Hydrologic and Water Quality Modelling of an Intensively Managed Landscape

Abstract:

Watershed management decisions are significantly reliant on the hydrologic models as they simulate both the natural and human elements of a watershed. Model calibration is often essential to ensure the reliability of these model outcomes. However, the calibration of a watershed hydrologic model is challenging, usually due to insufficient observation data, as well as existing deficits with the model structure. The challenge heightens particularly in a large watershed with existing tile drains, reservoirs, point source outlets, and other forms of human intervention. Here we present an experience on hydrologic model setup and calibration process followed to simulate a watershed in an intensively managed landscape of the US Corn Belt. We setup the Soil and Water Assessment Tool (SWAT) at the Upper Sangamon River Watershed in Central Illinois for 2000-2018. We calibrate the model in SWAT-CUP first with the observed daily flow and subsequently with the annual crop yields (corn and soy), monthly sediments, nitrate, and total phosphorus over multiple sites. The model performed well in the calibration period (2003-2012) with NSE >0.6 for flow (daily), sediments, and nutrients (monthly) and PBIAS within ±10% along with reasonable estimates in the validation period (2013-2018). Our analysis shows that precipitation has much influence on water quality variables, especially where non-point sources are dominant. As model performance depends on the quality of input data, we find significant improvement in performance after addressing probable errors in precipitation data related to snowfall together with dense weather stations. Increased weather stations are more relevant to larger watersheds where spatial variability of precipitation is prominent. We further improve the model performance by manually adjusting some of the reservoir and sediment parameters to overcome the modeling deficits. We employ the model to simulate the environmental response of different management practices and land-use change in the watershed. Such simulation will be essential for the Integrated Technology-Environment-Economics Modeling framework to assess the resilience of the Water Food Energy system.