Assessing water storage changes of Lake Poyang from multi-mission satellite data and hydrological models

Abstract

Understanding water storage changes in Lake Poyang, the largest freshwater lake in China, is essential for local hydro-ecological assessments and water resource management. The integration of multi-mission satellite data, hydrological models, and in situ measurements allows for a comprehensive estimate of Lake Poyang’s storage variations. We here estimated Lake Poyang water storage changes during the recent decade by using inundation areas mapped from optical satellite imagery and water levels measured by satellite radar altimetry and gauging stations. The amplitudes of seasonal variation from altimetry data are smaller than those from station measurements. This is likely attributed to their low temporal resolutions and limited footprint coverage, together with a complex surface gradient over Lake Poyang. The residual fields between land water storage changes assessed by the GRACE satellites and simulated by two hydrological models (GLDAS Noah and WGHM) were applied to estimate Poyang water storage changes. Leakage errors in the GRACE-model residuals are further corrected by a constrained forward modeling method, resulting in recovered water storage trends ~66 times of the uncorrected signals. Water level changes estimated by different methods are then compared. Results show that level changes inverted from recovered storage variations by GRACE-GLDAS and GRACE-WGHM are significantly larger than those from satellite altimetry and in situ measurements. This indicates that the combination of GRACE observation and global hydrological modeling is likely insufficient to estimate accurate water storage changes in Lake Poyang. Our methods and results provide a valuable example of using integrated methods for monitoring water storage changes in highly dynamic fluvial lake systems.