Subsurface Physical Property Estimation by a One-dimensional Land Subsidence Simulator using a Genetic Algorithm with Multiple Extensometer Data for Different Depths

Abstract:

This study tried to estimate the subsurface physical properties combining the multiple extensometer data and a one-dimensional land subsidence simulator using the modified Cam-clay model. We developed subsurface models from the geological columns and conducted the inversion analysis with a genetic algorithm to estimate parameters using the long-term land subsidence monitoring data at Kawajima and Koshigaya-higashi in the Kanto Plain, Japan. In these study sites, the seasonal groundwater level fluctuations have caused plastic compaction in summer and elastic expansion in winter every year. For both sites, the obtained set of subsurface physical properties satisfactorily reproduced the observed subsidence within the range of typical values in existing literatures. In addition, the estimated physical properties distribution at both sites identified the boundary of highly compressible Holocene and harder Pleistocene, suggesting that the parameter estimation in this study worked well.