부산대학교 도서관

The use of measurements from NASA's CYGNSS mission for the mapping of inland water bodies has been extensively demonstrated in past works. Despite this progress, the use of CYGNSS measurements for monitoring dynamic properties of water bodies, such as reservoir storage levels remains challenging given the need to combine observations over extended time periods or areas to improve water detection performance. This work explores the feasibility of an alternative sensing approach based on CYGNSS Level-1 products that has the potential to estimate reservoir water storage levels on intervals significantly finer compared to existing techniques. The method is explored using three example reservoirs with CYGNSS observations extending from October 2018 to December 2022. Retrieved water levels are found to be associated with relative water level root-mean-square-errors varying from 8–20% depending on the observed scenes' level of complexity. Correlations of up to $\approx\! \text{86}\%$ relative to in situ data are observed. The initial results presented in this work highlight the potential of spaceborne GNSS-R systems for reservoir water level monitoring.