Seasonal variation of land subsidence caused by the groundwater abstraction for snow removal in Nagaoka city , Japan

 This research primarily focuses on predicting and analyzing land subsidence and groundwater level caused by the groundwater abstraction for snow melting during winters in Nagaoka City in Niigata Prefecture , Japan. In this area, significant inconvenience caused by heavy snowfall during winters has led to the adoption of groundwater extraction for snow melting purposes. This countermeasure results in seasonal fluctuations in groundwater levels, contributing to plastic compaction occurrences during winter seasons every year. The uncertainty in modeling land subsidence due to the inability to accurately determine the distribution of subsurface physical property values poses a challenge. Additionally, inadequate data regarding the amount of water extracted for snow melting further complicates the analysis.Combining groundwater level data obtained through Kriging interpolation, we utilized numerical simulations by MODFLOW with subsidence package (IBS) and adjusted model parameters to reproduce the scenarios in this region. The model was calibrated to reproduce groundwater level and subsidence data. By referencing parameters in the existing literatures parameters and experimenting with various scenarios, the model successfully simulated changes in groundwater levels and land subsidence, demonstrating the effectiveness of the model. Moreover, by assigning different permeability coefficients in possible ranges, we determined the maximum extent of snowmelt water's contribution to groundwater replenishment. Through the analysis of various extraction and replenishment schemes, we clarified the main causes of land subsidence and the primary sources contributing to groundwater recovery. We also proposed possible extraction strategies to address the challenge of groundwater extraction under acceptable land subsidence considering the increasing demand for groundwater due to urbanization and possible climate change.