Search

Accurate water supply predictions are critical for robust water resources management. Over the past decades, multiple operational modeling tools and datasets have been employed to address this need. However, the net basin supply (NBS) forecasting for the Great Lakes remains challenging due to international boundaries and sparse ground gauge observations. In this context, this study aims to assess the potential for operational NBS forecasting by combining the North American Multi-Model Ensemble (NMME) and National Water Model (NWM). Specifically, daily precipitation and evaporation over the lakes are obtained by analyzing different models of NMME, runoff into the lakes is calculated by aggregating contributing streamflows of NWM. The NBS is hence equal to the precipitation plus the runoffs, and minus the evaporation. Analysis was performed for 2012 - 2020 that operational NWM and NMME outputs are publicly available. Models were first evaluated by comparing precipitation, evaporation, and runoff with the Large Lake Statistical Water Balance Model (L2SWBM), which takes a probabilistic approach to estimate the 95% credible interval of these components in the context of closing the water balance. NBS was then compared with the residual net basin supply, which is equal to the change in storage minus inflows to each lake through connecting channels, and plus outflows. Results of this study indicate the performance of operational NBS forecasting with NMME and NWM, which facilitate the development of new forecasting tools for water level management of the Great Lakes. [ABSTRACT FROM AUTHOR]