Reconstructing Groundwater Storage Changes in the North China Plain Using a Numerical Model and GRACE Data

Abstract

Groundwater has been extensively exploited in the North China Plain (NCP) since the 1970s, leading to various environmental issues. Numerous studies have utilized Gravity Recovery and Climate Experiment (GRACE) satellite data to analyze changes in groundwater storage in the NCP and provide valuable insights. However, the low spatial resolution of GRACE data has posed challenges for its widespread application, and there have been limited studies focusing on refining groundwater storage changes in the NCP. In addition, the lack of data on the gap period between GRACE and GRACE-FO hinders in-depth research on regional groundwater storage anomalies (GWSA). This paper applied a groundwater storage model called NGFLOW-GRACE to construct a groundwater storage change model in the NCP at spatial resolutions of both 1° and 0.05°. The groundwater storage change model was calibrated and driven using gratis data, with hydrogeological parameter values estimated using the shuffled complex evolution algorithm (SCE-UA). The model exhibited favorable performance, with correlation coefficients greater than 0.85 during the calibration period and 55% of coefficients greater than 0.50 during the validation period. Interestingly, the results indicate that different combinations of remote sensing data do not significantly impact the outcomes, while the hydraulic gradient coefficient demonstrates the highest sensitivity. Appropriate reconstructed data were selected within the empty window period, and by downscaling the model to a resolution of 0.05°, a complete cycle (January 2003 to December 2020) of GWSA was derived. Through comprehensive comparisons with previous research findings on both temporal and spatial scales, it can be concluded that the downscaled groundwater storage changes obtained from the established model demonstrated high reliability.

Camera and LiDAR Fusion for Urban Scene Reconstruction and Novel View Synthesis via Voxel-Based Neural Radiance Fields

“…3D reconstruction [5,6] can be used not only for data augmentation but also for direct 3D modeling of urban scenes [7]. Specifically, in the remote sensing mapping [8][9][10][11][12], it can generate high-precision digital surface models using multi-view satellite images [13,14] and combine the diversity of virtual environments with the richness of the real-world, generating more controllable and realistic data than simulation data.…”