Developing a new estimate for the water balance of the Laurentian Great Lakes

Abstract:

Understanding the water balance of large lakes – one of the main sources of unfrozen, fresh surface water – is essential for sustainable management of the world’s water resources. However, several components of the water balance of large lakes, such as over-lake precipitation and over-lake evaporation, are usually not reliably estimated due to a lack of in-situ measurements, or under-representation of land-lake-atmospheric interactions in hydrological models. To support water accounting for the Laurentian Great Lakes, the largest system of lakes on Earth, we develop a new estimate of monthly lake storage changes and water balance components for each of the Great Lakes from 1950 to 2019. Multiple independent data sources (including gauge-based aggregated data, model simulations and remote sensing products) were assimilated into a Bayesian inference framework (the Large Lakes Statistical Water Balance Model; hereafter referred to as L2SWBM). The L2SWBM used a conventional water balance model to constrain the new estimates, allowing for water balance closure over multiple time periods. The source code and inputs of the L2SWBM, as well as the new estimates of Great Lakes water balance, are made publicly available (https://doi.org/10.7302/tx97-nn12), and can be used as a basis for water resources management planning decisions, forecasting, and future improvements in water balance assessments.