Complex rupture during the 12 January 2010 Haiti earthquake

Abstract

This supplement describes the data collection and processing techniques for each 6 individual component of our multi-disciplinary analysis of the rupture process of 7 the 12 January 2010 Haiti earthquake (hereafter termed the 2010 Leogane 8 earthquake). First we describe details of the finite fault modeling technique, and 9 results for single-plane fault models using teleseismic data. Next we discuss InSAR 10 data sources and detailed processing, before presenting details of the geological 11 field deployment and data collection. Finally we present alternate joint inversions to 12 our preferred kinematic rupture model (figure 3), comparing these models to the 13 preferred solution, and discussing the relative merits of each. We present two 14 alternate rupture models; a three-plane model (figure S5) to show the effects of 15 north vs. south dip for fault B, the Leogane fault; and a three-plane model exploring 16 the effects of initiating rupture on fault B rather than fault A, the EPGF-like structure 17 ( figures S6 7). 18 19 Teleseismic Bodywave Finite fault modeling 20We invert for the rupture process of the earthquake using broadband teleseismic P-21 and SH-body waveforms recorded at GSN stations worldwide. Data were selected 22 based upon quality (high signal-to-noise ratios) and azimuthal distribution. 23Waveforms are first converted to displacement by removing the instrument 24 response and then used to constrain the slip history based on the finite fault 25 inversion algorithm of Ji et al. 1 . To improve our resolution of rupture onset, 26

Prediction of the area affected by earthquake-induced landsliding based on seismological parameters

“…The source is modelled as a wave emission line (or series of lines), defined by the location and length of the earthquake rupture, as determined by geophysical inversion of the slip distribution (or moment distribution for the Guatemala earthquake) on the seismogenic fault plane (Figs. 3, 4), (Kikuchi and Kanamori, 1991;Wald et al, 1996;Zeng and Chen, 2001;Hikima and Koketsu, 2005;Hayes et al, 2010;Suzuki et al, 2010;Fielding et al, 2013). The rupture length of such inversions agrees (within 30 % or less) with the a priori predicted length, based on the scaling proposed by Leonard (2010), for the Chi-Chi, Haiti, and Northridge earthquakes, but is smaller (by 40-50 %) for the Niigata and Iwate earthquakes, and larger for the Wenchuan and Guatemala earthquakes (Table 1).…”

“…Red perimeters show the area encompassing 95 % of the total area of landsliding defined by a uniform distance away from the wave emission line. For reference rupture slip distribution maps are shown for the Northridge and Haiti earthquakes (Wald et al, 1996;Hayes et al, 2010). For Haiti, in addition to the landslide inventory from Gorum et al (2013), we also show the one of Harp et al (2016) (cyan polygons) and its associated R 95 with red dashed line.…”

Two comparable earthquakes produced greatly different coseismic landslides: The 2015 Gorkha, Nepal and 2008 Wenchuan, China events

“…The results suggest that surface-rupture earthquakes trigger fewer but larger landslides in a smaller area than buried-rupture earthquakes of comparable magnitudes. In most cases, buriedrupture earthquakes trigger relatively more landslides than those caused by surface-rupture events, as exemplified by the 1994 Northridge, USA earthquake (Harp and Jibson, 1995;Hauksson et al, 1995Haiti (Xu et al, 2014cHayes et al, 2010), and the 2013 Lushan, China events . The relation between the surface-rupture Wenchuan Earthquake and the buried-rupture Gorkha Earthquake is the opposite.…”

Estimation of Peak Ground Acceleration from Horizontal Rigid Body Displacement: A Case Study in Port-au-Prince, Haiti

“…The epicenter is almost directly along the mapped trace of the Enriquillo Plaintain Garden fault (EPGF), the primary plate boundary fault in southern Haiti (e.g., Mann et al, 1991;Calais et al, 2002). Analysis of available data, including regional seismic, Global Positioning System (GPS), and Interferometric Synthetic Aperture Radar (InSAR) data, has yielded several different mainshock rupture models (Calais et al, 2010;Hayes et al, 2010). Although the models differ in detail, all involve moment release on a north-dipping fault adjacent to the EPGF and a predominantly unilateral rupture toward the west, although fault rupture with relatively minor moment release is inferred to have extended 5-10 km east.…”