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. 
 



SOIL WATER USE BY APPLE TREES 

A thesis presented in partial fulfillment 

of the requirements for 

the Master of Agricultural Science 

in Soil Science at 

Massey University 

New Zealand 

Pudjo RAHARDJO 

1989 



i 

ABSTRACT 

SOIL WATER USE BY APPLE TREES 

The study investigated the soil water use of an unirrigated tree 

and an irrigated apple tree in Hawke~ Bay, New Zealand in the middle 

of the summer of 1988/1989. A rainout shelter was used to eliminate 

any water inputs from both irrigation and rain to the unirrigated 

tree. ,The irrigated tree received water inputs from both irrigation 

and rain. The soil water content was measured by neutron probing and 

time domain reflectometry. The heat pulse technique was used to 

measure the sap-flow in the apple trunks. Both leaf water pressure 

potential and stomatal resistance were measured by the pressure 

chamber and porometer respectively. A measuring cylinder was used to 

monitor the apple growth during the study. 

The results of the water use measurements were that 

- the neutron probing and time domain reflectometry showed the soil 

water use was about 77 litres (4.3 mm) per day taken from 0 - 1900 

mm depth around the irrigated tree. However soil water extraction 

around the unirrigated tree was only 19 litres (1 mm) per day at the 

beginning of the study, and no water extraction was measured from 

the top 1900 mm later in the study. 

- the heat pulse technique showed that the unirrigated tree extracted 

slightly more soil water than the irrigated tree. The average sap­

flow measured was 66 litres per day. Probably the unirrigated tree 

extracted much of its water from below 1900 mm depth, or from beyond 

the covered area. 

- the amount of water use by the apple trees was similar to regional 

evaporation estimates obtained using the Priestley - Taylor formula, 

when 0.66 fractional canopy cover was assumed. 



ii 

The water stress monitoring showed that 

e pressure chamber technique was a more sensitive way to monitor 

ress than was porometry. 

e leaf water pressure potential values showed a significant 

fference between the irrigated and the unirrigated apple tree 

ring the latter part of the study. 

The readily available soil water storage capacity from Oto 400 

;pth (the most active part of the root zone), from O - 1000 mm 

h, and from Oto 1900 mm, was about 36 mm, 89 mm and 170 mm 

:ctively. When there was a lack of available soil water on the 

,il, the root system was forced to extract soil water from deep in 

,oil profile. 

The comparison of apple fruit growth showed that during the last 

days of the study, the apples on the unirrigated tree grew more 

Ly than those on the irrigated tree. 



iii 

A C K N O W L E D G M E N T S 

I am greatly indebted to my supervisors, Dr. B.E. Clothier and 

Dr. D.R. Scatter, who not only provided helpful guidance in this 

thesis, but also introduced me to Soil Physics. 

I express gratitude to the Ministry of External Relations and 

Trade, New Zealand for financial support, and to the Government of the 

Republic of Indonesia, who allowed me to study at Massey University, 

and still paid my salary during the course of the study. 

Thanks to Mr. Van Howard for providing the research site, to Mr. 

James Watt for providing the meteorological data, to Dr. Paul Gandar 

and Dr. Keith McNaughton for useful discussion, to Mr. John Julian and 

Mr. Brooke Tynan for helping me to install the rainout shelter, to Ms. 

Tina Baker for help with some field observations, and to Mr. Mark 

Roche for assistance with computing. 

Acknowledgement is also given to every member of the Department 

of Soil Science, Massey University, who provided a pleasant atmosphere 

in which to study. 

I am grateful for encouragement given by Soekinah - Doerjat (my 

parents), Siti Aminah - Isom Saebani (my parents in law), Farida 

Rahardjo (my wife), and my sons Danang, Ikhlas and Iqbal Rahardjo. 

Finally, thanks to Dachman, who sent · my office salary for 3 years . 



iv 

TABLE OF CONTENTS 

ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i 

ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii 

TABLE OF CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv 

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii 

LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii 

CHAPTER I 

THE WATER BALANCE OF APPLE TREES 

1. 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 

1 . 2 • THE WATER BALANCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 

1. 2. 1 . WATER INPUTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 

1. 2. 2. WATER OUTPUTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 

1. 3. THE STUDY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 

CHAPTER II · 

SITE DETAILS AND METHODOLOGY 

2.1. THE SITE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 

2. 1. 1. THE SOIL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 

2 . 1 . 2 . THE ORCHARD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 

2 .1. 3. ROOT DISTRIBUTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 

2 . 1.4. THE CLIMATE AND WEATHER.......... .... . . ....... 9 

2.1.5. THE RAINOUT SHELTER . . .. . ..•. . .. . ... . . . . . . . . . . . 12 

2.1.6. THE EXPERIMENTAL LAYOUT .............. . . ·... .. . . 12 



V 

2. 2. MEASUREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 

2.2.1. SOIL WATER CONTENT MEASUREMENT . . . . . . .. . . . . . . . . 17 

2.2.2. SOIL WATER PRESSURE POTENTIAL MEASUREMENT . . . . . 17 

2.2.3. SAP FLOW MEASUREMENT ......... .. . . . . . . . . . . .. . . . 18 

2.2.4. STOMATAL RESISTANCE MEASUREMENT .. . . . . . . . . . . . . . 18 

2.2.5. LEAF WATER PRESSURE POTENTIAL MEASUREMENT . . .. . 18 

2.2.6. APPLE FRUIT GROWTH MEASUREMENT . . . . . . . . . . . . . . . . 18 

CHAPTER III 

NEUTRON PROBE AND TIME DOMAIN REFLECTOMETER CALIBRATION 

AND TENSIOMETER DESCRIPTION 

3.1. NEUTRON PROBE 

3 .1.1. THEORY 

19 

19 

3.1.2 . METHODOLOGY AND CALIBRATION . ... . . . . . . . . . . . . . . . 20 

3.2. TIME DOMAIN REFLECTOMETER.......... ... . . . . . . . . . . . . . . . 22 

3. 2. 1. THEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 

3.2.2 . CALIBRATION METHOD .. . . . . . . . . ... . . . . . . . . . . . . . . . 27 

3.2.3. CALIBRATION RESULTS AND DISCUSSION . . .. .. . . . . . . 28 

3. 2. 4. APPLICATIONS .............. · . . . . . . . . . . . . . . . . . . . . 33 

3. 3. TENSIOMETER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 

3 . 3 . 1 . THEORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3 

3.3.2. TENSIOMETER USED IN THIS STUDY................ 35 



vi 

CHAPTER IV 

SOIL WATER MEASUREMENTS 

4.1. SOIL WATER CONTENT PROFILES . . . .. . .......•.. ...•. ... . . 36 

4.2. SOIL WATER STORAGE • . • . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . 39 

4.3. SOIL WATER CONTENT CHANGES .. •. .•. ... ... .. . ....• ... . . . 42 

4. 4. SOIL WATER USE . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . 44 

4.5. SOIL WATER PRESSURE POTENTIAL . . . . .... ... . . . . .. . . ... . . 48 

CHAPTER V 

THE ABOVE GROUND MEASUREMENTS 

5.1. HEAT PULSE TECHNIQUE . . . ... .. . . . . . ... . .. . . . . . .. . . . . . . . 50 

5.1.1. SAP-FLOW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 

5 .1 . 2. THE TECHNIQUE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 

5 .1. 3. INSTRUMENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 

5. 1. 4. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 6 

5. 2. STOMATAL RESISTANCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 

5.2.1. STOMATA . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 

5. 2. 2. THE POROMETER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 

5. 2. 3. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 

5. 3. LEAF WATER POTENTIAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 

5. 3 .1. WATER POTENTIAL . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . 69 

5.3.2. THE PRESSURE CHAMBER.......................... 70 

5.3.3. RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 

5 . 4 . APPLE FRUIT VOLUME . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 5 

5.4.1. MEASURING CYLINDER TECHNIQUE . . . . . . . .. . . . . . . . . . 75 

5.4.2. APPLE FRUIT GROWTH 75 



vii 

CHAPTER VI 

DISCUSSION, CONCLUSIONS AND PRACTICAL IMPLICATIONS 

6. 1. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 0 

6.2. EXPERIMENT DURING SUMMER 1987/1988 . .. . . . . . .. . . . . . . . . . 80 

6. 3. SOIL DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 

6. 4. PLANT DATA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 

6 . 5. GENERAL DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 

6. 6. CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 

6.7. SOME POSSIBLE PRACTICAL IMPLICATIONS OF THIS STUDY . . . 93 

6.7.1. THE TYPICAL WATER USE ... . . . ... . .... .. . . . . . . . . . 94 

6.7.2. THE POSSIBLE SOIL WATER SUPPLY FROM THE 

WATER TABLE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 

6.7.3. THE SOIL WATER RESERVOIR...................... 95 

6.7.4. THE TYPICAL WATER INPUT FROM IRRIGATION . . . . . . . 98 

6.7.5. THE TYPICAL APPLICATION RATE OF IRRIGATION . . . . 99 

6 . 8. SUGGESTIONS FOR FUTURE WORK.......................... 101 

REFERENCES 102 

APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 



viii 

L I S T 0 F F I G U R E S 

2.1 . . Apple trees in the orchard during the fruiting period ....... 7 

2.2. The average root length density measured with depth for all 

radii (a) and with depth in various radial classes (b) 

(K.A. Hughes, pers. comm.).................................. 8 

2.3. The average monthly rainfall (1959 to 1980) for Station 

D96689, Havelock North, 9 m above sea level, and average 

monthly Penrnann evaporation for Napier (NZ Met. Service, 

pers. comm.) ................................................ 10 

2.4. The rainout shelter around Tree U, open for measurements .... 13 

2.5. The difference between the soil within and outside the 

covered area around Tree U ..... ..... .... ...... . ... . ......... 13 

2.6. The layout of the experimental site in the apple orchard .... 14 

2.7. Layout around Tree I (above) and Tree U (bottom) ....... . .... 16 

3.1 (a). The soil water content -profile at site 0, as measured 

gravimetrically on exhumed soil cores, and by neutron 

probing irnrnediatelly afterwards ........................ 21 

3.1 (b). The soil water content profile at site 10, as measured 

gravimetrically on exhumed soil cores, and by neutron 

probing irnrnediatelly afterwards ...................... . . 21 

3.2. Comparison between the count ratio from the Troxler 1255 

probe with the soil water content measured by the 

Troxler 1255 ................................................ 23 

3.3. The factory and new calibration at Site A and Site B ........ 24 

3.4. A comparison of results of the TDR experiments on four 

different soils possesing a wide range of textures with 

the results of other experiments which used a variety of 

technique and soils (after Topp et al. 1980) ............... 26 



ix 

3.5. Laboratory TDR calibration sampling and measurement 

locations ................................... .... .. . ...... ... 27 

3.6. The three TDR instruments and a soil bucket as used in the 

calibration experiments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 9 

3.7. The relationships between TDR and corer sampling water 

content in the laboratory calibration .......... .. ........... 30 

3.8. The laboratory and field calibration of TDR no. 104968 

which was used in the experiment .................. .. ........ 32 

4.1. The volumetric soil water content profiles for Tree I 

measured on Day 1 (.), Day 18 (x) and Day 29 (o). For depth 

0 - 400 mm TDR apparatus was used, while for depth 

400 - 1800 mm a neutron probe was used ...................... 37 

4.2. The volumetric soil water content profiles for Tree U 

measured between on Day 1 (.) and Day 29 (x). For depth 

0 - 400 mm TDR apparatus was used, while for depth 

400 - 1800 mm a neutron probe was used ...................... 38 

4.3. Soil water storage from Oto 1900 mm depth with time for 

irrigated tree (Sites 1 - 4) and unirrigated tree 

(Sites 5 - 9), obtained by combining neutron probe and 

TDR data .............. .. ... ....... ........................ . . 41 

4.4. Change in soil water content profiles during two extraction 

periods for irrigated tree, (a) from Day 7 to Day 16 

(9 days), (b) from Day 23 to Day 29 (6 days) .......... .... . . 43 

4.5. Change in soil water content profile during two extraction 

periods for unirrigated tree, (a) from Day 7 to Day 16 

(9 days), (b) from Day 23 to Day 29 (6 days) ..... . . ..... . ... 45 

4.6. Soil water pressure potential measured by electronic 

tensiometer around Tree I (a) and around Tree U (b) ......... 49 

5.1 (a). The typical relationship between temperature and time 

for heat pulses upstream (Xu= 5 mm) and downstream 

(Xd = 10 mm) • • • • • • • • • • • • • • • • • • . • • • • • • • • • • • • • • • • • • . • • • • • 52 



5.1 (b). The typical difference between downstream and upstream 

temperature. t 0 is the time delay until the upstream 

and downstream temperatures are equal 

X 

(after Swanson, 1962) .................................. 52 

5.2 (a). A diagram showing the position of heater probe and 

thermistor beads used to monitor the heat pulse 

velocity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 

5.2 (b). Two of three sets of heater probes and thermistor beads 

in the trunk of Tree U ................................. 55 

5.3 (a). The heat pulse instruments in Tree U, consisting of 

3 sets of thermistor beads and heaters (1) which were 

controlled by heat pulse and heater circuits (2) and 

connected to a Campbell CR21X data logger (3) powered 

by a 12 Volts lead-acid battery (4) and connected to 

an audio cassette recorder (5) ......................... 57 

5.3 (b). Data saved by the cassette recorder (5) were 

transferred by a tape reader (6) to a computer (7) ..... 57 

5.4. The pattern of daily water use (litres/hour) from Day 2 to 

Day 24 as measured by the heat pulse technique for Tree I 

---) and Tree U (-----) .................................. 58 

5.5. Daily water use measured by the heat pulse technique from 

Day 2 to 24 for Tree I (---) and Tree U (-----) .......... 60 

5.6 (a). The Delta-T Device Porometer ........................... 63 

5.6 (b). Using the porometer to measure the stomatal resistance 

of the bottom of a leaf ................................ 63 

5.7. The correlation between diffusion time (s) and plate 

resistance (s/cm) for porometer calibration under various 

relative humidity and temperature conditions ................ 64 

5.8. Stomatal resistance of apple leaves from Trees I and U. 

Days 14, 16 and 15 showed a significant difference 

(indicated with*) in the stomatal resistance values between 

Trees I and U ............................................... 66 



xi 

5.9. Diagram of the pressure chamber apparatus ................... 71 

5.10. The pressure chamber apparatus used in the research . . ....... 71 

5.11. L~af water potential measured by pressure chamber apparatus 

on Days 8 to 30. The first and second numbers in the 

parenthes show the number of leaf samples of Trees I and U 

respectively being measured ................................. 73 

5.12. The measuring cylinder used to monitor apple growth ......... 76 

5.13. The .relationship between apple fruit volume (ml) and time 

(days) for Tree I (above) and Tree U (below) ... . .. . ......... 77 

5.14. The average apple fruit volume with time .................... 7 8 

6.1. Soil water extraction from 0 - 1100 mm and 1100 - 1900 mm 

depth around Trees I and U during the first and second 

period of extractions ................................. . ..... 83 

6.2. Water use per tree measured by heat pulse technique and 

regional evaporation of Priestley and Taylor method ......... 85 

6.3. The comparison between water use per tree measured by 

neutron probe (assuming full root distribution ), heat pulse 

tec hnique and regional e vaporation of Priestle y and Tay l or 

method (assuming 66 percent canopy c over) ................... 8 7 

6.4. The relationship between the Prie s t ley and Taylo r estimates 

and the heat pulse measurements ............................. 8 9 

6.5. The average "field capacity" and "stress point'' . .. .......... 90 

6.6. The relationship between meteorological data to the stomatal 

resistance and leaf water potential .. . ............ . ......... 97 

6.7. The wetting hydraulic conductivity, K(~), of Twyford sand 

loam from three cores, with disc and ring measurements for 

varying ~O (after Clothier et al., 1989) .................... 100 



xii 

L I S T 0 F T A B L E S 

2.1. Meteorological data for Havelock North (Station no. D 9668A) 

from 13 December 1988 to 10 January 1989 (supplied by NZ 

Meteorological Service) ..................................... 11 

3.1. The relationships between volumetric soil water contents 

from TDR using factory calibration (Y) and gravimetric 

sampling (X) •...•.•.•.•• • ..•• • ..•.••••••..•••.•.••••.•.....• 28 

4.1. Soil water storage change per unit land area during 

4.2. 

5 .1. 

5 . 2. 

5 . 3. 

5.4. 

extraction periods .. . . ...... . . . ...... . ...................... 47 

Soil water storage change per tree .......................... 47 

The diffusion resistance and time correction ................ 67 

Average and standard deviations of stomatal resistance 

(s/cm) va l ues a nd t -test parameter . . . . ........ . ............. 68 

Average l eaf water pressure p otential (MPa) .... . ... . ..... ... 7 4 

The average of apple fruit growth rate (ml/day) of both 

Trees I and U . .. . .. . .......... . . . ..... . ............. . ....... 79 



1 

CHAPTER I 

THE WATER BALANCE OF APPLE TREES 

1.1. INTRODUCTION 

Fruit and vegetables are in the top six New Zealand exports, 

after meat, wool, butter, forest products, and aluminium and alloys. 

The value of fruit and vegetables is about 7 percent of the national 

export receipts. Apples are the second most important commodity in 

the fruit export sector after kiwifruit (HEDC, 1982) . The national 

apple production is about 155 million tonnes/annum (Wong, 1987). Thus 

apples are an important New Zealand export commodity. 

Apple orchards usually use irrigation systems to overcome soil 

water deficits during dry periods when evaporation is greater than 

rainfall, and so to obtain the maximum yield and fruit quality. Using 

an irrigation system involves defining when and how the optimal amount 

of water should be applied in an orchard. Otherwise the orchard will 

received either over-irrigation or under-irrigation. 

has several disadvantages, namely : 

- higher irrigation expenses, 

- nutrient leaching which can affect ground water 

quality and increase fertiliser cost, 

- plant health problems due to water logging, 

- decreased yield and fruit quality 

Over-irrigation 

On the other hand, under-irrigation causes plants to become unhealthy 

due to water stress and low soil nutrient availability. 

important to investigate the amount of irrigation needed. 

Thus it is 

Irrigation is a water input, which is a component of the water 

balance. The understanding of the balance of the water inputs and 

outputs in an apple orchard is very important, because an unfavorable 

water balance can affect the apple tree development which can affect 

the export quantity and quality. 



2 

1.2. THE WATER BALANCE 

Mass conservation can be used to explain the soil water balance 

(Hillel, 1982). In the root zone of an orchard over any time interval 

~t, the change in storage equals the water inputs minus the outputs. 

The inputs are rainfall (R) and irrigation (I), and the outputs 

are evaporation (E), drainage below the root zone (D) and surface 

runoff (S) . In this thesis evaporation refers to all water vapour 

loss to the atmosphere, and so includes transpiration, evaporation 

from the soil and evaporation of intercepted water. So 

R + I - E - D - S ( 1.1 ) 

where ~Wis the change in the water storage in the root zone, and all 

terms have dimensions of length, being equivalent depths of water. 

1.2.1. WATER INPUTS 

Water inputs in the orchard are rainfall and irrigation water. 

Rainfall and irrigation are treated as independent variables and must 

be measured (Scotter et al., 1979). When water inputs bring the soil 

to "field capacity", then the soil water deficit is assumed to be 

zero (Taylor and Ashcroft, 1972). Excess water input leads to water 

redistribution and drainage beyond the root zone. But drainage losses 

during summer will be small if the irrigation system is well managed. 

In orchards infiltration with water ponded on the surface is 

rare. It usually only occurs during heavy rain and on less permeable 

soils. Most of the water falling on the land, as either rain or 

sprinkler irrigation, infiltrates as unsaturated flow (Philip, 1969). 



3 

1.2.2. WATER OUTPUTS 

Given no surface runoff, the water outputs in the orchard are 

evaporation, and drainage water, which only occurs when there is 

excess water input. The understanding of evaporation is very 

important in agriculture and horticulture because evaporation is a 

major term in the soil water balance. 

When the humidity in the atmosphere outside the leaf cuticle is 

lower than in the intercellular spaces within a leaf, there is 

molecular diffusion of vapour outwards through the stomata. The 

number and degree of opening of the stomata, and the humidity gradient 

control the rate of diffusion. The continual transpiration from 

leaves needs three physical conditions. Firstly, a supply of energy 

must be available to provide the quite large latent heat of 

vaporation. Secondly, there must be a lower vapour pressure in the 

surrounding air than at the evaporating surface. Thirdly, there must 

be a continuous supply of water . This is the rate limiting factor for 

transpiration in dry condition (Rose, 1966; Meidner and Sherif, 1976; 

Milburn, 1979) . 

Transpiration from plant leaves causes a water potential gradient 

between leaves and roots. The root water absorption and sap flow 

depend not only on the leaf water potential, but also on the soil 

water potential and hydraulic conductivity. On the other hand, the 

atmospheric environment largely determines the rate of evaporation 

from the leaves, because the opening of stomata depends on 

environmental variables such as the solar radiation received, and the 

humidity gradient between inside and outside the stomata. Thus, the 

whole soil-plant-atmosphere continuum affects the amount of water lost 

by evaporation (Philip, 1966). Often however the atmosphere has the 

dominant effect on the rate of evaporation as the process is usually 

energy limited . 



4 

When evaporation from bare soil can be ignored, such as in a 

region which is completely covered by vegetation, and soil water is 

always available, the root water extraction rate can be assumed to be 

equal to the evaporation rate. Then, provided adequate soil water is 

available, estimates of regional evaporation using climate data can be 

used to estimate root water extraction (Thornthwaite, 1948; Blaney and 

Criddle, 1950, Penman, 1948, Priestley and Taylor, 1972). The actual 

evaporation is usually measured only for research purposes. 

1.3. THE STUDY 

The aim of the study was to investigate the soil water use by two 

apple trees in Hawke's Bay. 

One apple tree was covered by a rainout shelter over the soil 

surface to eliminate any water input from irrigation and rainfall, and 

to prevent any water output from soil and grass evaporation. Thus 

transpiration is the only water use around this unirrigated tree. 

The other apple tree had no any cover. This tree received water 

inputs from both irrigation and rainfall. The water use consisted of 

transpiration and both soil and grass evaporations around the tree. 

The water use of both trees was investigated by using 

- neutron probing and time domain reflectometry to monitor spatial 

and temporal soil water content changes, reflecting the root water 

extraction, 

- the heat pulse technique to measure the sap flow in the tree, 

- meteorological data to estimate regional evaporation around the 

orchard. 

The unirrigated tree was expected to come under water stress, 

while the irrigated tree was expected to remain unstressed. To detect 

the level of plant water stress, a porometer was used to measure the 

stomatal resistance and a pressure chamber was used to measure the 

leaf water pressure potential. Soil matric potential was measured 

with tensiometers. Finally, a measuring cylinder was used to monitor 

the apple fruit growth on the two apple trees. 



5 

CHAPTER II 

SITE DETAILS AND METHODOLOGY 

2.1. THE SITE 

The research was carried out in the orchard of Mr. Van Howard, 

called the "RED APPLE", along Napier Road, Hastings, about 5 km from 

Havelock North. 

The soil of the orchard was a permeable alluvial Twyford soil. 

The Karamu Stream runs adjacent to the orchard. Excessive water 

inputs, due to heavy rainfall or over-irrigation of the orchard, could 

cause leaching losses of plant nutrients to below root zone depth, and 

eventually p ollute the stream. This would contribute to the serious 

weed and algal problems in the Karamu Stream. The difference in leve l 

between the orchard soil and the stream was about 4 m. 

The irrigation system used on the orchard was a micro jet type. 

The water was drawn from a well. The micro jet emitters could 

irrigate to a radial distance of about 2 m, and were located between 

alternate trees along each row. 

2.1.1. THE SOIL 

The alluvial Twyford sandy loam in the orchard is a Recent soil 

formed through regular deposition of materials eroded from soils and 

rocks in the catchment drained by the Karamu stream . Because the soil 

materials were transported and deposited by the stream during its 

formation (Gibbs, 1980), the soil shows substantial variation within a 

distance of a few metres. 

profile. 

Thus there is usually no well-defined soil 



6 

The classification of the Twyword sandy loam in the NZ Genetic 

system is a Recent soil from greywacke alluvium, while in US Soil 

Taxonomy it is a Fluventic Udic Ustochrept, coarse loamy mixed mesic. 

The soil occurs near most of the streams and rivers of mid Hawke's 

Bay. The soil's bulk density is about 1.25 to 1.4 Mg/m3. 

The Twyword s oils range from well drained to excessively drained. 

A humus rich topsoil and a more stable structure can develop when the 

soil has not been flooded for an extended period. 

The soil is a good soil for both agriculture and horticulture. 

To make the most of its potential productivity, the soil needs 

irrigation. On the other hand, irrigation with excess water can cause 

the accumulation of drainage water in lower areas (Pohlen et al., 

1947) . 

2.1.2. THE ORCHARD 

The apple trees (Malus sylvestris Var. Royal Gala) were planted 

about 18 years ago at a spacing of 5 m between rows and 3.6 m between 

trees in the rows. Figure 2 .1. shows typical apple trees in the 

orchard during the fruiting period. 

Grass strips between the rows were about 2.6 m wide, and bare 

soil strips were about 2 . 4 min width in the tree rows. The grass was 

regularly mown to keep it short. 

along the orchard border . 

2.1.3. ROOT DISTRIBUTION 

Shelter trees, poplar, we re planted 

Since the apple trees are mature (18 years old) the roots had 

fully explored the topsoil between them (K.A. Hughes, pers. comm.) . 

Rogers and Head (1969) also found that the roots of mature apple trees 

had a uniform distribution horizontally. 

Figure 2.2(a) shows the average root length density with depth 

averaged over all radii. The highest root density was found in the 

top 200 mm and the next highest between 200 and 400 mm depth. The 



7 

Figure 2.1. Apple trees in the orchard during the fruiting period. 



( a) 

0.00 

0.1 

0.3 

0.5 

Depth 0
·
7 

(m) o.9 

1.1 

(b) 

fl.A 

Root density ( 0.01 x mm/mm3) 
0.05 0.10 0.15 0.20 0.25 

RADIUS FRON THE THEE< m) 

0.30 

fl .S 1 .n 

r grass strir 

1.s 2.e 2.s 3.e 
0 .0 ,-..,...--,-,,--,--,-.,..--t~~---,t---.----,--,-,--r-..--.--+-.--,--.--. 

0.2 -~--,.......--........ ...,......r-t--t--t-t--t--t-t-t--t--~ 

0 .4 f----71r----?t-'::;,-'7t---t-1-t-t--t-r-t--i_."-+-r-+-F-i-+rl--< 

n = 33 

n = 31 

n = 29 

n = 34 

n = 27 

n = 27 

n = 22 

n = 27 

8 . 6 -t'----,,!'----orlc..,--,,c;..,-+-+--+--i-+--+-1-+--+-~~ LEGEND : 

8 

s.e. is standard 
error. 

n is number of 
observations. 

DEr TH 0 B -""-T-r-,-,~---,+_....,~....,__.~.._....,,._-+-1-+~,_.,.~ ( "' ) · ,- DENSITY ( fl . 81 x nl'l/ion3 

1.0 +--'--'--'-71~--,j~.....C:..-.,,.+-.:;..._-.,,.+-~L..L+-~--L.,J 

1.2 -l'-~-tc------,-!.:;..._-~::__---,1-<'---.,..::-___,. 

1.1 -t"--""7"'"-7'~--,,.j~--.,,.+-.:;..._--,,..+,<'---.,i,,c:::...__--,1 

1.6 ~--C.--JS::.----'<:.-..L..--Z'----""----"C.------1 

n=39 n°34 n-::71 P=-39 11=30 n=14 

0-0 .1 

fl .Hl .Z 

0 .2--ll.S 

) B .5 

Figure 2.2. The average root length density measured with depth for 
all radii (a) and with depth in various radial classes 
(b) (K.A. Hughes, pers. comm.) 



9 

high density between 800 - 1000 mm depth is probably due to the 

existence of a finer soil layer at that depth. The soil water data in 

Figure 4.1. also suggest this. Measurements were only made to 1500 mm 

depth, and a reasonable density of roots was still seen at this depth. 

So there was apparently root growth below 1500 mm depth. Radially, 

2.0 to 3.0 m radius from the tree showed the highest r oot density, 

probably due to overlapping roots from adjacent trees there (see 

Figure 2.2(b). 

2.1.4. THE CLIMATE AND WEATHER 

The rainfall in Hawke's Bay is usually fairly low. Average 

monthly rainfall for Havelock North is shown on Figure 2.3., based on 

data from 1950 to 1980 at station no. D96689, Havelock North, 9 m 

above sea level. The average Penman Regional Evaporation data (Er) 

for Napier are also shown. The distance between the Havelock No rth 

meteorological station and the experimental site is about 1 km, s o 

that rainfall at the station will·be similar to that at the site . The 

evaporation values show the regional evaporation around Havelock 

North . 

Figure 2.3. explains the typical monthly water balance in 

Havelock North. Water surplus usual occurs from April to September. 

On the other hand, the region usually has a soil water shortage from 

October to March. The annual average rainfall is 798 mm and annual 

regional potential evaporation is 967 mm. 

Table 2.1. shows the meteorological data, such as maximum and 

minimum temperatures (°C), sunshine (hours/day), rainfall (mm), from 

the beginning to the end of observations. The regional evaporation 

estimate of Priestley and Taylor was calculated using procedures 

explained in Appendix A. For convenience in later discussion the days 

of intensive field study are numbered from 1 to 29 as shown in 

Table 2.1. 



160± 
150 
140 \ 

130 t' \ \ 
120 \ 

110 

E . 1001 vaporat1on 90 
or ao 

rainfall 
70 

(mm) so 

50 

40 

30 

20 

10 

\ 

' ' ' ' \ 
\ 

\ 

' ' ' ' '-
'- / 

'- / - - -

/ 
/ 

/ 

/ 
/ 

I 
I 

I 
I 

/ 

I 

I 

/ 
/ 

/ 

/ 
/ 

/ 

--- RAINFALL 

- --- EVAPOR.:n.'l'ION 

0 -+--, ---------"---+---....;...._----"----+--------------'-----'----i 

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 

Figure 2.3. Average monthly rainfall (1959 to 1980) for Station 096689, 
· Havelock North, 9 m above sea level, and average monthly 

Penman evaporation for Napier (NZ Met. Service, pers. comm.). 

..... 
0 



Tab1e 2.1. Meteoro1ogica1 data for Have1ock North 
(Station no. D 9668A) from 13 December 1988 
to 10 January 1989 
(supp1ied by NZ Meteoro1ogica1 Service). 

TEMP. TEMP. SUNSHINE p & T* RAINFALL 
DAY: DATE: MAX. MIN. 

(OC) (OC) (hours) (mm) (mm) 

1 DEC 89,13 24.3 16.2 2.0 2.7 
2 14 25.0 8.2 10.2 5.5 
3 15 17.4 15.1 0.0 1. 8 
4 16 20.3 8.4 10.3 5.3 
5 17 22.7 5.5 13.6 6.4 
6 18 29.7 8.5 11. 5 6.2 
7 19 22.0 9.6 6.0 3.9 
8 20 22.5 16.6 7.8 4.9 
9 21 25.0 15.3 9.5 5.6 

10 22 26.1 16.1 3.2 3.3 
11 23 21. 0 14.5 9.1 5.2 
12 24 21. 5 13.5 13.6 6.8 
1 3 25 2 2 .9 9.9 13. 0 6.5 
14 26 23.2 13. 7 14.1 7 . 1 
15 27 23.5 15.9 14.0 7.2 
16 28 24 . 4 12.3 3.3 3.1 0.8 
17 29 21.2 18.3 0.0 2.0 14.0 
18 30 22.3 17.9 0.4 2.1 8.3 
19 31 23.4 17.6 2.4 2. 9 10.4 
20 JAN 90, 1 20.9 16.5 0.1 2.0 6.2 
21 2 22.9 17.6 2.9 3.1 0.3 
22 3 24.1 16.5 10.7 6.0 
23 4 25.5 14.5 10.6 6.0 
24 5 24.4 12.4 7.9 4.8 0.1 
25 6 23.8 14.5 5.5 4.0 
26 7 22.0 13. 6 1. 3 2.3 
27 8 22.8 16.9 0.3 2.0 0.7 
28 9 24.5 18.8 0.0 2.0 2.8 
29 10 27.0 17.2 2.5 3.0 

SUM 185.8 123.7 43.6 
MEAN 23.3 14.2 4.3 

{ 

* Priestley and Taylor estimate of evaporation 
{ see Appendix A). 

11 



12 

2 .1.5. THE RAINOUT SHELTER 

A rainout shelter was put in place around one tree to prevent all 

water inputs to this tree. Thus, the soil water extraction was 

assumed to be equal to the transpiration rate, as the shelter also 

prevented evaporation from the bar~ soil under the tree and the 

neighbouring grass strip. 

The rainout shelter, which was also used in the summer of 

1987/1988, has · a steel frame, plastic "Lusterlite" roofing sheets, and 

a heavy plastic skirt. The steel frame, which was built by DSIR 

technicians, was about 6 m long and 2.5 m wide. The height was about 

0.2 mat the centre and about 0.1 mat the edge. Inside the area 

covered by the shelter were tensiometer tubes, numerous TDR wave 

guides, neutron access tubes, and the heat pulse equipment. The 

roofing sheets were fixed on the steel frame and were opened when 

measurements were made. One roofing sheet had a hole as large as the 

trunk diameter to get the tree through the rainout shelter. The three 

pieces of heavy plastic skirt were attached on the edge of the steel 

frame and were joined to each other by zips. The grass covered by the 

skirt became yellowish due to a lack of solar radiation. Figure 2.4. 

show the rainout shelter when it was open for measurement. Figure 

2.5. shows the effect the rainout shelter had on soil water content 

and so colour. 

2 .1. 6. THE EXPERIMENTAL LAYOUT 

The research concentrated on two individual apple trees, Tree U 

was the covered unirrigated tree and Tree I was an ~rrigated tree. 

Both irrigated and unirrigated trees were in the 4th row from the west 

corner of the orchard. The Trees I and U were the 6th and the 10th in 

that row (see Figure 2.6.). Subsequent references in this thesis to 

"irrigated" and "unirrigated" trees will always refer to Trees I and U 

respectively. 



13 

Figure 2.4. The rainout shelter around Tree U, open for measurements . 

Figure 2.5. The difference between the soil within and outside the 
covered area around Tree U. 



20 m 

t • • 
Planting distance is 5 x 3. 6 m 

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 o is apple tree 

0 0 ~g~ 0 0 0 0 0 0 0 0 0 0 0 
O O O lt£?.i:'.r~L c.9.red grrig.Qd t,g O O O O O 0 
0 0 O W€lt 0 0 0 0 0 O O O O O 0 ... .,. .... ,.,.,.,.,,.,.. • g\e 0 

North 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
• g\e A 

\ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 'O O O O O O O O O O O 0 

irrigated tree o o o o~o o o o o o o o o o o 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 ~g~oo O O O O O O O O 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0----

Fence 
... _._.:_..:._:: .. · ~-: :..:.: ·:-~: :·, = ,: ... ;-: .. :.:_ ·:- .. ·:- .·.· ::. -.r=..., ._ .. :.-:, . . .. · .. : .: .. ~ -· .. : ':·-1.:-::-::.\:."~ 
-~ · - - · · ._ .. -~ :.·: • . .- .·. ·: ~ - . : . -:-Ka (Q 0:1 !J ::?~C~.q 0:1_:_:-:··:-~. :-.:.:-::--:: ·:: 77./__:~ 

Figure 2.6. The layout of the experimental site in the apple orchard. 
I-' 
,i,. 



15 

Around the irrigated tree there were 4 neutron probe access tubes 

(no. 1, 2, 3 and 4), 4 sets of TDR wave guides (no. 1, 2, 3 and 4), 

3 tensiometer tubes (no. 1, 2 and 6) · (Figure 2. 7.). Each set of TDR 

wave guides consisted of 200, 400 and 600 mm depth wave guides. 

Around the unirrigated tree there were 5 neutron probe access tubes 

(no. 5 to 9), 11 sets of TDR wave guides (no. 5 to 15) and 6 

tensiometer tubes (no. 3, 4, 5, 7, 8, and 9) (Figure 2.7.). In both 

trees the heat pulse technique equipment was installed to measure the 

plant sap flow. 

2.2. MEASUREMENTS 

The aim of the measurements was to e xplore the different behavior 

of the irrigated and unirrigated trees during the experiment. 

The field experiment was carried out between 16 November 1988 and 

10 January 1989. The rainout shelter was put over on 12 December 

1988. The micro jet emitters around the unirrigated tree had been 

blocked several days before the rainout shelter was installed. The 

blocking caused a lower initial soil water content around the 

unirrigated tree . Typically an irrigation of about 24 mm was applied 

as a single application each week, with an application rate of about 

2 mm/hour. 

Irrigations were carried out on 15 December, immediately after 

fixing the rainout shelter, and on 29,December 1988. The other water 

input was rainfall. These water inputs either through rainfall or 

irrigation affected only the soil water storage around Tree I, the 

irrigated tree. 



(m) 

0 2 TREE I (Tree 6) 

emitter:?t 
04 • • • 4 

• 01 · 
·rroe 5 • 1 • 2 • 1 02 

• 6 02 

e applo tree, 

0 neutron probe access tube, 

• tensiometer tube, 

• 

close up 

• • • 20 40 60 
• • • 
cm clepth 

0 

• Tree 9 

TOR probe 

TOR D 
probe 

(m) 

heat pulse instrument 

2 

OB 
OB 

Irrigation pipe"""'\, 

TREE U (Tree 10) 

09 • 05 06 
09 0 06 

• 5 

014 
DIS • 4 

• 3 • 0 
• 7 

• 10 

D • • 9 
13 

5 

O 11 

Irrigation pipe~ 

03 
D 3 

07 
07 

North 

/ 

o 12 

16 

Figure 2.7. Layout arowid Tree I (above) and Tree U (bottom). 

• --Tree 7 

. --
Tree 11 



17 

2.2.1. SOIL WATER CONTENT MEASUREMENT 

The soil moisture content was measured using neutron probing and 

time domain reflectometry from 16 November 1988. The neutron probe 

instrument was used to measure the in situ volumetric soil water 

content and its change in time and space. In this study the probe was 

used to measure soil water content (m3/m3) at 400, 600, 800, 1000, 

1200, 1400, 1500, 1600, and 1800 mm depth. Multiplying the soil water 

content (m3/m3) by the depth interval (m) gives the soil water storage 

in units of equivalent depth of water. The changes in soil water 

storage between 400 to 1800 mm depth could be monitored by repeated 

measurements of soil water content. 

Since the research needed precise information about soil water 

content near the soil surface where the roots were densest, a portable 

IRAM-Time Domain Reflectometer was used. The neutron probe was not 

used to measure soil water content near the soil surface due to the 

escape of fast neutrons from the soil surface. Time Domain 

Reflectometer (TOR) measures the average volumetric soil water 

content (m3/m3) over the length of the guides, based on the dielectric 

constant of water in the soil. This allowed the water content of the 

top 400 mm of soil to be found. Thus the soil water storage from Oto 

1900 mm depth could be found when the TDR and neutron probe data were 

combined. 

2.2.2. SOIL WATER PRESSURE POTENTIAL MEASUREMENT 

Tensiometers were used to measure the soil water pressure 

potential (kPa). The tensiometers used during the research consisted 

of nine tensiometer tubes and one electronic pressure transducer. The 

tensiometer tubes consisted of PVC pipe with a thin ceramic cup on the 

bottom and a rubber septum on the top. The tensiometer tubes were 

filled with boiled water to avoid gas bubbles. Each tube was immersed 

into a prepared hole, and its ceramic cup adjusted to ensure a good 

contact with soil around the cup. The electronic pressure transducer 

had a hypodermic needle attached that could pierce the rubber septums 

to measure the pressure potential inside the tube. 



18 

2.2.3. SAP FLOW MEASUREMENT 

Using heat as a tracer to measure the sap flow, the amount of 

plant water use can be found. This appears to offer a sensitive 

method for monitoring the dynamic water status of plants. This 

technique is called the heat pulse sap flow technique. The technique 

measured the sap flux density (m/day) directly in stems of intact 

apple trees with minimal interference. By multiplying the sap flux 

density (m/day) by the cross sectional area (m2 ), the plant water use 

(m3/day) can be calculated. 

2.2.4. STOMATAL RESISTANCE MEASUREMENT 

To monitor the level of plant stress, and to understand the 

diffusive resistance of leaves to water vapour, stomatal resistance 

was measured . A porometer, (made by DELTA-T DEVICES), was used to 

measure the stomatal resistance (s/cm) 

2.2.5. LEAF WATER PRESSURE POTENTIAL MEASUREMENT 

A portable pressure chamber was'used to measure leaf water 

pressure potential (MPa). The pressure chamber technique has become 

the standard method to evaluate plant water status in the field. 

2.2.6. APPLE FRUIT GROWTH MEASUREMENT 

The plant under water stress may experience detrimental effects, 

especially reduced apple fruit growth. 

was used to monitor apple fruit growth. 

A special measuring cylinder 



19 

CHAPTER III 

NEUTRON PROBE AND TIME DOMAIN REFLECTOMETER CALIBRATIONS 

AND TENSIOMETER DESCRIPTION 

3.1. NEUTRON PROBE 

3.1.1. THEORY 

The neutron scattering technique is quite popular for soil water 

studies, as the technique can be used to monitor the volumetric soil 

water content in the field without destructive sampling. The 

principle of the technique is that hydrogen nuclei in soil water are 

assumed the major soil component reducing the kinetic energy of fast 

neutrons, making them become slow neutrons in the thermal condition 

(Goodspeed, 1981). This occurs when the fast neutrons collide with 

the hyd~ogen nuclei in the soil. Energy transfer occurs in the 

collisions, so that the velocity of the fast neutrons decreases. 

After about 18 collisions, the fast neutrons become slow or 

thermalized neutrons (Visvalingam et al., 1972). The density of 

hydrogen nuclei counted is a reflection of the fraction of the soil 

volume consisting of water. 

The measurement can be carried out by lowering the probe down 

into a 50 mm diameter aluminium access tube. 

emits fast neutron which pass into the soil. 

The radioactive source 

The detector, which is 

situated near the source in the probe, measures the intensity of the 

slow neutron 'cloud' in the sphere of influence. The scaler counts 

the therrnalized neutrons detected. The count rate is proportional to 

the volumetric soil water content (Shirazi et al., 1976). The number 

of neutrons counted appears on the scaler screen after counting for a 

certain period of time. 

adjusted to 15, 30, 45, 

The period of time for counting can be 

60 or 120 seconds. A 30 second period 



20 

counting time was used during the experiment. The count of slow 

neutrons, divided by the standard count, can be used to calculate the 

volumetric soil water content by using a calibration equation 

(Greacen, 1981) . The standard count is a reading taken in the probe's 

shield. To reduce errors in measurements due to variations in 

electronic performance, radioactive decay, and due perhaps to 

temperature dependence in the electronics, the standard count has to 

be measured. 

Although the technique has many advantages, it has one 

significant disadvantage. It is that the technique cannot be used to 

measure soil water co'ntent near the soil surface. The reason is that 

the radius of the sphere of influence, particularly in dry soil, is 

quite large. Consequently, if the probe is used to measure soil water 

less than about 200 or 300 mm deep, some fast neutrons escape through 

soil surface and so cannot be thermalized and counted by the detector. 

So the soil water content will be under-estimated. The soil water 

contents near the soil surface were very important in this research, 

as most of the active plant roots were near the soil surface. Thus, 

to measure the soil water content near the soil surface another 

instrument was used, namely the Time Domain Reflectrometer, which will 

be discussed later. 

3.1.2. METHODOLOGY AND CALIBRATION 

The instrument used in the research consisted of a TROXLER 1255 

neutron probe with a 100 mCi 241Am-Be radioactive source, a detector 

placed near the source and a scaler. A calibration for this 

instrument was supplied, but it was decided to check its accuracy. 

Calibration was carried out in the beginning of 1987/1988 summer. 

The Troxler 1265 was also used during the calibration. The volumetric 

soil water content measured by the Troxler 1255 (assuming the factory 

calibration), the Troxler 1265, and core sampling were compared at 

Site O and Site 10 (Figure 2.6.). The results of this comparison are 

shown in Figure 3.l(a) and (b), and indicate that although the Troxler 

1255 and 1265 had similarly shaped profiles of soil water content with 



0.075 

o.or 
0.1 T (a) 
0.2 

0.3 

Depth o.4 
( m) o.5 

0.6 
0.7 

0.8 

0.9 

0.075 

Soil water content (m3/m3) 
0.100 0.125 0.150 0.175 0.200 

,,. 
,,. 

Soil water content (m1'm3) 
0.100 0.125 0.150 0.175 0.200 

0.225 

Gravimetric core 
samples 
at Site 0 

Troxler 1265 

-·-·- Troxler 1255 

0.225 

~:~ l ( b) I I I I I I 

0.2 

0.3 

Depth 0.4 

( m) o.5 
0.6 

0.7 
0.8 
0.9 

,,. 
' < 
' ' I 

Gravimetric core 
/ samples 

at Site 10 

"-

"" 
----- Troxler 1265 

/ ---·- Troxler 1255 

Figure 3.1(a). The soil water content profile at site 0, as measured 
gravimetrically on exhumed soil cores, and by neutron 
probing immediatelly afterwards. 

Figure 3.1(b). The soil water content profile at site 10, as measured 
gravimetrically on exhumed soil cores, and by neutron 
probing immediatelly afterwards. 

Iv 
t--' 



22 

depth, the Troxler 1265 profile was closer to the core sampling values 

than the Troxler 1255. Thus, the Troxler 1265, which had been 

calibrated in the field by Clothier (1977), was used to establish a 

new calibration equation for the Troxler 1255 which was always used 

during the experiment. 

A total of 299 contemporaneous measurements with both TROXLERs 

1255 and 1265 probes were carried out. The calibration equation of 

the TROXLER 1255 derived from regression between soil water contents 

measured by TROXLER 1265 and count ratios measured by TROXLER 1255, as 

shown in Figure 3.2. is 

e 0.376 Cr - 0.0675 (3 .1) 

where Cr is the count ratio of TROXLER 1255. 

Figure 3.3. confirms that the new calibration equation of the 

Troxler 1255 was a lot better than the factory calibration, when 

compared to core data from Site A and Site B (Figure 2.6.). 

-
3.2. TIME DOMAIN REFLECTOMETER (TDR) 

3.2.1. THEORY 

The Time Domain Reflectometer is a relatively new technique for 

measuring soil water content compared to the neutron probe. The 

instrument comprises a generator which releases an electrical pulse 

with a fast rise time step, a sampler which can transform a high­

frequency signal into a lower frequency output, and an oscilloscope or 

other devise showing the results of measurements. In the instrument 

used, the volumetric water content is displayed directly on an LCD 

screen. 

The pulse generator creates an electromagnetic pulse which then 

is transmitted into the soil through a parallel rod transmission line 

(Topp et al., 1984; Baker et al., 1982). From the length of the 



0.40 

0.35 

0.30 

Soil 0.25 

water 
ontent 0.20 

der1265 
n3/m3) 0.15 

0.10 

0.05 

. . . . 
/. /· .... .. . . . . 

. . . 

, . . 
. . . . .. 

_.•I ./ • 
.. :/_ .. 

... ✓• . . . .,,,._. . . . 

8 = 0.37603 Cr - 0.06756 
n = 299 
R2 = 0.95 
Syx = 0.019 m3 /m 3 

0.00 +------.c..+-----+-------t-----+-------+-------1 

0.00 0.20 0.40 0.60 0.80 1.00 1.20 

Count ratio (Cr) of Troxler 1255 

Figure 3.2. Comparison between the count ratio from the Troxler 1255 probe 
with the soil water content measured by the Troxler 1265. 

rv 
w 



0.25 

0.20 

!Utron probe 0.15 

"oxler 1255) 
iasurements 
(m3/m3) 0.10 

0.05 

----~--- -

+ 

* +©9 +/ 0 
0 

Factory calibration 

• FACTORY CALIBRATION SITE A 

+ FACTORY CALIBRATION SITE B 

- NEW CALIBRATION SITE A 

0 NEW CALIBRATION SITE B 

- 1:1 LINE 

= 0.0813 + 0.0268 Cr+ 0.234 Cr 2 

New calibration 
= 0.37603 Cr - 0.06756 

0.00 -----.-----4---~--+-----I 

0.00 0.05 0.10 0.15 0.20 0.25 

Corer sampling (m3/m3) 

Figure 3.3. The factory and new calibration at Site A and Site B. 
N 
,I>-



25 

transmission line (L) and the signal travel time (t) determined by the 

TDR receiver, the velocity (v) of the pulse travelling in the soil can 

be found. The velocity is used to measure the dielectric constant of 

soil (Topp and Davis, 1982; Topp et al., 1983). Topp et al. (1980) 

showed that 

V (3 .2) 

where C is the speed of an electromagnetic wave in free space (3xl0 8 

m/s) and K is the apparent dielectric constant. As v = 2L/t, K may be 

found from substitution into Equation (3.2.) and rearrangement (Topp 

and Davis, 1985) as 

K (Ct/ 2 L ) 2 (3. 3) 

The apparent dielectric constant (K) is strongly dependent on the 

volumetric water content of the soil. 

The frequency ranges used are between 20 MHz and 1 GHz. Recently 

frequency ranges up to 3 GHz have been used for the new generation of 

TDR such as the TRASE-TDR. 

The empirical relationship between dielectric constant (K) and 

volumetric water constant (0) is 

K 3.03 + 9.3 0 + 146.0 02 - 76.7 03 (3. 4) 

As the aim of using the TDR is to measure the volumetric water 

content, to find 0 Equation (3.4.) is solved to obtain (Topp, et al., 

1980) 

0 -5.3xl0-2 + 2.92xl0-2 K - 5.5xl0-4 K2 + 4.3xl0- 6 K3 (3. 5) 

Research carried out by Hoekstra and Delaney (1974) showed that 

the relationship between volumetric water content and the dielectric 

constant is relatively independent of soil type. Moreover, Topp et 

al. (1980) found that not only soil type but also soil density, soil 



26 

temperature and soluble salt content did not significantly affect the 

relationship between volumetric water content and dielectric constant 

as shown in Figure 3.4. However, it is prudent to check the 

calibration of any instrument before using it. This particularly 

applies to a new instrument which has just become commercially 

available, such as the TDR. So a calibration check was carried out. 

50 ·1-..,__...,_ __ ~--~~~-L ... .J. _ _ i_ 

0 40-
:,.: 

NIKODEM, 1966 1 

... ... ...... WOBSCHALL, 1978 

- ··- THOMAS, 1966 
---- LUNDIEN, 1971 
- ·- · DAVIS et al, 1966 

• - .... - KATSUBE, 1976 
1-
z 
~ 30· V) 

z 
0 
u 

u 
o:: 20 
1-
u 
w 
.J 
w 
0 

10 

!..-1.1.l TOR, 4 SOILS 

0 +--.--.-...--,---.--.--~-.-~-.--~-1-
0 .I .2 .3 .4 .5 .6 

WATER CONTENT Wvl 

Figure 3.4. A comparison of results of the TDR experiments on 
four different soils possessing a wide range of 
textures with the results of other experiments which 
used a variety of techniques and soils (after Topp 
et al. 1980) . 



27 

3.2.2. CALIBRATION METHOD 

TDR calibrations were conducted in the Agricultural Physics 

Laboratory, Plant Physiology Division, DSIR, Palmerston North. 

Disturbed coarse-textured C-horizon, and fine-textured A­

horizon of the Manawatu fine sandy loam soil were packed into 10 litre 

buckets at a certain known bulk density. Measurements were made in 

five buckets, three containing the coarse textured soil and two 

containing the fine-textured soil. Therefore, we had five soil water 

contents. The layout of 3 parallel pairs of 10 and 20 cm rod 

transmission lines and five corer samplings sites in the buckets is 

shown in Figure 3.5. The volumetric soil water content of the soil 

was found from the gravimetric soil water content of the cores 

multiplied by the bulk density of the soil in the bucket. 

0t---,------------125 cm 

10 litre bucket 

•G)• 
Qsoil core sampling site 

0 0 0 

CD •0• ® 0 10 cm TDR probe 

• 20 cm TDR probe 
0 0 0 

•G)• 

Figure 3.5. Laboratory TDR calibration sampling and measurement 

locations. 

The TDR used in the research was the Model 600 IRAMS System 

(Serial Number 104968) which is owned by PPD-DSIR. However, TDR 

Number 105799 (DSIR) and 105999 (Department of Agronomy, Massey 

University) were also calibrated at the same time. 



28 

Measurements using both the 100 mm and 200 mm probes were made 

with each TDR in each bucket. Three replicates were taken of every 

measurement, then five soil cores were taken. Figure 3.6. shows the 

3 TDRs used in the calibration experiments and one of the buckets. 

3.2.3. CALIBRATION RESULTS AND DISCUSSION 

The relationships between volumetric soil water contents as read 

on the TDRs (Y axis) and as found by corer sampling (X axis) are shown 

in Fig. 3.7. The linear regression equations from the data in Fig. 

3.7. are listed in Table 3.1. 

correlation coefficient (R). 

All of the equations have a high 

Figure 3.7. indicates that replicate 

corer samplings have less variability than replicate TDR measurements. 

It also shows that the TDR measurements tended to over-estimate the 

soil water content. 

Table 3.1. The relationships between volumetric soil 

water contents from TDR using factory 

calibration (Y) and g~avimetric sampling 

(X) 

TDR L Linear Regression Equation R2 ( %) 

Number (mm) 

104968 100 y 1.20361 X - 0.01941 97.0 
200 y 1.23622 X - 0.01140 99.6 

105799 100 y 1.15506 X - 0 . 01871 95.8 
200 y 1.19628 X - 0. 02727 99 . 5 

105999 100 y 1.51841 X - 0.07983 93.6 
200 y 1.17557 X - 0.00721 99.5 



29 

Figure 3.6. The three TOR instruments and a soil bucket as used in the 
calibration experiments. 



0.5 

0.4 

Volumetric 
water 

content o.3 

from TOR . 
using 

factory 0.2 

calibration 
(m3/m3) 

0.1 
+ 
+ 

* 
* 

... 

' ± 

8 
~ . 
=!!: 

0 _.....,_ __ ___._~-+----+----

0 0.1 0.2 0.3 0.4 0.5 

Volumetric water content 
from gravimetric sampling 

(m3 /m3) 

TOR no., Length 

104968, 100 mm 

+ 105799, 100 mm 

* 105999, 100 mm 

o 104968, 200 mm 

x 105799, 200 mm 

- 105999, 200 mm 

1 :1 Line 

Figure 3.7. The relationships between TOR and corer sampling 
water content in the laboratory calibration. 

w 
0 



31 

This suggests that the factory calibration was not good, as the data 

differ significantly from the 1:1 line. Furthermore, the slopes of 

the regression equation for each instrument were different. Thus, the 

different TOR instruments have different errors in their factory 

calibration. 

In this study only TDR No. 104968 was used in the field. 

Rahardjo (1988) also used this instrument in his experiment. He found 

the regression equation of his field data as 

1.062 ecore + 0.029 

n = numbers of observations 13 

R2= 0.95, 

syx = 0.026 m3 !m3 , 

(3. 6) 

where 0TDR is the soil water content measured by the TOR with 200 mm 

probes, 

Score is the soil water content measured by corer sampling to 

200 mm depth in the field, 

R2 is the correlation coefficient, and 

Syx is the standard error of the regression estimate. 

At-test of the data (Gomez and Gomez, 1984) showed there was no 

significant difference between laboratory and field calibration data 

for TDR No. 104968. Thus data from both the laboratory and field 

calibrations of TDR No.104968 were combined to create the final 

calibration equation used in the field study to be described. 

The final equation found was (see Figure 3.8.) 

1.18397 0field - 0.01387 (3. 7) 

R2 0.98 



0.51 

0.4 

TOR 
volumetric 

water o.3 
content 
using 

factory 0.2 

calibration 
(m3/m3) V, 

I · . 

;+ 
;\:. /+ . / 

h 
//. 
/ 

-t / 
/ 

/: 

0.1 0.2 0.3 0.4 

Volumetric water content 
from core sampling 

(m3 /m3) 

100 mm probe in the bucket 

+ 200 mm probe in the bucket 

X Field data {1988) 

0.5 

8TDR = 1.18397 8Fieid - 0.01387 

1 : 1 Line 

Figure 3.8. The laboratory and field calibration of TOR 
no: 104968, which was used in the experiment. w 

N 



33 

Thus, the corrected soil water content measured by TDR No. 104968 

(9field) is 

0.844616 0TDR + 0.01171 (3. 8) 

3.2.4. APPLICATIONS 

The applications of TDR in the research were, firstly to measure 

the soil water content near the soil surface from Oto 400 mm depth, 

and secondly to measure the soil water change around both Tree I and 

Tree U from Oto 200, 0 to 400 and Oto 600 mm depths. 

3.3. TENSIOMETER 

The tensiometer is an instrument for measuring the matric 

potential of soil in the field (Hagan et al, 1967; Michael, 1981). 

3.3.1. THEORY 

The matric potential of soil water is a measure of how strongly 

the soil water is bound to the soil matrix, which influences how easy 

it is for plant roots to extract soil water. 

The tensiometer consists of a porous cup, that is usually made by 

using ceramic material, connected through a tube to a pressure 

measuring device, with all parts filled with water (Hillel, 1980). 

The pressure difference between the tensiometer water and atmosphere 

pressure is measured. The pores of the cup must be small enough to 

stay water filled over the pressure range of interest (Kirkham and 

Powers, 1972) . 



From the formula 

- 2 y 
r 

cp 

where r is the pore radius (m), 

y is the surface tension between air and water 

7.34 X 10-2 N/m 

cp is pressure in Pa or N/m2 , relative to air pressure. 

34 

(3. 9) 

The diameter of pores used in tensiometer cups needs to be less than 

about 2.94xl0-3 mm (DeLeenheer and DeBoot, 1966). The length of tube 

used depends on research requirements. The pressure difference can be 

monitored by a manometer using mercury or by a vacuum gauge. Older 

tensiometers consist of a porous cup, a tube and a vacuum gauge. But 

recent versions of the tensiometer have a separate pressure 

transducer. This can be a Bourdon type gauge or an electronic 

pressure transducer. The tensiometer tube then consists of a porous 

cup, PVC pipe and a septum (rubber stopper) The pressure transducer 

is connected to the tensiometer tube by inserting a hypodermic needle 

through the rubber septum. 

withdrawn. 

This septum reseals when the needle is 

The tensiometer is reliable only over the range from about O to 

about - 85 kPa (Jensen, 1980; Hillel, 1971; Wither, et al., 1974, 

Michael, 1981, Curtis, et al., 1974). When the cup is in the soil 

where the soil water pressure potential is being measured, the water 

inside the cup will be attracted across the ceramic wall and tends to 

equilibrate with the soil water outside the cup. As the tensiometer 

system (in the tube) containing water is sealed completely, a negative 

pressure inside the system develops (Taylor and Ashcroff, 1972). It 

is very important to have good contact between the ceramic cup and the 

soil, so that the approach to equilibrium is not hindered by contact 

impedance. The time needed to achieve equilibrium will depend on the 

type of tensiometer, the soil and the soil water content. Old version 

tensiometers, which used a thicker ceramic cup with smaller pores 



35 

needed one day to reach equilibrium. Newer versions of the 

tensiometer, with thin ceramic cups with large pores, need less than 

one hour. 

3.3.2. TENSIOMETERS USED IN THIS STUDY 

Tensiometers used in the research used an electronic pressure 

transducer. The electronic tensiometer, which is made by Irrigation 

Technology and Management, Wanganui, New Zealand, consisted of a 

separated electronic pressure transducer and of a PVC tube connected 

to a fragile porous cup. Water was poured into the tensiometer tube 

to about 20 mm from the top and then the filled tube was sealed by a 

wet rubber septum. One of the advantages of using the electronic 

tensiometer is that only one electronic pressure transducer is needed 

to measure many tensiometer tubes. By inserting a hypodermic needle 

connected to the transducer into the rubber vacuum stopper, the soil 

water pressure potential value appears on the small LCD screen of the 

transducer unit. 

The tubes were installed in the soil at least 300 mm from a 

neutron probe access tube, otherwise the water in the tube could 

affect the neutron probing. There were two 300 mm depth tubes (TEN 1 

and 2), and a 600 mm depth tube (TEN 6) around Tree I. Around Tree U 

were three 300 mm depth tubes (TEN 3,4 and 5) and three 600 mm depth 

tubes (TEN 7, 8 and 9). All tensiometer tubes still had water in them 

on 26 Dec 1988. A correction of 3 kPa and 6 kPa was subtracted from 

the 300 mm and 600 mm depth readings respectively, to correct for the 

gravitation potential difference between the cup and the pressure 

transducer. 



36 

CHAPTER IV 

SOIL WATER MEASUREMENTS 

The soil water contents were measured at O - 400 mm using TDR, 

and at 400, 600, 800, 1000, 1200, 1400, 1500, 1600 and 1800 mm depths 

using a neutron probe. 

4.1. SOIL WATER CONTENT PROFILES 

The relationship between soil water content and depth gives the 

profile of soil water content at a particular site. Four profiles 

were investigated around Tree I (Sites 1 to 4), while around Tree U 

five profiles were measured at Sites 5 to 9. Data of Site 8 were not 

considered since the site had an unexpected leak. The soil around 

each tree had its own typical characteristic water content profile. 

Data for Sites 1 to 4, the profiles around Tree I, are shown in Figure 

4.1. Data for Sites 5 to 9, around Tree U are shown in Figure 4.2. 

Finer textured soil tends to contain more soil water than coarser 

textured soil, so bulges in the water contents profiles probably 

signify the presence of finer textured soil layers. The data in 

Figure 4.1. suggest that around Tree I there were two finer textured 

soil layers, one at about 1000 mm and the other at about 1600 mm depth 

at Sites 1,2 and 3, while there was just one finer textured soil layer 

at about 1500 mm at Site 4. In contrast Figure 4. 2. shows that there 

was just one finer textured soil layer around Tree U, at about 1500 mm 

depth, at Sites 5,6 and 9, but two finer textured soil layers were 

found at Site 7, at about 1400 and 1800 mm depth. The different 

profiles in the two figures illustrate the variability that can occur 

over a distance of only about 15 m particularly in illuvial soils. 



Soil water content (m1'm3) 
0 0.1 0.2 0.3 0 

01 ' , I 18 ),4 0-' 
! 200 tSite 2 200 Site '1 
* 400 400 

0.3 

600 600 

Depth aoo 800 

(mm) 1000 1000 

1200 1200 

1400 1400 

1600 1600 

1800 1800 

0 0.1 0.3 0 0.1 0.2 0.3 ,.... 
O· I 

I' 
0 

I : l 
200 Site 3 

200 tSite 4 
400 400 

I 
X 6 

600 600 

Depth 800 800 

(mm) 1000 1000 

1200 1200 

1400 1400 

1soo l 1600 

1800 1800 

Figure 4 . 1. The volumetric soil water content profiles for Tree I 
measured on Day 1 (.}, Day 18 (x) and Day 29 (o). For 
depth O - 400 mm TDR apparatus was used, while for depth 
400 - 1800 mm a neutron probe was used. 

w 
--.J 



Soil water content (m3/m3) Soil water content (m~m3) 
0 0.1 0.2 0.3 0 0.1 0.2 0.3 

0 0 • ] , , I " . 
200 

I / Site 5 200+ i Site 6 
400 X: 400 t ~ / . 
600 600 

Depth aoo 800 

(cm) 1000 1000 

1200 1200 

1400 1400 

1600 1600 

1800 1800 

0 0.1 0.2 0.3 0 0.1 0.2 0.3 
o· x . , 1 , o; .' . . 

200i [ I Site 7 
2001 11 Site 9 

400 \\ X · 400 ~~ X 

600 j~l 600 f, 
Depth 800 ~ 800 'i!.' 
(mm) 1000 . 1 10ooj 

1200 ~
1 

I 1200 

14001 ~ 14001 

1 soo~l 1600 I 
1800 , 1800- X 

Figure 4 . 2. The volumetric soil water content profiles for Tree U 
measured between on Day 1 ( . ) and Day 29 (x). For depth 
0 - 400 mm TDR apparatus was used, while for depth 400 -
1800 mm a neutr on orobe was used . 

w 
00 



39 

Throughout the study period, the soil around Tree I had a higher 

soil water content than that around Tree U. The uncovered Tree I 

received water inputs from irrigation water and rainfall, coupled with 

normal extraction by tree roots. Thus, the soil around Tree I showed 

soil water profile fluctuations which would be typical of other trees 

in the orchard. On the other hand Tree U, as a covered tree, did not 

have any water inputs during the study period, and the changes of soil 

water profile were caused only by root water extraction, as the soil 

was far too dry for any drainage to occur. 

The soil water content in the top 400 mm of soil around Tree I 

ranged from about 0.11 m3 /m3 on day 1 to 0.30 m3 /m3 on day 18 after 

the second irrigation. 

less pronounced, but 

Lower down in the profile the changes were 

substantial changes in water content were 

measured down to 1900 mm depth. This suggests either that there were 

some active apple roots from Oto 1900 mm depth, or that there was 

substantial drainage. 

In contrast to Tree I, the range of soil water contents from 0 to 

400 mm depth around Tree u was from about 0.08 m3/m3 on Day 1 to 0.06 

m3/m3 on Day 29. The observed changes in soil water content occurred 

relatively uniformly from Oto 1900 mm depth, indicating that active 

roots also existed from Oto 1900 mm depth around Tree U. 

4.2. SOIL WATER STORAGE 

As the radius of the sphere of influence of the neutron probe is 

about 100 - 150 mm (Greacen,1981), the neutron probe measurement at 

any depth can be used to estimate the soil water storage from 100 mm 

above to 100 mm below the probe depth. 



40 

Therefore the mathematical expression used to calculate soil water 

storage (W) from the surface to 1900 mm depth was 

W 4oo eTDR,0-400 + l0O eNP,400 + 200 eNP,600 + 

200 eNP'800 + 200 eNP,1000 + 200 8NP,1200 + 
150 eNP,1400 + lOO eNP,1500 + 150 eNP,1600 + 

2 00 eNP,1800 

where : 

w is soil water storage (mm)' 

eTDR is volumetric soil water content measured by 

eNP is volumetric soil water content measured by 

probing, 

(4 .1.) 

TDR, 

neutron 

Changes in soil water storage allow the water balance to be 

monitored. Changes indicate the balance between water inputs from 

irrigation and rainfall and water outputs from soil water extraction 

and drainage (Hillel, 1980) . Figure 4.3. shows the temporal changes 

in soil water storage from O to 190.0 mm depth at the monitored sites. 

Tree I received water inputs from both irrigation and rainfall. 

Irrigations were carried out on Days 3 and 17. The amount of each 

irrigation was about 24 mm of water applied at the rate of about 2 

mm/hour. Rainfall data were obtained from a meteorological station 

near the research site and are shown in Table 2.1. The effects of 

wetting can be recognized in Figure 4.3. as significant increases in 

profile water storage. They were measured between Day 2 and Day 4, 

and between Day 16 and Day 21. The first wetting around Tree I was 

caused solely by the first irrigation on Day 3. There was no rainfall 

from Day 1 to Day 15 and just 0.8 mm rainfall on Day 16. The second 

wetting occurred because of both the second irrigation and several 

days of significant rainfall after the second irrigation. There were 

also small amounts of rainfall between Days 20 and 29 but not enough 

to balance the root extraction. On the other hand, the data for Tree 

U show no fluctuations in profile storage since there was no wetting 

of the covered tree. 



425 
400 
375 
350 
325 
300 
275 

Soil 2so 
water 225 

storage 200 

(mm) 175 
150 
125 
100 
75 
50 
25 

I 

..,,., - .-•r = -~-==.. ~ -,··=·•-········. ·•·-···•·~"": ~T- . - - - - - - - --·=--=-= 

0 +----,,----,----,----,--+--,---,--,----,--!..-,----,---,---..----l-~---,----___!._----,-----l-~--,-~ 

0 5 10 15 20 25 30 

Time (days) 

- SITE 1 

-+- SITE 2 

~ SITE 3 

-e- SITE 4 

SITE 5 

SITE 6 

-- SITE 7 

-·- SITE 9 

Figure 4.3. Soil water storage from O to 1 ~9 m depth with time for 
irrigated tree (sites 1 - 4) and unirrigated tree (sites 5 - 9), 
obtained by combining neutron probe and TOR data. 

,,,. 
...... 



42 

Figure 4.3. shows that the amount of wetting was different for 

Sites 1 - 4. Site 3 gained about 100 mm during each period of 

wetting. Whereas Site 4 gained only about 20 mm and 60 mm at the 

first and second wettings respectively. This indicates that there was 

significant variability in the amount of wetting, presumably due to 

sprinkler non-uniformity. 

Because two significant periods of wetting occurred around Tree 

I, two periods of soil water extraction also can be recognized, one 

after each wetting period. The possible soil water extraction periods 

were from Days 4 to 16 and from Days 21 to 29. The soil water 

extraction can be estimated by the changes in profile soil water 

storage, providing there is no drainage. Drainage is negligible when 

the soil water content is below "field capacity". In case there had 

been significant drainage in the 3 days after irrigation or rainfall, 

the first and second periods of soil water extraction were taken from 

Day 7 to Day 16, and from Day 23 to Day 29 respectively. Figure 4.3. 

shows that during these periods the slope of soil water storage 

change, which indicates the amount of soil water extraction, was much 

higher around Tree I than around Tree U. 

4.3. SOIL WATER CONTENT CHANGES 

The amounts of soil water extraction from various depths, during 

the two extraction periods defined above, are shown in Figures 4.4. 

and 4.5. for Tree I and Tree U respectively. 

Figure 4.4. shows that the top 200 mm of soil around Tree I had 

by far the greatest water content changes. This suggests that the 

most active roots of Tree I (which had frequent water inputs), were in 

the top 200 mm depth. From 200 to 400 mm depth showed moderate soil 

water content changes, and smaller soil water changes occurred below 

400 mm. Extraction apparently occurred down to at least 1900 mm 

depth. Both extraction periods around Tree I show similar patterns of 

soil water extraction. If the availability of soil water is not a 

limiting factor, the r oot length density from Oto 1600 mm will be 



The soil water content change (A8) The soil water content change (A8) 

-0.05 0.00 0.05 0.10 -0.05 0.00 0.05 0.10 
0-1---~~--1---,~~~.+---.---.---..--~ 

200 

400 

600 

800 

(a) (b) 

SITE 1 

-+- SITE 2 

Depth 
(mm) 1 ooo _._ SITE 3 

1200 

1400 

1600 

1800 

-& SITE 4 

Figure 4.4. Change in soil water content profiles during two 
extraction periods for irrigated tree, (a) from Day 7 to 
Day 16 (9 days), (b) from Day 23 to Day 29 (6 days). 

.i,. 

w 



44 

correlated with the soil water content changes with depth. The water 

use profile of Tree I (Figure 4.4.) is somewhat similar to the root 

density profile (Figure 2.2.) The top of 400 mm depth had the 

greatest root density, and the most soil water extraction of Tree I 

occurred over that depth. 

Figure 4.5. shows the very different patterns of soil water 

extraction around Tree U. For this tree uptake depth was not 

correlated with root density. The extraction during the first period 

involved very small soil water content changes down to about 1300 mm 

depth, but somewhat greater soil water extraction between 1300 mm and 

1700 mm depth. Thus Tree U, which had a much lower initial soil water 

content than Tree I (see Figures 4.1. and 4.2.) extracted soil water 

from the subsoil rather than the topsoil, exactly the opposite of how 

Tree I behaved. The second period of soil water extraction showed 

little or no soil water being extracted, as the soil water content 

changes were approximately zero from the soil surface down to 1900 mm 

depth. So if water was taken up by Tree U during the second period, 

it must have come from below 1900 mm depth, or from shallower roots 

growing beyond the covered area. 

4.4. SOIL WATER USE 

The amount of soil water use per unit land area at each of the 

measuring sites during the first and second periods can be calculated 

from the changes in soil water storage between days 7 to 16 and 

between days 23 and 29 respectively. Table 4.1. shows the soil water 

use (mm) during the two periods of extraction. Tree I extracted 40.9 

+ 9.7 mm or an average of 4.5 ± 1.1 mm/day during the first period of 

soil water extraction and 24.1 ± 9.9 mm or an average of 4.0 ± 1.7 

mm/day during the second period of soil water extraction. However 

Tree U extracted only 10.4 ± 0.5 mm or 1.2 ± 0.1 mm/day during the 

first period and - 1.8 ± 2.2 mm or - 0.3 ± 0.4 mm/day during the 

second period of the extraction. 



The soil water content change (11.8) The soil water content change (68) 

-0.05 0.00 0.05 0.1 0 -0.05 0.00 0.05 0.10 
0 -t--.--,--.,..--.,..--L..-,---,--,,-,--l-~~~~--l 

200 

400 

600 

800 

(a) (b) 

SITE 5 

-+- SITE 6 

)epth 

:mm) 1000 _._ SITE 7 

1200 
-e- SITE 9 

1400 

1600 

1800 

Figure 4.5. Change in soil water content profile during two extraction 
periods for unirrigated tree, (a) from Day 7 to Day 16 
(9 days), (b) from Day 23 to Day 29 (6 days). 

~ 
U1 



46 

Rahardjo (1988) and Hughes pers. comm. (1988) found that the 

apple trees in the research orchard had a uniform horizontal root 

distribution between trees. Also the uptake from the four sites 

around each tree was similar (Table 4. 1.) . It seems reasonable 

therefore to assume that water use was uniform over the whole area 

around each tree. Thus, the soil water use per tree (litres) can be 

estimated by multiplying the average amount of soil water use per unit 

land area (mm) with the rooting area available to each tree (5.0 x 

3.6 m). Table 4.2. shows the calculated soil water use. Tree I used 

81.7 ± 19.4 litres per day and 7~.4 ± 29.8 litres per day during the 

first and second periods of extraction respectively. There was no 

significant difference at the 5 percent level in the rate of soil 

water extractions between the first and second periods. As the soil 

around Tree I had a similar and adequate soil water status during both 

periods (Figure 4.3.) this would be expected, if weather conditions 

were similar. On the other hand, Tree U, the covered tree, had 

significantly different soil water extraction at the 1 percent level, 

both between the first and the second periods of extraction, and also 

from Tree I. During the first period, extraction by Tree U was only a 

quarter of that by Tree I. While there was apparently no soil water 

extraction from around Tree U in the second period but Tree U still 

survived. This is an interesting finding. One possibility is that 

Tree U had deep roots going below 1900 mm depth, which could extract 

enough soil water from below 1900 mm to maintain the plant water 

status and metabolism. Another possible reason is that long 

horizontal roots extracted soil water from beyond the cover area. 

Thus, when there is adequate soil water in the topsoil, the apple 

tree will extract soil water from there, but when the there is 

inadequate soil water in the topsoil, the apple tree will extract soil 

water from deeper in the soil, perhaps from more than 1900 mm deep in 

this soil. 



47 

Table 4.1. Soil water storage change per unit land area during 
extraction periods. 

Period I Period II 
Days 7 to 16 Days 23 to 29 

Total Average Total Average 
(mm) (mm/day) (mm) (mm/day) 

Tree I Site 1 47.3 5 . 3 29.4 4.9 
Site 2 39.2 4.4 13.9 2 . 3 
Site 3 49.1 5.5 35.3 5.9 
Site 4 27 . 9 3.1 17.9 3.0 

Average for four sites 40.9 4.5 24.1 4.0 
Standard deviation 9.7 1.1 9.9 1. 7 

Tree u Site 5 6 . 3 0.7 0.0 0.0 
Site 6 10.2 1.1 -1. 7 - 0 . 3 
Site 7 10.9 1.2 0.4 0 .1 
Site 9 9. 9 1.1 -4.0 - 0 .7 

Average f o r f our sites 9. 3 1.0 -1. 3 - 0.2 
Standard deviation 2 .1 0.2 2 . 0 0 .3 

Table 4.2. Soil water storage change per tree. 

Period I Period II 
Days 7 to 16 Days 23 t o 29 

Total Average Total Average 
(L) (L/day) (L) (L/day) 

TREE I 
Average for four sites 735.4 81. 7 434.2 72. 4 
Standard deviation 174.3 19.4 178.6 29.8 

TREE U 
Average for four sites 168.2 18.7 -23.9 -4.0 
Standard deviation 37.3 4.1 35.5 5.9 



48 

4.5. SOIL WATER PRESSURE POTENTIAL 

Figure 4.6. shows the soil water pressure potential around both 

Tree I and Tree U at 300 and 600 mm depths, measured by tensiometers 

with an electronic pressure transducer. 

For Tree I the soil water pressure potential values, especially 

at 300 mm depth (Figure 4.6(a)), indicate wetting on Days 3 and 17, 

since the values on Days 4 and 18 were higher than the preceding 

values. 

Figure 4.6(a) shows that soil water potential around Tree I at 

300 mm depth was higher (less negative) than at 600 mm depth. This is 

in agreement with the soil water content measurement at 400 and 600 mm 

depths at Site 1, 2, 3 and 4 (Figure 4.1.) and would have been due to 

the irrigation and rainfall. Figure 4.6(b) does not show the complete 

soil water pressure potential from Days 1 to 25 because the soil 

became too dry for the tentiometers to function after Day 11. 

However, Figure 4 .6(b) shows that the soil water content at 300 mm 

depth was drier than at 600 mm depth around Tree U. The data in 

Figure 4.2 also show that the soil water content at 400 mm depth was 

drier than at 600 mm depth at Site 6, 7 and 9, but not at Site 5, 

probably due to the variability of the illuvial soil at the 

experimental site. In summary, the tensiometer data indicate the 

topsoil around Tree I to be adequately watered during the study 

period,but the topsoil around Tree U to be at potentials suggesting 

plant water stress. 



Time (days) 
0 3 6 9 12 15 18 21 24 27 30 

0 
I 

( a) \ 

Soil -10 

water -20 ! ' \ 
pressure ' ' ' ~,,,.-, potential -30 ,_ ' 

(kPa) -4o 
-- ......... / ' 

' - --- -...__ -
-50 

0 3 6 9 12 15 18 21 24 27 30 
0 

I 

(b) \ 

Soil -10 ! ~ \: -......_ 
water -20 \ ~- ------ - 300 mm 

pressure 
\ ~-"' 

-
potential -3o t -- 600 mm ·,.:----- -

(kPa) -40 

-50 

Figure 4.6. Soil water pressure potential measured by electronic 
tensiometer around Tree I (a) and around Tree U (b). 

,l>-

1.0 



50 

CHAPTER V 

THE ABOVE GROUND MEASUREMENTS 

5.1. HEAT PULSE TECHNIQUE 

The direct measurement of sap-flow in the stern of intact plants 

is an advanced and useful technique in soil water management. Using 

heat as a tracer for the sap-flow, which is equal to the amount of 

plant water use, offers a sensitive technique for monitoring the 

dynamic water status of plants (Swanson and Whitfield, 1981; Green and 

Clothier, 1988). The name of the technique is the heat-pulse sap-flow 

technique and it has been studied since 1932 when it was first 

suggested by Huber. 

5.1.1. SAP-FLOW 

Xylem is the system of "pipelines" within the plant stem or 

trunk. It serves the plant as a vascular water transport system 

(Huber, 1956; Milburn, 1979). The darker coloured core in the centre 

of a tree stern is the heartwood. This is not only dead, but also the 

oldest part of the tree stem. Heartwood cannot transport sap 

(Zimmermann, 1983). The outer wood is cambium and phloem. Cambium 

produces both xylem and phloem cells, while phloem functions to 

distribute the results of photosynthesis (Zimmermann and Brown, 1977). 

However, neither cambium nor phloem transport sap from roots to 

leaves. Thus sap flow only occurs in the xylem. Transpiration 

through stomata creates a potential difference between the leaves and 

roots, causing a decrease in the sapwood capillary potential (Siau, 

1984). This induces sap-flow in the sapwood. 



51 

The average sap flux density is proportional to the sap velocity, 

but also depends on the density and moisture content of the wood. The 

volumetric flow rate can be found by multiplying the average sap flux 

density and the cross-sectional area of the xylem annulus through 

which flow occurs. 

5.1.2. THE TECHNIQUE 

The key to the technique is heat transport in the sapwood. 

A pulse of heat is released by a heater and detected by nearby 

temperature sensors. The time delay (t 0 ) between releasing and 

receiving the heat pulse can be used to compute the heat pulse 

velocity. The heat pulse velocity can then be related to the sap 

velocity. The velocity of the heat pulse is not the same as the sap 

velocity because heat is absorbed by the wood as well as the sap. 

Hube r (1932) used a small heated wire loop , which wa s embedded 

under the bark of the stem, to release the heat pulse. As this 

technique could not detect heat pulse velocities below 1 m/h, Huber 

and Schmidt (1937) developed the compensation heat pulse technique 

where a heater is inserted radially into the stem. 

Heat enters the sap flow mainly by thermal conduction through the 

vessel walls (Huber, 1932; Huber and Schmidt, 1937; Marshall, 1958). 

When the sap velocity is very slow, most of heat is transferred away 

from the heater by conduction rather than convection (Closs, 1958). 

However, when the sap moves at greater speeds, the rate of movement of 

the heat pulse will depend more on convection and less on conduction 

(Marshall, 1958) . 

In the compensation heat pulse technique, a heater probe and two 

temperature sensors are used, one fixed at distance Xu upstream and 

the other a distance Xd downstream from the heater probe. The sensors 

and the heater are inserted radially into the stem. The heat pulse 

velocity upstream is faster than downstream, but the upstream heat 

pulse decays more quickly than the downstream pulse, as seen in Figure 

5 .1 (a). 



QI 10 Ill 
nl 
QI 
M 8 u 
~ 

•rl 6 ....... 
QI u 
MO 
::i~ 4 .µ 
nl 
M 
QI 2 ~· 

u E-t 

4 
p. 
;::l 2 

E-t 

(a) 

(b) 

f down 

- .. - - - - - - - - - - - - - - - -/i 
t 0 

+---t----1-----1---l----+---I--___.__ __ ___._ 
() 30 (10 90 120 150 180 210 240 

Time sin<.:e heat pulse (s) 

52 

Figure 5.1(a) . The typical relationship between temperature and time 
for heat pulses upstream (Xu= 5 mm) and downstream 
( x,d = 1 0 mm) . 

Figure 5.1(b) . The typical difference between downstream and upstream 
temperature. to is the time delay until the upstream 
and downstream temperatures are equal 
(after Swanson, 1962). 



53 

Swanson (1962) defined the heat pulse velocity (V) as 

V (5 .1) 

where t 0 is the time delay after pulse initiation when the 

temperatures at points Xu and Xd become equal. 

typical time delay. 

Figure 5.l(b) shows a 

The heat pulse velocity (V), calculated using equation (5.1), needs to 

be corrected to take account of the fact that the tissue in which the 

sap flows has been partly blocked by both heater and temperature probe 

insertion (Cohen et al., 1981). The probes are finite in size and 

have thermal properties different from the surrounding wood (Swanson 

and Whitfield, 1981). To calculate the corrected heat pulse velocity 

(V'), Green and Clothier (1988), who developed the heat pulse 

apparatus used in this research, used a numerical solution developed 

by Swanson (1983). 

The sap flux density (J) in a homogeneous stern can be calculated 

from V' and the properties of the wood as (Marshall, 1958) 

J ) V' (5. 2) 

Pw 

oven dried weight of wood 
where PT 

green volume 

the density of wood in kg rn- 3 

Pw the density of water (sap) in kg rn- 3 

green mass - oven dried mass 

oven dried mass 

the moisture content of wood in kg kg-l 



54 

CT the specific heat of oven dry wood 

1.39 X 10 3 J kg-1 K-1 

Cw the specific heat of water (sap) 

4.21 X 10 3 J kg-1 K-1 

Finally, by integrating the sap flux density over the sap-wood 

cross-sectional area, the volumetric flow rate (Q) is found as 

Q f 2nr J(r) dr (5.3) 

H 

where J(r) is the sap flux density at radial distance (r) in a stem 

with xylem between radii H (heartwood) and R (cambium) 

(see Figure 5.2a.). 

5.1.3. INSTRUMENTATION 

Each set of heat pulse probes used in the research consisted of 

two temperature sensors, controlled by a heat pulse circuit, and a 

heater which was controlled by a heater circuit. Tree I and Tree U 

each had three sets of the heat pulse probes installed in them and 

these were monitored by a Campbell CR21X data logger powered by a 12 

Volt lead-acid battery. 

The heater probes were made from insulated nichrome resistance 

wire (10 ohm/m) mounted inside a stainless steel tube of 1.8 mm 

outside diameter and 60 mm long. At one end the nichrome wire was 

soldered to the steel tube, while at the other end copper leads were 

soldered to both the stainless steel tube and the nichrome wire. The 

heater was connected in series with a 60 W, 0.4 Ohm power-resistor. 

Each heater was turned on for 1.5 seconds every measurement. 

Each temperature sensing probe consisted of four micro-bead 

thermistors spaced at intervals of 5 mm inside a length of teflon 

tubing, which was 1.85 mm in outside diameter and filled with epoxy 



55 

R ~---~ SAP FLUX DENSITY 

H 

GROWTH 

XYLEM 1ti)5;>}1::;trt-'t.(;D~ RING 

z 

I I u 

Figure 5.2(a). A diagram showing the position of heater probe and 
thermistor beads used to monitor the heat pulse 
velocity. 

Figure 5.2(b). Two of three sets of heater probes and thermistor beads 
in the trunk of Tree U. 



56 

resin. The leads from each thermistor were electrically insulated 

from each other. The sensor was very flexible and easily inserted and 

removed from the apple tree trunk (Green and Nicholson, 1987). Figure 

5.2(a) shows the position of a set of heat pulse probes and Figure 

5.2(b) shows two of the three sets of heat pulse probes installed in 

the trunk of Tree U. 

The Campbell CR21X data logger had programs to translate the 

analogue temperature inputs to digital outputs of the sap flow 

velocity from the three heat pulse probe sets (see Figure 5.3(a)). 

The sap velocities were measured every 30 minutes. 

As the memory of the data loggers was limited, to record all data 

audio cassette recorders were used to store the output from the data 

loggers in the field. The data were stored on normal audio cassette 

tape (see Figure 5.3(a)). To capture the sap flow velocity data from 

the cassette tape, a Tape Reader Cassette Interface (made by Campbell 

Scientific Inc.,Logan, Utah, USA) was used. The tape reader was 

connected to a computer. Figure 5.3(b) shows the cassette recorder, 

tape reader and computer. The calculations were carried out using 

MINITAB computer software at Plant Physiology Division, DSIR , 

Palmerston North. When the computer had read the values of the sap 

flow velocities, firstly the data were checked to remove inconsistent 

values . Then the average of the three sap flow velocities was 

calculated. By multiplying the cross-sectional area of the xylem 

annulus with the average sap flow velocity, the rate of sap flow was 

found. 

5. 1. 4. RESULTS 

The measured water use of Trees I and U from Day 2 (1 4 December 

1988) to Day 24 (5 January 1989) is shown in Figure 5.4. The figure 

shows that Tree U consistently used slightly more water than Tree I. 

This was not expected, as Tree U was not irrigated. A possible reason 

is that the canopy of Tree U might have been slightly larger than that 

of Tree I. The daily pattern of water use by Trees I and U was 

similar. Usually there was no sap flow at night, the exception to 



57 

l 

Figure 5.3(a). The heat pulse instruments in Tree U, consisting of 
3 sets of thermistor beads and heaters (1) which were 
controlled by heat pulse and heater circuits (2) and 
connected to a Campbell CR21X data logger (3) powered 
by a 12 Volts lead-acid battery (4) and connected to an 
audio cassette recorder (5). 

'T:'I.!---- C ~/L\ - -.L.- --



,....._ 
\... 

..c 
:::-
'-./ 

Q) 

"' :::> 
\... 

~ 
0 

3:: 

,....._ 
\... 
..c 
:::-
'-./ 

Q) 
U) 

:::> 
\... 
Q) 

0 
3:: 

10 

., :, • 8 ~ ~. .. I I ·~ ;,~ /\ ,, 

;1 
,., 

I I I 

I ' ' 
6 

~ 4 

1] 
2 

~ \ i 
l ..1 
I ! 

C 
2 3 4 5 6 7 

'l'ime (days) 
10 

8 
,, ' ,, i'(. ,, 

•• ,, 
'1 ~ I •' ! : I 

I I I 

6 I 

4 

2 

o '-'--1--'-4---"-.____._1__,._5-=..,_1..L..6-"-'_,_1~7~~1~8~~1~9~ 

Time (days) 

10 

8 

6 

4 

2 ~ 

t-

0 J 

10 

8 

6 

4 

2 

I ,, 
I 
I 

~: 

8 

~ 
'•' ~: 

9 

. ~. ,., 
I I 
I 

' I 
I 
I 

~ I 

I ( • 
' l I 

I 

~~ l I 

1 0 11 

Time (days) 

~ '. 
~: 

12 

.. ,_ 
I
I I 

l. I 

58 

-

t~ :,, 
h I 

r 

13 

o L_J_J..l-'-::2:'--:1--1....l.---'--=2"=2...)..J..-'-='2-=-3 ...L..L..--':2-='"4::-'-'--~ 

Time (days) 

Figure 5.4. The pattern of daily water use (litres/hour) from Day 2 to 
Day 24 as measured by the heat pulse technique for Tree I 
(--) and Tree U (-----). 



59 

this being Days 2, 
1
3, 9, 10, 14, and 15. Why water was lost on some 

nights but not others calls for some explanation. Stomata, which are 

small pores connecting the internal air spaces of a leaf to the 

outside air, usually are closed during the night and open during the 

day, depending on the available energy from sunshine (Jarvis and 

Mansfield, 1981). But in some dicotyledons, such as apple, stomata 

can remain open on windy nights with a low humidity ( Sharpe, 1973, 

Muchow et al., 1980; Turner et al., 1980; Judd et al .. 1986), leading 

to nocturnal transpiration (Green et al., 1989) 

Figure 5.5. shows the average daily water use of Trees I and U. 

Day 3 (December 15) was a cloudy day without rain, and s o had a l ower 

water use than most other days. However, cloudy days with rain (Days 

17, 18 and 20) had even lower water use values than Day 3. On cloudy 

days with rain, evaporation of the rain itself, and the presence 

intercepted water on the leaves, would lead to a high humidity and a 

sma ll saturatio n deficit. Thus the stomata would no t fully ope n and 

even if they did, little transpiration would occu r due t o the s ma l l 

v apour pressure deficit. The average water use from Day 7 t o Day 1 6 

for Tree I wa s 63.0 litres/day and for Tree U was 68.5 litres/day . 

5.2. STOMATAL RESISTANCES 

5.2.1. STOMATA 

In dicotyledon plants, such as apple, stomata are located on the 

bottom of leaves. The opening or closing of stomata will depend on 

the expansion and contraction of the two guard cells at the entrance 

of each stomate (Bidwell, 1974). 

If the rate of water loss from the plant through the stomata 

exceeds the rate of water uptake from the soil, the plant will suffer 

a water deficit. Higher plants can use different mechanisms to cope 



60 

80 
I 

I' 
I \ I \ I..,_ ......... ..... 

I I \ I U] \. - "\ 
I ,- ~, 

Q) \ 
\ I 

H I 
\ 

..µ 64 \ 
·rl 
.-I 
.......... 

Q) 48 U] 

~ 

H 
Q) 

..µ 32 cu 
3 

>-i 
.-I 
. rl 16 cu 
0 

6 : 10 14 18 22 24 
Time (days) 

Figure 5.5. Daily water use measured by the heat pulse technique from 
Day 2 to 24 for Tree I (--) and Tree U (-----). 



61 

with stress, such as slowing down transpirational water loss through 

stomata control , using water stored in a certain organ, or by 

developing deeper root systems to extract soil water (Speer et al. , 

1988) . The transpiration rate reflects in part the stomatal response 

to environmental conditions. Small reductions in epidermal turgor can 

result in increasing stomatal resistance (Davies et al., 1981) . As 

water stress develops in the plant, the degree of water deficit 

affects the stomatal response to photon flux density, temperature, 

carbon dioxide and vapour pressure deficit, giving a certain maximum 

stomatal opening which is lower than that when stress is absent. As 

the stress increases, the maximum degree of stomatal opening on any 

day will be less (Hall et al., 1976). Thus when there is enough water 

in both soil and plant, the guard cells surrounding stomata will open 

fully in sunlight, as the guard cells are turgid. On the other hand, 

when both soil and plant have less water, the turgor of the guard 

cells will decrease, and the stomata will reduce their opening period , 

or even stay closed all day. The closure o f stomata is r e lated to t he 

amount of the growth regulator abscisic acid (ABA) (Wright , 1 969 a nd 

1977) . During water stress, ABA is transported from mes ophyll t o the 

guard cells. The more ABA in the guard cells, the less the stomata 

open (Jones and Mansfield, 1972; Loveys, 1977). The degree o f 

stomatal opening affects CO 2 uptake, which will affect n o t on l y 

photosynthesis but also metabolism in general. Furthe rmore plant 

growth, which depends on photosynthesis, will be affected . 

When the humidity in the atmosphere outside the leaf cuticle is 

lower than in the intercellular spaces within a leaf, there is 

molecular diffusion of water vapour outwards through the stomata . The 

number and openness of the stomata, and the humidity difference, will 

affect the rate of diffusion. The porometer is an instrument to 

measure the diffusive resistance of leaves to water vapour (Stiles, 

1970; Montheith and Bull, 1970; Squire and Black, 1981). The stomatal 

resistance measured may be related to the transpiration rate from the 

leaves, and may detect whether the plant is under stress or not. High 

values of stomatal resistance indicate that the plant is under stress, 

as the opening of stomata is smaller due to their lower turgor. On 

the other hand, low stomatal resistances suggest a normal 

transpiration rate, as the stomata are fully open. 



62 

5.2.2. THE POROMETER 

The porometer used was Automatic Porometer Mk3 made by DELTA - T 

DEVICES, England (see Figure 5.6(a)). The clamp is attached to a cup, 

containing a relative humidity (RH) sensor and two thermistors, which 

are used to measure leaf and cup temperature. The leaf measured is 

held in the clamp with the cup facing the bottom of the leaf, as the 

stomata are there. A small electric diaphragm pump is used to blow 

dry air into the cup. 

containing silica gel. 

The source of dry air is a drying chamber 

When the RH in the cup has been lowered 5 

percent below the ambient level, the pump switches off. Water vapour 

then diffused through the stomata, increasing the RH in the cup. The 

device measures the time taken for the RH in the cup to rise from 5 

percent below to 5 percent above the ambient value, and a number of 

counts which is proportional to this time appears on the small LCD 

screen after each pumping cycle. The frequency of the oscillator used 

in the counting device is 200 Hz. The measurement is repeated until 

the response time is constant. The leaf stomatal resis t ance values 

can be found by comparing this time with similar time s measured using 

perforated plates with known diffusion resistance s in the apparatus 

instead of leaves. 

The sequences of events involved when the porometer was used were 

as follows. Firstly, the temperature and RH of the air were measured. 

The RH value switch was then set to the appropriate value. Secondly, 

the moulded polypropylene calibration plate, which was covered with 

blotting paper wet with distilled water, was clipped by the clamp. 

The plate has six sets of holes of varying size. The diffusion time 

associated with a certain hole size appeared on the right LCD screen 

of the device. By plotting the diffusion time and corrected 

resistance readings, a calibration graph was made, as shown in Figure 

5.7. This was done before and after every measurement. Thirdly, 

several randomly selected leaves from Trees I and U were clipped by 

the clamp and their diffusion times measured (see Figure 5.6(b)). The 

value of the temperature difference between the cup and leaf could be 

read on the left LCD screen of the device. 



63 

Figure 5.6(a) . The Delta- T Device Parameter. 

Figure 5.6(b). Using the parameter to measure the stomatal resistance 
of the bottom of a leaf. 



3.00 I 
2.75 l 
2.50 l 
2.25 l 
2.00 I 
1.75 -

Resistance 1.50 l 
(s/cm) 1 - I 

. . 2:, T 
1.0~ T 
o.7:, T 
o.50 T 
0.25-. 

t . GA .. 

1/

/, I/ " 
! /! 
I ; ! p I/!; . 

f !/H• i I : i }:' 

j
• j?.ll 

f • , ,: 

' . ,, ; 
': 

( _// 
I I It : ; 
/ / // ... j / 

t+ *•' ~ X 

, , , , 

, 

,x 
i 

, 
i 

/ , 
i 

i 
/ 

i 
i 

/ 
i 

, , , , 
_/ 

I , 
/ 

I , 
I , , 

I 
i 

a. oo l! ---'---'---'---'--+-----'------'-----'----'--+--'--'--'--'-~---'----'-----'-;;;---'---'-~~ 

0.0 C :: _.., LO : .5 2.0 2.5 

Diffu.sio~ ~i.me (s) 

Ambient Conditions 

RB, Temperature 

-- 30%, 28.4°c 

-:-- 20%, 2BA0c 
...... 20%, 21.sDc 
:: 38%, ':'= s0c _....,_ 

* 40%. 23.9°C 

-b- 40%, 25.30C 

-.::i- 40%, 2Q -0c --0 

...- 30%, 25.9°C 

-t- 20%, 33.8°c 

Figure 5.7. The correlation between diffusion time (s) and plate resistance (s/cm) 
for poromster calibratior.s under various relative humidity 
and temperature conditions. 0\ 

,I:> 



65 

The factory calibration was carried out at a temperature of 

20°C, so for measurement at temperatures other than 20°C, a correction 

of the known resistance of the plate was needed. This was made using 

Table 5.1. The correction was calculated by interpolating between air 

temperature and index values. The index calculated was then 

multiplied to the standard known resistance of the plate. 

During most of the measurements and calibrations, there was a 

small temperature difference (dT) between the cup and the leaf. When 

the leaf temperature was higher than the cup temperature, more water 

vapour would be driven into the cup, and this would decrease the 

diffusion time. So, the measured diffusion time needed to be 

corrected to take account of this also. 

(t') was calculated by 

The corrected diffusion time 

t' t + c
0 

t dT (5. 4) 

whe r e t is diffusion time me asured in the field 

( 1 c ount 1/200 second), 

co is coefficient of the temperature difference between cup 

and leaf or plate at a certain RH range in %/ OC 

(see Table 5. 1) , 

dT is temperature difference (cup - leaf) measured 

in oc 

5.2.3. RESULTS 

Figure 5.7. shows some typical calibration graphs found during 

the investigation. The hole sizes used were hol