A critical source area phosphorus index with topographic transport factors using high resolution LiDAR digital elevation models

A new phosphorus index (PI) tool is presented which aims to improve the identification of critical source areas (CSAs) of phosphorus (P) losses from agricultural land to surface waters. In a novel approach, the PI incorpo- rates topographic indices rather than watercourse proximity as proxies for runoff risk, to account for the dominant control of topography on runoff-generating areas and P transport pathways. Runoff propensity and hydrological connectivity are modelled using the Topographic Wetness Index (TWI) and Network Index (NI) respectively, util- ising high resolution digital elevation models (DEMs) derived from Light Detection and Ranging (LiDAR) to cap- ture the influence of micro-topographic features on runoff pathways. Additionally, the PI attempts to improve risk estimates of particulate P losses by incorporating an erosion factor that accounts for fine-scale topographic vari- ability within fields. Erosion risk is modelled using the Unit Stream Power Erosion Deposition (USPED) model, which integrates DEM-derived upslope contributing area and Universal Soil Loss Equation (USLE) factors. The PI was developed using field, sub-field and sub-catchment scale datasets of P source, mobilisation and transport factors, for four intensive agricultural catchments in Ireland representing different agri-environmental conditions. Datasets included soil test P concentrations, degree of P saturation, soil attributes, land use, artificial subsurface drainage locations, and 2 m resolution LiDAR DEMs resampled from 0.25 m resolution data. All factor datasets were integrated within a Geographical Information System (GIS) and rasterised to 2 m resolution. For each factor, values were categorised and assigned relative risk scores which ranked P loss potential. Total risk scores were calculated for each grid cell using a component formulation, which summed the products of weighted factor risk scores for runoff and erosion pathways. Results showed that the new PI was able to predict in-field risk variability and hence was able to identify CSAs at the sub-field scale. PI risk estimates and component scores were analysed at catchment and subcatchment scales, and validated using measured dissolved, particulate and total P losses at subcatchment snapshot sites and gauging stations at catchment outlets. The new PI provides CSA delineations at higher precision compared to conventional PIs, and more robust P transport risk estimates. The tool can be used to target cost-effective mitigation measures for P management within single farm units and wider catchments.