| Abstract: | A nutrient-transfer model was developed using a mass balance approach to explain the variations in the amounts of soil phosphorus (P) and sulphur (S) found across a range of slopes and aspects in four 10-ha farmlets located in summer-moist hill-country pastures in the southern North Island of New Zealand. The farmlets were continuously grazed by sheep for a period of 12 years, during which time they received a total of 115, 187, 387, 603 and 126, 201, 405, 630 kg ha−1 of P and S, respectively, as single superphosphate (SSP). The model takes account of the effect of topography and local climate on stock behaviour, herbage accumulation and its nutrient content, pasture utilization and the uneven nutrient return through excreta.The developed P model was reliable. The predictions of the transfer model explained 95% of variation in soil P amounts (0–150 mm depth) between farmlets (n=4), 89% of variation between slope units (n=12) and 79% between slope-aspect units (n=36) across all farmlets. When S input parameters were substituted for P parameters, the model was unable to predict the measured soil S levels in any of the four farmlets. Unlike P, S is subject to losses by leaching from these soils. The differences between predicted and measured soil S amounts were used to estimate S leaching losses and pinpoint areas from where they occurred.The concept of modelling the fate of P in P and S fertilized systems and then using the calculated nutrient-transfer functions of the model to evaluate the fate of S provides invaluable information for developing strategies for improving the efficiency of S fertilizer use. |
