Journal of Geophysical Research

Supporting Information for

Along-strike change in fault structural maturity due to fault growth controls location of largest earthquake slip and fast rupture

Clément Perrin1,2,*, Isabelle Manighetti1, Jean-Paul Ampuero3, Frédéric Cappa1,§, and Yves Gaudemer4

1 Université Nice Sophia Antipolis, Géoazur (UMR7329), CNRS, Observatoire de la Côte d’Azur, Sophia-Antipolis, France

2 Now at Lamont-Doherty Earth Observatory of Columbia University, New York, USA

3 Seismological Laboratory, California Institute of Technology, Pasadena, CA, USA

4 Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Université Paris Diderot, CNRS F-75005 Paris, France

§ Institut Universitaire de France, Paris, France

Contents of this file

Supplementary Document S1 Supplementary Figures S1 to S6 Supplementary Tables S1 to S5

Additional Supporting Information (Files uploaded separately)

Supplementary Figures S1 to S2 Supplementary Table S1

Introduction

The Supporting Information includes one document denoted Supplementary Document S1, five figures denoted Supplementary Figures S1, S2, S3, S4, S5 and S6 and five Tables denoted Supplementary Tables S1, S2, S3, S4 and S5, that provide detailed descriptions on the earthquake and fault datasets and additional information on a few specific points of the paper.

1

Supplementary Document S1: description of earthquake and fault data used in present analysis

In this document, for each couple of earthquake and causative fault, we describe the information that we found in literature and used to constrain the earthquake and fault parameters that we analyze in the present study (see section 3 of the paper). Because uncertainties on most of these parameters cannot be quantified with simple metrics, we estimate the quality of the principal parameters using criteria described below.

The principal earthquake parameters that we analyze in the paper are (i) the surface and depth slip-length profiles (which define the location of the zone of largest slip and hence the possible asymmetry of the slip profile), (ii) the position of the rupture trace within the entire causative long-term fault, (iii) the hypocenter position, and (iv) the rupture speed. • Uncertainties on hypocenter positions are rarely provided in literature. Because they might be large, especially for old earthquakes, we have assigned to all earthquakes a conservative, 5 km uncertainty on both horizontal components of the epicenter (as in Manighetti et al., 2005; see Figure S1). • Uncertainties on rupture speeds are rarely provided. We discuss them below on a case-by-case basis, and discuss their overall uncertainties in the text. • We estimate the robustness of the earthquake surface trace mappings and of the surface slip profiles by assigning arbitrary “quality factors”: data are considered of high quality (or quality 1, Q1) when they are constrained by at least two different sets of measurements from different authors, which are consistent; data are considered of good quality (or quality 2, Q2) when they are constrained by at least one series of well-constrained measurements; data are considered of low quality (or quality 3, Q3) when they are derived from incomplete, old, or disputable measurements. Note that, for surface slip profiles, we retained only those with at least one slip measurement per 1 and 10 kilometers of rupture on short (≤ 50 km) and long ruptures, respectively. • We use 25 published finite source rupture models (available for 13 out of the 27 analyzed earthquakes), most were compiled by Mai and Thingbaijam (2014), while others were provided to us directly by their authors. Because uncertainties on source inversion models cannot be appropriately provided with the present modeling procedures, we do not discriminate the available models; we suggest however which model is based on more complementary data and hence might be best constrained.

The principal fault parameters that we analyze in the paper are (i) the overall map-view traces and geometry of the fault zones, and (ii) the direction of long-term propagation of the faults. • We estimate the robustness of the fault zone maps by assigning arbitrary “quality factors”: fault zone maps are considered of high quality (or quality 1, Q1) when they have been produced consistently in at least two different studies by different authors; fault maps are considered of good quality (or quality 2, Q2) when they have been produced in only one, yet rigorous study. • As explained in the paper, the lateral propagation of a fault over the long-term produces a gradient of ages and net slips along the fault. Therefore having multiple fault age and/or net slip measurements along a fault is the most warranted situation to properly constrain the direction of long-term propagation of the fault. Other criteria exist however, such as degree of fault segment connection along the fault, 2

along-strike decrease in fault slip rate, development of typical tip splay networks (see text). Therefore, we consider that the direction of long-term propagation of a fault is robustly assessed (high quality or quality 1, Q1) when it is derived from at least age and/or net slip data along the fault; it is fairly well assessed (quality 2, Q2) when it is derived from at least two other sorts of evidence; and it is more weakly constrained (quality 3, Q3) when it is derived from observation of tip splay networks only.

We indicate these quality factors in the descriptions below (as “Q1, Q2, or Q3” by the end of each related section). We also compile them all in Supplementary Table S2, so as to provide a comprehensive view of the robustness of the entire dataset that we analyze.

------

1957 Mw ~8.1 BOGD EARTHQUAKE Earthquake: 04/Dec/1957, Mw 8.1, Mongolia, Left-lateral strike-slip Causative fault: Bogd fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured, both on the field soon after the earthquake and on satellite imagery (e.g., Florensov and Solonenko 1963; Baljinnyam et al., 1993; Bayarsayhan et al., 1996; Kurushin et al., 1997; Rizza et al., 2011; Choi et al., 2012). These available measurements are consistent. The rupture trace and length are thus robustly constrained (length 250-260 km; e.g., Kurushin et al., 1997; Choi et al., 2012). Q1. - Rupture extent within causative fault: The 1957 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., references above). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured both on the field and from satellite images (maximum lateral slip at surface ~7 m, whereas vertical slip occurred at local sites of the fault, with maximum amplitude of 3-5 m; Kurushin et al., 1997; Choi et al., 2012). The slip measurements are consistent among studies, and therefore, the surface slip profile is robust. The lateral slip is the actual dominant displacement on the fault, whereas the vertical slip is only local and occurring off- the master Bogd fault. Therefore, the slip profile that characterizes the Bogd earthquake rupture is the one that describes the lateral slip distribution. We chose to represent in Supplementary Figure S1 the slip profile determined by Choi et al. (2012) because these authors combined both new data and prior measurements from Kurushin et al. (1997). Q1 - Hypocenter: The hypocenter location is not precisely constrained but a robust fact is that the earthquake initiated near the western tip of the rupture (Chen and Molnar, 1977) and then propagated mainly unilaterally towards the east.

Causative fault parameters - General parameters Slip mode and strike: left lateral strike slip, ~N100°E trending fault Initiation age: Between 2 and 8 Ma (e.g., Vassalo et al., 2007a) as a result of northward propagation of Indo-Eurasia Cenozoic transpressive deformation (Molnar and Tapponnier 1975; Tapponnier and Molnar 1977, 1979). The initiation age is not measured however, and thus is poorly constrained. Length: ~430 km Net slip: unknown Long-term slip rate: ~1 mm/yr (lateral) and 0.1-0.2 mm/yr (vertical) over the Quaternary (Ritz et 3

al., 1995, 2006; Vassalo et al., 2005, 2007b; Rizza et al., 2011), consistent among available studies, and hence well constrained. Current slip rate: ~1.2 mm/yr (Calais et al., 2003), yet poorly constrained (based on two GPS stations only). - Mapping of fault trace: Several precise fault maps have been done and well describe the fault zone architecture at surface (e.g., Tapponnier and Molnar 1979; Kurushin et al., 1997; Cunningham 2013). We choose to represent in Supplementary Figure S1 the fault map from Tapponnier and Molnar (1979) because it presents the fault along its entire length. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Bogd Fault is intermediate. - Direction of long-term propagation: Q2 Bayasgalan et al. (1999a-b) have shown that the age of off-fault thrusts associated with the Bogd fault decreases eastward. The degree of connection of the Bogd fault segments also decreases eastward (Choi et al., 2012). At its western tip, the Bogd fault abuts and terminates against the ~N115° trending Gichgeniyin Nuruu Thrust. At its eastern tip, the Bogd fault splays into multiple oblique branches (e.g., North Baga Bogd; Supplementary Figure S1) which form a fan-shape network enlarging in the East direction. Altogether these evidences show that the Bogd fault has propagated from west to east during its lifetime.

1983 Mw ~6.9 BORAH PEAK EARTHQUAKE Earthquake: 08/Oct/1983, Mw 6.9, Idaho, Normal fault Causative fault: Thousand Springs-Warm Spring Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured, on the field, soon after the earthquake (Crone and Machette 1984; Crone et al., 1987). The measurements are consistent among available studies and therefore the rupture trace and length are robustly constrained (length ~34 km). Q1. - Rupture extent within causative fault: The 1983 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while precise rupture maps are available (e.g., Crone and Haller 1991; U.S. Geological Survey and Idaho Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field soon after the earthquake and is consistent among studies (maximum vertical slip of ~2.7 m, Crone and Machette 1984; Crone et al., 1987). The earthquake surface slip profile is thus robust. Q1 - Earthquake slip profile at depth: One source inversion model was published on the 1983 Borah Peak earthquake, whose grid data is available (Mendoza and Hartzell, 1988). The model is based on teleseismic data only however, and hence is not well constrained. In the model, the rupture length is slightly longer (~40 km) than observed at surface due to slip inferred south of the southern tip of the surface rupture. The maximum displacement inferred at depth (~1.5 m at 20-25 km) is lower than the maximum slip measured at the surface, what confirms that the model is not fully correct. Despite of its limitations, the model provides a slip distribution in fair agreement with the surface slip distribution, with a maximum slip located in the southern part of the rupture (Supplementary Figure S2). - Hypocenter: The hypocenter of the Borah Peak earthquake is poorly located, with errors of 5-10 km (Richins et al., 1987). It is however inferred that the earthquake initiated at ~16 km depth (Doser and Smith 1985), at the southern rupture tip near the intersection with the Mackay fault (Susong et al., 1990). The rupture then propagated unilaterally to the north.

4

Causative fault parameters - General parameters Slip mode and strike: normal fault, NW-trending fault Initiation age: Likely between 2 and 7 Ma ago, based on structural correlation between faults, sedimentary deposits and dated volcanic rocks (Scott et al., 1985; Christiansen 1986 and references therein). Length: ~34 km (but the fault possibly extends slightly further to the north; e.g., Crone and Haller 1991). Net slip: ~ 2.7 km, estimated from elevation differences between the highest point of Idaho ranges (Borah Peak) and the lowest point in the Thousand Spring Valley, plus the thickness of sediments in the basin (0.6-0.9 km) (e.g., Crosthwaite et al., 1970; Crone and Machette 1984; Scott et al., 1985). Long-term slip rate: < 0.3 mm/yr over the Holocene (Scott et al., 1985 and references therein), yet not robustly constrained. Current slip rate: 0.5 to 1 mm/yr of horizontal extension is measured at a few GPS stations across the fault (Payne et al., 2008). - Mapping of fault trace: Several precise fault maps have been done which are all consistent (e.g., Crone et al., 1987; Crone and Haller 1991; compilation in U.S. Geological Survey and Idaho Geological Survey, 2006). Q1. - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Thousand Springs-Warm Spring fault is immature. - Direction of long-term propagation: Q1 Many studies have shown that the cumulative slip (approached by range elevation) on the Thousand Springs-Warm Spring fault is largest in the southern part of the fault and decreases towards the north (Anders and Schlische, 1994; Densmore et al., 2005). Moreover, the Holocene and current fault slip rates also decrease from south to north (Scott et al., 1985; Payne et al., 2008). At its southern tip, the Thousand Springs-Warm Spring fault abuts and terminates abruptly against the strongly oblique Mackay fault. At its northern tip, the Thousand Springs-Warm Spring fault splays into oblique branches (i.e., Lone Pine fault, e.g., Crone and Haller 1991) which form a fan-shape network enlarging in the North direction. Altogether these evidence show that the Thousand Springs-Warm Spring fault has propagated from south to north during its lifetime.

1968 Mw ~6.4 BORREGO MOUNTAIN EARTHQUAKE Earthquake: 09/Apr/1968, Mw 6.5, California, Right lateral strike slip Causative fault: San Jacinto fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field, soon after the earthquake (Clark, 1972). The rupture trace and length are thus fairly well constrained (length ~33 km). Q2. - Rupture extent within causative fault: The 1968 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (Clark 1972; U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been densely measured on the field, yet in only one study (Clark, 1972; maximum lateral slip of ~0.4 m). Q2 - Hypocenter: The location of the hypocenter is not precisely constrained but a robust fact is that the earthquake initiated in the northern third of the rupture trace, apparently at about ~11 km depth (Allen and Nordquist, 1972), and then propagated mainly to the southeast.

5

Causative fault parameters - General parameters Slip mode and strike: right lateral strike-slip, ~N140°E-trending fault Initiation age: The San Jacinto fault is a major splay fault of the Southern San Andreas fault, which formed ~5 Ma ago (Powell and Weldon, 1992 and references therein; Sims, 1993). The southernmost part of the San Jacinto fault has formed less than ~2 Ma ago, based on stratigraphic and structural evidence in the Pleistocene Borrego formation (e.g., Lutz et al., 2006; Kirby et al., 2007; Dorsey et al., 2012). The initiation age of the San Jacinto fault is thus fairly well constrained. Length: ~260 km. Net slip: ~25 km based on offsets of intrusive igneous rocks in the central part of the fault (Sharp 1967). Long-term slip rate: constrained from multiple, consistent measurements, from ~15 mm/yr in the north to ~4 mm/yr in the south (Hudnut and Sieh, 1989; Rockwell et al., 1990; Gurrola and Rockwell 1996; Kendrick et al., 2002; Blisniuk et al., 2010; Janecke et al., 2010; http://www.data.scec.org/Module/links/sratemap.html). Current slip rate: constrained from multiple, consistent GPS and InSAR measurements, of 20 to 25 mm/yr (InSAR; Fialko 2006; Lundgren et al., 2009; Lindsey and Fialko 2013). - Mapping of fault trace: Several consistent precise fault maps have been published (e.g., Sharp, 1967; compilation in U.S. Geological Survey and California Geological Survey, 2006; Dorsey 2002). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the San Jacinto fault is intermediate. - Direction of long-term propagation: Q1 Many studies have shown that both the net slip (Lutz et al., 2006; Dorsey et al., 2012) and the long- term slip rate (Hudnut and Sieh, 1989; Rockwell et al., 1990; Gurrola and Rockwell 1996; Kendrick et al., 2002; Blisniuk et al., 2010; Janecke et al., 2010; http://www.data.scec.org/Module/links/ sratemap.html) decrease from north to south along the San Jacinto fault. Fault age also decreases from north to south (Lutz et al., 2006; Kirby et al., 2007). Moreover, fault segments become increasingly disconnected towards the southeast (Marliyani et al., 2013). At its northwestern tip, the San Jacinto fault terminates by connecting to the master San Andreas fault. The San Jacinto fault splays into multiple oblique branches along its trace and especially near its southeastern tip (see Supplementary Figure S1); there, the branches form a fan-shape network enlarging in the southeast direction. Altogether these evidence show that the San Jacinto fault has propagated unilaterally from northwest to southeast during its lifetime.

1999 Mw ~7.6 CHICHI EARTHQUAKE Earthquake: 21/Sep/1999, Mw 7.6, Taiwan, left-lateral and reverse Causative fault: Chelungpu fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured both on the field (e.g., Chen et al., 2001) and from SPOT satellite images (Dominguez et al., 2003). The various measurements are consistent, and therefore, the rupture trace and length are robustly constrained (length ~90 km). Q1. - Rupture extent within causative fault: The 1999 surface rupture is still fairly clear on high- resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Chen et al., 2001; Dominguez et al., 2003; Yue et al., 2005 and references therein). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured soon after

6

the earthquake both on the field (maximum vertical slip of ~10 m, Chen et al., 2001), and from SPOT images (maximum net slip of ~11 m, Dominguez et al., 2003), and these measurements are consistent. The earthquake surface slip profile is thus well-constrained. We chose to represent in Supplementary Figure S1 the slip profile determined by Dominguez et al. (2003) because it provides continuous slip measurements along the rupture. We have calculated the net slip profile (horizontal component) by combining the N-S and E-W measured components (the calculation has been validated by S. Dominguez, personal communication, 2013). Q1 - Earthquake slip profile at depth: Three source inversion models were published on the 1999 Chichi earthquake, whose grid data are available. The models differ principally from the data they used (Ma et al., 2001: Teleseismic, strong motions, GPS; Jonhson et al., 2001: GPS; Wu et al., 2001: GPS and strong motions). All three models use a large amount of various data and hence are likely robust. As a matter of fact, they all produce a similar shaped slip profile, showing slip decreasing fairly regularly from a maximum value in the north of the rupture down to the southern termination of the rupture. While the maximum slip zone is co-located in all three models, the amplitude of the maximum slip differs from ~10 to ~25 m. All three models find most of slip distributed in the depth range 0-20 km, as underlined by the aftershocks (Chang et al., 2000). The model of Ma et al. (2001) being the one using more data, both local and regional, it is the one we most consider in Supplementary Figure S2. The zone of largest displacements at depth is co-located with that observed at surface (compare Supplementary Figure S1 and S2). This further supports that the models are correct overall. - Hypocenter: The location of hypocenter is well constrained in the southern half of the rupture, at ~8 km depth (Chang et al., 2000). The rupture thus mainly propagated towards the north. - Rupture speed: The rupture speed was suggested to be similar along the rupture length (~2.5 km/s). However the slip velocity was faster in the northern part of the rupture (Ma et al., 2001, 2003; Wu et al., 2001; Zeng and Chen, 2001).

Causative fault parameters - General parameters Slip mode and strike: left lateral and reverse fault, N-S-trending Initiation age: The fault is located in the Western Foothills that are a young fold-and-thrust belt that is still developing in the Pliocene-Quaternary sediments (Lu and Hsü, 1992). Its initiation age is thus considered to be young, < 1 Ma, yet is not constrained by dating (Suppe 1980; Lee et al., 2001). Length: ~90 km Net slip: 10-15 km, based on offsets estimated on geological cross sections and seismic lines (Suppe, 1980; Lee et al., 2001; Yue et al., 2005). Long-term slip rate: from 16 mm/yr in the northern fault part to 10 mm/yr in the southern fault part (Simoes et al., 2014). Current slip rate: from 2 to 10 mm/yr estimated near the central part of the fault based on GPS measurements before the 1999 Chichi earthquake (Loevenbruck et al., 2001; Yue et al., 2005 and references therein). - Mapping of fault trace: Several precise fault maps have been done, which are consistent (e.g., Chen et al., 2001; Dominguez et al., 2003; Yue et al., 2005 and references therein). We chose to represent the fault map from Dominguez et al. (2003) because these authors mapped the fault by using both existing data and maps and new tectonic observations (which for instance revealed the splay faults suggested in dashed lines in Fig. S1). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Chelungpu fault is intermediate. - Direction of long-term propagation: Q2 Simoes et al. (2014) have shown that the long-term slip rate of the Chelungpu fault is largest in the

7

northern part of the fault (~16 mm/yr) and progressively decreases towards the south down to ~10 mm/yr. Simoes and Avouac (2006) have shown that the subsidence of the foreland has propagated southward. At its northern tip, the Chelungpu fault terminates abruptly, either bending into or connecting to a sub-perpendicular fault. At its southern tip, the Chelungpu fault splays into small secondary branches (Supplementary Figure S1) which form a fan-shape network enlarging in the South direction. Altogether these evidences suggest that the Chelungpu fault has propagated from north to south during its lifetime.

2002 Mw ~7.9 DENALI EARTHQUAKE Earthquake: 03/Nov/2002, Mw 7.9, Alaska, Right lateral strike slip Causative fault: Denali fault (+ secondary Totschunda and Susitna glacier faults) Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured, both on the field and from remote sensing analysis, and these measurements are consistent (Eberhart- Philipps et al., 2003; Crone et al., 2004; Haeussler et al., 2004; Taylor et al., 2008; Haeussler, 2009; Schwartz et al., 2012). The rupture trace and length are thus robustly constrained (length ~300 km). Q1. - Rupture extent within causative fault: The 2002 surface rupture is clear on high-resolution satellite images (e.g., Google Earth) and available Lidar data (opentopography.org), except at a few locations obscured with glaciers. Several precise rupture maps are also available, which are consistent (e.g., Haeussler et al., 2004; Haeussler, 2009). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field in several studies, and the measurements are consistent. They reveal a maximum lateral coseismic slip at surface of 8.8 ± 0.5 m on the eastern termination the Denali fault sensus stricto, and about 2 m of slip on the Totschunda fault (Eberhart-Philipps et al., 2003; Haeussler et al., 2004). A large vertical slip was also measured on the Susitna Glacier reverse fault, of at most ~5.4 m (Crone et al., 2004). The overall Denali surface slip profile (including the related Totschunda and Susitna Glacier rupture zones) is thus robustly defined. We chose to represent in Supplementary Figure S1 the slip profile determined by Haeussler et al. (2004) because it results from continuous measurements along the entire rupture length. Q1 - Earthquake slip profile at depth: Two source inversion models were published on the 2002 Denali earthquake, whose grid data are available (Oglesby et al., 2004; Asano et al., 2005). The two models are based on near-field seismological and geodetic data, and hence are likely well constrained. They produce a similar slip profile showing slip decreasing from a maximum value (~10 m) in the easternmost part of the Denali rupture (i.e., on Denali fault sensus stricto) down to the western termination of the entire rupture. The largest slip is co-located on both profiles. The two models also find most of slip distributed in the depth range 0-30 km. Because the model from Oglesby et al. (2004) is based on more data including surface slip, it is the one we most consider in Supplementary Figure S2. The zone of largest displacements at depth is co-located with that observed at surface (compare Supplementary Figure S1 and S2). This further supports that the models are robust. - Hypocenter: The hypocenter is fairly well constrained due to permanent and temporary seismological networks installed by the Alaska Earthquake Information Center in response to the Nenana Mountain foreshock that occurred ten days before and ~20 km west of the Denali mainshock. The earthquake initiated in the western part of the rupture zone, on the reverse Susitna Glacier fault, likely at ~4 km depth (e.g., Ratchkovski et al., 2003) or at the intersection between the Susitna Glacier and the Denali faults. The rupture then propagated unilaterally to the east onto the Denali fault.

8

- Rupture speed: The rupture speed was estimated along the rupture length, varying from ~3.3 km/s in the western part of the rupture to 5.0-5.5 km/s (supershear velocity) in the eastern part of the Denali rupture zone (Frankel, 2004; Oglesby et al., 2004; Dunham and Archuleta, 2004; Ellsworth et al., 2004; Asano et al., 2005; Walker and Shearer, 2009).

Causative fault parameters - General parameters Slip mode and strike: right-lateral strike slip, arcuate fault strike Initiation age: The Denali fault is considered to have localized on an inherited Late Paleozoic continental suture. It is taken to have initiated about 55-70 Ma ago (e.g., Nokleberg et al., 1985 and references therein, Plafker and Berg, 1994; Marechal, 2015 and references therein). The initiation age is consistent among different authors, thus is fairly well constrained. Length: ~2300 km Net slip: 300-400 km of maximum lateral slip located in the central part of the Denali fault (see synthesis in Lowey, 1998). Long-term slip rate: from 12-13 mm/yr in central fault section (Matmon et al., 2006; Mériaux et al., 2009) to ~7 mm/yr in the western half of the Denali fault (Matmon et al., 2006; Mériaux et al., 2009) and < 8 mm/yr in the eastern half of the Denali fault (Richter and Matson, 1971; Kalbas et al., 2008 and references therein). The various estimates are consistent and thus the available long- term slip rates are well constrained. Current slip rate: 6-10 mm/yr, measured at several sites along the fault (Biggs et al., 2007; Fletcher 2002). - Mapping of fault trace: Precise maps of the Denali Fault are available, that are consistent overall (e.g., Plafker and Berg, 1994 and references therein). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Denali fault is mature. - Direction of long-term propagation: Q1 Several studies have shown that the cumulative slip, the Quaternary slip rate and the current slip rates all decrease from the fault center (between about -145°E and -143°E) towards both the eastern and the western tips of the Denali fault (Richter and Matson, 1971; Kalbas et al., 2008 and references therein; Mériaux et al., 2009; Matmon et al., 2006; Lowey 1998; Plafker and Berg, 1994). Moreover, both at its western and its eastern tips, the Denali fault splays into multiple oblique branches (Chatham Strait in the east, EKF and Togiak-Tikchik in the west, Supplementary Figure S1), which form a fan-shape network enlarging towards the west at the western fault tip and towards the east at the eastern fault tip. Altogether these evidence show that the Denali fault has propagated bilaterally both eastward and westward from its center over its lifetime.

1954 Mw ~6.7 DIXIE VALLEY EARTHQUAKE Earthquake: 16/Dec/1954, Mw 6.7, Nevada, Normal fault Causative fault: Dixie Valley fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field, soon after the earthquake, and the measurements are consistent (e.g., Slemmons, 1957; dePolo et al., 1991; Caskey et al., 1996; Zhang et al., 1999). The rupture trace and length are thus well constrained overall (length ~46 km). An ambiguity remains however in the mapping of the off-fault rupture traces south of the main rupture, due to the Fairview Peak earthquake having occurred 4 minutes before the Dixie Valley earthquake. Q1. - Rupture extent within causative fault: The 1954 surface rupture is still clear on high-resolution

9

satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Caskey et al., 1996; U.S. Geological Survey and Nevada Bureau of Mines and Geology, 2006). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been measured on the field by Caskey et al. (1996), yet some time after the earthquake; however they took into account the dozen of measurements and the fault trace mapping that had been done by Slemmons (1957) right after the earthquake. Altogether the measurements reveal a maximum vertical slip of ~2.5 m in the central part of the rupture. Triangulations and leveling data recorded a similar maximum vertical slip of ~2.1 m (U.S. Coast and Geodetic Survey in Slemmons, 1957). Assuming that the Dixie Valley normal fault dips eastward by ~50° (Caskey et al., 1996), we infer a maximum on-fault slip of ~3.3 m. In Supplementary Figure S1, we represent both the original vertical slip profile determined by Caskey et al. (1996), and the net slip profile calculated using the ~50° fault dip provided by Caskey et al. (1996). Q1 - Hypocenter: The hypocenter location is poorly constrained because of the old age of the earthquake but a fairly robust fact is that the earthquake initiated in the central part of the rupture, likely at ~12 km depth (Doser 1986).

Causative fault parameters - General parameters Slip mode and strike: normal, NNE to NE-trending fault Initiation age: unknown Length: ~55 km Net slip: ~5 km (vertical component) estimated from refraction imaging, gravity and magnetic measurements (Thompson et al., 1967; Smith et al., 1968; Thompson and Burke, 1973). It is thus fairly well constrained. Long-term slip rate: ~0.5 mm/yr (vertical component over the last 12ka), based on paleoseismological trenches in the central part of the fault (Bell and Katzer, 1990; Bell et al., 2004) and on measurements of offset shorelines and terraces in the northern part of the fault (Thompson and Burke, 1973). Current slip rate: unknown - Mapping of fault trace: Several fault maps have been published, which are consistent (e.g., Bell and Katzer, 1990; compilation in U.S. Geological Survey and Nevada Bureau of Mines and Geology, 2006). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Dixie Valley fault is immature. - Direction of long-term propagation: Q1 Wallace and Whitney (1984) have shown that the cumulative slip on the Dixie Valley fault decreases progressively from the fault center towards both its northern and its southern tips. At its southern tip, the Dixie Valley fault splays into multiple small branches (Supplementary Figure S1) and interacts with the Fairview Peak fault. Splay faults are not described at the northern fault tip. The net slip variation along the fault strike suggests that the Dixie Valley fault initiated near its present central part and propagated bilaterally towards the south and north. The southern tip splay network confirms the southward propagation.

1980 Mw ~7.1 EL ASNAM EARTHQUAKE Earthquake: 10/Oct/1980, Mw 7.1, Algeria, Reverse fault Causative fault: Oued Fossa fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the 10

displacements measured directly on the field soon after the earthquake (Yielding et al., 1981; Philip and Meghraoui, 1983) and from using geodetic data (Ruegg et al., 1982). The measurements are consistent. The rupture trace and length are thus robustly constrained (length ~25 km). Q1. - Rupture extent within causative fault: The 1980 surface rupture is still fairly clear on high- resolution satellite images (e.g., Google Earth), while several consistent rupture maps are available (Yielding et al., 1981; Philip and Meghraoui, 1983). The rupture zone is thus well located within the causative fault. The rupture broke entirely the Oued Fossa fault and also possibly a few km-long part of the Cheliff fault. Q1. - Surface earthquake slip profile: The surface vertical slip profile has been densely measured on the field, yet only the vertical component of slip could be measured (Yielding et al., 1981; Philip and Meghraoui, 1983). The measurements reveal a maximum vertical coseismic slip of ~5 m, confirmed by geodetic measurements (~5.2 m; Ruegg et al., 1982). Q2 - Hypocenter: A robust fact is that the earthquake initiated at the southern tip of the Oued Fossa fault, at ~10 km depth, and then propagated unilaterally to the north (Deschamps et al., 1982).

Causative fault parameters - General parameters Slip mode and strike: reverse (with minor left lateral strike slip component), NE-trending fault Initiation age: 2 to 5 Ma, supposedly as a result of regional N-S compression (Guiraud, 1977; Philip and Thomas, 1977). Thus the fault initiation age is not well constrained. Length: ~25 km Net slip: unknown Long-term slip rate: unknown Current slip rate: unknown - Mapping of fault trace: The Oued Fossa fault has been mapped in several studies and these mapping are consistent (Yielding et al., 1981, 1989; Philip and Meghraoui, 1983). We use the fault map from Yielding et al. (1989) because it represents the entire Oued Fossa fault and its connection with the Cheliff fault. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Oued Fossa fault is immature. - Direction of long-term propagation: Q2 Avouac et al. (1992) and Boudiaf et al. (1998) showed that the Oued Fossa fault long-term growth was accompanied with the development of so-called “growing folds”. These growing faults become younger towards the southwest, in keeping with the Oued Fossa fault having propagated southwestwards over its lifetime. The Oued Fossa fault is actually a major tip splay fault of the master Cheliff fault. Along with other splay faults developed at the western tip of the Cheliff fault, the Oued Fossa fault form a fan-shape tip splay network enlarging in the southwest direction. This architecture is indicative of the western long-term propagation of the Cheliff fault and thus also, likely, of the Oued Fossa fault.

2010 Mw ~7.2 EL MAYOR CUCAPAH EARTHQUAKE Earthquake: 04/Apr/2010, Mw 7.2, Baja California (Mexico), Right lateral strike slip (with normal additional component) Causative fault: Elsinore fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped in the northern half of the rupture, and the displacements measured on the field (Fletcher, 2010; Teran et al., 2011, 2015; Fletcher et al., 2014), and from LiDAR imagery (Oskin et al., 2012; Gold et al.,

11

2013). The measurements are consistent overall. The southern half of the rupture was more difficult to map on the field due to its location in an anthropized region. However, Wei et al. (2011) fairly well constrained the entire rupture trace using InSAR and SPOT images (rupture length ~120 km). Q2. - Rupture extent within causative fault: The northern half of the 2010 surface rupture is very clear on high-resolution satellite images (e.g., Google Earth) and available Lidar data (opentopography.org), whereas several precise rupture maps are available (e.g., Fletcher et al., 2014). The rupture zone is thus fairly well located within the causative fault (good location in the north, more uncertain in the south, for reasons discussed above). Q2. - Surface earthquake slip profile: The surface coseismic slips were densely measured on the field (Fletcher et al., 2014) and from Lidar data (Gold et al., 2013), yet only in the northern part of the rupture. The complete surface slip profile could only be determined by using SPOT/InSAR images (Wei et al., 2011). The different studies are consistent overall, and especially find the largest coseismic slip in the northern part of the rupture (3.1-3.5 m of lateral slip and ~2 m of vertical slip, see Wei et al., 2011). Since the normal vertical component is found along most of the rupture length and not only locally at step-overs, it is a representative slip component. Therefore the composite net slip should be considered. This net slip reaches a maximum of 3-4 m (Fletcher 2010; Teran et al., 2011; Fletcher et al., 2014). We chose to represent in Supplementary Figure S1 the surface slip profile determined by Wei et al. (2011) because it presents continuous measurements along the entire rupture length. However, the profile does not take into account the vertical component of the rupture. This is why we also represent, for comparison, the surface slip profile derived from the source inversion model of Wei et al. (2011). Q2 - Earthquake slip profile at depth: As far as we know, only one source inversion model published on the 2010 El Mayor Cucapah earthquake has its grid data available (Wei et al., 2011). The model is based on numerous different and complementary data (SPOT images, GPS, InSAR and teleseismic). The model is well constrained in the northern part of the rupture where accurate data are available (GPS and SPOT images in US), but is more poorly constrained in the southern half of the rupture where data are few or lacking. Surprisingly, although measured surface slips are used to constrain the model, the displacements eventually modeled near the surface are lower than those actually measured on the field (see Supplementary Figure S1). At depth, the model produces a slip profile (as before, maximum slip in each column of the model grid) showing coseismic slip decreasing from a maximum value (6-7 m) in the northwest part of the rupture down to ~zero at the southeastern termination of the rupture. An isolated high slip patch is inferred near the southeastern end of the rupture (see Supplementary Figure S2) but this is where InSAR signal is noisy and the model resolution very low; therefore this slip patch is likely an artifact. If so, the maximum slip occurred in the northern part of the rupture, in agreement with observations at surface (compare Supplementary Figures S1 and S2). According to the model, most of the slip occurred between 0 and 15-18 km depth. - Hypocenter: The hypocenter location is well constrained (see precise earthquake relocations from Hauksson et al., 2012). The earthquake initiated in the center of the rupture on a small oblique normal fault at ~16 km depth, and then propagated bilaterally on the Elsinore fault. - Rupture speed: Uchide et al. (2013) described the “expansion speed” of their modeled rupture. We use these modeled values as a proxy of the rupture speed; the derived speed varies from ~1 km/s in the southeastern part of the rupture to 3 km/s in its northwestern part.

Causative fault parameters - General parameters Slip mode and strike: right-lateral strike slip, N130°-trending fault Initiation age: The Elsinore fault is taken by some authors to have initiated 1-2.5 Ma ago (e.g., Dorsey et al., 2012; Hull and Nicholson, 1992). However, the Elsinore fault developed as a major splay fault of the Southern San Andreas fault (Scholz, 1977, 2010; Crowel, 1979). Since the Southern San Andreas fault initiated ~12 Ma ago (Powell and Weldon, 1992 and references therein;

12

Sims, 1993), at least the northern part of the Elsinore fault must have formed several Ma ago, possibly up to the 12 Ma age of the Southern San Andreas fault, and more likely ~5 Ma ago (Powell and Weldon, 1992, Hull and Nicholson, 1992; Lutz et al., 2006; Kirby et al., 2007; Dorsey et al., 2012). Length: ~400 km Net slip: 10-15 km measured in the northwestern part of the fault, consistently among different studies (Weber, 1977; Morton and Miller, 1987; Dorsey et al., 2012). It is thus fairly well constrained. Long-term slip rate: 3-6 mm/yr over the Holocene-Pliocene in the northern part of the fault (Millman and Rockwell, 1986; Vaughan and Rockwell, 1986; Hull and Nicholson, 1992; Rockwell et al., 2000a), to 1-2 mm/yr over the Pleistocene in its southern part (Fletcher et al., 2011; Dorsey et al., 2012). Current slip rate: It is considered to be low (Fialko, 2006). Meade and Hager (2005) (GPS) and Lundgren et al. (2009) (InSAR) estimate a maximum lateral slip rate of 2 ± 3 and 3 ± 1 mm/yr, respectively. The large uncertainties show that the current slip rate is not well constrained. - Mapping of fault trace: Several precise fault maps have been published (e.g., U.S. Geological Survey and California Geological Survey, 2006; Dorsey et al., 2012). We chose to represent the fault map from Dorsey et al. (2012) because it shows the entire fault trace up to the California Gulf in Mexico. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Elsinore fault is intermediate. - Direction of long-term propagation: Q1 The Elsinore fault being a major splay of the Southern San Andreas fault, its initiation age in its northernmost part which splays from the San Andreas fault is similar or younger to that of the Southern San Andreas fault, i.e., 12 to 5 Ma according to the references cited above. At its southern end, the Elsinore fault is interacting with the Gulf of California faults. Since the Gulf of California opened ~5 Ma ago (e.g., Spence and Normark, 1979; Lonsdale, 1989), the southern part of the Elsinore fault is likely younger than 5 Ma. As a matter of fact, Dorsey et al. (2012) have shown that the southernmost segments of the Elsinore faults are extremely juvenile and offset recent stratigraphic levels. Therefore, the Elsinore fault age decreases from north to south. Meanwhile, several studies have shown that both the cumulative slip and the long-term slip rate decrease from north to south along the fault (Weber, 1977; Morton and Miller, 1987; Hull and Nicholson, 1992; Dorsey et al., 2012 and references therein). At its northwestern tip, the Elsinore fault abuts against the San Gabriel Mountains thrust, but the San Gabriel fault continues the Elsinore fault up to its connection with San Andreas. At its southeastern tip, the Elsinore fault splays into multiple oblique branches that interact with the Gulf of California faults (Supplementary Figure S1). The splay faults form a fan-shape network enlarging in the southeast direction. Altogether these evidences show that the Elsinore fault has propagated from northwest to southeast over its lifetime.

1939 Mw ~7.8 ERZINCAN EARTHQUAKE Earthquake: 26/Dec/1939, Mw 7.8, Turkey, Right lateral strike slip Causative fault: North Anatolian fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field soon after the earthquake (Pamir and Ketin, 1941; Parejas et al., 1942; compiled in Barka, 1996). The rupture trace and length are thus robustly constrained (length ~335 km). Q1. - Rupture extent within causative fault: The 1939 surface rupture is still visible on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (Barka, 1996;

13

Emre et al., 2013). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile was measured on the field soon after the earthquake (Pamir and Ketin, 1941; Parejas et al., 1942; compiled in Barka, 1996). Yet the measurement density is not very high. Therefore, while the overall features of the surface slip profile are well constrained, details of the slip distribution are not recovered in the available data. The maximum reported lateral coseismic slip at surface is ~7.5 m near the eastern tip of the rupture. Q2 - Hypocenter: The hypocenter location is not well constrained, but a robust fact is that the earthquake initiated at the eastern tip of the rupture, and then propagated unilaterally to the west (McKenzie, 1972).

Causative fault parameters - General parameters Slip mode and strike: right-lateral strike slip, E-W-trending fault Initiation age: The North Anatolian fault system initiated 11-13 Ma ago, based on precise dating of offset features (e.g., Armijo et al., 1996, 1999; Schindler, 1998; Hubert Ferrari et al., 2002; Sengör et al., 2005), possibly localizing on a paleosuture (e.g., Sengör et al., 2005). Length: ~1400 km Net slip: ~85 km, determined from a few offset features (e.g., Armijo et al., 1999; Hubert-Ferrari et al., 2002) but it is likely underestimated (see pattern of net slip profile in Bohnhoff et al., 2016). Long-term slip rate: ~18 mm/yr over the Quaternary measured in the central part of the fault (e.g., Hubert-Ferrari et al., 2002). Current slip rate: 20-22 mm/yr (e.g., McClusky et al., 2000; Hubert Ferrari et al., 2002) measured near the eastern end of the fault. - Mapping of fault trace: Several precise fault maps have been published, that are consistent (e.g., Barka, 1992; Armijo et al., 2002; see a synthesis in Sengör et al., 2005). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the North Anatolian fault is mature. - Direction of long-term propagation: Q1 Many studies have shown that the fault offsets become younger from the eastern part of the fault (~13 Ma) to its western end (the fault entered in the Aegean Sea ~1 Ma ago) (e.g., Sengör, 1979; Armijo et al., 1996, 1999; Schindler, 1998; Hubbert-Ferrari et al., 2002, 2003; Sengör et al., 2005; synthesis in Bohnhoff et al., 2016). The net slip also decreases from east (~85 km) to west (4-70 km, with a preferred value of ~50 km) (Barka, 1992; Hubbert-Ferrari et al., 2002, 2003; Sengör et al., 2005; Akbayram et al., 2016 ; Bohnhoff et al., 2016). At its eastern tip, the North Anatolian fault terminates and abuts against the East Anatolian fault. At its western tip, the North Anatolian fault splays into multiple oblique branches (the main one are the Northern and the Southern strands; Supplementary Figure S1; e.g., Sengör et al., 2005), which form a fan-shape network enlarging towards the west. Altogether these evidence show that the North Anatolian fault has propagated unilaterally from east to west over its lifetime.

1954 Mw ~7.2 FAIRVIEW PEAK EARTHQUAKE Earthquake: 16/Dec/1954, Mw 7.2, Nevada, Normal fault (with an additional right-lateral component of slip) Causative fault: Fairview Peak fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped on the field and from aerial photo analysis, and the displacements were measured on the field, all soon after the

14

earthquake (e.g., Slemmons, 1957; dePolo et al., 1991; Caskey et al., 1996; Zhang et al., 1999). The rupture trace is thus well constrained overall. However, the broad distribution of the rupture traces in the north and the occurrence of the Dixie Valley earthquake 4 minutes later in the northern prolongation of the Fairview Peak rupture, make the total rupture length difficult to measure precisely (~46 km, see Supplementary Figure S1). Note that in Fig. S1, we only represent the rupture trace on the Fairview Peak fault, that is, we ignore the small slip that was triggered on another fault (Philipp Walsh fault, small slip described over less than 5 km long, Caskey et al., 1996), located 10 km further south. Q2 - Rupture extent within causative fault: The 1954 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., dePolo et al., 1991; Caskey et al., 1996; Zhang et al., 1999). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been measured on the field both soon after the earthquake and later in time (e.g., Slemmons, 1957; Caskey et al., 1996; Zhang et al., 1999). The measurements are consistent, and reveal maximum vertical and lateral coseismic slips at surface of ~3.8 and ~2.5 m, respectively. Since both slip components characterize the actual fault motion, the net slip must be calculated. Assuming that the Fairview Peak fault dips eastwards by 50-70° (Caskey et al., 1996; see details in Supplementary Figure S1), we infer a maximum combined on-fault slip of ~5.2 m, located in the southern part of the rupture. Q2 - Hypocenter: The hypocenter location is poorly constrained but the earthquake seems to have initiated in the central part of the rupture, at ~15 km depth (Doser, 1986).

Causative fault parameters - General parameters Slip mode and strike: normal and right lateral strike slip, NNE-trending fault Initiation age: unknown Length: ~46 km Net slip: unknown. Long-term slip rate: 0.1-0.2 mm/yr based on the measurement of a single ~8 m cumulative vertical slip whose age is estimated of 35-100 ka (Bell et al., 2004). Thus poorly constrained. Current slip rate: ~2 mm/yr of extension rate, poorly estimated from only two GPS sites either side of the northern Fairview Peak Fault (Thatcher et al., 1999; Bennett et al., 2003). - Mapping of fault trace: Several precise fault maps have been published, that are consistent (e.g., Caskey et al., 1996; U.S. Geological Survey and Nevada Bureau of Mines and Geology, 2006). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Fairview Peak fault is immature. - Direction of long-term propagation: Q3 Fault age and net slip data are presently insufficient to constrain the direction of long-term propagation of the Fairview Peak fault. At its southern tip, the Fairview Peak fault terminates where it meets the Walker Lane shear zone (e.g., Stewart, 1980; Supplementary Figure S1). At its northern tip, a tip splay network is observed, whose architecture suggests that the Fairview Peak fault has been propagating northward.

1857 Mw ~7.9 FORT TEJON EARTHQUAKE Earthquake: 09/Jan/1857, Mw 7.9, California, right-lateral strike-slip Causative fault: San Andreas fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped and the

15

displacements measured, essentially late after the earthquake, from measurements of lateral offsets across recent geomorphic features, performed on the field and from Lidar data analysis (Sieh, 1978; Grant Ludwig et al., 2010; Zielke et al., 2010, 2012). The measurements are dense and precise, whereas they are basically consistent (see more details below). The rupture trace and length are thus robustly constrained (~330 km). Q1 - Rupture extent within causative fault: The 1857 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Sieh, 1978; compilation in U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been determined long after the earthquake, from measurements of lateral offsets across recent geomorphic features, performed on the field and from Lidar data analysis (Sieh, 1978; Grant Ludwig et al., 2010; Zielke et al., 2010, 2012). The slip measurements are dense and precise. The field- and Lidar data-based slip profiles are basically consistent on the overall features of the slip distribution (both show largest surface slips in the northern part of the rupture), but differ in largest slip amplitude (~9 m from Sieh, 1978 (field measurements); ~6 m from Zielke et al., 2012 (Lidar measurements)). We represent both profiles in Supplementary Figure S1, but we use the slip profile from Sieh (1978) in our analysis because the approach to measure slip in Lidar data results in smoothing out the slip distribution (Zielke et al., 2012). Q2 - Hypocenter: the hypocenter location is not well constrained but a robust fact is that the earthquake initiated near the northern tip of the rupture, and then mainly propagated towards the south (Sieh et al., 1978).

Causative fault parameters - General parameters Slip mode and strike: right lateral strike slip, NW-SE-trending fault Initiation age: Dextral slip initiated 24-29 Ma ago in the northern part of the San Andreas fault (e.g., Atwater and Stock, 1998; Liu et al., 2010) whereas it initiated ~12 Ma ago in the southern part of the San Andreas fault (e.g., Powell and Weldon, 1992; Sims, 1993). The San Jacinto and Elsinore splay faults formed later, ~5 Ma ago (e.g., Powell and Weldon, 1992 and references therein; Sims, 1993; Nicholson et al., 1994; Dorsey et al., 2012). They are less than 2 Ma old in their southernmost part (e.g., Hull and Nicholson, 1992; Lutz et al., 2006; Kirby et al., 2007; Dorsey et al., 2012). Length: ~1300 km Net slip: maximum cumulative lateral slip of ~600 km in the northern half part of the fault (north of the creeping section), and < 300 km in its southern half part (e.g., Hill and Dibblee, 1953; Wallace; 1970; Suppe et al., 1970; Crowell, 1979; Sims, 1993; Revenaugh and Reasoner 1997; Darin and Dorsey, 2013) Long-term slip rate: 20-35 mm/yr over the Holocene (Sieh and Jahns, 1984; Weldon and Sieh, 1985; Perkins et al., 1989; Sims 1993; van der Woerd et al., 2006; Titus et al., 2011), measured along the creeping and the southern section of the San Andreas Fault. Current slip rate: ~25 mm/yr, but varying along the fault from 4 to 25 mm/yr (Segal and Harris, 1986; Meade and Hager, 2005; Titus et al., 2011; Lindsey and Fialko, 2013; Tong et al., 2013). - Mapping of fault trace: Many fault maps have been published, that are consistent (e.g., Schwartz and Coppersmith 1984; compilation of U.S. Geological Survey and California Geological Survey, 2006). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the San Andreas fault is mature. - Direction of decreasing maturity or long-term propagation: Q1

16

Different scenarios have been proposed to recover where the San Andreas fault (SAF) initiated originally and how it subsequently grew, but none of these scenario is entirely validated and the growth history of the San Andreas fault is thus still debated (e.g., Atwater, 1970, 1989; Powell and Weldon, 1992; Sims, 1993; Nicholson et al., 1994; Dickinson et al., 2005; Sharman et al., 2013). The most broadly accepted scenario is that of Atwater and Stock (1998; provided in the so-called “California-20 movie” at http://emvc.geol.ucsb.edu; note that similar reconstructions are provided by Nicholson et al., 1994). It shows the following points: First, the SAF formed at the expense of a pre-existing subduction fault. In its early stages, which occurred 20-29 Ma ago in the northern part of the present SAF, the “proto-SAF” did not propagate laterally. Rather, the transition from subduction to strike-slip occurred through the northward lateral migration of a sliver of the Pacific plate along the existing fault line, driven by the northward migration of the Mendocino Triple Junction. The northward migration of the Pacific sliver made the northernmost section of the present Northern SAF slipping laterally continuously over the entire duration of the migration, i.e., at least 20 Ma. Second, the present SAF is a composite of multiple original faults and has progressively migrated to fault branches towards the East. A western branch first formed ~20 Ma ago and an eastern branch formed ~16 Ma ago. While slip was likely distributed on the two branches till ~4 Ma, it later concentrated mainly on the eastern branch. This eastern branch and the northernmost section of the western branch constitute the present trace of the Northern-Central SAF. Once the Northern-Central SAF was well developed in the continental crust, the southern part of the Southern San Andreas fault formed, likely around 4-5 Ma, and developed by propagating southwards (see below). Therefore, the growth history of the San Andreas fault is complex. However, the reconstructions suggest that the northernmost section of the present Northern San Andreas fault is the section of the entire fault zone that has been slipping over the longest time, 20 Myrs and possibly more. All other parts of the San Andreas fault zone have either stopped or slowed down slipping long before present (ex: San Gregorio fault) or have formed later (ex: the eastern branch that forms the rest of the Northern-Central San Andreas fault, the Southern San Andreas fault, the Hayward-Calaveras faults). Therefore, in terms of total slip duration, the northernmost section of the present Northern San Andreas fault is the longest-lived section of the entire San Andreas fault. It is thus the most mature section of the San Andreas fault, and more specifically, of the Northern San Andreas fault. Several complementary evidence corroborate the above inferences: The cumulative slip (Hill and Dibblee, 1953; Wallace; 1970; Suppe et al., 1970; Crowell, 1979; Sims, 1993; Revenaugh and Reasoner 1997; Darin and Dorsey, 2013), the long-term slip rate (van der Woerd et al., 2006), and the degree of connection of the fault segments (Aviles et al., 1987), all decrease from north to south along the San Andreas fault. Bilham and Williams (1985) dated one of the southernmost fault segments (Durmid segment) and found it very young, with an age of ~30 ka. At its southern tip, the San Andreas fault splays into multiple fault branches which include the San Jacinto and the Elsinore faults (formed ~5 Ma ago, e.g., Powell and Weldon, 1992; Sims, 1993; Nicholson et al., 1994; Dorsey et al., 2012), but also many other smaller faults (Aviles et al., 1987; Ando et al., 2009; Supplementary Figure S1). These splaying faults form a fan-shape network enlarging towards the south. Altogether these evidences support that the Northern San Andreas fault is older and hence more mature than the Southern San Andreas fault, and that the Southern San Andreas fault has propagated southward over its lifetime.

1931 Mw ~7.9 FUYUN EARTHQUAKE Earthquake: 19/Aug/1931, Mw 7.9, China, right-lateral strike-slip Causative fault: Fuyun fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured both on the field (Jianbang et al., 1984; Lin and Lin, 1998), and on high- 17

resolution Quickbird satellite images (Etchebes, 2011; Klinger et al., 2011), yet long time after the earthquake. The measurements are consistent however, suggesting that the rupture trace and length are well constrained (~165 km). Q2 - Rupture extent within causative fault: The 1931 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Lin and Lin, 1998; Klinger et al., 2011). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been determined long after the earthquake from measurements of preserved recent offset features, both on the field (Jianbang et al., 1984; Lin and Lin, 1998) and from satellite images. The measurements are consistent overall, and all show that largest surface slip occurred in the northern part of the rupture. The measurements only differ on the maximum lateral slip amplitude: ~14 m from field measurements (Jianbang et al., 1984; Lin and Lin 1998; plus ~3 m of vertical slip), and ~9 m from satellite image measurements (Etchebes, 2011; Klinger et al., 2011). We use the slip profile from Lin and Lin (1998) because these authors performed the most complete study where they describe both the lateral and the vertical displacements along the entire rupture length. Vertical displacements were only measured in the northern part of the rupture, and this is why we do not represent them in Supplementary Figure S1, so as to better compare the available lateral slip profiles. Q2 - Hypocenter: The location of the hypocenter is not constrained but one robust fact is that the earthquake initiated in the northern part of the rupture, close to the location of the eventual maximum coseismic displacement (Chen and Molnar, 1977). The rupture then propagated bilaterally toward both the north and south.

Causative fault parameters - General parameters Slip mode and strike: right-lateral strike slip, NNW-trending fault Initiation age: unknown but the transpressive deformation in the Gobi-Altai region started between 2 and 8 Ma (e.g., Vassalo et al., 2007) as a result of India-Eurasia convergence (e.g., Tapponnier and Molnar, 1977). Length: ~450 km Net slip: maximum cumulative lateral slip estimated of ~20 km, based on the lateral offset of Devonian formations in the northern part of the fault. Long-term slip rate: unknown. Current slip rate: unknown. However, Calais et al. (2003; GPS data) estimate a mean lateral slip rate of ~5 mm/yr across the whole Altay domain (with respect to Eurasia), what suggests that the lateral slip rate of the Fuyun Fault is low, less than a few mm/yr. - Mapping of fault trace: The entire fault length has been precisely mapped in a prior study from Tapponnier and Molnar, (1979). Q2 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Fuyun fault is intermediate. - Direction of long-term propagation: Q3 The available fault age and net slip data are insufficient to constrain the direction of long-term propagation of the Fuyun fault. However the main fault trace is flanked by well-developed secondary splay faults which form a fan shape enlarging towards the south. This architecture suggests that the Fuyun fault has propagated southward over the long-term (Perrin et al., 2016). However, the existence of the perpendicular major Barkol Tagh range fault at the present southern Fuyun fault tip makes it unlikely that the Fuyun fault can presently continue propagating further southward. The fault might thus be instead in the process of resuming propagating northwards, as suggested by a network of dense small splay faults developed at the northern Fuyun fault tip.

18

1959 Mw ~7.5 HEBGEN LAKE EARTHQUAKE Earthquake: 17/Aug/1959, Mw 7.5, Montana, normal fault Causative fault: Hebgen fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured, soon after the earthquake (Witkind et al., 1962; Myers and Hamilton, 1964; Witkind, 1964; Zhang et al., 1999), and the measurements are consistent. The rupture trace and length are thus robustly constrained (~25 km). Q1 - Rupture extent within causative fault: The 1959 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Myers and Hamilton, 1964; Witkind, 1964; Zhang et al., 1999; U.S. Geological Survey and Montana Bureau of Mines and Geology, 2006). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been measured on the field soon after the earthquake (Myers and Hamilton, 1964). The measured surface slip profile is fairly dense along the rupture and well constrained. It reveals a maximum vertical slip value of ~5.5 m. Q2 - Hypocenter: The hypocenter position is not well defined and much varies among the various studies (see Ryall, 1962). In all available analyses however, the hypocenter locates nearby the eastern end of the rupture, at about 12 km depth (Murphy and Brazee, 1964; Doser, 1985b). The rupture thus mainly propagated toward the west.

Causative fault parameters - General parameters Slip mode and strike: normal, N120°-trending fault Initiation age: 0.6-2 Ma ago (Scott et al., 1985; Christiansen, 1986) deduced from structural correlation between faults and age of “present” Yellowstone Caldera. Length: ~45 km Net slip: unknown Long-term slip rate: 0.8-2.5 mm/yr (Doser et al., 1985a; Anderson et al., 1996), but poorly constrained. Current slip rate: difficult to estimate, especially due to the proximity of the Yellowstone Caldera (magma-driven uplift and subsidence motions; e.g., Puskas et al., 2007). - Mapping of fault trace: Several precise fault maps are available (e.g., Myers and Hamilton 1964; compilation in U.S. Geological Survey and Montana Bureau of Mines and Geology, 2006) but the zone is complex due to the presence of volcanic products. Q2 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Hebgen fault is immature. - Direction of long-term propagation: Q3 Available fault age and net slip data are insufficient to constrain the direction of long-term propagation of the Hebgen fault. At its western tip, the Hebgen fault stops against an oblique mountain range. At its eastern tip, the fault splays into multiple branches (Supplementary Figure S1), which form a fan-shape network enlarging towards the east. This architecture suggests that the Hebgen fault has propagated eastward over its lifetime (Perrin et al., 2016).

19

1999 Mw ~7.1 HECTOR MINE EARTHQUAKE Earthquake: 16/Oct/1999, Mw 7.1, California, Right lateral strike slip Causative fault: Lavic Lake-Pisgah-Bullion-Mesquite Lake Fault zone Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured both on the field (e.g., USGS et al., 2000; Treiman et al., 2002) and from radar imagery (Sandwell 2000, 2002; Peltzer et al., 2001; Jónsson et al., 2002; Simons et al., 2002). InSAR data well agree with surface mapping and give additional insights on slip at the tips of the rupture (see dashed line on the slip profile in Supplementary Figure S1). The rupture trace and length are thus robustly constrained (length 50-55 km, Sandwell 2000, 2002; Peltzer et al., 2001; Jónsson et al., 2002; Simons et al., 2002). Q1. - Rupture extent within causative fault: The 1999 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured both on the field (maximum lateral slip of ~5.3 m, USGS et al., 2000; Treiman et al 2002), and from InSAR data and modeling (maximum lateral slip of 6-6.5 m, Peltzer et al., 2001; Jónsson et al., 2002; Sandwell 2002; Simons et al., 2002). The on-field measured and Insar-inferred surface slip profiles are similar and hence are robust. We chose to represent in Supplementary Figure S1 the slip profile determined by Jónsson et al., 2002 because it gives additional insights on slip at the tips of the rupture (dashed line on the slip profile in Supplementary Figure S1). Q1 - Earthquake slip profile at depth: We use four source inversion models that have been published on the 1999 Hector Mine earthquake and whose grid data are available. The models differ principally from the data they used (Ji et al., 2002: Teleseismic, strong motions, GPS and surface slip data; Jónsson et al., 2002: InSAR and GPS data; Kaverina et al., 2002: InSAR, GPS, local and regional seismicity and surface slip data; Salichon et al., 2004: InSAR, GPS, teleseismic and surface slip data). All four models thus use a large amount of various data and hence are likely robust. As a matter of fact, they all produce a similar slip profile showing slip decreasing fairly regularly from a maximum value (range 6-9 m) in the north of the rupture (same location of zone of largest slips on all 4 profiles) down to the southern termination of the rupture. All four models also find most of slip distributed in the depth range 0-15 km. The model of Kaverina et al. (2002) being the one using more data, both local and regional, it is the one we most consider in Supplementary Figure S2. The zone of largest displacements at depth is at a different location than that observed at surface (compare Supplementary Figure S1 and S2). The zone where the largest deep slip is modeled is where distributed splay faults, including the main Bullion Mountains Blind fault, were revealed by the aftershocks (Hauksson et al., 2002). Together these suggest that the modeled larger deep slip might result from having summed the entire distributed slip onto the fault. It so, the models are not accurate in the northern rupture part. - Hypocenter: The hypocenter location is well constrained, and shows that the earthquake initiated close to the central part of the rupture at ~7 km depth (see refine earthquake catalog from Hauksson et al., 2012), then propagated bilaterally toward both the north and south. - Rupture speed: The rupture speed was estimated along the rupture length, and shown to vary from 1.8 km/s in the northern part of the rupture to 2.2-2.8 km/s in its southern part (Ji et al., 2002; Kaverina et al., 2002).

Causative fault parameters - General parameters Slip mode and strike: right lateral strike slip, NW-trending fault Initiation age: < 10 Ma, recognized in multiple studies (Dibblee 1961; Garfunkel 1974; Dokka

20

1983; Dokka and Travis, 1990 and references therein). Length: ~90 km Net slip: 6-14 km on the northern Pisgah fault section (field: Dokka 1983; Dokka and Travis 1990; magnetic anomalies: Jachens et al., 2002) to 20 km on the southern Bullion-Mesquite fault section (Jachens et al., 2002). Even if some uncertainties remain on the Pisgah fault, the measurements are consistent between the different studies. Long-term slip rate: 0.8 mm/yr, measured on the northern Pisgah fault section (Hart et al., 1988; Oskin et al., 2007 and references therein), based on dated offset markers (Hart et al., 1988). Current slip rate: not defined. However, the total current slip rate across the southern part of the East California Shear Zone (ECSZ) is ~12 mm/yr (Sauber et al., 1994; McClusky et al., 2001). If one assumes that the ECSZ is composed of six major sub-parallel faults, including the Lavic Lake- Pisgah-Bullion-Mesquite Lake Fault zone (e.g., Jennings et al., 1994; Jachens et al., 2002), each might be currently slipping at ~2 mm/yr. - Mapping of fault trace: Several precise fault maps have been published by the USGS (Jennings et al., 1994; compilation in U.S. Geological Survey and California Geological Survey, 2006), which are consistent. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Lavic Lake-Bullion-Mesquite Fault zone is immature. - Direction of long-term propagation: Q1 Several studies have shown that the net slip on the Lavic Lake-Pisgah-Bullion-Mesquite Lake fault is largest in the southern part of the fault (Bullion-Mesquite Lake fault section; Jachens et al., 2002) and decreases towards the north (Pisgah fault section; Dibblee, 1961; Garfunkel, 1974; Dokka, 1983; Dokka and Travis, 1990). At its southern tip, the Lavic Lake-Pisgah-Bullion-Mesquite Lake fault abuts and terminates abruptly against the ~E-W trending Pinto Mountain fault. Near its northern tip, the Lavic Lake-Pisgah-Bullion-Mesquite Lake fault splays into multiple oblique branches (Pisgah, Lavic Lake, “Bullion Mountains Blind”, e.g., Jachens et al., 2002; Supplementary Figure S1) which form a fan-shape network enlarging towards the north. Altogether these evidence suggest that the Lavic Lake-Pisgah-Bullion-Mesquite Lake fault has propagated from south to north during its lifetime.

1940 Mw ~7.0 IMPERIAL VALLEY EARTHQUAKE Earthquake: 18/May/1940, Mw 7.0, USA-Mexico border, right-lateral strike-slip Causative fault: Imperial Valley fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped (Sharp 1982), and the displacements measured soon after the earthquake both on the field (e.g., from J.P Buwalda, 1940, unpublished but compiled in Sharp, 1982) and from aerial imagery (measurements not along entire rupture length; Rockwell and Klinger, 2013). The rupture trace and length are thus robustly constrained (length ~58 km). Q1. - Rupture extent within causative fault: The 1940 surface rupture cannot be easily discriminated, on Google Earth and Landsat 7 satellite images, from the cumulative fault trace, which is subtle (yet visible). However, several precise rupture maps are available (e.g., Sharp 1982; U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field after the earthquake (J.P Buwalda, 1940, unpublished but compiled in Sharp, 1982), whereas only measurements in the central part of the rupture could be performed using satellite imagery (Rockwell and Klinger, 2013). The measurements are consistent overall, and show a maximum lateral slip of 6 to 7 m (Sharp 1982; Rockwell and Klinger, 2013). We chose to represent in

21

Supplementary Figure S1 the slip profile determined by Sharp (1982) because it provides surface measurements along the entire rupture length. Q2 - Hypocenter: The hypocenter location is not well constrained but a robust fact is that the earthquake initiated in the northern part of the rupture, then propagated mainly towards the southeast (Richter, 1958; Trifunac and Burne, 1970; Trifunac, 1972).

Causative fault parameters - General parameters Slip mode and strike: right lateral strike slip, NW-trending fault Initiation age: possibly 2 Ma ago, at the same time as the southern part of the San Jacinto fault (e.g., Lutz et al., 2006; Kirby et al., 2007). But this is not well constrained. Length: ~60 km Net slip: unknown Long-term slip rate: 15-20 mm/yr (over the last 300-500 yr), based on offset of penultimate lake shoreline deposits in the central part of the fault (Thomas and Rockwell, 1996). Current slip rate: 5-9 mm/yr of shallow creep in the north (GPS; e.g., Lyons et al., 2002; Crowell et al., 2013; InSAR, Lindsey and Fialko, 2016) and 25-35 mm/yr of deep slip rate (estimated by far field geodetic data; Lindsey and Fialko, 2016 and references therein). However, the latter value might be over-estimated due to noisy InSAR data and presence of blind faults off the Imperial fault which may accommodate part of the slip (see discussion in Lindsey and Fialko, 2016). - Mapping of fault trace: Several precise fault maps have been published which are consistent (e.g., Sharp, 1982; compilation in U.S. Geological Survey and California Geological Survey, 2006). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Imperial fault is possibly intermediate. - Direction of long-term propagation: Q3 Larsen and Reilinger (1991) showed that the northernmost fault section is young, formed 3 ka ago. In the south, the fault seems to connect with the California Gulf faults that formed ~5 Ma ago (e.g., Dorsey et al., 2012). At its northwestern tip, the Imperial fault splays into two fault branches (Supplementary Figure S1), which form a fan-shape network enlarging towards the north. Altogether these evidence suggest that the Imperial fault has propagated northwestward during its lifetime.

1979 Mw ~6.5 IMPERIAL VALLEY EARTHQUAKE Earthquake: 15/Oct/1979, Mw 6.5, USA-Mexico border, right-lateral strike-slip Causative fault: Imperial Valley fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field soon after the earthquake (e.g., Sharp et al., 1982). The rupture trace and length seem well constrained, although only one study exists (~31 km). However the earthquake initiated south of the rupture which has been mapped (see Supplementary figure S1) and therefore, even though no surface ruptures were reported in this southern region, it is likely that the rupture length is a few km longer than reported from the surface traces. Q2 - Rupture extent within causative fault: The 1979 surface rupture cannot be discriminated, on Google Earth and Landsat 7 satellite images, from the cumulative fault trace, which is subtle (yet visible). As said above, there is an uncertainty on the rupture length. However, the available rupture maps (Sharp et al., 1982; U.S. Geological Survey and California Geological Survey, 2006) make the rupture zone fairly easy to locate within the causative fault. Q2. - Surface earthquake slip profile: The complete surface slip profile has been well measured on the

22

field, soon after the earthquake. It shows a maximum lateral slip of ~0.6 m, in the southern part of the rupture (Sharp et al., 1982). Q2 - Earthquake slip profile at depth: One source inversion model has been published for the 1979 Imperial Valley earthquake, whose grid data are available (Hartzell and Heaton, 1993). The model is based on a moderate number of strong motions and teleseismic data, and hence is moderately well constrained. It suggests a ~1.8 m maximum slip in the center of the rupture, at 8-10 km depth. The zone of largest slip modeled at depth is thus at a different location than that observed at surface (compare Supplementary Figure S1 and S2). This suggests that the model is not fully correct, and that, possibly, part of the large slip modeled at depth is the sum of more distributed slip near the surface. - Hypocenter: The hypocenter location was constrained from Global CMT and hence has large uncertainties (Hartzell and Heaton 1983). A robust fact however is that the earthquake initiated near the southern end the rupture, at 10-12 km depth. The rupture then propagated unilaterally towards the northwest. - Rupture speed: The rupture speed was estimated along the rupture length varying from 2.5-2.9 km/s in the southern part of the rupture to 2-8.8 km/s in its northern part (Archuleta, 1984; Spudich and Cranswick, 1984). These estimations are uncertain however, for they were constrained from limited data.

Causative fault parameters See a description of the Imperial fault in the “1940 Imperial Valley earthquake” section.

1999 Mw ~7.5 IZMIT EARTHQUAKE Earthquake: 17/Aug/1857, Mw 7.5, Turkey, right-lateral strike-slip Causative fault: Northern strand (called Northern Branch fault) of the North Anatolian fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been mapped on the field (e.g., Barka et al., 2002), and the displacements measured on the field (e.g., Barka et al., 2002) and from SPOT images (Michel and Avouac, 2002). In the west, part of the rupture might have extended offshore (Yalova segment, see discussion and references in Manighetti et al., 2005). In the east, the earthquake also broke a small secondary fault (dotted line in Supplementary Figure S1) off the main Northern Branch fault ruptured by the Izmit earthquake. Thus the total rupture length is not completely well defined, and it furthermore depends on whether the small off-fault rupture to the east is included or not. Q2 - Rupture extent within causative fault: The 1999 surface rupture is in part clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Barka et al., 2002). Even if there is an uncertainty in the rupture length, the rupture zone is thus fairly well located within the causative fault. Q2 - Surface earthquake slip profile: The surface slip profile has been measured almost entirely both on the field (maximum lateral slip of ~5 m, Barka et al., 2002), and from SPOT images (maximum lateral slip of 5-5.5 m, Michel and Avouac, 2002), soon after the earthquake. The measurements are consistent. We chose to represent in Supplementary Figure S1 the slip profile determined by Michel and Avouac (2002) because it provides continuous measurements along the rupture. In the east where the SPOT coverage is lacking, we have added the measurements made by Barka et al. (2002) (dashed line in Supplementary Figure S1). Q1 - Earthquake slip profile at depth: We use four source inversion models that have been published on the 1999 Izmit earthquake and whose grid data are available. The models differ principally from the data they used (Delouis et al., 2002: Teleseismic, strong motions, GPS and InSAR data; Cakir et al., 2003: InSAR and surface offsets data; Bouchon et al., 2002: strong motions data; Reilinger et al., 2004: InSAR and GPS data). Yet, overall, data were few to constrain the Izmit earthquake. The 4 models find most of the slip distributed in the depth range 0-20 km, with similar maximum slip

23

values in the range 6-8 m. However they produce very different overall slip patterns (see Supplementary Figure S2). These major differences prevent us to using any of them for we ignore which model might be more appropriate, if any (since data were few). - Hypocenter: The earthquake initiated in the central part of the rupture at a depth of ~17 km (USGS), then propagated bilaterally toward the west and the east. - Rupture speed: The rupture speed was estimated along the rupture length, varying from 3 km/s in the western part of the rupture to 4.8-5.8 km/s (supershear rupture velocity) in its eastern part (Bouchon et al., 2001, 2002; Sekiguchi and Iwata, 2002).

Causative fault parameters - General parameters Slip mode and strike: right lateral strike slip, E-W-trending fault Initiation age: 2-5 Ma ago (e.g., Schindler, 1998; Armijo et al., 1999; Sengör et al., 2005). Length: ~500 km Net slip: debated between 4 and 70 km (e.g., Armijo et al., 1999; Sengör et al., 2005; Bohnoff et al., 2016; Akbayram et al., 2016 and references therein). Long-term slip rate: ~16 mm/yr measured on the eastern part (onshore, Rockwell et al., 2009) and the western part (offshore, Grall et al., 2013). Current slip rate: 16-28 mm/yr, but the more accurate value seems to be ~17 mm/yr (e.g., Hergert and Heidbach, 2010 and references therein). - Mapping of fault trace: Several precise fault maps have been published that are consistent (e.g., Barka et al., 2002; Armijo et al., 2002; see a synthesis in Sengör et al., 2005). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Northern Branch of the North Anatolian Fault is intermediate. - Direction of long-term propagation: Q1 The Northern Branch fault is one of the major splay faults of the North Anatolian fault (Supplementary Fig. S1). It has thus propagated westward over the long-term as its master fault (see “1939 Erzincan earthquake section”).

2001 Mw ~7.8 KUNLUN EARTHQUAKE Earthquake: 14/Nov/2001, Mw 7.8, China, left-lateral strike-slip Causative fault: Kunlun fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured both on the field and from satellite imagery (e.g., Lin et al., 2002, 2003; Fu and Lin, 2003; Fu et al., 2005; Klinger et al., 2005, 2006; Xu et al., 2006). The measurements are consistent. The rupture trace and length are thus robustly constrained (~450 km). Q1 - Rupture extent within causative fault: The 2001 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Klinger et al., 2005, 2006). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been measured both on the field (maximum lateral slip of ~8 m, e.g., Xu et al., 2006), and from SPOT images (maximum lateral slip of ~8.3 m, Klinger et al., 2006). The field and SPOT images-inferred surface slip profiles are consistent. We represent them together in Supplementary Figure S1. Q1 - Earthquake slip profile at depth: We use one source inversion model whose grid data are available (Lasserre et al., 2005). The model is based on InSAR data only (InSAR decorrelates in the vicinity of the fault). The model produces a slip profile showing slip decreasing from a maximum value (~ 8 m) in the east of the rupture down to zero in the western termination of the rupture. Most of slip is

24

distributed in the depth range 0-20 km. The slip profile at depth is similar to that observed at surface (compare Supplementary Figure S1 and S2). - Hypocenter: Although there was a lack of near-field stations, the hypocenter was fairly well located from far-field data. The earthquake initiated at the western tip of the rupture, at 9-12 km depth (e.g., Antolik et al., 2004), then propagated unilaterally toward the east. - Rupture speed: The rupture speed was estimated along the rupture length, varying from 2.6-3.3 km/s in the western part of the rupture to 5-6.7 km/s (supershear rupture velocity) in its eastern part (Bouchon and Vallée, 2003; Antolik et al., 2004; Robinson et al., 2006; Vallée et al., 2008; Walker and Shearer, 2009).

Causative fault parameters - General parameters Slip mode and strike: left lateral strike slip, ~E-W-trending fault Initiation age: The Kunlun fault localized onto a Paleozoic suture (e.g., Tapponnier et al., 2001), about ~10 Ma ago (e.g., Fu and Awata, 2007). Yet older initiation ages have been proposed (23-34 Ma, Tapponnier et al., 2001 and references therein). Length: ~1500 km Net slip: 85-120 km of maximum lateral slip measured at two different locations near the central part of the fault (Gaudemer et al., 1989; Fu and Awata, 2007). Long-term slip rate: 12 mm/yr over the Quaternary in the central part of the Kunlun fault, decreasing to ~10 mm/yr and 1-6 mm/yr towards the west and the east, respectively (Van der Woerd et al., 1998, 2000, 2002; Li et al., 2005; Kirby et al., 2007; Harkins et al., 2010). Current slip rate: 7-11 mm/yr (e.g., Zhang et al., 2004; Meade, 2007) measured in the central part of the fault (and also deduced from block models). - Mapping of fault trace: Several precise fault maps have been published that are consistent (e.g., Van der Woerd et al., 2002; Kirby et al., 2007). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Kunlun fault is mature. - Direction of long-term propagation: Q1 The net slip decreases from the central part of the fault towards both the east and the west (Gaudemer et al., 1989; Fu and Awata, 2007). The long-term slip rate also decreases from the fault center to its eastern and western tips (Van der Woerd et al., 1998, 2000, 2002; Li et al., 2005; Kirby et al., 2007; Harkins et al., 2010). Moreover, Jolivet et al. (2003) described that the recent westward lengthening of the Kunlun fault occurred ~3 Ma ago. Both at its western and eastern tips, the Kunlun fault splays into several oblique branches (see Supplementary Figure S1) which form a fan- shape network enlarging in the west and east directions, respectively. Altogether these evidence show that the Kunlun fault has propagated bilaterally from its central part over its lifetime.

1992 Mw ~7.3 LANDERS EARTHQUAKE Earthquake: 28/Jun/1992, Mw 7.3, California, right-lateral strike-slip Causative fault: The earthquake actually broke three distinct fauts disposed en echelon: the Camp Rock-Emerson, the Homestead Valley and the Johnson Valley faults Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured both on the field (Sieh et al., 1993) and from InSAR imagery (Massonet et al., 1993, 1994; Zebker et al., 1994; Peltzer et al., 1994), immediately after the earthquake. Postseismic deformation rapidly took place due south of the Landers rupture along the Eureka Peak fault (rupture trace reported by Sieh et al., 1993). The surface trace reported in some studies on the

25

Eureka Peak fault is thus of post-seismic origin, not coseismic. The small northern part of the rupture (dotted line in Supplementary Figure S1) was only a surficial rupture of the ground, not a rupture of the fault at depth as shown by the lack of aftershocks in this northernmost zone (Hauksson et al., 2012). The rupture trace and length are thus well constrained (~67 km), but the rupture length slightly differs among authors for the two reasons discussed above. Q2 - Rupture extent within causative fault: The 1992 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (Sieh et al., 1993; U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field (maximum lateral slip of ~6.6 m, e.g., Sieh et al., 1993), as with InSAR (Massonet et al., 1993) and GPS (Hudnut et al., 1994). The measurements are consistent and in particular show a co-located maximum lateral slip of 6-6.6 m. Q1 - Earthquake slip profile at depth: We use four source inversion models that have been published on the 1992 Landers earthquake and whose grid data are available. The models differ principally from the data they used (Wald and Heaton, 1994: Teleseismic, strong motions, GPS and surface offsets data; Cohee and Beroza, 1994: strong motions and surface offsets data; Cotton and Campillo, 1995: strong motions data; Hernandez et al., 1999: InSAR and GPS data). All four models thus use a large amount of various data and hence are likely robust. They all produce a fairly similar slip profile showing slip decreasing from a maximum value (range 5-8 m) in the north of the rupture (similar location of zone of largest slips on all four profiles) down to zero at the southern termination of the rupture. All four models also find most of the slip distributed in the depth range 0-15 km. The model of Wald and Heaton (1994) being the one using more data, both local and regional, it is the one we most consider in Supplementary Figure S2. The zone of largest displacements at depth is co-located with that observed at surface (compare Supplementary Figure S1 and S2). This further supports that the models are appropriate. - Hypocenter: The hypocenter location is well constrained, and shows that the earthquake initiated near the southern end of the rupture at ~7 km depth (e.g., refine earthquake catalog from Hauksson et al., 2012), then propagated unilaterally toward the north. - Rupture speed: The rupture speed was estimated along the rupture length, and shown to vary from 2.5-3.6 km/s in the southern part of the rupture to ~4 km/s (supershear rupture velocity) in its northern part (Wald and Heaton, 1994; Peyrat et al., 2001).

Causative fault parameters - General parameters Slip mode and strike: right lateral strike slip, NW to N-S-trending faults Initiation age: < 10 Ma recognized in multiple studies (Dibblee, 1961; Garfunkel, 1974; Dokka, 1983; Dokka and Travis, 1990 and references therein). The fault age is thus fairly well constrained. Length: ~95 km Net slip: ~4.6 km on the northern Camp Rock fault (field: Dokka, 1983; Dokka and Travis, 1990; magnetic anomalies: Jachens et al., 2002) to less than 3.5 km on the southern Homestead Valley and Johnson Valley faults (Jachens et al., 2002). Long-term slip rate: 0.2-0.7 mm/yr (late Cenozoic), based on data from many paleoseismic trenches across the Emerson, Homestead Valley and Johnson Valley faults (Rubin and Sieh, 1997; Rockwell et al., 2000b). Current slip rate: not defined. However the total current slip rate across the southern part of the East California Shear Zone (ECSZ) is ~12 mm/yr (Sauber et al., 1994; McClusky et al., 2001). If one assumes that the ECSZ is composed of six major sub-parallel rotating faults including the Camp Rock-Emerson-Homestead Valley-Johnson Valley fault system (e.g., Jennings et al., 1994; Jachens et al., 2002), the fault system might be currently slipping at ~2 mm/yr. - Mapping of fault trace: Several precise fault maps have been published by the USGS (Jennings et

26

al., 1994; compilation in U.S. Geological Survey and California Geological Survey, 2006), which are consistent. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Camp Rock-Emerson-Homestead Valley-Johnson Valley fault system is immature. - Direction of long-term propagation: Q1 Jachens et al., (2002) have shown that the net slip on the Camp Rock-Emerson, Homestead Valley and Johnson Valley faults is largest in the northern part of the fault zone (Camp Rock fault) and decreases progressively towards the south (Johnson Valley fault). Furthermore, Zachariasen and Sieh (1995) have shown that the northern Emerson fault is older than the southern Homestead Valley fault. At its northern tip, the Camp Rock-Emerson-Homestead Valley-Johnson Valley fault zone terminates in a large step-over with the Harper Lake fault. In the south, the Homestead Valley and Johnson Valley faults splay off the Camp Rock-Emerson fault, and, together with other oblique fault branches, they form a fan-shape network enlarging in the south direction. Altogether these evidence show that the Camp Rock-Emerson-Homestead Valley-Johnson Valley fault system has propagated from north to south during its lifetime.

1997 Mw ~7.5 MANYI EARTHQUAKE Earthquake: 09/Nov/1997, Mw 7.6, California, left-lateral strike-slip Causative fault: Manyi fault Earthquake parameters - Surface rupture trace: There has been no detailed field study after the earthquake due to the difficulty to access the area. However the surface trace of the earthquake was precisely mapped, and the displacements measured, using high quality radar images (arid climate, no clouds; Peltzer et al., 1999; Funning et al., 2007). Thus the rupture trace and length are well constrained (length ~165 km). Q2. - Rupture extent within causative fault: The 1997 surface rupture is still clear on high resolution satellite images (e.g., Google Earth), and one precise rupture map is available (e.g., Peltzer et al., 1999). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The complete surface slip profile has been precisely measured from InSAR data (arid climate, no clouds, thus precise measurements; maximum lateral slip of 6-7 m, e.g., Peltzer et al., 1999). Q2 - Earthquake slip profile at depth: We use one source inversion model that has been published on the 1997 Manyi earthquake and whose grid data is available. The model uses InSAR data only (Funning et al., 2007). The slip profile inferred at depth is similar to that measured at surface, as expected from Insar data being used in both cases (Peltzer et al., 1999; Funning et al., 2007). The model finds most of slip distributed in the depth range 0-20 km, with a maximum slip of ~7.6 m close to the surface, in the eastern part of the rupture. - Hypocenter: The hypocenter location is no well constrained. However one robust fact is that the earthquake initiated in the central part of the rupture at ~16 km depth (SCARDEC, Vallée et al., 2011; Global CMT), then it propagated bilaterally both to the east and to the west. - Rupture speed: The rupture speed was estimated along the rupture length, and suggested to vary from 3-3.2 km/s in the eastern part of the rupture to 3 km/s in its western part (Funning et al., 2014).

Causative fault parameters - General parameters Slip mode and strike: left lateral strike slip, N76°E-trending fault Initiation age: unknown, but the Manyi fault is considered to have possibly initiated at about the same time than the Kunlun fault (10-34 Ma ago; see 2001 Kunlun earthquake section). Length: ~275 km (mainly based on our mapping, see below).

27

Net slip: unknown Long-term slip rate: unknown Current slip rate: 2.5 ± 2.5 mm/yr, yet not directly measured but inferred from a large-scale modeled velocity field (Holt et al., 2007). - Mapping of fault trace: Since few and incomplete information was available (Taylor and Yin, 2009, in Bell et al., 2011), we mapped the entire fault trace and its associated secondary faults based on the analysis of Google Earth, Landsat 7 and ASTER GDEM v2 images. The main fault and largest secondary fault traces are clear on these images, and therefore our mapping is robust. Q2 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Manyi fault would be intermediate. - Direction of long-term propagation: Q3 There is no published data that provides insights on the direction of long-term propagation of the Manyi fault. However, based on our fault mapping and available fault maps (Taylor and Yin, 2009, in Bell et al., 2011), we find that the Manyi fault splays into multiple oblique branches at both its western and eastern tips (connection with the Kunlun fault to the east), forming a fan-shape network enlarging in the East and West directions (Supplementary Figure 1). These tip splay networks suggest that the Manyi fault has propagated bilaterally towards the East and West during its lifetime (Perrin et al., 2016).

1967 Mw ~7.0 MUDURNU EARTHQUAKE Earthquake: 22/Jul/1967, Mw 7.0, Turkey, right-lateral strike-slip Causative fault: North Anatolian fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field, soon after the earthquake (e.g., Ambraseys and Zatopek, 1969; Barka, 1996). The rupture trace and length are thus robustly constrained (length ~75 km). Q1. - Rupture extent within causative fault: The 1967 surface rupture is fairly clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., Ambraseys and Zatopek, 1969; Barka, 1996). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field soon after the earthquake. (Ambraseys and Zatopek, 1969; also compiled in Barka, 1996). Yet the measurement density is not very high. The maximum reported lateral coseismic slip at surface is ~2 m near the eastern tip of the rupture. Q2 - Hypocenter: The hypocenter location was not well constrained, but a robust fact is that the earthquake initiated in the eastern part of the rupture at shallow depth (< ~10 km; Ambraseys and Zatopek, 1969), and then propagated mainly to the west.

Causative fault parameters See description of the North Anatolian fault in the “1939 Erzincan earthquake” section

2005 Mw ~7.6 MUZAFFARABAD EARTHQUAKE Earthquake: 08/Oct/2005, Mw 7.6, Pakistan, reverse fault Causative fault: Balakot-Bagh fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the 28

displacements measured both on the field (Kaneda et al., 2008) and from ASTER/InSAR imagery (Avouac et al., 2006; Pathier et al., 2006; Yan et al., 2013) soon after the earthquake. The rupture trace and length are thus well constrained (~65-70 km). Q1 - Rupture extent within causative fault: The 2005 surface rupture is still partly clear on high- resolution satellite images (e.g., Google Earth), while precise rupture maps are available (e.g., Kaneda et al., 2008). The rupture zone is thus fairly well located within the causative fault. Q2. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field (maximum vertical slip of ~7 m, Kaneda et al., 2008) and ASTER images (maximum horizontal slip of ~7 m, Avouac et al., 2006). The on-field measured and ASTER-inferred surface slip profiles are consistent. We chose to represent in Supplementary Figure S1 the slip profile determined by Kaneda et al. (2008) because it provides more accurate measurements than the slip profile measured by Avouac et al. (2006) (for the study was based on 15 m resolution ASTER images). Q1 - Hypocenter: The hypocenter location was not well constrained, but a robust fact is that the earthquake initiated near the central part of the rupture at ~12 km (NEIC) to ~26 km (USGS) depth, then propagated bilaterally both to the northwest and to the southeast.

Causative fault parameters - General parameters Slip mode and strike: reverse, ~NW-trending fault Initiation age: <1 Ma year ago (Hussain et al., 2009), based on offset sediment deposits (Siwalik deposits). Length: ~90 km. However the fault trace is unclear near its southeastern tip (Kaneda et al., 2008; Hussain et al., 2009). Net slip: ~0.12 km of horizontal shortening measured locally in the central part of the fault (Kaneda et al., 2008; Hussain et al., 2009). Long-term slip rate: ~3 mm/yr of shortening estimated in the central part of the fault (Kondo et al., 2008; range 1.4-4.1 mm/yr, Hussain et al., 2009 and references therein). Current slip rate: not constrained, but several studies measured a current slip rate of ~10 mm/yr across the Riassi thrust (Vignon 2011; Jouanne et al., 2014), which is possibly connected to the Balakot-Bagh fault (see below). - Mapping of fault trace: Several precise fault maps have been published that are consistent (e.g., Kaneda et al., 2008; Hussain et al., 2009). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Balakot-Bagh fault is immature. - Direction of long-term propagation: Q3 Fault age and net slip data are presently insufficient to constrain the direction of long-term propagation of the Balakot-Bagh fault. However the fault is intercepted to the west by the Main Boundary Thrust, which thus likely prevents its northwestward propagation (Supplementary Fig. S1). At its southeastern tip, the Balakot-Bagh fault seems to being connecting with the Riassi thrust fault (Jayangondaperumal and Thakur, 2008; Kaneda et al., 2008; Hussain et al., 2009; Thakur et al., 2010; Jouanne et al., 2014; Supplementary Figure S1). Together these evidences suggest that the Balakot Bagh fault has propagated southeastward over the long-term.

1915 Mw ~7.0 PLEASANT VALLEY EARTHQUAKE Earthquake: 03/Oct/1915, Mw 7.0, Nevada, normal fault Causative fault: Pleasant Valley fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the

29

displacements measured on the field, soon after the earthquake and later (Jones, 1915; Wallace, 1984; dePolo et al., 1991; Zhang et al., 1999). During the field investigations, no surface trace was observed on the northern China Mountain fault section (Jones, 1915), and therefore, we do not include the small, likely post-seismic slip that was subsequently measured on this section (dashed line in Supplementary Figure S1) (Rupture length ~45 km). Q2 - Rupture extent within causative fault: The 1915 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (e.g., dePolo et al., 1991; Zhang et al., 1999; U.S. Geological Survey and Nevada Bureau of Mines and Geology, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field yet long after the earthquake from measurements of well-preserved scarps, while previous descriptions made by Jones (1915) were integrated (Wallace, 1984; Zhang et al., 1999). The measurements reveal a maximum vertical slip value of ~6 m near the central part of the rupture. Q2 - Hypocenter: The hypocenter location is not constrained. Yet its most likely position is at the northern tip of the rupture at ~9 km depth (e.g., Doser, 1988). If this position is correct, the rupture has propagated unilaterally toward the south.

Causative fault parameters - General parameters Slip mode and strike: normal, ~NNE-trending fault Initiation age: unknown. But Ferrill et al. (1999) analyzed the architecture of the Pleasant Valley fault zone and deduced that the oldest part of the fault is its central Pearce section. Length: ~70 km. Net slip: unknown, but a minimum apparent cumulative vertical throw of ~2 km can be measured in present topography. Long-term slip rate: 0.1-0.2 mm/yr across the Pearce central fault section, based on the measurement of a ~8 m cumulative vertical slip whose age is estimated in the range 35-100 ka (Anderson and Machette, 2003; Bell et al., 2004). Current slip rate: unknown - Mapping of fault trace: Several precise fault maps have been published that are consistent (e.g., dePolo et al., 1991; Zhang et al., 1999; U.S. Geological Survey and Nevada Bureau of Mines and Geology, 2006). Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the Pleasant Valley fault is immature. - Direction of long-term propagation: Q1 The fault seems older in its central part (Ferrill et al., 1999). Anders and Schlische (1994) have shown that the net slip decreases from the central part of the fault towards both the north and the south. Jackson and Leeder (1994) found similar results from their geomorphological analysis of the drainage networks. At its northern and southern tips, the Pleasant Valley fault splays into several oblique branches (Supplementary Figure S1), forming a fan-shape network enlarging in the north and south directions. Altogether these evidence suggest that the Pleasant Valley fault has propagated bilaterally towards the south and the north during its lifetime.

1906 Mw ~7.7 SAN FRANCISCO EARTHQUAKE Earthquake: 18/Apr/1906, Mw 7.7, California, right-lateral strike-slip Causative fault: San Andreas fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been mapped, and the

30

displacements measured both on the field (Lawson, 1908) and from triangulation networks (Thatcher et al., 1997), both soon after the earthquake and later. The northern part of the rupture (north of Point Arena) occurred offshore, but slip could be well constrained from geodetic observations made before and after the earthquake (Thatcher et al., 1997). Furthermore, the very northern tip of the rupture was observed on land, at Shelter Cove (Lawson, 1908). The total rupture trace is thus well known (rupture length ~475 km). Q1 - Rupture extent within causative fault: The 1906 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available, including offshore (e.g., Lawson 1908; U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The surface slip profile has been measured on the field soon after the earthquake yet in the southern onland part of the rupture only (maximum lateral slip value of ~6 m measured near San Francisco; Lawson 1908). By constrast, the coseismic slip could not be measured in the northern offshore part of the rupture. However, using triangulation networks, Thatcher et al. (1997) inferred that the maximum lateral slip occurred at the northern end of the rupture and was at most ~8.6 m. The on-field measured and triangulation-inferred surface slip profiles are similar in the southern onland part of the rupture. The complete earthquake surface slip profile is well represented by combining the profile measured on the field by Lawson (1908) in the southern part of the rupture, and that inferred by Thatcher et al., (1997) in the northern part of the rupture (Supplementary Figure S1). Q3 - Hypocenter: The hypocenter location is poorly constrained, but a robust fact is that the earthquake initiated in the central part of the rupture (Bolt, 1968), and then propagated bilaterally to the south and the north. - Rupture speed: From the review of available seismological data, the rupture speed has been recently estimated along the entire rupture length, and shown to have varied from ~2.9 km/s in the southern part of the rupture to ~3.9 km/s (supershear rupture velocity) in its northern part (Song et al., 2008).

Causative fault parameters See description of the San Andreas fault in the “1857 Fort Tejon earthquake” section

2008 Mw ~7.9 SICHUAN EARTHQUAKE Earthquake: 12/May/2008, Mw 7.9, China, reverse and right-lateral strike-slip Causative fault: Beichuan fault (Longmen Shan thrust belt) Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been mapped, and the displacements measured both on the field (Liu Zeng et al., 2009; Xu et al., 2009; Zhang et al., 2010; Lin et al., 2009, 2010, 2012) and from radar imagery (Hashimoto et al., 2009; Kobayashi et al., 2009; Shen et al., 2009; Chini et al., 2010; Feng et al., 2010; deMichele et al., 2010a, 2010b; Tong et al., 2010; Wang et al., 2011; Zhang et al., 2011; Fielding et al., 2013) soon after the earthquake. The rupture trace and length are thus well constrained (~280 km) Q1. - Rupture extent within causative fault: The 2008 surface rupture is not clear on satellite images (e.g., Google Earth, Landsat 7), but several precise rupture maps are available (e.g., Liu Zeng et al., 2009; Lin et al., 2009, 2010, 2012). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The surface slip profile has been measured by various authors, on the field, soon after the earthquake. We have represented the envelope curves derived from the numerous available measurements, distinguishing the vertical and horizontal components (Supplementary Figure S1 and references therein). Maximum vertical slips of 6-6.5 m and ~3.5 m were measured in the southern part of the rupture on the Beichuan and Pengguan faults, respectively. A maximum lateral slip of ~4 m was also measured in the northern part of the rupture

31

(Beichuan fault). Note that we chose to ignore the large vertical slip of 8-11 m suggested at Beichuan (city situated in the central part of the rupture) by Liu Zeng et al. (2009) and Zhang et al. (2010), because this large value is local and departing from all other slip measures, including geodetic estimates (e.g., de Michele et al., 2010). Therefore, this large value is likely an outlier resulting from gravity sliding (Aiming Lin, personal communication, 2013). We have calculated the net surface slip profile by using the fault dips provided by Jia et al. (2010). The largest net slip is 10-12 m. Q1 - Earthquake slip profile at depth: We use one source inversion model that has been published on the 2008 Sichuan earthquake and whose grid data is available (Fielding et al., 2013). The model uses GPS, InSAR and teleseismic data. The model finds most of the slip distributed in the depth range 0-20 km, with a maximum slip of 16 m in the southern part of the rupture. The slip profiles inferred at depth and measured at the surface are similar. - Hypocenter: The hypocenter location is well constrained, and shows that the earthquake initiated at the southern tip of the rupture at ~17 km depth (An et al., 2010); it then propagated unilaterally to the north. - Rupture speed: The rupture speed was estimated along the rupture length, and found to vary from 2.5-3 km/s in the northern part of the rupture to 3-3.2 km/s in its southern part (Wen et al., 2012; Zhang et al., 2012).

Causative fault parameters - General parameters Slip mode and strike: reverse and right lateral strike slip, N50°E-trending fault Initiation age: The Longmen Shan fault system is supposed to have reactivated 5-12 Ma ago (during the India-Asia collision) an ancient suture inherited from the Triassic Indosinian orogeny (e.g., Burchfiel et al., 1995; Kirby et al., 2002). Length: ~380 km Net slip: ~20 km accumulated since the Cenozoic, estimated in the southern part of the fault from calculations based on slip rates, denudation rates and topographic relief (Burchfiel et al., 2008). This value was recently validated from P-T paths of metamorphic rocks (Airaghi et al., 2015). Long-term slip rate: < 1.1 mm/yr (reverse component) and <1.5 mm/yr (strike-slip component) over the Cenozoic, based on deformed geomorphic markers across the southern part of the Longmen Shan fault system (Zhou et al., 2007). The long-term slip rate of the Beichuan fault is inferred to be very low. Current slip rate: the compression rate across the Longmen Shan fault system is < 3 mm/yr and the current vertical slip rate on the fault system is considered to be very low (< 1 mm/yr; King et al., 1997; Chen et al., 2000, Shen et al., 2005, Gan et al., 2007; Burchfiel et al., 2008, Zhang et al., 2008). - Mapping of fault trace: Several precise fault maps have been published along the entire Longmen Shan fault system (e.g., Burchfiel et al., 1995; Shen et al., 2009; Jin et al., 2010), which are consistent. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Beichuan fault would be intermediate. - Direction of long-term propagation: Q1 Several studies have shown that the net slip on the Beichuan fault decreases progressively from southwest to northeast (Burchfiel et al., 2008; Zhang et al., 2011; Airaghi et al., 2015). At its southwestern tip, the Beichuan fault is surrounded by a dense network of secondary reverse faults but those were likely formed long ago (e.g, Burchfiel et al., 1995). The Beichuan fault actually terminates abruptly against a zone of sub-perpendicular thrust faults (in grey in Supplementary Figure S1). At its northeastern tip, the Beichuan fault splays into several oblique branches whose morphological signature suggests they are recent (Supplementary Figure S1), although their

32

absolute age is unknown (Burchfiel et al., 1995). Altogether these evidence suggest that the Beichuan fault has propagated from southwest to northeast during its lifetime.

1987 Mw ~6.6 SUPERSTITION HILLS EARTHQUAKE Earthquake: 24/Nov/1987, Mw 6.6, California, right-lateral strike-slip Causative fault: San Jacinto fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been precisely mapped, and the displacements measured on the field, soon after the earthquake (Sharp et al., 1989; Williams and Magistrale 1989). The rupture trace and length are thus robustly constrained (length ~27 km). Q1. - Rupture extent within causative fault: The 1987 surface rupture is still clear on high-resolution satellite images (e.g., Google Earth), while several precise rupture maps are available (Sharp et al., 1989; U.S. Geological Survey and California Geological Survey, 2006). The rupture zone is thus well located within the causative fault. Q1. - Surface earthquake slip profile: The complete surface slip profile has been measured on the field soon after the earthquake (Sharp et al., 1989; Williams and Magistrale 1989). The different available surface slip profiles are similar. They suggest a maximum lateral slip of ~0.5 m in the central part of the rupture (Sharp et al., 1989; Williams and Magistrale 1989). We chose to represent in Supplementary Figure S1 the slip profile determined by Sharp et al. (1989) because it provides more measurements along the rupture. Q1 - Earthquake slip profile at depth: We use one source inversion model that has been published on the 1987 Superstition Hills earthquake and whose grid data are available (Wald et al., 1990). The model is based on a few strong motions data only, and therefore is not well constrained. As a matter of fact, the modeled rupture length is shorter (~20 km) than observed at surface. A maximum displacement of ~4 m is inferred at depth (at 8-9 km depth), yet in the northwestern part of the rupture, whereas maximum slip was observed at surface in its central part. This likely confirms that the model is not fully correct. - Hypocenter: The hypocenter location was fairly well constrained. It shows that the earthquake initiated at the northwestern tip of the rupture at ~13 km depth (refine earthquake catalog from Hauksson et al., 2012), and then propagated to the southeast. The main shock might have been triggered by the Elmore Ranch earthquake which occurred 12 hours before (Hudnut et al., 1989).

Causative fault parameters See description of the San Jacinto fault in the “1968 Borrego Mountain earthquake” section.

2010 Mw ~7.1 YUSHU EARTHQUAKE Earthquake: 13/Apr/2010, Mw 7.1, China, left-lateral strike-slip Causative fault: Garze-Yushu fault Earthquake parameters - Surface rupture trace: The surface trace of the earthquake has been mapped, and the displacements measured on the field (Chen et al., 2010; Lin et al., 2011; Li et al., 2012), from optical satellite imagery (Li et al., 2012) and from InSAR data (Zhang et al., 2010; Li et al., 2011; Liu et al., 2011; Tobita et al., 2011; Qu et al., 2012; Zhang et al., 2013), soon after the earthquake. However, because part of the rupture did not reach the surface, the rupture trace and length are debated. Recently, the field studies of Li et al. (2012) provided new surface measurements in the northwestern part of the rupture which allow to better constrain the rupture length, and find it in fair agreement with prior InSAR data (rupture length ~75 km). Q2. - Rupture extent within causative fault: The 2010 surface rupture is clear on high-resolution satellite

33

images (e.g., Google Earth), while several precise rupture maps are available (e.g., Lin et al., 2011; Li et al., 2012). The rupture zone is thus well located within the causative fault. Q1 - Surface earthquake slip profile: The surface slip profile has been measured on the field soon after the earthquake but measurements are incomplete in the central part of the rupture (Supplementary Figure S1). The maximum lateral slip was measured in the southeastern part of the rupture, where it varies depending on authors, between ~1.8 m (Chen et al., 2010; Zhang et al., 2010; Li et al., 2012) and ~3.2 m (Lin et al., 2011). The lower estimate is in better agreement with the slip values inferred from InSAR data and modeling (~1.3 m from Liu et al., 2011 (yet noisy correlation in the northwest), ~1.5 m from Li et al., 2011, ~1.9 m from Tobita et al., 2011, and ~2.2 m from Zhang et al., 2013). Thus the surface slip profile is not fully constrained. Q3 - Earthquake slip profile at depth: We use one source inversion model that has been published on the 2010 Yushu earthquake and whose grid data is available. The model uses InSAR data (ASAR and PALSAR) and also teleseismic data and SPOT 5 images to constrain the slip distribution and the fault segmentation at surface (Li et al., 2012). The model finds most of slip distributed in the depth range 0-20 km, with a maximum lateral slip of 1.5 m. This value is lower than the maximum slip measured at surface, what suggests that the model is not fully constrained. However the slip profile inferred at depth is similar to that measured at the surface. - Hypocenter: The hypocenter location is well constrained, and shows that the earthquake initiated in the northwestern part of the rupture at ~13 km depth (relocated by Wang et al., 2013), then propagated mainly toward the southeast. - Rupture speed: The rupture speeds was estimated along the rupture length, and found to vary from 1.6-2.5 km/s in the northern part of the rupture to 4-5 km/s (supershear rupture velocity) in its southern part (Zhang et al., 2010; Wang and Mori, 2012).

Causative fault parameters - General parameters Slip mode and strike: left lateral strike slip, N120°E-trending fault Initiation age: ~13 Ma ago (e.g., Tapponnier et al., 2001) based on K/Ar and apatite fission-track ages of offset geological features in the southeastern part of the Garze-Yushu fault (Wang et al., 2009). Length: ~450 km Net slip: ~ 80 km, based on the lateral offset of a granitic pluton in the southeastern part of the Garze-Yushu fault (Wang and Burchfiel 2000; Wang et al., 2009). Long-term slip rate: 9-12 mm/yr (over the Quaternary) measured in central and southeastern parts of the Garze-Yushu Fault (Wen et al., 2003; Zhou et al. 2013) and 15 ± 5mm/yr on the Xianshui He Fault further south (Allen et al., 1991). Current slip rate: ~20 mm/yr (e.g., Meade 2007) yet poorly constrained for it is inferred from a large-scale block model of the India-Asia collision zone (Geodetic data). - Mapping of fault trace: Several precise fault maps have been done (e.g., Wen et al., 2003, Wang et al., 2008), which are consistent. Q1 - Degree of overall structural maturity: in the classification proposed by Manighetti et al. (2007), the structural maturity of the Garze-Yushu fault would be intermediate. - Direction of long-term propagation: Q1 Wang et al., (2008) have shown that the net lateral slip on the Garze-Yushu fault decreases from southeast to northwest, and that faulting is younger in the northwest part of the fault; the oldest age of the fault was found in its southern part. At its southwestern tip, the Garze-Yushu fault terminates in a large step-over with the Xianshui He fault. At its northwestern tip, the Garze-Yushu fault splays into several oblique branches (Supplementary Figure S1), forming a fan shape enlarging in the northwest direction. Altogether these evidence show that the Garze-Yushu fault has propagated from southeast to northwest during its lifetime.

34

References Airaghi, L., de Sigoyer, J., Vidal, O., Lanari, P., Tan, X. B., Xu, X., Guillot, S. Total Vertical Offset for the Beichuan Fault (Longmen Shan, Sichuan, China) Deduced from Metamorphic Minerals. T41B-2880, AGU Fall Meeting, 2015. Akbayram, K., Sorlien, C. C., and Okay, A. I. (2016). Evidence for a minimum 52±1km of total offset along the northern branch of the North Anatolian Fault in northwest Turkey. Tectonophysics 668-669, 35-41, doi: 10.1016/j.tecto.2015.11.026 Allen, C. R., and Nordquist, J. M. (1972). Foreshock, main shock, and larger aftershocks of the Borrego Mountain earthquake. US Geol. Surv. Profess. Paper, 787, 16-23. Allen, C.R., Qian, H., Wen, X., Zhou, H., and Huang, W. (1991). Field study of a highly active fault zone: The Xianshuihe Fault of southwestern China. Geol. Soc. Am. Bulletin, 103, 1178–1199. Ambraseys, N. N., and Zatopek, A. (1969). The Mudurnu Valley, West Anatolia, Turkey, earthquake of 22 July 1967. Bull. Seism. Soc. Am., 59(2), 521-589. An, M., Feng, M., and Long, C. (2010). Deep ruptures around the hypocenter of the 12 May 2008 Wenchuan earthquake deduced from aftershock observations. Tectonophysics, 491(1), 96-104. Anders, M. H., and Schlische, R. W. (1994). Overlapping faults, intrabasin highs, and the growth of normal faults. J. Geol., 165-179. Anderson, R. E., and M. N. Machette (Compilers) (2003). Fault number 1136a-d, Pleasant Valley fault zone, Pearce section, in Quaternary Fault and Fold Database of the United States, ver. 1.0, U.S. Geol. Surv. Open-File Rpt. 03-417, http://qfaults.cr.usgs.gov. Anderson, J. G., Wesnousky, S. G., and Stirling, M. W. (1996). Earthquake size as a function of fault slip rate. Bull. Seism. Soc. Am., 86(3), 683-690. Ando, R., Shaw, B. E., and Scholz, C. H. (2009). Quantifying natural fault geometry: statistics of splay fault angles. Bull. Seism. Soc. Am., 99(1), 389-395. Antolik, M., R. E. Abercrombie, and G. Ekström 2004. The 14 November, 2001 Kokoxili (Kunlunshan), Tibet earthquake: rupture transfer through a large extensional step-over, Bull. Seism. Soc. Am. 94, 1173–1194. Archuleta, R. J. (1984). A faulting model for the 1979 Imperial Valley earthquake. J. Geoph. Res., 89(B6), 4559-4585 Armijo, R., Meyer, B. G. C. P., King, G. C. P., Rigo, A., and Papanastassiou, D. (1996). Quaternary evolution of the Corinth Rift and its implications for the Late Cenozoic evolution of the Aegean. Geoph. J. Int., 126(1), 11-53. Armijo, R., Meyer, B., Hubert, A., and Barka, A. (1999). Westward propagation of the North Anatolian fault into the northern Aegean: Timing and kinematics. Geology, 27(3), 267-270. Armijo, R., Meyer, B., Navarro, S., King, G., and Barka, A. (2002). Asymmetric slip partitioning in the Sea of Marmara pull-apart: a clue to propagation processes of the North Anatolian Fault?. Terra Nova, 14(2), 80-86. Asano, K., Iwata, T. and Irikura, K. (2005). Estimation of Source Rupture Process and Strong Ground Motion Simulation of the 2002 Denali, Alaska, Earthquake. Bull. Seism. Soc. Am., 95, 5, 1701–1715, doi: 10.1785/0120040154 Atwater, T. (1970) Implications of Plate Tectonics for the Cenozoic Tectonic Evolution of Western North America Geol. Soc. Am. Bulletin, 81, 3513-3536, doi:10.1130/0016- 7606(1970)81[3513:IOPTFT]2.0.CO;2 Atwater, T. (1989). Plate tectonic history of the northeast Pacific and western North America. The eastern Pacific Ocean and Hawaii: Boulder, Colorado, Geol. Soc. Am., Geology of North America, v. N, 21-72. Atwater, T., and Stock, J. (1998). Pacific-North America plate tectonics of the Neogene southwestern United States: an update. Int. Geol. Rev., 40(5), 375-402. Aviles, C. A., Scholz, C. H., and Boatwright, J. (1987). Fractal analysis applied to characteristic segments of the San Andreas fault. J. Geoph. Res., 92(B1), 331-344. Avouac, J. P., Meyer, B., and Tapponnier, P. (1992). On the growth of normal faults and the existence of flats and ramps along the El Asnam active fold and thrust system. Tectonics, 11(1), 1-11. Avouac, J. P., Ayoub, F., Leprince, S., Konca, O., and Helmberger, D. V. (2006). The 2005, M w 7.6 Kashmir earthquake: Sub-pixel correlation of ASTER images and seismic waveforms analysis. Earth Planet. Sc. Lett., 249(3), 514-528. Baljinnyam, I., et al. (1993), Ruptures of major earthquakes and active deformation in Mongolia and its surroundings, Mem. Geol. Soc. Am., 181, 62 pp. Barka, A. (1992). The north Anatolian fault zone. In Annales tectonicae (Vol. 6, No. Suppl, pp. 164-195). Barka, A. (1996). Slip distribution along the North Anatolian fault associated with the large earthquakes of the period 1939 to 1967. Bull. Seism. Soc. Am., 86(5), 1238-1254. Barka, A. A., and KadinskyCade, K. (1988). Strikeslip fault geometry in Turkey and its influence on earthquake activity. Tectonics, 7(3), 663-684. Barka, A., Akyüz, H. S., Altunel, E., Sunal, G., Cakir, Z., et al. (2002). The surface rupture and slip distribution of the 17 August 1999 Izmit earthquake (M 7.4), North Anatolian fault. Bull. Seism. Soc. Am., 92(1), 43-60. Bayarsayhan, C., Bayasgalan, A., Enhtuvshin, B., Hudnut, K. W., Kurushin, R. A., Molnar, P., and Ölziybat, M. (1996). 1957 Gobi-Altay, Mongolia, earthquake as a prototype for southern California's most devastating earthquake. Geology, 24(7), 579-582. Bayasgalan, A., Jackson, J., Ritz, J. F., and Carretier, S. (1999a). Field examples of strikeslip fault terminations in Mongolia and their tectonic significance. Tectonics, 18(3), 394-411. Bayasgalan, A., Jackson, J., Ritz, J. F., and Carretier, S. (1999b). Forebergs', flower structures, and the development of large intra-continental strike-slip faults: the Gurvan Bogd fault system in Mongolia. J. Struct. Geol., 21(10), 1285- 1302. Bell, J. and T. Katzer (1990). Timing of late Quaternary faulting in the 1954 Dixie Valley earthquake area, central Nevada, Geology 18, 622-625.

35

Bell, J. W., Caskey, S. J., Ramelli, A. R., and Guerrieri, L. (2004). Pattern and rates of faulting in the central Nevada seismic belt, and paleoseismic evidence for prior beltlike behavior. Bull. Seism. Soc. Am., 94(4), 1229-1254. Bell, M. A., Elliott, J. R., and Parsons, B. E. (2011). Interseismic strain accumulation across the Manyi fault (Tibet) prior to the 1997 Mw 7.6 earthquake. Geophys. Res. Lett., 38(24). Bennett, R. A., B. P. Wernicke, N. A. Niemi, A. M. Friedrich, and J. L. Davis (2003). Contemporary strain rates in the northern Basin and Range province from GPS data, Tectonics, 22(2), 1008, doi:10.1029/2001TC001355. Biggs, J., Wright, T., Lu, Z. and Parsons, B. (2007). Multi-interferogram method for measuring interseismic deformation—Denali Fault, Alaska. Geoph. J. Int., 170, 1,165–1,179. Bilham, R., and Williams, P. (1985). Sawtooth segmentation and deformation processes on the southern San Andreas fault, California. Geoph. Res. Lett., 12(9), 557-560. Blisniuk, K., T. Rockwell, L. A. Owen, M. Oskin, C. Lippincott, M. W. Caffee, and J. Dortch. 2010. Late Quaternary slip rate gradient defined using high-resolution topography and 10 Be dating of offset landforms on the southern San Jacinto Fault zone, California. J. Geophys. Res., 115, B08401, doi:10.1029/2009JB006346. Bohnhoff, M., Martínez-Garzón, P., Bulut, F., Stierle, E., and Ben-Zion, Y. (2016). Maximum earthquake magnitudes along different sections of the North Anatolian fault zone. Tectonophysics, 674, 147-165. Bolt, B. A. (1968). The focus of the 1906 California earthquake, Bull. Seism. Soc. Am. 58, 457-471. Bouchon, M. and Vallée, M., (2003). Observation of long supershear rupture during the magnitude 8.1 Kunlunshan earthquake, Science, 301, 824–826. Bouchon, M., Bouin, M. P., Karabulut, H., Toksöz, M. N., Dietrich, M., and Rosakis, A. J. (2001). How fast is rupture during an earthquake? New insights from the 1999 Turkey earthquakes. Geoph. Res. Lett., 28(14), 2723-2726. Bouchon, M., Toksöz, M. N., Karabulut, H., Bouin, M. P., Dietrich, M., et al. (2002). Space and time evolution of rupture and faulting during the 1999 Izmit (Turkey) earthquake. Bull. Seism. Soc. Am., 92(1), 256-266. Boudiaf, A., Ritz, J. F., and Philip, H. (1998). Drainage diversions as evidence of propagating active faults: example of the El Asnam and Thenia faults, Algeria. Terra Nova-Oxford, 10(5), 236-244. Burchfiel, B.C., Chen, Z., Liu, Y. and Royden, L.H. (1995). Tectonics of the Longmen Shan and Adjacent regions, central China. Int. Geol. Rev. 37, 661–735. Burchfiel, B.C., Royden, L.H., van der Hilst, R., Hager, B.H., Chen, Z., et al. (2008). A geological and geophysical context for the Wenchuan earthquake of 12 May 2008, Sichuan, People's Republic of China, GSA Today, 18, 4-11. Çakir, Z., de Chabalier, J. B., Armijo, R., Meyer, B., Barka, A., and Peltzer, G. (2003). Coseismic and early post-seismic slip associated with the 1999 Izmit earthquake (Turkey), from SAR interferometry and tectonic field observations. Geophys. J. Int., 155(1), 93-110. Calais, E., M. Vergnolle, V. Sankov, A. Lukhnev, A. Miroshnitchenko, S. Amarjargal, and J. Déverchère (2003). GPS measurements of crustal deformation in the Baikal-Mongolia area (1994 – 2002): Implications for current kinematics of Asia, J. Geophys. Res., 108 (B10), 2501, doi:10.1029/2002JB002373. Caskey, S.J., Wesnousky, S.G., Zhang, P., Slemmons, D.B. (1996). Surface faulting of the 1954 Fairview Peak (Ms 7.2) and Dixie Valley (Ms 6.9) earthquakes, Central Nevada. Bull Seismol. Soc. Am. 86, 761–787. Chang, C. H., Wu, Y. M., Shin, T. C., and Wang, C. Y. (2000). Relocation of the 1999 Chi-Chi earthquake in Taiwan. Terrestrial Atmospheric and Oceanic Sciences, 11(3), 581-590. Chen, W. P., and P. Molnar (1977). Seismic moments of major earthquakes and the average rate of slip in Central Asia, J. Geophys. Res., 82, (20), 2945-2969. Chen, Z., Burchfiel, B.C., Liu, Y., King, R.W., Royden, L.H., Tang, W., Wang, E., Zhao, J. and Zhang, X. (2000). Global Positioning System measurements from eastern Tibet and their implications for India/Eurasia intercontinental deformation, J. Geophys. Res., 105, 16215-16227 Chen, Y. G., Chen, W. S., Lee, J. C., Lee, Y. H., Lee, C. T., et al. (2001). Surface rupture of 1999 Chi-Chi earthquake yields insights on active tectonics of central Taiwan. Bull. Seism. Soc. Am., 91(5), 977-985. Chen, L., H. Wang, Y. Ran, X. Sun, G. Su, J. Wang, X. Tan, Z. Li, and X. Zhang (2010). The MS7.1 Yushu earthquake surface rupture and large historical earthquakes on the Garzê-Yushu Fault. Chinese Sci Bull, 55, 3504−3509, doi: 10.1007/s11434-010-4079-2 Chini, M., Atzori, S., Trasatti, E., Bignami, C., Kyriakopoulos, C., Tolomei, C., and Stramondo, S. (2010). The May 12, 2008,(Mw 7.9) Sichuan earthquake (China): multiframe ALOS-PALSAR DInSAR analysis of coseismic deformation. Geoscience and Remote Sensing Letters, IEEE, 7(2), 266-270. Christiansen, R. L. (1986) The Quaternary and Pliocene Yellowstone Plateau Volcanic Field of Wyoming, Idaho,and Montana. U.S. Geol. Surv. Professional Paper 729-G, 145 pp. Choi, J.-H., K. Jin, D. Enkhbayar, B. Davvasambuu, A. Bayasgalan, and Y.-S. Kim (2012). Rupture propagation inferred from damage patterns, slip distribution, and segmentation of the 1957 MW 8.1 Gobi-Altay earthquake rupture along the Bogd fault, Mongolia, J. Geophys. Res., 117, B12401, doi:10.1029/2011JB008676. Clark, M.M. (1972). Surface rupture along the Coyote Creek fault, in the Borrego Mountain Earthquake of April 9, 1968, U.S. Geol. Sur. Prof. Paper 787, 55-86. Cohee, B. P., and G. C. Beroza (1994). Slip distribution of the 1992 Landers earthquake and its implications for earthquake source mechanics. Bull. Seis. Soc. Am., 84, 692-712. Cotton, F., and M. Campillo (1995). Frequency-Domain Inversion of Strong Motions - Application to the 1992 Landers Earthquake. J. Geophys. Res.,100, 3961-3975. Crone, A. J. and Machette, M. N. (1984). Surface faulting accompanying the Borah Peak earthquake, central Idaho. Geology, 12, 664-667 Crone, A. J., and Haller, K. M. (1991). Segmentation and the coseismic behavior of Basin and Range normal faults: examples from east-central Idaho and southwestern Montana, USA. J. Struct. Geol., 13(2), 151-164.

36

Crone, A. J., Machette, M. N., Bonilla, M. G., Lienkaemper, J. J., Pierce, K. L., et al. (1987). Surface faulting accompanying the Borah Peak earthquake and segmentation of the Lost River fault, central Idaho. Bull. Seismol. Soc. Am. 77, 730–770. Crone, A. J., S. F. Personious, P. A. Craw, P. J. Haeussler, and L. A. Staft. (2004). The Susitna Glacier thrust fault— characteristics of surface ruptures on the fault that initiated the 2002 Denali Fault earthquake, Bull. Seism. Soc. Am. 94, no. 6B, S5–S22. Crosthwaite, E. G., C. A. Thomas, and K. L. Dyer (1970). Water resources in the Big Lost River basin, southcentral Idaho, U.S. Geol. Surv., Open-File Rept. 70-93, 109 p. Crowell, J. C. (1979). The San Andreas fault system through time. J. Geol. Soc. London, 136(3), 293-302. Crowell, B. W., Bock, Y., Sandwell, D. T., and Fialko, Y. (2013). Geodetic investigation into the deformation of the Salton Trough. J. Geoph. Res., 118(9), 5030-5039. Cunningham, D. (2013). Mountain building processes in intracontinental oblique deformation belts: Lessons from the Gobi Corridor, Central Asia. J. Struct. Geol., 46, 255-282. Darin, M. H., and Dorsey, R. J. (2013). Reconciling disparate estimates of total offset on the southern San Andreas fault. Geology, 41(9), 975-978. Delouis, B., Giardini, D., Lundgren, P., and Salichon, J. (2002). Joint inversion of InSAR, GPS, teleseismic, and strong- motion data for the spatial and temporal distribution of earthquake slip: Application to the 1999 Izmit mainshock. Bull. Seism. Soc. Am., 92(1), 278-299. de Michele, M., Raucoules, D., De Sigoyer, J., Pubellier, M. and Chamot-Rooke, N. (2010a). Three-dimensional surface displacement of the 2008 May 12 Sichuan earthquake (China) derived from Synthetic Aperture Radar: evidence for rupture on a blind thrust, Geophys. J. Int., 183, 1097-1103. de Michele, M., Raucoules, D., Lasserre, C., Pathier, E., Klinger, Y., Van Der Woerd, J., de Sigoyer, J. and Xu, X. (2010b). The Mw 7.9, 12 May 2008 Sichuan earthquake rupture measured by sub-pixel correlation of ALOS PALSAR amplitude images. Earth Planets and Space, 62(11), 875. Densmore, A. L., Dawers, N. H., Gupta, S., and Guidon, R. (2005). What sets topographic relief in extensional footwalls?. Geology, 33(6), 453-456. dePolo, C. M., Clark, D. G., Slemmons, D. B., and Ramelli, A. R. (1991). Historical surface faulting in the Basin and Range province, western North America: implications for fault segmentation. J. Struct. Geol., 13(2), 123-136. Deschamps, A., Gaudemer, Y., and Cisternas, A. (1982). The El Asnam, Algeria, earthquake of 10 October 1980: multiple-source mechanism determined from long-period records. Bull. Seism. Soc. Am., 72(4), 1111-1128. Dibblee, T. W., Jr. (1961). Evidence of strike-slip movement on northwest-trending faults in the western Mojave Desert, California: U.S. Geological Survey Professional Paper, 424-B, p. B197-B199. Dokka, R. K. (1983). Displacements on late Cenozoic strike-slip faults of the central Mojave Desert, California. Geology, 11, 305-308. Dokka, R. K., and C. J. Travis (1990). Late Cenozoic strike-slip faulting in the Mojave Desert, California, Tectonics, 9, 311 –340. Dickinson, W. R., Ducea, M., Rosenberg, L. I., Greene, H. G., Graham, S. A., Clark, J. C., Weber, G. E., Kidder, S., Ernst, W. G. and Brabb, E. E. (2005). Net dextral slip, Neogene San Gregorio–Hosgri fault zone, coastal California: Geologic evidence and tectonic implications. Geol. Soc. Am. Special Papers, 391, 1-43. Dokka, R. K. (1983). Displacements on late Cenozoic strike-slip faults of the central Mojave Desert, California. Geology, 11, 305-308. Dokka, R. K., and C. J. Travis (1990). Late Cenozoic strike-slip faulting in the Mojave Desert, California, Tectonics, 9, 311 –340. Dominguez, S., Avouac, J. P., and Michel, R. (2003). Horizontal coseismic deformation of the 1999 ChiChi earthquake measured from SPOT satellite images: Implications for the seismic cycle along the western foothills of central Taiwan. J. Geophys. Res., 108(B2). Dorsey, R. J. (2002). Stratigraphic record of Pleistocene initiation and slip on the Coyote Creek fault, lower Coyote Creek, southern California. SPECIAL PAPERS-GEOL. SOC. AM., 251-270. Dorsey, R. J., Axen, G. J., Peryam, T. C., and Kairouz, M. E. (2012). Initiation of the southern Elsinore fault at 1.2 Ma: evidence from the Fish Creek–Vallecito Basin, southern California. Tectonics, 31(2), p. TC2006.1-TC2006.21. Doser, D. I. (1985a). The 1983 Borah Peak, Idaho and 1959 Hebgen Lake, Montana earthquakes: models for normal fault earthquakes in the Intermountain seismic belt. In Proceedings of Workshop XXVIII—On the Borah Peak, Idaho, Earthquake: US Geological Survey Open-File Report (pp. 85-290). Doser, D. I. (1985b) Source Parameters and faulting processes of the Hebgen Lake, Montana, earthquake sequence. J. Geoph. Res., 90, B6, 4537-4555 Doser, D. I. (1986). Earthquake processes in the Rainbow Mountain-Fairview Peak-Dixie Valley, Nevada, Region 1954- 1959. Jour. Geophys. Res., 91, 12,572-12,586 Doser, D. I. (1988). Source Parameters of Earthquakes in the Nevada Seismic Zone , 1915-1943. Jour. Geophys. Res., 93, 15,001-15,015. Doser, D. I. and Smith, R. B. (1985). Source parameters of the 28 October 1983 Borah Peak, Idaho, earthquake from body wave analysis. Bull. Seism. Soc. Am., 75, 1041-1051 Dunham, E. M. and Archuleta, R. J. (2004). Evidence for a Supershear Transient during the 2002 Denali Fault Earthquake. Bull. Seism. Soc. Am., 94, No. 6B, pp. S256–S268, Eberhart-Phillips, Donna, Haeussler, P.J., Freymueller, J.T., Frankel, A.D., Rubin, C.M., Craw, P.A., Ratchkovski, N.A., Anderson, Greg, Carver, G.A., Crone, A.J., Dawson, T.E., Fletcher, Hilary, Hansen, Roger, Harp, E.L., Harris, R.A., Hill, D.P., Hreinsdottir, Sigrun, Jibson, R.W., Jones, L.M., Kayen, R.E., Keefer, D.K., Larsen, C.F., Moran, S.C.,

37

Personius, S.F., Plafker, George, Sherrod, B.L., Sieh, Kerry, Sitar, Nicholas, and Wallace, W.K. (2003). The 2002 Denali fault earthquake, Alaska—A large magnitude, slip-partitioned event. Science, 300, 1,113–1,118. Ellsworth, W. L., Celebi, M., Evans, J. R., Jensen, E. G., Kayen, R., Metz, M. C., Nyman, D. J., M.EERI, Roddick, J.W., Spudich, P., M.EERI, and Stephens, C. D. (2004). Near-field ground motion of the 2002 Denali Fault, Alaska, earthquake recorded at Pump Station 10. Earthquake Spectra, 20(3), 597-615. Etchebes, M. (2011). Paléosismologie spatiale : segmentation et scénarios de ruptures sismiques, Applications à deux failles actives décrochantes asiatiques : La faille de Fuyun et la faille de Kunlun, Chine. Thèse. Université Paris Diderot, Paris, 387 pp. Feng, G., E. A. Hetland, X. Ding, Z. Li, and L. Zhang (2010). Coseismic fault slip of the 2008 Mw 7.9 Wenchuan earthquake estimated from InSAR and GPS measurements, Geophys. Res. Lett., 37, L01302, doi:10.1029/2009GL041213 Ferrill, D. A., Stamatakos, J. A., and Sims, D. (1999). Normal fault corrugation: Implications for growth and seismicity of active normal faults. J. Struct. Geol., 21(8), 1027-1038. Fialko, Y. (2006). Interseismic strain accumulation and the earthquake potential on the southern San Andreas fault system, Nature, 441, 968–971, doi:10.1038/nature04797 Fielding, E. J., Sladen, A., Li, Z., Avouac, J. P., Bürgmann, R., and Ryder, I. (2013). Kinematic fault slip evolution source models of the 2008 M7.9 Wenchuan earthquake in China from SAR interferometry, GPS and teleseismic analysis and implications for Longmen Shan tectonics. Geoph. J. Int., 194(2), 1138-1166. Fletcher, H.J., 2002, Crustal deformation in Alaska measured using the Global Positioning System. Fairbanks, Alaska, University of Alaska Fairbanks, Ph.D. Thesis, 135 p. Fletcher, J.M., Rockwell, T.K., Hudnut, K., Teran, O., Masana, E., Faneros, G., and Galetzka, J.E. (2010). Kinematic and Dynamic Analysis of the Mayor-Cucapah Earthquake: A Case for 3-D Strain Accommodation in a Single Earthquake Cycle. In AGU Fall Meeting Abstracts (Vol. 1, p. 07). Fletcher, J. M., and 20 others (2014). Assembly of a large earthquake from a complex fault system: Surface rupture kinematics of the 4 April 2010 El Mayor–Cucapah (Mexico) Mw 7.2 earthquake. Geosphere, 10(4), 797-827. Florensov, N. A., and Solonenko, V. P. (1963). Gobi-Altayskoye zemletrayasenie. Izadtel. Akad. Nauk SSSR; also 1965 The Gobi-Altai Earthquake. Frankel, A. (2004). Rupture Process of the M 7.9 Denali Fault, Alaska, Earthquake: Subevents, Directivity, and Scaling of High-Frequency Ground Motions. Bull. Seism. Soc. Am., 94, 6B, S234–S255, Fu, B., and Lin, A. (2003). Spatial distribution of the surface rupture zone associated with the 2001 Ms 8.1 Central Kunlun earthquake, northern Tibet, revealed by satellite remote sensing data. International Journal of Remote Sensing, 24(10), 2191-2198. Fu, B., Awata, Y., Du, J., Ninomiya, Y., and He, W. (2005). Complex geometry and segmentation of the surface rupture associated with the 14 November 2001 great Kunlun earthquake, northern Tibet, China. Tectonophysics, 407(1), 43- 63. Fu, B., and Awata, Y. (2007). Displacement and timing of left-lateral faulting in the Kunlun Fault Zone, northern Tibet, inferred from geologic and geomorphic features. J. Asian Earth Sc., 29(2), 253-265. Funning, G. J., B. Parsons, and T. J. Wright (2007). Fault slip in the 1997 Manyi, Tibet earthquake from linear elastic modelling of InSAR displacements, Geophys. J. Int., 169, 988–1008, doi:10.1111/j.1365- 246X.2006.03318.x. Funning, G. J., Fukahata, Y., Yagi, Y., and Parsons, B. (2014). A method for the joint inversion of geodetic and seismic waveform data using ABIC: application to the 1997 Manyi, Tibet, earthquake. Geoph. J. Int., 196(3), 1564-1579. Gan, W., P. Zhang, Z.-K. Shen, Z. Niu, M. Wang, Y. Wan, D. Zhou, and J. Cheng (2007), Present-day crustal motion within the Tibetan Plateau inferred from GPS measurements, J. Geophys. Res., 112, B08416, doi:10.1029/2005JB004120 Gaudemer, Y., Tapponnier, P., and Turcotte, D. L. (1989). River offsets across active strike-slip faults. Ann. Tecton, 3(2), 55-76. Gold, P.O., Oskin, M.E., Elliott, A.J., Hinojosa-Corona, A., Taylor, M.H., Kreylos, O. and Cowgill, E. (2013). Coseismic slip variation assessed from terrestrial lidar scans of the El Mayor–Cucapah surface rupture. Earth Planet. Sci. Lett., 366, 151–162. Grall, C., Henry, P., Thomas, Y., Westbrook, G. K., Çağatay, M. N., Marsset, B., Saritas, H., Çifçi, G. and Geli, L. (2013). Slip rate estimation along the western segment of the Main Marmara Fault over the last 405–490 ka by correlating mass transport deposits. Tectonics, 32(6), 1587-1601. Grant Ludwig, L., S. O. Akciz, G. N. Noriega, O. Zielke, and J. R. Arrowsmith (2010). Climate modulated channel incision and rupture history of the San Andreas fault in the Carrizo Plain, Science, 327, 5969, 1117– 1119, doi 10.1126/science.1182837. Guiraud, R. (1977). Sur la néotectonique des régions ouest constantinoises. Bull. Soc. Géol. Fr., 7, 645-650. Gurrola, L. D., and T. K. Rockwell. 1996. Timing and slip for prehistoric earthquakes on the Superstition Mountain fault, Imperial Valley, southern California, J. Geophys. Res., 101, 5977–5985, doi:10.1029/95JB03061. Haeussler, P.J., Schwartz, D.P., Dawson, T.E., Stenner, H.D., Lienkaemper, J.J., et al. (2004). Surface rupture and slip distribution of the Denali and Totschunda faults in the 3 November 2002 M7.9 earthquake, Alaska. Bull. Seism. Soc. Am., 94, 6B, S23–S52. Haeussler, P.J. (2009). Surface Rupture Map of the 2002 M7.9 Denali Fault Earthquake, Alaska; Digital Data. U.S. Geological Survey Data Series 422 [http://pubs.usgs.gov/ds/422/]. Harkins, N., Kirby, E., Shi, X., Wang, E., Burbank, D., and Chun, F. (2010). Millennial slip rates along the eastern Kunlun fault: Implications for the dynamics of intracontinental deformation in Asia. Lithosphere, 2(4), 247-266.

38

Hart, E. W., W. A. Bryant, J. E. Kahle, M. W. Manson, and W. J. Bortugno (1988), Summary report: Fault Evaluation Program, 1986– 1987, Mojave Desert region and other areas, Open File Rep. 88-1, 40 pp., Calif. Dep. of Conserv., Div. of Mines and Geol., Sacramento. Hartzell, S. H., and Heaton, T. H. (1983). Inversion of strong ground motion and teleseismic waveform data for the fault rupture history of the 1979 Imperial Valley, California, earthquake. Bull. Seism. Soc. Am., 73(6A), 1553-1583. Hashimoto, M., Enomoto, M., and Fukushima, Y. (2010). Coseismic deformation from the 2008 Wenchuan, China, earthquake derived from ALOS/PALSAR images. Tectonophysics, 491(1), 59-71. Hauksson, E., L. M. Jones, and K. Hutton (2002). The 1999 Mw 7.1 Hector Mine, California, earthquake sequence: complex conjugate strike-slip faulting, Bull. Seism. Soc. Am. 92, 1154–1170 Hauksson, E., W. Yang, and P.M. Shearer (2012). Waveform Relocated Earthquake Catalog for Southern California (1981 to June 2011); Bull. Seismol. Soc. Am., 102, doi:10.1785/0120120010. Hernandez, B., F. Cotton, and M. Campillo (1999). Contribution of radar interferometry to a two-step inversion of the kinematic process of the 1992 Landers earthquake. J. Geophys. Res.,104, 13,083-13,099. Hergert, T., and Heidbach, O. (2010). Slip-rate variability and distributed deformation in the Marmara Sea fault system. Nature Geoscience, 3(2), 132-135. Hill, M. L. and T. W. Dibblee, E. Jr. (1953). San Andreas, Garlock and Big Pine faults, California, a study of the character, history and tectonic significance of their displacements. Geol. Soc. Am. Bull. 64, 4, 443-458, doi:10.1130/00167606(1953)64[443:SAGABP]2.0.CO;2 Holt, W. E., Chamot-Rooke N., Le Pichon, X., Haines, A. J., Shen-Tu, B. and Ren, J. (2000). Velocity field in Asia inferred from Quaternary fault slip rates and Global Positioning System observations. J. Geoph. Res., 105, 19,185- 19,209. Hubert-Ferrari, A., Armijo, R., King, G., Meyer, B., and Barka, A. (2002). Morphology, displacement, and slip rates along the North Anatolian Fault, Turkey. J. Geoph. Res., 107(B10), ETG-9. Hubert-Ferrari, A., King, G., Manighetti, I., Armijo, R., Meyer, B., and Tapponnier, P. (2003). Long-term elasticity in the continental lithosphere; modelling the Aden Ridge propagation and the Anatolian extrusion process. Geoph. J. Int., 153(1), 111-132. Hudnut, K. W., and K. Sieh. (1989). Behavior of the Superstition Hills fault during the past 330 years, Bull. Seismol. Soc. Am., 79, 304–329. Hudnut, K.W., L. Seeber, T. Rockwell, J. Goddmacher, R. Klinger, S. Lindvail, and R. McElwain. (1989). Surface ruptures on cross-faults in the 24 November 1987 Superstition Hills, California, earthquake sequence, Bull. Seismol. Soc. Am., 79, 282-296 Hudnut, K. W., Y. Bock, M. Cline, P. Fang, Y. Feng, J. Freymueller, X. Ge, W. K. Gross, D. Jackson, M. Kim, N. E. King, J. Langbein, S. C. Larsen, M. Lisowski, Z.-K. Shen, J. Svarc, and J. Zhang (1994). Co-seismic displacements of the 1992 landers earthquake sequence. Bull. Seism. Soc. Am., 84, 625-645 Hull, A. G., and C. Nicholson (1992), Seismotectonics of the northern Elsinore fault zone, southern California, Bull. Seismol. Soc. Am., 82, 800–818. Hussain, A., Yeats, R. S. and MonaLisa (2009). Geological setting of the 8 October 2005 Kashmir earthquake. J. Seism., 13(3), 315-325. Jennings, C. W. (1994). Fault activity map of California and adjacent areas, California Department of Conservation, Division of Mines and Geology California Geologic Data Map Series Map 6, scale 1:750,000. Jachens, R. C., Langenheim, V. E., and Matti, J. C. (2002). Relationship of the 1999 Hector Mine and 1992 Landers fault ruptures to offsets on Neogene faults and distribution of late Cenozoic basins in the Eastern California Shear Zone. Bull. Seism. Soc. Am., 92(4), 1592-1605. Jackson, J., and Leeder, M. (1994). Drainage systems and the development of normal faults: an example from Pleasant Valley, Nevada. J. Struct. Geol., 16(8), 1041-1059. Janecke, S.U., Dorsey, R.J., Forand, D., Steely, A.N., Kirby, S.M., Lutz, A.T., Housen, B.A., Belgarde, B., Langenheim, V.E., and Rittenour, T.M. (2010), High Geologic Slip Rates since Early Pleistocene Initiation of the San Jacinto and San Felipe Fault Zones in the San Andreas Fault System: Southern California, USA: Geol. Soc. Am. Special Paper 475, 48 p., doi: 10.1130/2010.2475. isbn 9780813724751 Jayangondaperumal, R., and Thakur, V. C. (2008). Co-seismic secondary surface fractures on southeastward extension of the rupture zone of the 2005 Kashmir earthquake. Tectonophysics, 446(1), 61-76. Ji, C., J. D. Wald, and D. V. Helmberger (2002). Source description of the 1999 Hector Mine, California earthquake. II. Complexity of slip history, Bull. Seism. Soc. Am., 92, no. 4, 1208–1226. Jianbang, S., F. Xianyue, G. Shumo, Y. Zhang, B. Meixiang, H. Jun et al. (1984). The Fuyun earthquake fault zone in Xinjiang, China, presented at International symposium on continental seismicity and earthquake prediction (ISCSEP), Seismological Press, Beijing, 325-346 Jin, W., Tang, L., Yang, K., Wan, G., and Lü, Z. (2010). Segmentation of the Longmen Mountains thrust belt, western Sichuan foreland basin, SW China. Tectonophysics, 485(1), 107-121. Johnson, K. M., Hsu, Y. J., Segall, P., and Yu, S. B. (2001). Fault geometry and slip distribution of the 1999 ChiChi, Taiwan Earthquake imaged from inversion of GPS data. Geoph. Res. Lett., 28(11), 2285-2288. Jolivet, M., Brunel, M., Seward, D., Xu, Z., Yang, et al. (2003). Neogene extension and volcanism in the Kunlun Fault Zone, northern Tibet: New constraints on the age of the Kunlun Fault. Tectonics, 22(5), p7.1-7.20 Jones, J. C. (1915). The Pleasant Valley, Nevada, earthquake of October 2,1915. Bull. Seismol. Soc. Am., 5, no. 4, p. 190- 205. Jónsson, S., H. Zebker, P. Segall, and F. Amelung (2002). Fault slip distribution of the 1999 Mw 7.1 Hector Mine, California, earthquake, estimated from satellite radar and GPS measurements, Bull. Seismol. Soc. Am., 92, 1377– 1389, doi:10.1785/0120000922

39

Jouanne, F., Awan, A., Pêcher, A., Kausar, A., Mugnier, J. L., Khan, I., Khan, N. A. and Van Melle, J. (2014). Present day deformation of northern Pakistan from Salt Ranges to Karakorum Ranges. J. Geoph. Res., 119(3), 2487-2503. Kalbas, J. L., Freed, A. M., and Ridgway, K. D. (2008). Contemporary fault mechanics in southern Alaska. Active Tectonics and Seismic Potential of Alaska, 321-336. Kaneda, H., Nakata, T., Tsutsumi, H., Kondo, H., Sugito, et al. (2008). Surface rupture of the 2005 Kashmir, Pakistan, earthquake and its active tectonic implications. Bull. Seism. Soc. Am., 98(2), 521-557. Kaverina, A., Dreger, D., and Price, E. (2002). The combined inversion of seismic and geodetic data for the source process of the 16 October 1999 Mw 7.1 Hector Mine, California, earthquake. Bull. Seism. Soc. Am., 92(4), 1266- 1280. Kendrick, K. J., D. M. Morton, S. G. Wells, and R. W. Simpson. 2002. Spatial and temporal deformation along the northern San Jacinto fault, southern California: Implications for slip rates, Bull. Seismol. Soc. Am., 92, 2782– 2802, doi:10.1785-/0120000615. King, G., and Yielding, G. (1984). The evolution of a thrust fault system: processes of rupture initiation, propagation and termination in the 1980 El Asnam (Algeria) earthquake. Geoph. J. Int., 77(3), 915-933. Kirby, E., Reiners, P.W., Krol, M.A., Whipple, K.X., Hodges, K.V., Farley, K.A., Tang, W., and Chen, Z. (2002). Late Cenozoic evolution of the eastern margin of the Tibetan Plateau: inferences from 40Ar/39Ar and (U– Th)/He thermochronology. Tectonics 21, 1001. doi:10.1029/2000TC001246. Kirby, S. M., Janecke, S. U., Dorsey, R. J., Housen, B. A., Langenheim, V. E., McDougall, K. A., and Steely, A. N. (2007). Pleistocene Brawley and Ocotillo formations: Evidence for initial strike-slip deformation along the San Felipe and San Jacinto fault zones, southern California. The Journal of geology, 115(1), 43-62. Kirby, E., Harkins, N., Wang, E., Shi, X., Fan, C. and Burbank, D. (2007). Slip rate gradients along the eastern Kunlun fault, Tectonics, 26, TC2010, doi:10.1029/2006TC002033. Klinger, Y., Xu, X., Tapponnier, P., Van Der Woerd, J., Lasserre, C. and King, G. (2005). High-resolution satellite imagery mapping of the surface rupture and slip distribution of theMw 7.8, November 14, 2001Kokoxili earthquake (Kunlun Fault, Northern Tibet, China). Bull. seism. Soc. Am., 95, 1970–1987, doi:10.1785/0120040233. Klinger, Y., Michel, R., and King, G. C. (2006). Evidence for an earthquake barrier model from Mw 7.8 Kokoxili (Tibet) earthquake slip-distribution. Earth Planet. Sc. Lett., 242(3), 354-364. Klinger, Y., M. Etchebes, P. Tapponnier and C. Narteau (2011), Characteristic slip for five great earthquakes along the Fuyun fault in China. Nat. Geosci. 4, 389–392, doi:10.1038/ngeo1158 Kobayashi, T., Takada, Y., Furuya, M. and Murakami, M. (2009). Locations and types of ruptures involved in the 2008 Sichuan earthquake inferred from SAR image matching, Geoph. Res. Lett., 36, L07302, doi:07310.01029/02008GL036907. Kurushin, R. A., A. Bayasgalan, M. Olziybat, B. Enhtuvshin, P. Molnar, C. Bayarsayhan, K. W. Hudnut, and J. Lin (1997). The surface rupture of the 1957 Gobi-Altay, Mongolia, earthquake, Spec. Pap. Geol. Soc. Am., 320, 143 pp. Larsen, S., and Reilinger, R. (1991). Age constraints for the present fault configuration in the Imperial Valley, California: Evidence for northwestward propagation of the Gulf of California rift system. J. Geophys. Res., 96(B6), 10339- 10346. Lasserre C., Peltzer, G., Klinger, Y., Van Der Woerd, J. and Tapponnier, P. (2005). Coseismic deformation of the 2001 Mw = 7.8 Kokoxili earthquake in Tibet, measured by synthetic aperture radar interferometry, J. geophys. Res., 110, B12408, doi:10.1029/2004JB003500. Lawson, A. C. -Chairman- (1908) The California Earthquake of April 18, 1906, Report of the State Earthquake Investigation Commission vol. I and II, Carnegie Inst. of Washington, Washington, D.C. Lee, J. C., Chen, Y. G., Sieh, K., Mueller, K., Chen, et al. (2001). A vertical exposure of the 1999 surface rupture of the Chelungpu fault at Wufeng, western Taiwan: structural and paleoseismic implications for an active thrust fault. Bull. Seism. Soc. Am., 91(5), 914-929. Li Haibing, Van der Woerd, J., Tapponnier, P., Klinger, Y., Qi Xuexiang, Yang Jingsui and Zhu Yintang. (2005). Slip rate on the Kunlun Fault at Hongshui Gou, and recurrence time of great events comparable to the 14/11/2001, Mw 7.9 Kokoxili earthquake, Earth Planet. Sci. Lett., 237, 285 – 299. Li, Z., J. R. Elliott, W. Feng, J. A. Jackson, B. E. Parsons, and R. J. Walters (2011). The 2010 MW 6.8 Yushu (Qinghai, China) earthquake: Constraints provided by InSAR and body wave seismology, J. Geophys. Res., 116, B10302, doi:10.1029/2011JB008358. Li, C.-Y., Pang, J.-Z. and Zhang Z.-Q. (2012). Characteristics, Geometry, and Segmentation of the Surface Rupture Associated with the 14 April 2010 Yushu Earthquake, Eastern Tibet, China. Bull. Seism. Soc. Am., 102, 4, 1618– 1638, doi: 10.1785/0120110261 Lin, A., and S. Lin (1998). Tree damage and surface displacement: The 1931 M 8.0 Fuyun earthquake, J. Geol., 106, 751- 757. Lin, A., Fu, B., Guo, J., Zeng, Q., Dang, G., He, W., and Zhao, Y. (2002). Co-seismic strike-slip and rupture length produced by the 2001 Ms 8.1 Central Kunlun earthquake. Science, 296(5575), 2015-2017. Lin, A., Kikuchi, M., and Fu, B. (2003). Rupture segmentation and process of the 2001 Mw 7.8 central Kunlun, China, earthquake. Bull. Seism. Soc. Am., 93(6), 2477-2492. Lin, A.M., Ren, Z.K., Jia, D. and Wu, X.J. (2009). Co-seismic thrusting rupture and slip distribution produced by the 2008 M-w 7.9 Wenchuan earthquake, China, Tectonophysics, 471, 203-215. Lin, A.M., Ren, Z.K., and Jia, D. (2010). Co-seismic ground-shortening structures produced by the 2008 Mw 7.9 Wenchuan earthquake, China. Tectonophysics 491, 21–34 Lin, A., G. Rao, D. Jia, X. Wu, and Z. Ren (2011). Co-seismic strike-slip surface rupture and displacement produced by the 2010 Mw 6.9 Yushu earthquake, China, and implications for Tibetan tectonics, J. Geodyn. 52, 249–259.

40

Lin, A., Rao, G., and Yan, B. (2012). Field evidence of rupture of the Qingchuan Fault during the 2008 Mw7.9 Wenchuan earthquake, northeastern segment of the Longmen Shan Thrust Belt, China, Tectonophysics, 522–523, 5, 243-252, ISSN 0040-1951, 10.1016/j.tecto.2011.12.012. Lindsey, E. O., and Y. Fialko (2013). Geodetic slip rates in the southern San Andreas Fault system: Effects of elastic heterogeneity and fault geometry, J. Geophys. Res. Solid Earth, 118, 689–697, doi:10.1029/2012JB009358. Lindsey, E. O., and Fialko, Y. (2016). Geodetic constraints on frictional properties and earthquake hazard in the Imperial Valley, Southern California. J. Geoph. Res.. Liu, M., Wang, H., and Li, Q. (2010). Inception of the eastern California shear zone and evolution of the Pacific‐North American plate boundary: From kinematics to geodynamics. J. Geophys. Res., 115(B7). Liu, Y. H., Shan, X. J., Qu, C. Y., and Zhang, G. F. (2011). Earthquake deformation field characteristics associated with the 2010 Yushu Ms7.1 earthquake. Sci China Earth Sci, 54, 571–580, doi: 10.1007/s11430- 010-4116-7 Liu-Zeng, J., Zhang, Z., Wen, L., Tapponnier, P., Sun, J., et al. (2009). Co-seismic ruptures of the 12 May, 2008, Ms 8.0 Wenchuan earthquake, Sichuan: East-west crustal shortening on oblique, parallel thrusts along the eastern edge of Tibet, Earth Planet. Sc. Lett., 286, 355-370. Loevenbruck, A., Cattin, R., Le Pichon, X., Courty, M. L., and Yu, S. B. (2001). Seismic cycle in Taiwan derived from GPS measurements. Comptes Rendus de l'Académie des Sciences-Series IIA-Earth and Planetary Science, 333(1), 57- 64. Lonsdale, P. (1989). Geology and tectonic history of the Gulf of California. The Geology of North America, 499-521. Lowey, G.W. (1998). A new estimate of the amount of displacement on the Denali fault system based on the occurrence of carbonate megaboulders in the Dezadeash Formation (Jura–Cretaceous), Yukon, and the Nutzotin Mountains Sequence (Jura–Cretaceous), Alaska. Bull. Canad. Petrol. Geol., 94, 4,333–4,359. Lu, C. -Y., and Hsü, K. -J., 1992, Tectonic evolution of the Taiwan mountain belt : Petroleum Geology of Taiwan, v. 27, p. 21-46. Lundgren, P., E. A. Hetland, Z. Liu, and E. J. Fielding. (2009). Southern San Andreas-San Jacinto fault system slip rates estimated from earthquake cycle models constrained by GPS and interferometric synthetic aperture radar observations, J. Geophys. Res., 114, B02403, doi:10.1029/2008JB005996 Lutz, A. T., Dorsey, R. J., Housen, B. A., and Janecke, S. U. (2006). Stratigraphic record of Pleistocene faulting and basin evolution in the Borrego Badlands, San Jacinto fault zone, Southern California. Geol. Soc. Am. Bull., 118(11-12), 1377-1397. Lyons, S. N., Bock, Y., and Sandwell, D. T. (2002). Creep along the Imperial Fault, southern California, from GPS measurements. J. Geoph. Res., 107(B10). Ma, K. F., Mori, J., Lee, S. J., and Yu, S. B. (2001). Spatial and temporal distribution of slip for the 1999 Chi-Chi, Taiwan, earthquake. Bull. Seism. Soc. Am., 91(5), 1069-1087. Ma, K. F., Brodsky, E. E., Mori, J., Ji, C., Song, T. R. A., and Kanamori, H. (2003). Evidence for fault lubrication during the 1999 Chi-Chi, Taiwan, earthquake (Mw7. 6). Geoph. Res. Lett., 30(5). Mai, P. M., and Thingbaijam, K. K. S. (2014). SRCMOD: An online database of finite‐fault rupture models. Seismological Research Letters, 85(6), 1348-1357. Manighetti, I., Campillo, M., Bouley, S., and Cotton, F. (2007). Earthquake scaling, fault segmentation, and structural maturity. Earth Planet. Sci. Lett., 253(3), 429-438. Marechal, A. (2015). Active tectonics of the Yakutat - North America collision zone : GPS and geomorphology contribution to the study of strain partitioning. Université Montpellier 2. PhD Thesis. 285pp. Marliyani, G. I., Rockwell, T. K., Onderdonk, N. W., and McGill, S. F. (2013). Straightening of the Northern San Jacinto Fault, California, as Seen in the FaultStructure Evolution of the San Jacinto Valley Stepover. Bull. Seism. Soc. Am., 103(3), 2047-2061. Matmon, A., Schwartz, D.P., Haeussler, P.J., Finkel, R., Lienkaemper, J.J., et al. (2006). Denali fault slip rates and Holocene–late Pleistocene kinematics of central Alaska. Geology, 34, 645–648. McClusky, S.,et al., (2000). Global Positioning System constraints on plate kinematics and dynamics in the eastern Mediterranean and Caucasus. J. Geoph. Res., 105(B3), 5695-5719. McClusky, S. C., S. C. Bjornstad, B. H. Hager, R. W. King, B. J. Meade, M. M. Miller, F. C. Monastero, and B. J. Souter (2001). Present day kinematics of the Eastern California Shear Zone from a geodetically constrained block model, Geophys. Res. Lett., 28, 3369–3372. Meade, B. (2007). Present-day kinematics at the India-Asia collision zone. Geology, 35, 81—84, doi: 10.1130/G22924A.1 Meade, B. J., and B. H. Hager (2005), Block models of crustal motion in southern California constrained by GPS measurements, J. Geophys. Res., 110, B03403, doi:10.1029/2004JB003209. Mendoza, C., and S. H. Hartzell (1988). Inversion for slip distribution using teleseismic P waveforms; North Palm Springs, Borah Peak, and Michoacan earthquakes. Bull. Seis. Soc. Am. 78 (3),1092-1111. Mériaux, A.-S., K. Sieh, R. C. Finkel, C. M. Rubin, M. H. Taylor, A. J. Meltzner, and F. J. Ryerson (2009). Kinematic behavior of southern Alaska constrained by westward decreasing postglacial slip rates on the Denali Fault, Alaska, J. Geophys. Res., 114, B03404, doi:10.1029/2007JB005053 Meyer, B., Tapponnier, P., Bourjot, L., Metivier, F., Gaudemer, Y., Peltzer, G., Shunmin, G. and Zhitai, C. (1998). Crustal thickening in Gansu-Qinghai, lithospheric mantle subduction, and oblique, strike-slip controlled growth of the Tibet plateau. Geoph. J. Int., 135(1), 1-47. Michel, R., and Avouac, J. P. (2002). Deformation due to the 17 August 1999 Izmit, Turkey, earthquake measured from SPOT images. J. Geophys. Res., 107(B4), ETG-2.

41

Millman, D. E., and Rockwell, T. K. (1986). Neotectonics of the Elsinore fault in Temescal Valley, California. Guidebook and Volume on Neotectonics and Faulting in Southern California, P. Ehlig, ed., Cordilleran Section, Geol. Soc. Am., California State University, Fresno, CA, 159-166. Molnar, P. and Tapponnier, P. (1975). Cenozoic Tectonics of Asia : Effects of a Continental Collision. Science, 189, 419– 426. Morton, D. M., and F. C. Miller (1987), K/Ar apparent ages of plutonic rocks from the northern part of the Peninsular Ranges batholith, southern California, Geol. Soc. Am. Abstr. Programs, 19, 435. Murphy, L. M. and Brazee, R. J. (1964). Seismological investigations of the Hebgen Lake earthquake. U.S. Geol. Surv. Professional Paper 435-I, 13–17. Myers, W.B. and Hamilton, W. (1964). Deformation accompanying the Hebgen Lake earthquake of August 17, 1959. U.S. Geol. Surv. Professional Paper 435-I, 55–98. Nicholson, C., Sorlien, C. C., Atwater, T., Crowell, J. C., and Luyendyk, B. P., (1994). Microplate capture, rotation of the western Transverse Ranges, and initiation of the San Andreas transform as a low-angle fault system. Geology, 22(6), 491-495. Nokleberg, W.J., Jones, D.L., and Silberling, N.J. 1985. Origin and tectonic evolution of the Maclaren and Wrangellia terranes, eastern Alaska Range, Alaska. Geol. Soc. Am. Bulletin, 96, 1,251–1,270 Oglesby, D. D., D. S. Dreger, R. A. Harris, N. Ratchkovski, and R. Hansen. (2004). Inverse kinematic and forward dynamic models of the 2002 Denali, Alaska earthquake, Bull. Seismol. Soc. Am., 94(6B), 214–233. Oskin, M., L. Perg, D. Blumentritt, S. Mukhopadhyay, and A. Iriondo (2007). Slip rate of the Calico fault Implications for geologic versus geodetic rate discrepancy in the Eastern California Shear Zone, J. Geophys. Res., 112 , B03402, doi:10.1029/2006JB004451. Oskin, M.E., Arrowsmith, J.R., Corona, A.H., Elliott, A.J., Fletcher, J.M., Fielding, E.J., Gold, P.O., Garcia, J.J.G., Hudnut, K.W., Liu-Zeng, J., and Teran, O.J. (2012). Near-Field Deformation from the El Mayor–Cucapah Earthquake Revealed by Differential LIDAR. Science, 335, 702–705. Pamir, H. N., and Ketin, I. (1941). Das anatolische Erdbeben ende 1939. Geologische Rundschau, 32(3), 279-287. Parejas, E., Akyol, I. H., and Altinli, E. (1942). Le tremblement de terre d’Erzincan du 17 Decembre 1939. Rev. Fac. Sci. Univ. Istanbul, 177-222. Pathier, E., Fielding, E. J., Wright, T. J., Walker, R., Parsons, B. E., and Hensley, S. (2006). Displacement field and slip distribution of the 2005 Kashmir earthquake from SAR imagery. Geoph. Res. Lett., 33(20). Payne, S. J., McCaffrey, R., and King, R. W. (2008). Strain rates and contemporary deformation in the Snake River Plain and surrounding Basin and Range from GPS and seismicity. Geology, 36(8), 647-650. Peltzer, G., K. W. Hudnut and K. Feigl (1994). Analysis of coseismic surface displacement gradients using radar interferometry: New insights into the Landers earthquake, J. Geophys. Res., 99, 21,971-21,981. Peltzer, G., F. Crampé, and G. King (1999). Evidence of nonlinear elasticity of the crust from Mw7.6 Manyi (Tibet) earthquake, Science, 286, 272–276, doi:10.1126/science.286.5438.272. Peltzer, G., Crampé, F., and Rosen, P. (2001). The Mw 7.1 Hector Mine, California earthquake, surface rupture, surface displacement field and fault slip solution from ERS SAR data. Academie des Sciences, Paris, Comptes Rendus, 333, 545–555 Perkins, J. A., Sims, J. D. and Sturgess, S. S. (1989). Late Holocene Movement Along the San Andreas Fault at Melendy Ranch : Implications for the Distribution of Fault Slip in Central California. J. of Geoph. Res., 94, 10,217-10,230. Perrin, C., Manighetti, I., and Gaudemer, Y. (2016). Off-fault tip splay networks: A genetic and generic property of faults indicative of their long-term propagation. Comptes Rendus Geoscience. 348, 52-60. Peyrat, S., Olsen, K., and Madariaga, R. (2001). Dynamic modeling of the 1992 Landers earthquake. J. Geoph. Res., 106(B11), 26467-26482. Philip, H., and Meghraoui, M. (1983). Structural analysis and interpretation of the surface deformations of the El Asnam earthquake of October 10, 1980. Tectonics, 2(1), 17-49. Philip, H. and Thomas, G. (1977). Détermination de la direction de raccourcissement de la phase de compression quaternaire en Oranie (Algérie). Rev. Géogr. Phys. Géol. Dynam., 19, 315-324. Plafker, G. and Berg, H. C. (1994), The Geology of Alaska. Geol. Soc. Am., 1068 p. Powell, R.E., and Weldon, R.J., II (1992), Evolution of the San Andreas fault: Annual Review of Earth and Planetary Sciences, 20, 431–468, doi:10.1146/annurev.ea.20.050192.002243. Puskas, C. M., R. B. Smith, C. M. Meertens, and W. L. Chang (2007), Crustal deformation of the Yellowstone–Snake River Plain volcano-tectonic system: Campaign and continuous GPS observations, 1987–2004, J. Geophys. Res., 112, B03401, doi:10.1029/2006JB004325. Qu, C., Zhang, G., Shan, X., Zhang, G., Song, X., Liu, Y. (2012). Coseismic deformation derived from analyses of C and L band SAR data and fault slip inversion of the Yushu Ms7.1 earthquake, China in 2010. Tectonophysics 584, 119– 128 Ratchkovski, N. A., R. A. Hansen, J. C. Stanching, T. Cox, O. Fox, L. Rio, E. Clark, M. Lafevers, S. Estes, J. B. MacCormack, and T. Williams (2003). Aftershock sequence of theMw 7.9 Denali fault, Alaska, earthquake of 3 November 2002 from regional seismic network data, Seism. Res. Lett. 74, 743–752. Reilinger, R. E., Ergintav, S., Bürgmann, R., McClusky, S., Lenk, O., et al. (2000). Coseismic and postseismic fault slip for the 17 August 1999, M= 7.5, Izmit, Turkey earthquake. Science, 289(5484), 1519-1524. Revenaugh, J., and C. Reasoner (1997) Cumulative offset of the San Andreas fault in central California: A seismic approach, Geology, 25, 123-126. Richins, W. D., Pechmann, J. C., Smith, R. B., Langer, C. J., Goter, S. K., Zollweg, J. E., and King, J. J. (1987). The 1983 Borah Peak, Idaho, earthquake and its aftershocks. Bull. Seism. Soc. Am., 77(3), 694-723. Richter, C. F. (1958). Elementary seismology. Freeman and co. San Francisco

42

Richter, D. H., and Matson, N. A. (1971). Quaternary faulting in the eastern Alaska Range. Geol. Soc. Am. Bulletin, 82(6), 1529-1540. Ritz, J.-F., E. T. Brown, D. L. Bourles, H. Philip, A. Schlupp, G.M. Raisbeck, F. Yiou, and B. Enkhtuvshin (1995). Slip rates along active faults estimated with cosmic-ray-exposure dates: Application to the Bogd Fault, Gobi-Altai, Mongolia, Geology, 23, 1019–1022, doi:10.1130/0091-7613(1995)023<1019:SRAAFE>2.3.CO;2. Ritz, J.-F., R. Vassallo, R. Braucher, E. T. Brown, S. Carretier, and D. L. Bourles (2006). Using in situ produced 10 Be to quantify active tectonics in the Gurvan Bogd mountain range (Gobi-Altay, Mongolia), Spec. Pap. Geol. Soc. Am., 415, 87–110. Rizza, M., J.-F. Ritz, R. Braucher, R. Vassallo, C. Prentice, S. Mahan, S. McGill, A. Chauvet, S. Marco, M. Todbileg, S. Demberel and D. Bourlès (2011). Slip rate and slip magnitudes of past earthquakes along the Bogd left-lateral strike- slip fault (Mongolia). Geophys. J. Int., 186, 897–927, doi:10.1111/j.1365- 246X.2011.05075.x. Robinson, D. P., C. Brough, and S. Das (2006). The Mw 7.8, 2001 Kunlunshan earthquake: Extreme rupture speed variability and effect of fault geometry, J. Geophys. Res., 111, B08303, doi:10.1029/2005JB004137 Rockwell, T. K., and Klinger, Y. (2013). Surface rupture and slip distribution of the 1940 Imperial Valley earthquake, Imperial Fault, Southern California: Implications for rupture segmentation and dynamics. Bull. Seism. Soc. Am., 103(2A), 629-640. Rockwell, T. K., C. Loughman, and P. Merifield. (1990). Late Quaternary rate of slip along the San Jacinto fault zone near Anza, southern California, J. Geophys. Res., 95, 8593–8605, doi:10.1029/JB095iB06p08593. Rockwell, T., Bergmann, M., and Kenney, M. (2000a). Holocene slip rate of the Elsinore fault in Temecula Valley, riverside county, California. Geology and Enology of the Temecula Valley, 105-118. Rockwell, T. K., S. Lindvall, M. Herzberg, D. Murbach, T. Dawson, and G. Berger (2000b). Paleoseismology of the Johnson Valley, Kickapoo, and Homestead Valley faults: Clustering of earthquakes in the Eastern California Shear Zone, Bull. Seismol. Soc. Am., 90, 1200–1236. Rockwell, T., Ragona, D., Seitz, G., Langridge, R., Aksoy, M. E., Ucarkus, G., Ferry, M., Meltzner, A. J., Klinger, Y., Meghraoui, M., Stair, D., Barka, A. and Akbalik, B. (2009). Palaeoseismology of the North Anatolian Fault near the Marmara Sea: implications for fault segmentation and seismic hazard. Geological Society, London, Special Publications, 316(1), 31-54. Rubin, C. M., and K. Sieh (1997). Long dormancy, low slip rate, and similar slip-per-event for the Emerson fault, eastern California shear zone, J. Geophys. Res., 102, 15,319– 15,333. Ruegg, J. C., Kasser, M., Tarantola, A., Lepine, J. C., and Chouikrat, B. (1982). Deformations associated with the El Asnam earthquake of 10 October 1980: Geodetic determination of vertical and horizontal movements. Bull. Seism. Soc. Am., 72(6A), 2227-2244. Ryall, A. (1962) The Hebgen lake, Montana, earthquake of august 18, 1959 : P Waves. Bull. Seism. Soc. Am., 52, 2, 235- 271. Salichon, J., P. Lundgren, B. Delouis, and D. Giardini (2004). Slip History of the 16 October 1999 Mw 7.1 Hector Mine Earthquake (California) from the Inversion of InSAR, GPS, and Teleseismic Data. Bull. Seis. Soc. Am., 94, 2015- 2027. Sandwell, D. T., L. Sichoix, D. Agnew, Y. Bock, and J-B. Minster (2000). Near-real-time radar interferometry of the Mw 7.1 Hector Mine Earthquake, Geophys. Res. Lett. 27, 3101–3104. Sandwell, D. T., L. Sichoix and B. Smith (2002). The 1999 Hector Mine Earthquake, Southern California: Vector Near- Field Displacements from ERS InSAR Bull. Seism. Soc. Am., 92, 1341-1354, doi:10.1785/0120000901 Sauber, J., W. Thatcher, S. C. Solomon, and M. Lisowski (1994). Geodetic slip rate for the eastern California shear zone and the recurrence time of Mojave desert earthquakes, Nature, 367, 264– 266. Schindler, C. (1998), Geology of NW Turkey: Results of the Marmara polyproject, in Active Tectonics of Northwestern Anatolia: The Marmara Poly-Project, a Multidisciplinary Approach by Space-Geodesy, Geology, Hydrology, Geothermics and Seismology, edited by C. Schindler and M. Pfister, pp. 329–375, Verlag der Fachvereine, Zurich, Switzerland. Scholz, C. H. (1977). Transform fault systems of California and New Zealand: similarities in their tectonic and seismic styles. J. Geol. Soc., 133(3), 215-228. Scholz, C. H., Ando, R., and Shaw, B. E. (2010). The mechanics of first order splay faulting: The strike-slip case. J. Struct. Geol., 32(1), 118-126. Schwartz, D. P., and Coppersmith, K. J. (1984). Fault behavior and characteristic earthquakes: Examples from the Wasatch and San Andreas fault zones. J. Geoph. Res., 89(B7), 5681-5698. Schwartz, D. P., P. J. Haeussler, G. G. Seitz, and T. E. Dawson. (2012). Why the 2002 Denali fault rupture propagated onto the Totschunda fault: Implications for fault branching and seismic hazards, J. Geophys. Res., 117, B11304, doi:10.1029/2011JB008918 Scott, W.E., Pierce, K. L. and Hait, M. H. Jr. (1985) Quaternary tectonic setting of the 1983 Borah Peak earthquake, Central Idaho, Bull. Seism. Soc. Am., 75, 1053-1066. Segall, P. and Harris, R. (1986). Slip Deficit on the San Andreas Fault at Parkfield, California, as Revealed by Inversion of Geodetic Data. Science, New Series, 233, 4771, 1409-1413. Sekiguchi, H., and Iwata, T. (2002). Rupture process of the 1999 Kocaeli, Turkey, earthquake estimated from strong- motion waveforms. Bull. Seism. Soc. Am., 92(1), 300-311. Sengör, A. M. C. (1979). The North Anatolian transform fault: its age, offset and tectonic significance. J. Geol. Soc., 136(3), 269-282. Sengör, A. M. C., Tüysüz, O., Imren, C., Sakinç, M., Eyidogan, H., et al. (2005). The North Anatolian fault: A new look. Annu. Rev. Earth Planet. Sci., 33, 37-112.

43

Sharman, G. R., Graham, S. A., Grove, M., and Hourigan, J. K. (2013). A reappraisal of the early slip history of the San Andreas fault, central California, USA. Geology, 41(7), 727-730. Sharp, R. V. (1967). San Jacinto Fault zone in the Peninsular Ranges of southern California, Geol. Soc. Am. Bull., 78, 705–730, doi:10.1130/ 0016-7606(1967)78[705:SJFZIT]2.0.CO;2. Sharp, R. V. (1982). Comparison of 1979 surface faulting with earlier displacements in the Imperial Valley. US Geol. Surv. Prof. Pap, 1254, 213-221. Sharp, R. V., Lienkaemper, J. J., Bonilla, M. G., Burke, D. B., Fox, B. F., et al. (1982). Surface faulting in the central Imperial Valley. The Imperial Valley, California, Earthquake of October, 15(1979), 119-143. Sharp, R. V., Budding, K. E., Boatwright, J., Ader, M. J., Bonilla, M. G., et al. (1989). Surface faulting along the Superstition Hills fault zone and nearby faults associated with the earthquakes of 24 November 1987, Bull. Seism. Soc. Am. 79, 252- 281. Shen, Z.K., Lü, J., Wang, M. and Bürgmann, R. (2005). Contemporary crustal deformation around the southeast borderland of the Tibetan Plateau, J. Geophys. Res, 110, B11409, doi:11410.11029/12004JB003421. Shen, Z.K., Sun, J.B., Zhang, P.Z., Wan, Y.G., Wang, M., Burgmann, R., Zeng, Y.H., Gan, W.J., Liao, H. and Wang, Q.L. (2009). Slip maxima at fault junctions and rupturing of barriers during the 2008 Wenchuan earthquake, Nature Geoscience, 2, 718-724. Sieh, K. E. (1978). Slip along the San Andreas fault associated with the great 1857 earthquake, Bull. Seismol. Soc. Am. 68, 5, 1421–1448. Sieh, K. E., and R. H. Jahns (1984). Holocene activity of the San Andreas fault at Wallace Creek, California, Geol. Soc. Am. Bull. 95, 883–896 Sieh, K., L. M. Jones, E. Hauksson, K. Hudnut, D. Eberhart-Phillips, et al. (1993). Near-field investigations of the Landers earthquake sequence, Science 260, 171–176. Simoes, M., and Avouac, J. P. (2006). Investigating the kinematics of mountain building in Taiwan from the spatiotemporal evolution of the foreland basin and western foothills. J. Geophys. Res., 111(B10). Simoes, M., Chen, Y. G., Shinde, D. P., and Singhvi, A. K. (2014). Lateral variations in the longterm slip rate of the Chelungpu fault, Central Taiwan, from the analysis of deformed fluvial terraces. J. Geophys. Res., 119(4), 3740-3766. Sims, J. (1993). Chronology of displacement on the San Andreas fault in central California : Evidence from reversed positions of exotic rock bodies near Parkfield, California, in Powell, R. E., Weldon, R. J., Matti, J. C. eds, The San Andreas Fault System : Displacement, Palinspastic Reconstruction and Geologic Evolution : Boulder, Colorado, Geol. Soc. Am. Memoir 178. Slemmons, D. B. (1957). Geological effects of the Dixie Valley-Fairview Peak, Nevada, earthquakes of December 16, 1954. Bull. seism. Soc. Am. 47, 353-375. Smith, T. E. (1968). Aeromagnetic measurements in Dixie Valley, Nevada; implications on basin-range structure. Jour. Geophys. Res., 73, 1321-1331. Song, S. G., Beroza, G. C., and Segall, P. (2008). A unified source model for the 1906 San Francisco earthquake. Bull. Seism. Soc. Am., 98(2), 823-831. Spencer, J. E., and Normark, W. R. (1979). Tosco-Abreojos fault zone: A Neogene transform plate boundary within the Pacific margin of southern Baja California, Mexico. Geology, 7(11), 554-557. Spudich, P., and Cranswick, E. (1984). Direct observation of rupture propagation during the 1979 Imperial Valley earthquake using a short baseline accelerometer array. Bull. Seism. Soc. Am., 74(6), 2083-2114. Stewart, J. H. (1980). Regional tilt patterns of late Cenozoic basin-range fault blocks, western United States. Geol. Soc. Am. Bulletin, 91(8), 460-464. Suppe, J. (1970). Offset of late Mesozoic basement terrains by the San Andreas fault system. Geol. Soc. Am. Bull., 81(11), 3253-3258. Suppe, J. (1980). A retrodeformable cross section of northern Taiwan, Proc. Geol. Soc. China 23, 46–55. Susong, D. D., Janecke, S. U. and Bruhn, R. L. (1990). Structure of a fault segment boundary in the Lost River fault zone, Idaho, and possible effect on the 1983 Borah Peak earthquake rupture. Bull. Seism. Soc. Am., 80, 57-68. Tapponnier, P. and P. Molnar (1977). Active faulting and tectonics in China. J. Geophys. Res. 82, 2905–2930. Tapponnier, P. and P. Molnar (1979). Active faulting and Cenozoic tectonics of the Tien Shan, Mongolia and Baikal regions, J. Geophys. Res., 84, 3425–3459. Tapponnier, P., Xu, Z., Roger, F., Meyer, B., Arnaud, N., Wittlinger, G. and Yang, J. (2001). Oblique stepwise rise and growth of the Tibet plateau, Science, 294, 1671–1677. Taylor, M.H., Leprince, S., Avouac, J.P., and Sieh, K. (2008). Detecting co-seismic displacements in glaciated regions— An example from the great November 2002 Denali earthquake using SPOT horizontal offsets. Earth Planet. Sci. Lett., 270, 209–220 Taylor, M., and A. Yin (2009), Active structures of the HimalayanTibetan orogen and their relationships to earthquake distribution, contemporary strain field, and Cenozoic volcanism, Geosphere, 5(3), 199–214. Teran, O. J., Fletcher, J. M., Rockwell, T. K., Hudnut, K., Oskin, et al. (2011). Structural Controls on the Surface Rupture Associated with the Mw7.2 El Mayor-Cucapah Earthquake of 4 April 2010: A Comparative Analysis of Scarp Array Kinematics, Orientation, Lithology and Width. In AGU Fall Meeting Abstracts (Vol. 1, p. 2348). Teran, O. J., Fletcher, J. M., Oskin, M. E., Rockwell, T. K., Hudnut, K. W., Spelz, R. M., et al. (2015). Geologic and structural controls on rupture zone fabric: A field-based study of the 2010 Mw 7.2 El Mayor–Cucapah earthquake surface rupture. Geosphere, 11(3), 899-920. Thakur, V. C., Jayangondaperumal, R., and Malik, M. A. (2010). Redefining Medlicott–Wadia's main boundary fault from Jhelum to Yamuna: An active fault strand of the main boundary thrust in northwest Himalaya. Tectonophysics, 489(1), 29-42.

44

Thatcher, W., G. Marshall, and M. Lisowski (1997), Resolution of fault slip along the 470-km-long rupture of the great 1906 San Francisco earthquake and its implications, J. Geophys. Res., 102(B3), 5353–5367, doi:10.1029/96JB03486. Thatcher, W., G. R. Foulger, B. R. Julian, J. Svarc, E. Quilty, and G. W. Bawden (1999). Present-day deformation across the Basin and Range province, western United States, Science, 283, 1714 – 1718. Thomas, A. P., and Rockwell, T. K. (1996). A 300-to 550-year history of slip on the Imperial fault near the US–Mexico border: Missing slip at the Imperial fault bottleneck. J. geophys. Res, 101(B3), 5987-5997. Thompson, G. A., Meister, L. J., Herring, A. T., Smith, T. E., Burke, D. B., Kovach, R. L., Burford, R. O., Salehi, A., and Wood, M. D. (1967). Geophysics, study of basin-range structure, Dixie Valley region, Nevada. U.S. Air Force Cambridge Research Labs. Spec. Rept. 66-848. Thompson, G. A. and Burke, D. B. (1973). Rate and Direction of Spreading in Dixie Valley, Basin and Range Province, Nevada. Geol. Soc. Am. Bulletin, 84, 627-632. Titus, S. J., M. Dyson, C. DeMets, B. Tikoff, F. Rolandone, and R. Bürgmann (2011). Geologic versus geodetic deformation adjacent to the San Andreas fault, central California. GSA Bulletin, 123, no. 5/6, 794–820; doi: 10.1130/B30150.1; Tobita, M., T. Nishimura, T. Kobayashi, K. X. Hao, and Y. Shindo (2011). Estimation of coseismic deformation and a fault model of the 2010 Yushu earthquake using PALSAR interferometry data, Earth Planet. Sci. Lett. 307, 430–438. Tong, X., Sandwell, D.T. and Fialko, Y. (2010). Coseismic slip model of the 2008 Wenchuan earthquake derived from joint inversion of interferometric synthetic aperture radar, GPS, and field data, J. Geophys. Res., 115, B04314. Tong, X., D. T. Sandwell, and B. Smith-Konter (2013). High-resolution interseismic velocity data along the San Andreas Fault from GPS and InSAR, J. Geophys. Res. Solid Earth, 118, doi:10.1029/2012JB009442. Treiman, J. A., Kendrick, K. J., Bryant, W. A., Rockwell, T. K., and McGill, S. F. (2002). Primary surface rupture associated with the mw 7.1 16 october 1999 hector mine earthquake, san bernardino county, california. Bull. Seism. Soc. Am., 92(4), 1171-1191. Trifunac, M. (1972). Tectonic stress and the source mechanism of the Imperial Valley, California, earthquake of 1940. Bull. Seism. Soc. Am., 62(5), 1283-1302. Trifunac, M. D., and J. N. Brune (1970). Complexity of energy release during the Imperial Valley, California, earthquake of 1940, Bull. Seismol. Soc. Am. 60, 137–160. Uchide, T., Yao, H., and Shearer, P. M. (2013). Spatiotemporal distribution of fault slip and highfrequency radiation of the 2010 El MayorCucapah, Mexico earthquake. J. Geophys. Res., 118(4), 1546-1555. U.S. Geological Survey, the Southern California Earthquake Center, and the California Division of Mines and Geology (2000). Preliminary report on the 16 October 1999 M7.1 Hector Mine, California, earthquake, Seism. Res. Lett. 71, 11–23. Vallée, M., Landès, M., Shapiro, N. M. and Klinger, Y. (2008). The 14 November 2001 Kokoxili (Tibet) earthquake: high-frequency seismic radiation originating from the transitions between sub-Rayleigh and supershear rupture velocity regimes, J. geophys. Res., 113, B07305, doi:10.1029/2007JB005520. Vallée, M., J. Charléty, A.M.G. Ferreira, B. Delouis, and J. Vergoz, (2011). SCARDEC : a new technique for the rapid determination of seismic moment magnitude, focal mechanism and source time functions for large earthquakes using body wave deconvolution, Geophys. J. Int., 184, 338-358. van Der Woerd, J., Ryerson, F.J., Tapponnier, P., Gaudemer, Y., Finkel, R., Meriaux, A.S., Caffee, M., Guoguang, Z. and Qunlu, H. (1998). Holocene left slip-rate determined by cosmogenic surface dating on the Xidatan segment of the Kunlun Fault (Qinghai, China), Geology, 26, 695–698. van der Woerd, J., Ryerson, F.J., Tapponnier, P., Meriaux, A.S., Gaudemer, Y., Meyer, B., Finkel, R., Caffee, M.W., Guoguang, Z. and Zhiqin, X. (2000). Uniform slip-rate along the Kunlun Fault: Implications for seismic behavior and large-scale tectonics, Geophys. Res. Lett., 27, 2353 – 2356. van der Woerd, J., Tapponnier, P., Ryerson, F. J., Meriaux, A. S., Meyer, B., et al. (2002). Uniform postglacial slip-rate along the central 600 km of the Kunlun Fault (Tibet), from 26Al, 10Be, and 14C dating of riser offsets, and climatic origin of the regional morphology. Geoph. J. Int., 148(3), 356-388. van der Woerd, J., Y. Klinger, K. Sieh, P. Tapponnier, F. J. Ryerson, and A.-S. Mériaux (2006), Long-term slip rate of the southern San Andreas Fault from 10Be-26Al surface exposure dating of an offset alluvial fan, J. Geophys. Res., 111 B04407, doi:10.1029/2004JB003559 Vassallo, R., J.-F. Ritz, R. Braucher, and S. Carretier (2005). Dating faulted alluvial fans with cosmogenic 10Be in the Gurvan Bogd mountain range (Gobi-Altay, Mongolia): Climatic and tectonic implications, Terra Nova, 17, 278–285, doi:10.1111/j.1365-3121.2005.00612.x. Vassallo, R., Jolivet, M., Ritz, J. F., Braucher, R., Larroque, C., Sue, C., Todbileg, M. and Javkhlanbold, D. (2007a). Uplift age and rates of the Gurvan Bogd System (Gobi-Altay) by apatite fission track analysis, Earth planet. Sci. Lett., 259, 333–346. Vassallo, R., et al. (2007b), Transpressional tectonics and stream terraces of the Gobi-Altay, Mongolia, Tectonics, 26, TC5013, doi:10.1029/2006TC002081. Vaughan, P., and Rockwell, T. (1986). Alluvial stratigraphy and neotectonics of the Elsinore fault zone at Agua Tibia Mountain, Southern California. Neotectonics and Faulting in Southern California, 177-191. Vignon, V. (2011), Activité hors séquence des chevauchements dans la syntaxe nord-ouest himalayenne: Apports de la modélisation analogique et quantification quaternaire par analyse morphotectonique, PhD thesis Université de Grenoble. Walker, K. T., and Shearer, P. M. (2009). Illuminating the nearsonic rupture velocities of the intracontinental Kokoxili Mw 7.8 and Denali fault Mw 7.9 strikeslip earthquakes with global P wave back projection imaging. J. Geoph. Res., 114(B2).

45

Wald, D. J., and Heaton, T. H. (1994). Spatial and temporal distribution of slip for the 1992 Landers, California, earthquake. Bull. Seism. Soc. Am., 84(3), 668-691. Wald, D.J., D.V. Helmberger, and S.H. Hartzell. (1990). Rupture process of the 1987 Superstition Hills earthquake from the inversion of strong motion data. Bull, Seismol. Soc. Am., 80, 1079-1098 Wallace, R. E. (1970). Earthquake Reccurrence Intervals on the San Andreas Fault. Geol. Soc. Am. Bulletin, 81, 2875- 2890. Wallace, R.E. (1984). Fault scarps formed during the earthquake of October 2, 1915, Pleasant Valley, Nevada and some tectonic implications. U.S. Geol. Surv. Prof. Pap. 1274-A, 1–33. Wallace, R.E. and Whitney, R.A. (1984). Late Quaternary history of the Stillwater seismic gap. Bull. Seismol. Soc. Am., 74, 301–314. Wang, E., and Burchfiel, B.C. (2000). Late Cenozoic to Holocene deformation in southwestern Sichuan and adjacent Yunnan, China, and its role in formation of the southeastern part of the Tibetan Plateau: Geol. Soc. Am. Bulletin, 112, 413–423. Wang, D., and Mori, J. (2012). The 2010 Qinghai, China, earthquake: A moderate earthquake with supershear rupture. Bull. Seism. Soc. Am., 102(1), 301-308. Wang, S., Fan, C., Wang, G., and Wang, E. (2008). Late Cenozoic deformation along the northwestern continuation of the Xianshuihe fault system, Eastern Tibetan Plateau. Geol. Soc. Am. Bull., 120(3-4), 312-327. Wang, S., Fang, X., Zheng, D. and Wang, E. (2009). Initiation of slip along the Xianshuihe fault zone, eastern Tibet, constrained by K/Ar and fission-track ages. International Geology Review, 51, 12, 1121-1131 Wang, Q., Qiao, X., Lan, Q., Freymueller, J., Yang, S., Xu, C., Yang, Y., You, X., Tan, K. and Chen, G., (2011). Rupture of deep faults in the 2008 Wenchuan earthquake and uplift of the Longmen Shan, Nature Geosci, 4, 634-640. Wang W L, Wu J P, Fang L H and Wang C Z (2013) Relocation of the Yushu MS7.1 earthquake and its aftershocks in 2010 from HypoDD. Science China: Earth Sciences, 56, 182–191, doi: 10.1007/s11430-012-4450-z Weber, F. H. (1977), Seismic hazards related to geologic factors, Elsinore and Chino fault zones, northwestern Riverside county, California, Calif. Div. Mines Geol. Open File Rep., 77-4, 96 pp. Wei, S.J., Fielding, E., Leprince, S., Sladen, A., Avouac, J.P., Helmberger, D., Hauksson, E., Chu, R.S., Simons, M., Hudnut, K., Herring, T., Briggs, R. (2011). Superficial simplicity of the 2010 El Mayor–Cucapah earthquake of Baja California in Mexico. Nat. Geosci. 4, 615–618. Weldon, R. J., and K. E. Sieh (1985), Holocene rate of slip and tentative recurrence interval for large earthquakes on the San Andreas fault, Cajon Pass, southern California, Geol. Soc. Am. Bull., 96, 793–812. Wen, X., Xu, X., Zheng, R., Xie, Y., and Wan, C. (2003). Average slip-rate and recent large earthquake ruptures along the Ganzi–Yushu Fault: Science in China Series D, 46, 276–288 Wen, Y. Y., Ma, K. F. and Oglesby, D. D. (2012). Variations in rupture speed, slip amplitude and slip direction during the 2008 Mw 7.9 Wenchuan Earthquake. Geoph. J. Int., 190(1), 379-390. Williams, P.L., and H.W. Magistrale. (1989). Slip along the Superstition Hills Fault associated with the 24 November 1987 Superstition Hills, California, earthquake. Bull. Seismol.. Soc. Am., 79, 390- 410 Witkind, I. J. (1964) Reactivated faults north of Hebgen Lake. U.S. Geol. Surv. Professional Paper 435-I, 37–50. Witkind, I. J., Myers, W.B., Hadley, J. B., Hamilton, W. and Fraser, G. D. (1962). Geologic features of the earthquake at Hebgen Lake, Montana, August 17, 1959. Bull. Seism. Soc. Am., 52, 2, 163-180 Wu, C., Takeo, M., and Ide, S. (2001). Source process of the Chi-Chi earthquake: A joint inversion of strong motion data and global positioning system data with a multifault model. Bull. Seism. Soc. Am., 91(5), 1128-1143. Xu, X., Yu, G., Klinger, Y., Tapponnier, P. and Van Der Woerd, J. (2006). Re-evaluation of surface rupture parameters and faulting segmentation of the 2001 Kunlunshan earthquake (Mw 7.8), northern Tibetan Plateau, China, J. geophys. Res., 111, B05316, doi:10.1029/2004JB003488. Xu, X.W., Wen, X.Z., Yu, G.H., Chen, G.H., Klinger, Y., Hubbard, J. and Shaw, J., (2009). Coseismic reverseand oblique-slip surface faulting generated by the 2008 Mw 7.9 Wenchuan earthquake, China, Geology, 37, 515-518. Yan, Y., Pinel, V., Trouvé, E., Pathier, E., Perrin, J., Bascou, P., and Jouanne, F. (2013). Coseismic displacement field and slip distribution of the 2005 Kashmir earthquake from SAR amplitude image correlation and differential interferometry. Geoph. J. Int., 193(1), 29-46. Yielding, G., Jackson, J. A., King, G. C. P., Sinvhal, H., Vita-Finzi, C., and Wood, R. M. (1981). Relations between surface deformation, fault geometry, seismicity, and rupture characteristics during the El Asnam (Algeria) earthquake of 10 October 1980. Earth Planet. Sc. Lett., 56, 287-304. Yielding, G., Ouyed, M., King, G. C. P., and Hatzfeld, D. (1989). Active tectonics of the Algerian Atlas Mountains— evidence from aftershocks of the 1980 El Asnam earthquake. Geoph. J. Int., 99(3), 761-788. Yue, L. F., Suppe, J., and Hung, J. H. (2005). Structural geology of a classic thrust belt earthquake: The 1999 Chi-Chi earthquake Taiwan (M w= 7.6). J. Struct. Geol., 27(11), 2058-2083. Zachariasen, J., and K. Sieh (1995). The transfer of slip between two en echelon strike-slip faults: A case study from the 1992 Landers earthquake, southern California, J. Geophys. Res., 100, 15281–15301, doi:10.1029/95JB00918. Zebker, H. A., P. Rosen, R.M. Goldstein, A. Gabriel, and C.L. Werner (1994) On the derivation of coseismic displacement fields using differentiai radar interferometry: the Landers earthquake, J. Geophys. Res., 19,617- 19,634. Zeng, Y., and Chen, C. H. (2001). Fault rupture process of the 20 September 1999 Chi-Chi, Taiwan, earthquake. Bull. Seism. Soc. Am., 91(5), 1088-1098. Zhang, P., Mao, F. and Slemmons, D. B. (1999). Rupture terminations and size of segment boundaries from historical earthquake ruptures in the Basin and Range Province. Tectonophysics, 308, 37–52. Zhang, P.-Z., Shen, Z., Wang, M., Gan, W., Bürgmann, R., Molnar, P., Wang, Q., Niu, Z., Sun, J., Wu, J., Hanrong, S., and Xinzhao, Y. (2004). Continuous deformation of the Tibetan Plateau from global positioning system data, Geology, 32, 809 – 812

46

Zhang, P.Z., Xu, X.W., Wen, X.Z. and Ran, Y.K. (2008). Slip rates and recurrence intervals of the Longmen Shan active fault zone, and tectonic implications for the mechanism of the May 12 Wenchuan earthquake, 2008, Sichuan, China, Chinese J Geophys-Ch, 51, 1066-1073. Zhang, P.-Z., Wen, X.-z., Shen, Z.-K. and Chen, J.-h. (2010). Oblique, High-Angle, Listric-Reverse Faulting and Associated Development of Strain: The Wenchuan Earthquake of May 12, 2008, Sichuan, China, Ann. Rev. Earth Planet. Sci., 38, 353-382. Zhang, Y., Yao, X., Xiong, T., Ma, Y., Hu, D., Yang, N. and Guo, C. (2010). Rapid Identification and Emergency Investigtion of Surface Ruptures and Geohazards Induced by the Ms 7.1 Yushu Earthquake. Acta Geologica Sinica (English Edition), 84, 6, 1315-1327 Zhang, H. P., Zhang, P. Z., Kirby, E., Yin, J. H., Liu, C. R., and Yu, G. H. (2011). Along-strike topographic variation of the Longmen Shan and its significance for landscape evolution along the eastern Tibetan Plateau. Journal of Asian Earth Sciences, 40(4), 855-864. Zhang, G., Qu, C., Shan, X., Song, X., Zhang, G., et al. (2011). Slip distribution of the 2008 Wenchuan Ms 7.9 earthquake by joint inversion from GPS and InSAR measurements: a resolution test study. Geoph. J. Int., 186(1), 207-220. Zhang, G., Vallée, M., Shan, X., and Delouis, B. (2012). Evidence of sudden rupture of a large asperity during the 2008 Mw7. 9 Wenchuan earthquake based on strong motion analysis. Geoph. Res. Lett., 39(17). Zhang, G., Shan, X., Delouis, B., Qu, C., Balestra, J., Li, Z., Liu, Y. and Zhang, G. (2013). Rupture history of the 2010 Ms 7.1 Yushu earthquake by joint inversion of teleseismic data and InSAR measurements. Tectonophysics 584, 129– 137 Zhou, R., Yong, L., Densmore, A. L., Ellis, M. A., He, Y., Li, Y. and, Li, X. (2007). Active tectonics of the Longmen Shan region on the eastern margin of the Tibetan plateau. Acta Geologica Sinica English Edition, 81(4), 593-604. Zhou, R., Li, Y., Liang, M., Xu, X., He, Y., Wang, S., Ma, C., and Liu, Y. (2013). Determination of mean recurrence interval of large earthquakes on the Garzê-Yushu Fault (Dengke Segment) on the eastern margin of the Qinghai– Tibetan Plateau, Quaternary International, Available online 12 December 2013, ISSN 1040-6182, http://dx.doi.org/10.1016/j.quaint.2013.11.010. Zielke, O., J R. Arrowsmith, L. Grant Ludwig, and S. O. Akciz (2010). Slip in the 1857 and earlier large earthquakes along the Carrizo Plain, San Andreas fault, Science 327, 1119–1122, doi 10.1126/science.1182781. Zielke, O., J R. Arrowsmith, L. Grant Ludwig, and S. O. Akciz (2012). High-Resolution Topography-Derived Offsets along the 1857 Fort Tejon Earthquake Rupture Trace, San Andreas Fault, Bull. Seism. Soc. Am., 102, 3, 1135–1154, doi: 10.1785/0120110230

: Uploaded separately because larger than one page

Supplementary Figure S1: All earthquakes analyzed in this study. The figures emphasize the relations between surface rupture traces, surface slip-length profiles, architecture and direction of long-term propagation of causative faults (most fault maps are taken from the compilation by Perrin et al., 2016; all data are described in Document S1 and compiled in Table S1). For each earthquake, the causative fault and off-fault splays and associated secondary faults are shown in black (unless ruptured), the rupture surface trace in red, the zone of largest coseismic surface slips (≥ 80% of maximum slip; indicated with horizontal dotted line in slip profiles) in green, the hypocenter (H) as a red square with a 5 km uncertainty, and rupture speeds indicated in red where available along the ruptures. Blue rectangles locate zooms that are shown for some earthquakes. Slip mode and overall structural maturity of causative faults are indicated (see Table S1). References for surface slip profiles and maps are cited; references for direction of long- term propagation are in Document S1 and Table S1.

< Supplementary Figure S2>: Uploaded separately because larger than one page

Supplementary Figure S2: All earthquakes analyzed in this study for which finite source inversion models are available. The figures are done as in Figure S1. However, the slip-length profiles are here derived at depth from source inversion models (references indicated; models described in Document S1). They depict the maximum slip values along the rupture length. Because subsurface slip models are less precise than surface measurements, we define the largest slips as those ≥90% of the maximum slip value. Blue rectangles locate zooms that are shown for some earthquakes.

47

Supplementary Figure S3: Same as Fig. 2 but largest surface slips are here taken to be ≥70% of the maximum surface slip value. Superposition of dark green and pale green symbols indicates largest best-constrained slips in dark green, and local peaks of large slip in pale green (see Fig. S1).

Denali (2002), SS, Mature Kunlun (2001), SS, Mature Zone of largest Izmit (1999), SS, Interm. coseismic slips at Sichuan (2008), RE, Interm. depth (≥90%*Dmax) Yushu (2010), SS, Interm. Hypocenter El Mayor Cucapah (2010), SS, Interm. Manyi (1997)*, SS, Interm. Superstition Hills (1987), SS, Interm. Landers (1992), SS, Immature Hector Mine (1999), SS, Immature ChiChi(1999), RE, Interm. Oriented in direction length of causative fault length of causative of long-term fault Imperial Valley (1979)*, SS, Interm. propagation Borah Peak (1983), N, Immature Earthquakes ordered by increasing increasing by Earthquakes ordered 0 0.2 0.4 0.6 0.8 1 Normalized rupture length

Supplementary Figure S4 ; Perrin et al.

Supplementary Figure S4: Same as Fig. 2 but for slip derived at depth from source inversion models (Fig. S2). As in Fig. S2, we define the largest slips as those ≥90% of the maximum slip value. Additional comments on the model data can be found in Document S1.

48

(a) Strike slip ruptures (67%) (b) Dip slip ruptures (33%) 4 4 Direction of long-term Direction of long-term 3.5 fault propagation 3.5 fault propagation 3 3 2.5 2.5 mean 2 mean 2 D/D 1.5 D/D 1.5 1 1 0.5 0.5 0 0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Normalized rupture length Normalized rupture length

(c) Rupture on mature fault sections (d) Rupture on immature fault sections 3.5 4 Direction of long-term Direction of long-term 3 3.5 fault propagation fault propagation 2.5 3 2.5 2

mean mean 2 1.5

D/D D/D 1.5 1 1 0.5 0.5 0 0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Normalized rupture length Normalized rupture length Supplementary Figure S5; Perrin et al.

Supplementary Figure S5: Relations between earthquake slip asymmetry and fault propagation for earthquakes distinguished by slip mode and by location on mature or immature sections of causative faults. Slip profiles are those measured at the surface. All have been oriented so as that the direction of long-term fault propagation is to the right. (a) Earthquakes on strike-slip faults, (b) Earthquakes on dip-slip faults. The relationship between earthquake slip asymmetry and direction of fault propagation is observed independent of the slip mode, (c) Earthquakes that broke a mature section of their causative fault (Denali, Erzincan, San Francisco, Muzaffarabad, Hebgen Lake, Fuyun). Note that for the Denali earthquake, we consider only the rupture of the main Denali fault, (d) Earthquakes that broke an immature section of their causative fault (Borrego Mountain, El Mayor Cucapah, Fort Tejon, Hector Mine, Imperial Valley 1979, Izmit, Mudurnu, Superstition Hills, Yushu). Note that for the Izmit earthquake, we consider only the rupture of the main fault whose maturity is described. The degree of asymmetry is on average 85% for earthquakes having ruptured mature fault sections, and 65% for earthquakes having ruptured immature fault sections (see apex of yellow average curves).

49

Supplementary Figure S6: Along-strike evolution of off-fault damage, fault segmentation and fault planarity with increasing maturity along fault strike, for a dip-slip (normal) fault. The figure is similar to that for a strike-slip fault (Fig. 7) because the properties that are described are mainly related to the lateral fault lengthening. A significant difference is the smoothing effect of increasing slip on fault roughness: because slip is parallel to fault inter-segments on dip-slip faults, roughness may be more persistent than in a strike-slip fault on which each slip event propagates through some of its inter-segments. Furthermore, because dip-slip ruptures generally propagate from crustal depth to surface, the fault plane roughness might also differ along the fault dip, being greater in the shallowest fault parts.

: Uploaded separately because larger than one page

Supplementary Table S1: Principal characteristics of the earthquakes analyzed and of their causative faults. The various parameters are described in more details in Document S1. SS for strike-slip (LL left-lateral, RL right-lateral), N for normal, RE for reverse; LT for long-term slip rate and C for current, geodetic slip rate. Overall structural maturity of causative faults is derived from classifications by Manighetti et al. (2007). Number of major segments along faults is derived from the literature (references indicated). References listed in Table can be found in Document S1.

50

Supplementary Table S2: Relative qualities attributed to the principal earthquake and fault parameters described in detail in Document S1.

Earthquake parameters Fault parameters Rupture Surface extent Earthquake Mapping of Direction of Earthquake rupture within slip profile fault trace long-term trace causative at surface propagation fault Bogd 1957 1 1 1 1 2 Borah Peak 1983 1 1 1 1 1 Borrego Mountains 1968 2 1 2 1 1 ChiChi 1999 1 1 1 1 2 Denali 2002 1 1 1 1 1 Dixie Valley 1954 1 1 1 1 1 El Asnam 1980 1 1 2 1 2 El Mayor Cucapah 2010 2 2 2 1 1 Erzincan 1939 1 1 2 1 1 Fairview Peak 1954 2 1 2 1 3 Fort Tejon 1857 1 1 2 1 1 Fuyun 1931 2 1 2 2 3 Hebgen Lake 1959 1 1 2 2 3 Hector Mine 1999 1 1 1 1 1 Imperial Valley 1940 1 1 2 1 3 Imperial Valley 1979 2 2 2 1 3 Izmit 1999 2 2 1 1 1 Kunlun 2001 1 1 1 1 1 Landers 1992 2 1 1 1 1 Manyi 1997 2 1 2 2 3 Mudurnu 1967 1 1 2 1 1 Muzaffarabad 2005 1 2 1 1 3 Pleasant Valley 1915 2 1 2 1 1 San Francisco 1906* 1 1 3 1 1 Sichuan 2008 1 1 1 1 1 Superstition Hills 1987 1 1 1 1 1 Yushu 2010 2 1 3 1 1

Supplementary Table S2: Relative qualities attributed to the principal earthquake and fault parameters described in detail in Document S1. Qualities are defined by criteria provided in Document S1, and provide a general view of the robustness of the analyzed data. Quality 1, 2 and 3 for high, fair and low quality, respectively. Asterisk for San Francisco earthquake indicates that its causative Northern San Andreas fault has not grown as the other faults (see Document S1 and text).

51

Supplementary Table S3: Rupture width and depth of largest coseismic slip, derived from the available finite source inversion models for 13 of the analyzed earthquakes

Borah ChiChi Denali El Mayor Hector Imperial Izmit Kunlun Landers Manyi Sichuan Superstition Yushu Mean Earthquake Peak 1999 2002 Cucapah Mine Valley 1999 2001 1992 1997 2008 Hills 1987 2010 value 1983 2010 1999 1979 Rupture ~25 ~40 ~30 ~18 ~20 ~12 ~20 ~20 ~15 ~20 ~28 ~12 ~20 ~22 width (km) Depth of ~18 ~6 ~8 ~4 2-6 ~7 ~5 ~3 ~5 ~2 ~8 ~8 ~4 ~6 largest coseismic slip (km) Mendoza Ma et al., Oglesby Wei et al., Kaverina Hartzell Delouis Lasserre Wald and Funning Fielding Wald et al., Li et Principal and 2001 et al., 2011 et al., and et al., et al., Heaton, et al., et al., 1990 al., references Hartzell, 2004 2002 Heaton, 2002 2005 1994 2007 2013 2012 1988 1983

Supplementary Table S3: Rupture width and depth of largest coseismic slip, derived from the available finite source inversion models for 13 of the analyzed earthquakes. Detailed descriptions of the inversion models and their references can be found in Document S1.

52

Supplementary Table S4: Rupture speeds and ranges estimated for 12 of the analyzed earthquakes

Earthquake Chichi Denali 2002 El Mayor Hector Imperial Izmit Kunlun Landers Manyi San Sichuan Yushu 1999 Cucapah Mine Valley 1979 1999 2001 1992 1997 Francisco 2008 2010 2010 1999 1906 Rupture speed ~2.5 3.3 1 ? 1.8 ~2 to 7-8.8 3 2.6-3.3 ~2.5-3.6 < 3 2.9 2.5-3 1.6-2.5 in most (supershear) immature half of rupture (km/s) Rupture speed ~2.5 5-5.5 3 ? 2.2-2.8 2.5-2.9 4.8-5.8 5-6.7 ~4 3-3.2 3.9 3-3.4 4-5 in most mature (super (super (super (super (super (super half of rupture shear) shear) shear) shear) shear) shear) (km/s) Faster slip Three similar Additional velocity in peak slip rates information most along rupture mature half length

Ma et al., Frankel 2004; Uchide et Kaverina Archuleta, Bouchon et Bouchon Wald and Funning Song et al., Wen et Zhang et 2001, 2003; Oglesby et al., al., 2013 et al., 1984; Spudich al., 2001, and Vallée Heaton, et al., 2008 al., 2012; al., 2010; References Wu et al., 2004; Dunham 2002; Ji and Cranswick, 2002; 2003; 1994; 2014 Zhang et Wang 2001; Zeng and Archuleta, et al., 1984 Sekiguchi Antolik et Peyrat et al., 2012 and and Chen, 2004; 2002 and Iwata, al., 2004; al., 2001 Mori, 2001. Ellsworth et 2002 Robinson et 2012 al., 2004; al., 2006; Asano et al., Vallée et al., 2005; Walker 2008; and Shearer, Walker and 2009 Shearer, 2009

Supplementary Table S4: Rupture speeds and ranges estimated for 12 of the analyzed earthquakes. We discriminate the rupture speeds along the most mature and along the most immature halves of the ruptures (maturity defined from direction of long-term propagation of causative fault; see text). All ruptures (but possibly the 1979 Imperial Valley; yet its rupture speed is not well constrained) have propagated faster, sometimes with supershear speed, along the most mature part of the ruptures. Rupture speeds are presented in Fig.5. 53

Supplementary Table S5: Available measurements of seismic velocity reduction and of compliance along six of the analyzed faults

Northern San Andreas fault San Jacinto Elsinore North Anatolian Garze-Yushu Longmen Shan fault Fault (southward decrease in maturity) fault fault fault fault system (southward (southward (westward (northwestward (northward decrease Northern part of Region due north decrease in decrease in decrease in decrease in in maturity) SA (around of central maturity) maturity) maturity) maturity) 38°30’) creeping zone Seismic velocity 50-60% 35-50% ~40 % 25-40 % ~22% --- ~10% (or rigidity) of rigidity (reduction in Vp) (3-6 km large, fairly (several km large, (~100 km large, (broad zone of low reduction off reduction continuous, Vs continuous, Vs continuous, Vs Vp and Vs velocity, most mature anomaly north of anomaly between anomaly between of crustal scale) fault half Hot Springs inter- 118° and ~34 an 39°E, over segment, from at 118°30’W, from 1 20 km depth) least 2 to 8 km to 7 km depth) depth) Seismic velocity -- ~20% < ~27% No clear anomaly, No clear 1-8% No clear anomaly (or rigidity) (reduction in Vp) (few km large, but local reductions anomaly in (reduction in Vp) reduction off discontinuous, Vs of Vs Fichtner et al., most immature anomaly off Anza < 20% 2013 fault half segment, from at < 3-6% Vp least 2 to 8 km velocity contrast depth) found in more detailed recent studies Jolivet et al., 2009 McGuire and Allam and Ben- Fichtner et al., Yang et al., 2015 Lei and Zhao, 2009 References Ben-Zion, 2005 Zion, 2012 (Figs. 4 Zigone et al., 2015 2013 (Fig. 6a); to 7); Allam et al., (Fig. 12 and 13) Bulut et al., 2012; 2014 (Figs. 5 and Najdahmadi et al., 7); Zigone et al., 2016 2015 (Fig. 12 and 13)

Supplementary Table S5: Available measurements of seismic velocity reduction and of compliance along six of the analyzed faults. We discriminate the velocity reductions and compliance that were measured along the most mature and along the most immature halves of the faults (maturity defined from direction of long-term propagation of causative fault; see text). References listed in Table are:

54

Allam, A. A., and Ben-Zion, Y. (2012). Seismic velocity structures in the Southern California plate-boundary environment from double-difference tomography. Geophys. J. Int., 190(2), 1181-1196. Allam, A. A., Ben-Zion, Y., Kurzon, I., and Vernon, F. (2014). Seismic velocity structure in the Hot Springs and Trifurcation Areas of the San Jacinto Fault Zone, California, from double-difference tomography. Geophys. J. Int., 198(2), 978-999. Bulut, F., Ben-Zion, Y., and Bohnhoff, M. (2012). Evidence for a bimaterial interface along the Mudurnu segment of the North Anatolian Fault Zone from polarization analysis of P waves. Earth Planet. Sci. Lett., 327, 17-22. Fichtner, A., Saygin, E., Taymaz, T., Cupillard, P., Capdeville, Y., and Trampert, J. (2013). The deep structure of the North Anatolian fault zone. Earth Planet. Sci. Lett., 373, 109- 117. Jolivet, R., Bürgmann, R., and Houlié, N. (2009). Geodetic exploration of the elastic properties across and within the northern San Andreas Fault zone. Earth Planet. Sci. Lett., 288(1), 126-131. Lei, J., and Zhao, D. (2009). Structural heterogeneity of the Longmenshan fault zone and the mechanism of the 2008 Wenchuan earthquake (Ms 8.0). Geochemistry, Geophysics, Geosystems, 10(10). McGuire, J., and Ben-Zion, Y. (2005). High-resolution imaging of the Bear Valley section of the San Andreas Fault at seismogenic depths with fault-zone head waves and relocated seismicity. Geophys. J. Int., 163(1), 152-164. Najdahmadi, B., Bohnhoff, M., and Ben-Zion, Y. (2016). Bimaterial interfaces at the Karadere segment of the North Anatolian Fault, northwestern Turkey. J. Geophys. Res.. doi: 10.1002/2015JB012601. Yang, W., Peng, Z., Wang, B., Li, Z., and Yuan, S. (2015). Velocity contrast along the rupture zone of the 2010 Mw6. 9 Yushu, China, earthquake from fault zone head waves. Earth Planet. Sci. Lett., 416, 91-97. Zigone, D., Ben-Zion, Y., Campillo, M., and Roux, P. (2015). Seismic tomography of the Southern California plate boundary region from noise-based Rayleigh and Love waves. Pure and Applied Geophysics, 172(5), 1007-1032.

55