One-dimensional land subsidence modelling constrained by electrical logging and its uncertainty analysis

Abstract

. Due to the limited availability of core sample test data, the land subsidence modeling is often highly uncertain. On the other hand, the electrical logging data are frequently accessible and might give some information to constrain the spatial distribution of physical properties in land subsidence modeling. Therefore, this study tried to constrain land subsidence model using electrical logging data. The estimated physical properties, based on the combination of existing empirical relations between the resistivity and physical properties, were used as initial values for the model inversion. A calibration process was then conducted by adjusting the physical parameters to reproduce the observed land subsidence. As a result, the obtained sets of physical properties were within the range of typical values in the existing literatures and satisfactorily reproduced the observed subsidence. Furthermore, numerous possible parameter realizations were generated using the Null Space Monte Carlo method to analyze the uncertainty in both physical properties and future subsidence predictions. The results also suggested the potential to reduce the uncertainty of land subsidence predictions by easily available geophysical logging data.