No abstract

No abstract

Hydro-meteorological datasets are key components for understanding physical hydrological processes, but the scarcity of observational data hinders their potential application in poorly gauged regions. Satellite-retrieved and atmospheric reanalysis products exhibit considerable advantages in filling the spatial gaps in in-situ gauging networks and are thus forced to drive the physically lumped hydrological models for long-term streamflow simulation in data-sparse regions. As machine learning (ML)-based techniques can capture the relationship between different elements, they may have potential in further exploring meteorological predictors and hydrological responses. To examine the application prospects of a physically constrained ML algorithm using earth observation data, we used a short-series hydrological observation of the Hanjiang River basin in China as a case study. In this study, the prevalent modèle du Génie Rural à 9 paramètres Journalier (GR4J-9) hydrological model was used to initially simulate streamflow, and then, the simulated series and remote sensing data were used to train the long short-term memory (LSTM) method. The results demonstrated that the advanced GR4J9–LSTM model chain effectively improves the performance of the streamflow simulation by using more remote sensing data related to the hydrological response variables. Additionally, we derived a reservoir operation model by feeding the LSTM-based simulation outputs, which further revealed the potential application of our proposed technique.

Abstract. Reservoirs and dams are essential infrastructure in water management; thus, information of their surface water area (SWA), water surface elevation (WSE), and reservoir water storage change (RWSC) is crucial for understanding their properties and interactions in hydrological and biogeochemical cycles. However, knowledge of these reservoir characteristics is scarce or inconsistent at the national scale. Here, we introduce comprehensive reservoir datasets of 338 reservoirs in China, with a total of 470.6 km3 storage capacity (50 % Chinese reservoir storage capacity). Given the scarcity of publicly available gauged observations and operational applications of satellites for hydrological cycles, we utilize multiple satellite altimetry missions (SARAL/AltiKa, Sentinel-3A and Sentinel-3B, CroySat-2, Jason-3, and ICESat-2) and imagery data from Landsat and Sentinel-2 to produce a comprehensive reservoir dataset on the WSE, SWA, and RWSC during 2010–2021. Validation against gauged measurements of 93 reservoirs demonstrates the relatively high accuracy and reliability of our remotely sensed datasets. (1) Across gauge comparisons of RWSC, the median statistics of the Pearson correlation coefficient (CC), normalized root mean square error (NRMSE), and root mean square error (RMSE) are 0.89, 11 %, and 0.021 km3, with a total of 91 % validated reservoirs (83 of 91) having good RMSE from 0.002 to 0.31 km3 and NRMSE values smaller than 20 %. (2) Comparisons of WSE retracked by six satellite altimeters and gauges show good agreement. Specifically, the percentages of reservoirs having good and moderate RMSE values smaller than 1.0 m for CryoSat-2 (validated in 30 reservoirs), SARAL/AltiKa (9), Sentinel-3A (34), Sentinel-3B (25), Jason-3 (11), and ICESat-2 (26) are 77 %, 75 %, 79 %, 87 %, 81 %, and 82 %, respectively. By taking advantages of six satellite altimeters, we are able to densify WSE observations across spatiotemporal scales. Statistically, around 96 % of validated reservoirs (71 of 74) have RMSE values below 1.0 m, while 57 % of reservoirs (42 of 74) have good data quality with RMSE values below 0.6 m. Overall, our study fills such a data gap with regard to comprehensive reservoir information in China and provides strong support for many aspects such as hydrological processes, water resources, and other studies. The dataset is publicly available on Zenodo at https://doi.org/10.5281/zenodo.7251283 (Shen et al., 2021).

No abstract