Deriving Land Management Practices for Reduced Nutrient Movement from an Agricultural Watershed Using the AGNPS Model

Abstract

The effect of nutrient management practices and the land-use system on nutrient enrichment in water resources of a hilly watershed was assessed with an event-based agricultural non-point source (AGNPS) model. The model intended to assess the runoff, sediment and nutrient loads in a typical hilly agricultural watershed. The model was calibrated, evaluated and applied in integration with GIS to predict the soil and nutrient loss. Two nutrient management scenarios were simulated with 25 and 50% reductions in the nutrient application from the present nutrient application. The third scenario was simulated by converting 25% of the agricultural land-use to tea plantations. A total of 15 simulations were run for the different rainfall intensities of the year 2017. The existing land-use scenario simulated the maximum soil loss and Nitrogen and phosphorus load of 8.23 t ha−1, 22.8 and 5.0 kg ha−1. The 50% nutrient reduction scenario reduced 48 and 36% of the Nitrogen and phosphorus load compared to the existing farmers’ practice. The same nutrient dose was compared with the STCR equation, developed for major crops such as potato and carrot in Nilgiris soil, and confirmed the sufficient nutrient supply to produce a sustainable yield. The conversion of 25% of the agricultural land-use to tea plantations reduced the soil loss by 10% from the current land-use. The Nitrogen and phosphorus load was reduced to 56 and 48%, respectively. Hence, the farmers may convert 25% of the land area to tea plantations and reduce 50% of the present fertilizer dose for the major vegetable crops with INM to reduce the nutrient enrichment in the surface water bodies. This study demonstrated the applicability of the AGNPS model in similar watersheds for deriving possible management strategies to reduce soil loss and nutrient movement. Further, the hydrological models can provide valuable insights for promptly prioritizing and making policy decisions in ungauged/data-scarce watersheds.