Unraveling elastic and inelastic storage of aquifer systems by integrating fast independent component analysis and a variable preconsolidation head decomposition method

Land subsidence is a geological process mainly caused by groundwater overdraft. Numerical modeling of land subsidence is the main method used for its simulation and prediction. The elastic skeletal storage coefficient (Ske), inelastic skeletal storage coefficient (Skv), and the related specific values (Sske and Sskv) are fundamental parameters to quantify land subsidence. In this paper, a novel approach integrating fast independent component analysis (Fast-ICA) with variable preconsolidation head decomposition method is proposed to disentangle Sske and Sskv at various depth and over time from piezometric and extensometer records. The proposed method is applied to areas affected by severe land subsidence in the North China Plain (Tianzhu, Pinggezhuang and Cangzhou stations). The elastic and inelastic parameters of the aquifer systems are quantified at different depths. It is found that Sske and Sskv decrease with depth. The finer the sediment grain size is, the smaller of the ratio Sske/Sskv. Moreover, Ske remains almost unchanged over time, while Skv decreases as compaction and land subsidence increase.