Optimizing the spatial pattern of land use types in a mountainous area to minimize non-point nitrogen losses

Abstract

Spatial patterns of land use types (mixed forest: MF, tea garden: TG, bamboo forest: BF) in a mountainous area with rapid agricultural land development were optimized to reduce NO3−-N leaching and N2O emission in this study. Firstly, a process-oriented biogeochemical model (Denitrification Decomposition, DNDC) was calibrated and validated on representative MF, TG and BF hillslopes. To upscale the simulations, hydropedological function (HPF) units were generated by overlapping maps of land use, soil organic carbon, rock fragment content and slope, which were recognized as critical factors affecting soil N cycle. The calibrated DNDC models were then adopted to simulate the soil N cycles in different HPF units and assess temporal and spatial variations of NO3−-N leaching and N2O emission risks. Lastly, spatial allocations of TG were determined respectively for minimizing NO3−-N leaching (MLN_LU) and N2O emission (MEN_LU), and balancing the reductions of both (BRN_LU). Results showed that the DNDC model had acceptable accuracies on these representative hillslopes (R2 > 0.50, NSE > 0.30). Land use of TG had the greatest N loss risks, the mean annual NO3−-N leaching and N2O flux in TG HPF units (72.11- and 3.63- kg N ha−1, respectively) were respectively 2.61 and 2.50 times of those of the entire study area. Temporal variations of NO3−-N leaching and N2O flux were both controlled by the timing of precipitation, and their spatial patterns were both primarily controlled by land use and then respectively by soil hydraulic properties (NO3−-N leaching) and soil carbon and N concentrations (N2O flux). Due to the relatively small spatial variations of soil and terrain properties and rational land use spatial patterns in this study area, the MLN_LU only reduced 7.6% of NO3−-N leaching, the MEN_LU only reduced 6.0% of N2O flux, and the BRN_LU reduced both by 3.3% and 4.1%, respectively. This study emphasized the role of land use pattern optimization in reducing the non-point N losses in mountainous area, and could provide scientific guidelines for future agricultural land developments.