Phosphorus leaching losses from an irrigated free-draining soil under dairying

Abstract

Recent expansion of dairy farming in New Zealand and the associated increases in phosphorus (P) inputs may have a significant impact on P loss by leaching from free-draining grassland soils under irrigation. This in turn has important implications for the long-term sustainability of dairy farming. The major objective of this study was to investigate the impacts of continued P and nitrogen (N) inputs in the form of farm dairy effluent (FDE) and mineral fertilizer on the amounts, forms and mechanisms of P loss by leaching. This study was carried out on a Lismore stony silt loam soil (Orthic Brown Soil) and involved detailed analyses of P in leachate from 32 soil monolith lysimeters (70 cm depth by 50 cm diameter) over a two year period. The monolith lysimeters were flood irrigated at 100 mm every three weeks during summer.

Annual P accumulation in the soil was higher for the treatments receiving FDE plus P fertilizer compared with P fertilizer alone, with the latter showing a negative P balance. Although the amounts of total P applied over the two year period were similar for the FDE and P fertilizer treatments (ca. 180 kg P ha⁻¹), significantly higher P losses occurred from the FDE treatment (1.4-2.5 kg ha⁻¹ y⁻¹) compared with the P fertilizer treatment (0.6-1.3 kg ha⁻¹ y⁻¹). However, these losses of P were found to be equivalent to less than 3 % of the total P applied. The application of P fertilizer at a high rate (90 kg P ha⁻¹ y⁻¹) did not significantly increase P concentrations in leachate compared with the lower P fertilizer treatment (45 kg P ha⁻¹ y⁻¹), which in turn reflects the capacity of the Lismore soil profile to adsorb soluble inorganic P. Phosphorus losses were higher from the treatments amended with both P and N fertilizer than P fertilizer alone during the second year of trial. This is possibly due to the stimulation of microbial activity that resulted in greater mineralization of organic P and increased inorganic P in soil solution.

Of the TP losses that occurred during 51 drainage events over two years from the FDE treatments, 44-61 % (1.9-2.6 kg ha⁻¹) was lost from 8 leaching events that occurred within 24 hours following FDE application. Concentrations of TP determined in leachate were commonly 100 - 400 µg L⁻¹, although TP concentrations immediately following FDE application were often much higher (>2500 µg L⁻¹ ). This was mainly attributed to preferential flow that resulted in transfer of unreactive P from the upper soil profile. Therefore, short-term strategies for reducing P loss must aim to increase the residence time of P within the soil such as more frequent application of smaller amounts of FDE. This will facilitate greater interaction of applied FDE with the soil matrix thereby improving plant utilization of inorganic P, allowing greater mineralization of added organic P, and increased retention of P. Results also indicated that a change from flood to spray irrigation may be effective in reducing short-term P loss.

Examination of the seasonal pattern of P forms in the leachate indicated that total particulate P (TPP) was the dominant form in the leachate (77 % of TP) compared with total dissolved P (TDP) (23 % of TP) during irrigation. This was attributed to greater preferential flow during the irrigation season, which increased physical dislocation of particulate P. During the non-irrigation seasons, the proportions of TPP (51 % of TP) and TDP (49 % of TP) were similar.

Physicochemical fractionation of P showed that unreactive forms of P (particulate unreactive P (PUP)- 58 %; and dissolved unreactive P (DUP)- 29 %) were the dominant forms present in leachate compared with reactive P (particulate reactive P (PRP) - 8 %; and dissolved reactive P (DRP) - 5 %). The predominance of unreactive P suggests that these forms of P are less strongly adsorbed onto soil colloids than reactive P. ³¹P NMR analysis confirmed that unreactive P in the leachate was mainly comprised of orthophosphate monoester (67 % of TP) and diester (20 % of TP) forms of organic P. The addition of specific phosphatases to leachate revealed that most of the monoester organic P was present as labile monoester P and inositol hexakisphosphate indicating their greater mobility and potential to contaminate water bodies compared with reactive P.