Earthquake slip-length scaling: new insights from an expanded data base

Student: Pauline Billion
Supervisors: Isabelle Manighetti

ABSTRACT

One of the goals of this dissertation of three months was to firstly, collect all possible sets of data of earthquakes occurring in New Zealand and in Iran in order to, secondly, expand the database and the model of the worldwide earthquakes starts by I. Manighetti and coauthors in a paper published in 2007. All this work will help to understand better, reduce uncertainties, bring new insights of the scaling law DMAX-L and provide a better basis for seismic hazard with the integration of the fault properties, all this thank to the expanded database.

This paper first sets a summary of the previous work on the numerous scaling laws. In this first part the work of many authors in the previous 30years is examined and summarized (Kanamori and Anderson, 1975; Bonilla et al., 1984; Rundle and Romanowicz, 1993; Bodin and Brune, 1996). The scaling laws we focus on in this research, link the observed length at the ground surface of the rupture (i.e. L) with the maximum displacement along this length (DMAX). As found by Manighetti et al., 2007, these laws can be really dependent on several factors as for instance the age of the earthquake and the number of broken segments during this quake.

The database was so composed of about 60 more earthquakes from two different but precise regions of the world: New Zealand and Iran.

The earthquakes occurring on subduction zones have been carefully excluded to not interfere with other data, especially about the width (counted in vertical or along the fault plane) of the earthquakes. Only data of surface ruptures appear on this database to not obstruct with some values of displacement at depths from inversion models.

The useful values for us (DMAX and L among others) for the two regions of the world specified earlier have so been added to the previous database and in the model established by Manighetti et al., 2007 (described in their paper). Contrarily to Manighetti et al. paper, the question of maturity had not really been asked here due to the short time of research.

We propose new DMAX-L parameterizations based on the idea of multiple segments-ruptures. Each segment roughly scales as a single crack but the multiple segments-rupture does not.

The results with the enriched database and the conclusions of this dissertation work show that a part of the curves and hypothesis are confirmed by these new data whereas the rest seem different due to the data from Iran. However, the variations on the values of the curves are similar and stay in the same order of magnitude than the ones found by the work of Manighetti and her co-authors in 2007.

You may download a digital version of this MSc dissertation here.