Assessment of the current state of empirical watershed models to support adaptive management

Session: Applications of Simulation Models in Watershed Science and Lake Ecology (1)

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

Abstract

Our project aims to impartially evaluate the capacity of existing empirical watershed models to operate as adaptive management and early warning system tools for the Great Lakes basin. Firstly, we identify the strengths and weaknesses of known empirical modelling frameworks, such as SPARROW, GREEN, IROWC-P. Secondly, we compared the deliverables of empirical models against known outputs of more complex, physically-based, mechanistic watershed models developed for the same watersheds in the Great Lake basin. Unlike hydrological models, which have demonstrated significant advancement in their model performance and predictive power due to corroborated physical laws, scaling up watershed models from hydrology to nutrient concentrations has resulted in significant deviations from recorded watershed responses. We argue that this finding casts doubt on the adoption of existing mechanistic models as reliable decision-making tools. Thirdly, we pinpoint the modelling practices that cause high uncertainty on watershed nutrient predictions in the Great Lakes basin. Finally, we discuss a set of potential recommendations to guide further watershed model augmentations, thereby improving their suitability as operational adaptive management tools.