Dynamic spatially explicit mass-balance modeling for targeted watershed phosphorus management: I. Model development

Abstract

Surface waters are frequently impaired by excessive phosphorus (P) from nonpoint sources, especially in regions of intensive livestock agriculture. Despite concerted efforts to apply new management measures, reductions in nonpoint source P loads have been difficult to accomplish. Watershed management to reduce P export could be more cost-effective if treatments were targeted to critical source areas at high risk for excessive P export. These critical source areas can be defined as the intersection of P source areas and active runoff contributing areas; such areas vary in space and time due to watershed characteristics and management practices. We developed an approach to identify, analyze, and map high-risk areas for P export by integrating spatial data with land use and agronomic data. We evaluated changes over time and space in soil P concentration and P export in response to changes in inputs and outputs with a dynamic mass-balance simulation model running in grid cells across a watershed. The temporal and spatial relationships that define the risk of P export are captured simultaneously using a raster-based distributed dynamic modeling approach and related to management interventions. Simulated responses to management interventions are analyzed and displayed spatially through a geographic information system (GIS). This approach allows the spatial distribution of P runoff risk to be tracked through time in response to long-term P input/output balance, evolving from either continuation of current practices or from management changes specifically targeted to areas of high P loss risk. Baseline simulations show that if present-day management continues, both soil test P and P export will increase dramatically in some parts of a test watershed; critical P source areas in a watershed will evolve over time and are likely to occur in limited areas that can be identified and tracked. Model results can contribute to improved targeting of scarce resources by focusing management interventions on those areas at highest risk of nutrient loss. This paper describes the underlying principles of the model, discusses the process of model development, and presents the final modeling system. Application of the model to alternative management scenarios is discussed in a subsequent paper.