Work

A Large-Scale Evaluation of SWOT-Derived Water Surface Elevations: Precision Drivers and Strategies to Enhance Data Availability

Year: 2026

Type: article

Abstract: Accurate water surface elevation (WSE) measurements are critical in hydrological applications, such as river discharge estimation, hydrological modeling, and flood forecasting. We evaluated the performance of WSE measurements from the Surface Water and Ocean Topography (SWOT) mission using in situ records from 132 lakes across Brazil, comparing vector (LakeSP) and raster (Raster_250m, Raster_100m) products. Median errors were low (11 to 13 cm), though high variability indicated outliers. A Random Forest analysis revealed key factors influencing WSE accuracy, including the quality flags, lake geometry, cross-track distance and the incidence angle. Although these flags effectively categorize WSE data, Flag 0 might be overly restrictive, containing only ∼22% of observations, while many reliable measurements reside within Flag 1, which, on the other hand, has higher variability. Hence, we proposed targeted spatial filtering, combining all Flag 0 observations with Flag 1 data constrained to a 29-59 km cross-track distance window. This approach significantly improved data availability without compromising WSE overall accuracy, showing that a supplementary approach to the flagging system can be beneficial, adding to the existing framework and enhancing the practical applicability of SWOT for hydrological monitoring and offering the first systematic evaluation of WSE measurements from SWOT for lakes covering the spatial extent of a country as large and diverse as Brazil.

Source: Remote Sensing

DOI: 10.3390/rs18101609