 and briefly: % 35s0 oxidized taken up = [I. % of 35s taken up] * 1 00 1 (100 - (% residuai 35s0)] 2 average of seven replications Appendix 7. 7 Treatments RAM/HA Top ss LL SS/PR SS/gr Middle ss LL SS/PR SS/gr Bottom ss LL SS/PR SS/gr F-test Layer Lsd 5% within layer C.V. % TOKOMARU Top ss MM LL SS/PR SS/gr Middle ss MM LL SS/PR SS/gr Bottom ss MM LL SS/PR SS/gr F-test Layer Lsd 5% within laye? C.V. % 325 Percentage recovery of extractable 3Ss (CaP-S) in three soil layers of Ramiha and Tokomaru soils at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications). Days after fertilizer application Field 30 60 90 1801 180 .. . . . % recovered .... . 21 .1 1 1 .5 13 .1 17.3 5.1 1 .0 1 .4 1 .3 3 .2 2.0 12.7 12.4 13.2 23.8 6.0 7.6 10.9 5.3 14.8 6. 1 0.2 0.3 0.3 1 .0 1 .9 0.3 0. 1 0.1 0.1 0.2 0.1 0.3 1.3 1 .4 2.9 0.1 0. 1 0.3 0.5 2.3 0.1 0.1 0.5 0.2 3.9 0. 1 0.2 0.1 0.1 0.1 0.1 0.1 0.7 0.3 4.0 0.8 0. 1 0.3 0.2 4.4 ** ** ** ** ** 5.4 3 .5 3.7 5.2 1 . 1 78.1 41 .2 41 .0 44.4 55. 1 ... . . % recovered .... . 10.1 8.7 12. 1 1 1 .3 4.2 1 .0 3 .0 2.0 2.3 1 .0 1 . 1 1 .0 1 .0 0.7 0.8 14.3 10.7 6.1 1 1 .6 5.9 10.6 2.5 8.7 7.8 4.2 0.1 0.1 0.5 0.6 1 .3 0.0 0. 1 0.2 0. 1 0.2 0.0 0.0 0.1 0.1 0.1 0.0 0.2 0.4 0.8 1 .9 0.1 0 .1 0.8 0.5 1 .2 0.1 0.2 0.9 0.9 0.9 0.2 0. 1 0.1 0.1 0.1 0.1 0. 1 0.1 0.1 0.1 0.1 0.2 0.3 0.3 1 .6 0.1 0.2 0.7 0.4 0.9 ** ** ** ** ** 4.1 3.2 1 .8 3 .1 1 .2 87.5 76.7 40.4 45.3 40.8 1 average of seven replications; 2 Lsd at 5% level Appendix 7.8 Treatments RAMI HA Top ss LL SS/PR SS/gr PR Ctrl Middle ss LL SS/PR SS/gr PR Ctrl Bottom ss LL SS/PR SS/gr PR Ctrl F-test Layer Lsd 5% within laye? C.V. % 326 Concentration of soil extractable S (CaP-S) in three layers of Ramiha soil at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications). Days after fertilizer application Field 30 60 90 1801 180 k -1 ... . . mg g .. . . . 54.4 54.4 46.9 36.1 27.7 20.8 26.6 19.2 21 .0 19.7 57. 1 56.0 44.3 44.8 22.9 41 .6 64.5 29.3 54. 1 25.6 32.5 32.3 19.6 21 .7 24.0 26. 1 29.3 17.6 22.6 19.2 14.4 18 .1 14.4 21 .9 20.1 14.4 18.7 9.6 18.5 19.7 16.0 20.2 13.7 24.0 24.0 16.0 17. 1 15.5 17.6 19.7 13.3 18.6 13 .8 19.9 20.1 15.5 19.2 16.5 19.9 19.8 14.9 18 .1 19.7 25.8 26.3 15.4 22.4 18 .1 22.1 21.8 16.5 22.9 17.1 29.0 29.3 16.0 19.7 18.1 23.3 24.0 14.4 24.0 17.6 24.9 21.9 20.2 21 .8 20.8 24.6 20.3 *** *** *** *** *** 15.1 9.2 13.5 14.2 5.6 35.5 16.2 37.3 25. 1 14.5 1 average of seven replications; ??? = significant at D. 1% level; 2 Lsd at 5% level Appendix 7.9 Treatments TOKOMARU Top ss MM LL SS/PR SS/gr PR Ctrl Middle ss MM LL SS/PR SS/gr PR Ctrl Bottom ss MM LL SS/PR SS/gr PR Ctrl F-test Layer Lsd 5% within laye? C.V. % 327 Concentration of soil extractable S (CaP-S) in three layers of Tokomaru soil at six sampling times in glasshouse trials and after 180 days in the field trials (average of three replications). Days after fertilizer application Field 30 60 90 1801 180 k - 1 ... . . mg g .... . 33.1 39.5 45.9 33.8 15.5 17.6 29.3 15.5 15.4 12.3 17.1 20.3 14.9 1 1 .2 10.7 41 . 1 38.4 20.8 35.4 19.2 33.3 25.6 30.4 28.1 17.7 25.6 19.2 19.2 12.6 17.7 14.9 18.1 1 1 .7 13.7 1 1 .7 6.4 7.5 8.5 12.0 13.3 9.6 8.5 6.4 10.1 1 1 .2 9.6 6.9 5.3 9.1 13 .3 12.8 10.3 4.8 9.8 14.9 1 1 .2 9.6 6.4 9.4 16.0 10.1 8.0 5.3 7.7 13.8 8.0 8.5 5.3 9.8 14.9 7.4 6.9 9.6 9.8 1 1 .2 9.7 10.7 7.5 10.7 8.5 8.5 8.0 7.5 1 1 .6 1 1 .8 8.5 8.0 6.9 1 1.4 1 1 .2 8.0 8.5 6.9 12.8 1 1 .7 7.4 9.1 6.9 13.7 13.3 6.4 6.9 6.9 14.8 1 1 .3 *** *** *** *** *** 13 .5 9.1 6.2 6.5 5.6 47.1 31 .8 26.5 24.5 23.4 1 average of seven replications; ... = significant at 0. 1% level; 2 Lsd at 5% level Appendix 7. 10 Treatments RAM/HA Top ss LL SS/PR SS/gr Middle ss LL SS/PR SS/gr Bottom ss LL SS/PR SS/gr F-test Layer Lsd 5% Within layer C.V. % TOKOMARU Top ss MM LL SS/PR SS/gr Middle ss MM LL SS/PR SS/gr Bottom ss MM LL SS/PR SS/gr F-test Layer Lsd 5% Within Laye? C.V. % 328 Percentage recovery of total 35s in three soil layers in glasshouse trials and after 180 days in the field trials (average of three replications). Days after fertilizer application Field 30 60 90 1801 180 .. . . . % recovered .. . . . 1 1 1 .2 89.2 86.9 69. 1 5 1 . 1 129.3 1 18.1 109.6 96.3 8 1 .3 99.4 100.2 92.3 93.1 52.8 96.6 85.5 85.3 77.3 5 1 .9 1 .2 1.7 3 . 1 5.4 13 .4 0.8 0.5 0.2 1 . 1 5.5 1.3 0.4 4.2 5.9 18.4 0.4 0.9 0.9 2.1 15.5 2.5 0.7 1 .4 0.9 8.9 1 .9 1.2 0.2 0.4 0.9 3.8 1 .2 1 .9 1 . 1 10.0 1.7 1 .9 0.8 0.4 1 1 .3 *** *** *** *** *** 16.5 14.9 5.8 10.7 13.9 25. 1 26.7 10. 1 21 .9 30.7 .... . % recovered .. . . . 93.8 98.6 86.4 73.2 67.4 1 16.8 121.2 100.4 89.6 73.5 1 18 .3 120.1 139.6 88.5 83.1 105.4 106.5 103.2 81 .4 68.8 108.1 105.5 76.6 70.6 85.3 10.3 0.9 1 .4 14.7 20.7 6.4 1 . 1 1 . 8 5.7 18.1 3.8 1 .4 1 .5 12.3 16.9 4.9 1.9 2.4 13.3 19.8 1 .9 0.5 1 .3 1 1 .3 25.2 10.9 1.5 0.6 6.9 6.8 13.8 9.8 0.8 6.5 8.8 4.5 1 . 1 0.3 3 .9 5 .5 6.2 2.7 0.8 5.2 12.8 10.6 1 .4 1.3 5.5 8 .2 *** *** *** *** *** 18.5 15.6 12.6 15.6 13 .9 27.9 24.8 21 .1 29.9 30.7 1 average of seven replications; ... = significant at 0. 1% level; 2 Lsd at 5% level Appendix 7. 1 1 Sulphur concenlration, cumulative sulphur uptake and dry matter yield of pastures on two soils at three samplings in glasshouse trials. Sulphur Cumulative Cumulative Treatments concentration s dry mauer uptake yield Days 30 60 90 30 60 90 30 60 90 ... % ... -2 .. g m .. -2 .. g m .. Ra.miha GP 0.44a OAr 0.45a 0.57a L23a 1 .67a 130 267 39 1 SSP o.wa 0.46a o.sr 0.64a L2la 1 .67a 105 23 1 348 Ctrl 0.34b 0.17b 0.15b 0.40b 0.56b o.nb 1 13 214 325 F-test *** *** *** ** *** *** ns ns ns c.v. % 1 0. 1 1 1 .7 9.2 1 8.9 18.2 17. 1 18.2 16.9 16.4 Element.al S (SS) 0.47 0.23 0.20 0.40 0.66 0.93 88 193 333 Tokomaru GP 0.44a OAW 0.4 13 0 .54a 1 .083 1 .45a 125 269a 347a SSP 0.44a 0.4Sa OAif o.5if 1 .45a 1 .69"1 1 12 248a 382a Ctrl 0.29b 0. 14b O. l4b 0.24b 0.32b 0.5 1b 86 168b 272b F-test *** *** *** *** *** *** ns * * c.v. % 9.6 6.8 9.3 28.2 26.2 27.6 27.5 22.6 23.7 Element.al S (SS) 0.36 0.18 0.20 0.39 0.58 0.80 105 210 3 19 ?? ?? and ??? = signifK;ant at 5, 1 and 0. 1% level, respectively; ns = not signif1Ca.f1t; mean separation by DMRT at 5% level denoted by letters Vl IV \0 330 Appendix 7. 12 Cumulative 35s taken up by pasture and percentage of plant sulnhur derived from labelled fertilizers (%SDFF) after application of 35s labelled gypsum and superphosphate in glasshouse trials. Cumulative 35s %SDFF Treatments uptake Days 30 60 90 30 60 90 ... % ... .. . % . . . Rami ha GP 9.4 17.5 22.6 47.7 42.4 37.9 SSP 7. 1 18.5 19.6 34.3 37.4 35.9 F-test ns ns ns ns ns ns C.V. % 1 1 .6 1 1 .8 10.8 9.9 6.9 7.4 Elemental S (SS) 3 . 1 8.6 12.5 23.8 39.0 40.2 Tokomaru GP 8.3 14.3 19.4 49.8 48.6 42.3 SSP 7.9 16.9 22.4 48.6 44.5 40. 1 F-test ns ns ns ns ns ns C.V. % 14.2 20.3 16.7 27.9 24.1 17.5 Elemental S (SS) 1 .8 4.0 6.3 14. 1 20.6 23.7 ns = not significant at 5% level Appendix 7.13 Treatments GP Top Middle Bottom SSP Top Middle Bottom Ctrl Top F-test Treatments Middle Bottom Within treatments1 C.V. % 331 Phosphate extractable S concentration and amounts present in three soil depths of two soils 90 days after application of gypsum and superphosphate. Concentration Amount Rami ha Tokomaru Rami ha Tokomaru mg kg-1 mg layer?1 95.2 66.8 29.2 28.6 14.4 1 1.2 8.3 7.6 12.4 6.0 16.2 10.7 91 .2 71 .2 26.5 28.7 18.8 6.6 10.8 4.2 14.8 5.2 20. 1 9 .1 17.7 1 1 .7 5.8 5.6 16.5 5 .3 10.6 3.3 20.8 6.9 29.9 10.0 ** ** ** ** 20.5 30. 1 7.2 7.5 27.5 35.3 21.7 25.4 ?? = significant at 0. 1% level; 1 Lsd at 5% level 332 Appendix 7.14 Sulphate retention (%) and simple relationship between amounts of adsorbed sulphate (mg S kg-1) and sulphate concentration in solution (mg S r1) for three layers ofRamiha and Tokomaru soils. Soils Layers s Relationships R2 retention C"f = a + bX)a % % Rami ha Top 12. 1 3 .8 + 0.6X 95.8 Middle 28.0 2.2 + 1 .4X 89.0 Bottom 29.1 2.6 + 1 .9X 98.3 Tokomaru Top 24. 1 3.4 + 1 .5X 98.1 Middle 28.2 4.0 + 1 .8X 99.4 Bottom 31 .1 2.1 + 2.3X 94.6 a Y = adsorbed S (mg S kg-1 ); X= solution concentration (mg S r1) Appendix 7.15 Particle sizes <0.150 mm 0.250-0.500 mm 333 Recovery of 35s0 from two sizes elemental S at day 0 ?rough NaHC03 dry digestion (total 35s) and acetone extraction (total 5so activity and S0) after application onto the surface of soil cores. Recovery percentage Rate Total Total Total applied 35s 35so so mg S core-1 % % % 12.5 71 .6 82.3 96.9 25.0 74.2 77.7 91.3 5o.oa 76.7 80.7 87.2 Mean 74.2 80.2 91 .9 F-test ns ns ns C.V. % 17.1 16.7 14.3 12.5 91 . 1 93.5 83.7 25.0 101 .8 90.8 91 .4 5o.oa 88.6 78.2 90.6 Mean 95.8 87.2 88.6 F-test ns ns ns C.V. % 21 .3 19.1 13.2 a, equivalent to approximately 30 kg S ha -1 remark: 35s0 activity applied at 3.3 MBq g So and negligible amount ot35sO were detected in the phosphate extractable S ns = not significant at 5% level Temperature Ra i nfa l l (mm ) sa 6a 4a 2a ?,:,.?.?J -?- ? -,j _ I I U .J1 n ..-y .JJ,._. a I' ,,?h rrnl ,,, \' ,,,,,,, ,,, \1 1!1 '1' ,,,...,, !Uf I I' I!' 1!1 1?!'1'11 1!' 11 !I !1 l=! I Dra inage ( mm ) 2a 4a 6a Appendix 7.16 a i start i end 4 8 12 16 20 24 28 32 Weeks Weekly rainfall (Rain) and drainage water (Drainage), average maximum (Max) and minimum (Min) temperature during November 1987 - June 1988. D Ra infa l l -?- Max ?C -+- M in oc [}] Dra inage Vl Vl ?