[S-CG10_2AM2] Microcracks preceding ruptures in the crust related to earthquakes, volcanic eruptions and landslide

[SCG10-06] The Development of self-potential tomography to estimate the ground water condition

Hiroshi OTSUBO¹, Katsumi HATTORI¹, *Tomohiro YMAZAKI¹ (1.Graduate school of science, Chiba University)

Keywords:landslide, self-potential

Landslides are one of the most severe natural disasters in the world and there are two types; rainfall induced landslides and landslides triggered by an earthquake. In this research, basic study on early warning system for landslides will be performed to understand rainfall-induced landslide process by hydrological and electromagnetic changes. The final goal of this research is to develop a simple technology for landslide monitoring/forecasting using self potential method. The advantages of this method are lower cost and easier to set up than the hydrological approaches using pore pressure sensors. The laboratory experiments show that the self-potential variation has relationship with the water and soil displacements. But, we can not estimate the ground water condition by self-potential yet. So, in this study, we developed self-potential tomography to estimate the ground water condition.Measured self-potential value under the ground and charge distribution to estimate is given by the Coulomb's law. Therefore, this is inverse problem. To solve the inverse problem, we adapt Phillips-Tikhonov regularization with Generalized Cross Validation (GCV). To evaluate the reconstructed charge distribution and investigate the relationship with the ground water condition, computational simulations and applications to practical data by using the sandbox experiment has been examined.It is found that the developed algorithm is effective through numerical simulations. Results of application to sandbox experiments show good performance but there are some problems to solve.The details will be given in our presentation.