Abstract's details
Global Daily Reservoir Evaporation Modeling Augmented by SWOT PIXC Data
Event: 2025 SWOT Science Team Meeting
Session: Hydrology: SWOT Lakes, Estuaries and Wetlands (SLEW)
Presentation type: Oral
Open-water evaporation can exceed 10% of annual reservoir storage, imposing significant constraints on water-supply reliability and hydropower generation, especially in arid and semi-arid regions. Robust and effective reservoir management practices require evaporation estimates at daily resolution; however, conventional approaches still rely predominantly on monthly Class A pan measurements, which do not account for critical factors like heat storage and wind-fetch effects.
To address this gap, we integrate the physically based Daily Lake Evaporation Model (DLEM)—an advanced Penman energy-balance formulation explicitly accounting for heat storage and wind fetch—with dynamic reservoir geometry derived from the Surface Water and Ocean Topography (SWOT) mission’s KaRIn pixel-cloud (PIXC) product. This integrated approach will potentially allow us to produce daily evaporation estimates for more than 21,000 reservoirs globally, as cataloged in the GeoDAR v1.1 dataset, for the period 2020–2024.
For every SWOT overpass, reservoir surface area and elevation are retrieved from PIXC observations and subsequently used to estimate mean reservoir depth. These morphological parameters are interpolated to a daily resolution and, along with meteorological forcing from ERA5-Land reanalysis data, drive DLEM simulations. Validation of the DLEM outputs against eddy-covariance station measurements demonstrates good model accuracy, with coefficient of determination (R²) values ranging from 0.56 to 0.79 and root mean square errors (RMSE) between 1.27- and 2.52-mm day⁻¹.
To evaluate the role of SWOT data in enhancing the reservoir evaporation modeling, this dataset is also compared with another version which is not augmented by SWOT observations.
This comprehensive evaporation dataset holds promise for refining operational water management strategies and informing adaptive policies in response to evolving hydro-climatic conditions. It can also benefit other SWOT projects related to reservoirs and flow regulations.
Back to the list of abstractTo address this gap, we integrate the physically based Daily Lake Evaporation Model (DLEM)—an advanced Penman energy-balance formulation explicitly accounting for heat storage and wind fetch—with dynamic reservoir geometry derived from the Surface Water and Ocean Topography (SWOT) mission’s KaRIn pixel-cloud (PIXC) product. This integrated approach will potentially allow us to produce daily evaporation estimates for more than 21,000 reservoirs globally, as cataloged in the GeoDAR v1.1 dataset, for the period 2020–2024.
For every SWOT overpass, reservoir surface area and elevation are retrieved from PIXC observations and subsequently used to estimate mean reservoir depth. These morphological parameters are interpolated to a daily resolution and, along with meteorological forcing from ERA5-Land reanalysis data, drive DLEM simulations. Validation of the DLEM outputs against eddy-covariance station measurements demonstrates good model accuracy, with coefficient of determination (R²) values ranging from 0.56 to 0.79 and root mean square errors (RMSE) between 1.27- and 2.52-mm day⁻¹.
To evaluate the role of SWOT data in enhancing the reservoir evaporation modeling, this dataset is also compared with another version which is not augmented by SWOT observations.
This comprehensive evaporation dataset holds promise for refining operational water management strategies and informing adaptive policies in response to evolving hydro-climatic conditions. It can also benefit other SWOT projects related to reservoirs and flow regulations.