Integrating GLEAMS sedimentation into RZWQM for pesticide sorbed sediment runoff modeling

Sediment transport from agricultural fields to native waterways is a significant pollution vector, not just for the bulk sediment, but also for additional mass of pesticides traveling offsite that are sorbed to soil particles. Existing field scale models that track plant growth as well as the fate and transport of applied pesticides lack an integrated sediment transport component. This study sought to address this lack of available modeling tools for researchers and regulators by integrating the sediment and surface flow components of Groundwater Loading Effects of Agricultural Management Systems (GLEAMS) model into the mature Root Zone Water Quality Model (RZWQM) to create a derivative model named RZWQM-Sed. Previous research identified RZWQM as a quick running, agricultural field scale model that accurately estimated offsite transport of solutes. Unlike other well performing field scale agricultural models, the full source code of RZWQM was available for modification and extension. However, RZWQM lacked a sediment component and thus could not measure all pollutants moving offsite. GLEAMS sedimentation was chosen for integration due to its well documented history, compatibility with the RZWQM codebase, and source code availability. Sensitivity analysis of the RZWQM runoff variables showed that the residual water content, saturated water content, and bubbling pressure from the Brooks-Corey equation had the highest influence for RZWQM followed by the saturated hydraulic conductivity and the non-Brooks-Corey bubbling pressure. Analysis of GLEAMS variables showed the most significant variables are the USLE parameters (cfact, pfact, ksoil) and Manning's N. The latter variable only showed sensitivity at very high surface roughness while the USLE parameters had a linear relationship over the entire domain. The integrated model was calibrated and validated using multiple real-world datasets spanning ranges of space and time. The final model performed well in the primary task of predicting the mass of sorbed chemicals in the tailwater (Nash-Sutcliffe coefficient > 0.3).