Tracking shallow and deep groundwater storage changes in North China Plain with improved fusion method and hybrid spectral analysis approach

Abstract

The North China Plain (NCP) is a critical agricultural hub in China, confronting a significant issue of groundwater depletion. However, different terrestrial water storage (TWS) change products from the Gravity Recovery and Climate Experiment (GRACE) satellites differ considerably in estimating groundwater depletion rates in the NCP. It is urgent to optimize these different products to obtain an elevated understanding of regional groundwater storage (GWS) changes. The generalized three-cornered hat (GTCH) method is used to quantify uncertainties of different GRACE products, and this prior information is integrated with the Bayesian model averaging (BMA) method to fuse GRACE products. Results demonstrate that the GTCH + BMA method can generate uniform GWS products with heightened precision, enhanced coherence, and diminished uncertainty in the NCP. After that, the shallow and deep GWS are meticulously differentiated based on the high-coverage in-situ data, and the shallow GWS (2005.1–2016.8) (-13.7 ± 4.6 mm/yr) declined faster than deep GWS (-6.05 ± 3.9 mm/yr) in the NCP. To further explore the primary factors driving these changes, the latest Hankel Spectrum Analysis (HSA) method combined with ridge regression simulation is employed to decouple the change components of human and natural factors in the shallow and deep GWS. The first attempt to give the contribution of human and natural factors to shallow GWS is 77.18 %/22.82 %, and to deep GWS is 87.59 %/12.41 %.