Use of turnips to reduce potassium accumulation on areas receiving farm dairy effluent : a thesis presented in partial fulfilment of the requirements for the degree of Master of Applied Science in Soil Science at Massey University, Palmerston North, New Zealand

Abstract

Land treatment of farm dairy effluent (FDE) on small areas of intensive of dairy farms has enriched soils with nutrients particularly K. Solving the problem solely by increasing the area allocated for land treatment requires large investment in pump, pipes and irrigator infrastructure. A less costly strategy, of sowing and grazing a summer turnip on the land treatment area in order to redistribute K to the pasture area is evaluated in this thesis. A survey (February 2006) showed that in the Manawatu region turnip crop yields (8 to 17t DM.ha⁻¹) provided profitable feed for dairy cows, were a suitable re-grassing strategy and if harvested, removed 350 to 700 kg K ha⁻¹ from the soil. In the summer of 2005/06, a turnip (Brassica rapa cv. Barkant) trial was established after permanent pasture on a Pallic soil (pH 6.5, Olsen P 35.2 ug. g⁻¹, exchangeable K⁺ 0.7, Ca²⁺ 6.3, Mg²⁺ 1.4 me/ 100 g soil). The following treatments pre-plant fertiliser only (38 kg N ha⁻¹, 25 kg P ha⁻¹) and 25 kg K ha⁻¹), pre-plant fertiliser plus side-dressed urea at 40 DAS (46 kg N ha⁻¹) and pre-plant fertiliser plus 5 x 10 mm FDE applications (57kgha⁻¹) all produced similar final dry matter yields (8 t DM ha⁻¹) at 100 days after sowing (DAS). Leaf was the largest component of dry matter and had higher K concentrations (4.6 and 6.8% K in the control and FDE treatments respectively) than bulb (3 and 4 %K in the control and FDE treatments respectively). The ratio of leaf to bulb dry matter however varied for each different treatment. Side-dressed urea and FDE treatments produced the largest leaf biomass and reached maximum yields earlier by 75 DAS and 64 DAS, respectively and generated more K removal at harvest (339, 428 & 537 kg K ha-⁻ at 75 DAS and 316, 372 & 490 kg K ha-⁻ at 100 DAS for pre-plant only, urea & FDE treatments, respectively). The lack of yield response to N partially resulted from crop uptake of between 107 and 114 kg N ha-⁻ from mineralisable soil N. The dynamic N crop model N-able predicted that extra side-dressed N would not increase turnip yield but in the absence of pre-plant N (38 kg N ha-⁻) the turnips would yield 7.4 t DM ha-⁻ at 100 DAS. The use of the N-able model demonstrated a need for a decision support model to assist farmers in choosing appropriate N fertiliser application rates. A simple model was created to simulate how the grazing cow can transfer K from turnip paddocks (part of a FDE treatment block) to other parts of the farm. The model simulation of 490 cows on a mixed diet of 4kg DM turnips and 12 kg DM pasture predicted that the grazing of turnips (8t DM ha-⁻ crop) would result in the net transfer of significant quantities (>170 kg K ha-⁻) of K from land growing turnips to other parts of the farm. To cause net transfer to occur the allocated turnip dry matter must be grazed in the shortest time possible and the cows returned to pasture after short milking times.