Evaluation of process-based watershed models in the Great Lakes basin

Session: Poster session

Dong-Kyun Kim, University of Toronto Scarborough, [email protected]
George Arhonditsis, University of Toronto Scarborough, [email protected]
Feifei Dong, University of Toronto, [email protected]
Alex Neumann, University of Toronto Scarborough, [email protected]
Aisha Javed, University of Toronto Scarborough, [email protected]

Abstract

Watershed models have been extensively used in hydrological science and environmental management for a number of important tasks, including estimating non-point source pollutant inputs to receiving waterbodies and their source areas and predicting the effects of climate and land-use change on water quality. However, the adequacy of watershed models for informing decision making has been questioned. Concerns of overparameterization and equifinality have brought to the forefront of modelling efforts the development of methodologies that will obtain "the right answers for the right reasons". In this study, we evaluate the current status of widely-used watershed models and discussed their strengths and weaknesses with respect to data requirements, model complexity, and best management practices (BMP) assessment. Specifically, we review all the Soil and Water Assessment Tool (SWAT) applications in the Lake Erie watershed. SWAT models developed by different groups were evaluated on the basis of spatial discretization, input data sources, subroutines used, model parameterization, and calibration strategies adopted. We report moderate model performance in reproducing flow, sediment, total phosphorus, and dissolved reactive phosphorus (DRP) time-series. We conclude by advocating a number of new directions for distributed watershed modeling, including in-depth uncertainty analysis and the use of additional information to constrain parameter estimation.

1. Keyword
watersheds

2. Keyword
Great Lakes basin

3. Keyword
phosphorus