TSUNAMI INVERSION OF FAULT DISPLACEMENTS BASED ON SPARSE MODELING

Abstract

A study on tsunami inversion analysis was conducted using Lasso regression, which is a typical approach to sparse modeling. Characteristics of spatiotemporal distribution of the fault slips were then analyzed. Sparse modeling is characterized by yielding obtained solutions that contribute significantly to the output of the model. This feature is useful for investigating the development of prediction models. This paper applies a tsunami inversion analysis based on sparse modeling to the 2011 Tohoku Earthquake Tsunami and analyzes the characteristics of the fault slips.