Cross-validation of water elevation measurements performed by UAS and satellite altimetry in unmonitored areas

Water levels of rivers and lakes provide essential information regarding hydrological phenomena, such as floods and droughts. Currently, the most accurate water level data comes from in-situ gauging stations (standard error less than 1 cm). However, installation and maintenance of in-situ stations is expensive and, unlike satellite earth observation, they are not suitable for remote areas. Depending on river size and mission characteristics, water levels from satellite altimetry have standard errors of between 10 and 50 cm, while water levels from UAS altimetry have stadard errors of 5 cm or better . Validation of water level retrievals from satellite EO is performed by comparing satellite and UAS altimetry datasets. The performance is quantified using common statistics such as the root mean square error (RMSE), the bias, the Median Absolute Deviation (MAD), which provides an assessment of the errors by removing the impacts of potential outlier data points, and the Pearson coefficient (R), that gives an indication on the correlation of the time series.