Earthquake-Induced Chains of Geologic Hazards: Patterns

Home , 1920 Haiyuan earthquake, 2002 Denali earthquake, 2014 Ludian earthquake

REVIEW ARTICLE Earthquake‐Induced Chains of Geologic Hazards: 10.1029/2018RG000626 Patterns, Mechanisms, and Impacts Key Points: Xuanmei Fan1 , Gianvito Scaringi1,2 , Oliver Korup3, A. Joshua West4 , • Coupled surface processes initiated 5 5 3 6 by strong seismic shaking are Cees J. van Westen , Hakan Tanyas , Niels Hovius , Tristram C. Hales , important hazards in mountain Randall W. Jibson7 , Kate E. Allstadt7 , Limin Zhang8, Stephen G. Evans9, Chong Xu10, landscapes Gen Li4 , Xiangjun Pei1, Qiang Xu1, and Runqiu Huang1 • Earthquake‐induced landslides pose challenges to hazard and risk 1State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, assessment, management, and 2 mitigation Chengdu, China, Institute of Hydrogeology, Engineering Geology and Applied Geophysics, Faculty of Science, Charles • Multidisciplinary approaches University, Prague, Czech Republic, 3Institute of Earth and Environmental Science, University of Potsdam, Potsdam, further the understanding of the Germany, 4Department of Earth Sciences, University of Southern California, Los Angeles, CA, USA, 5Faculty of Geo‐ earthquake hazard cascade, yet Information Science and Earth Observation (ITC), University of Twente, Enschede, Netherlands, 6School of Earth and challenges remain Ocean Sciences, Cardiff University, Cardiff, UK, 7U.S. Geological Survey, Golden, CO, USA, 8Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Hong Kong, 9Department of Earth Supporting Information: 10 • Supporting Information S1 and Environmental Sciences, University of Waterloo, Waterloo, Ontario, Canada, Key Laboratory of Active Tectonics and Volcano, Institute of Geology, China Earthquake Administration, Beijing, China

Correspondence to: Q. Xu and R. Huang, Abstract Large earthquakes initiate chains of surface processes that last much longer than the brief [email protected]; moments of strong shaking. Most moderate‐ and large‐magnitude earthquakes trigger landslides, ranging [email protected] from small failures in the soil cover to massive, devastating rock avalanches. Some landslides dam rivers and impound lakes, which can collapse days to centuries later, and flood mountain valleys for hundreds of Citation: kilometers downstream. Landslide deposits on slopes can remobilize during heavy rainfall and evolve into Fan, X., Scaringi, G., Korup, O., West, debris flows. Cracks and fractures can form and widen on mountain crests and flanks, promoting increased A. J., van Westen, C. J., Tanyas, H., et al. fl (2019). Earthquake‐induced chains of frequency of landslides that lasts for decades. More gradual impacts involve the ushing of excess debris geologic hazards: Patterns, downstream by rivers, which can generate bank erosion and floodplain accretion as well as channel mechanisms, and impacts. Reviews of avulsions that affect flooding frequency, settlements, ecosystems, and infrastructure. Ultimately, earthquake Geophysics, 57. https://doi.org/10.1029/ 2018RG000626 sequences and their geomorphic consequences alter mountain landscapes over both human and geologic time scales. Two recent events have attracted intense research into earthquake‐induced landslides and Received 1 OCT 2018 their consequences: the magnitude M 7.6 Chi‐Chi, Taiwan earthquake of 1999, and the M 7.9 Wenchuan, Accepted 22 APR 2019 China earthquake of 2008. Using data and insights from these and several other earthquakes, we analyze Accepted article online 7 MAY 2019 how such events initiate processes that change mountain landscapes, highlight research gaps, and suggest pathways toward a more complete understanding of the seismic effects on the Earth's surface. Plain Language Summary Strong earthquakes in mountainous regions trigger chains of events that modify mountain landscapes over days, years, and millennia. Earthquake shaking can cause many tens of thousands of landslides on steep mountain slopes. Some of these sudden slope failures can block rivers and form temporary lakes that can later collapse and cause huge floods. Other landslides move more slowly, in some cases in a stop‐start fashion during heavy rains or earthquake aftershocks. Debris from these landslides can clog channels, and during heavy rainfall, the debris can be transported downstream for many kilometers with catastrophic consequences. New landslides tend to happen more frequently than usual for months to years following an earthquake because the strong ground shaking has fractured and weakened the slopes. Other effects of large earthquakes can last, in various forms, over geologic time scales. Over the past two decades, our understanding of these issues has advanced because of the detailed study of the 1999 Chi‐Chi earthquake in Taiwan and the 2008 Wenchuan earthquake in China. We compile and discuss the results of research on these and other earthquakes and explain what we have learned, what we still need ©2019. The Authors. to know, and where we should direct future studies. This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial‐NoDerivs 1. Introduction License, which permits use and distribution in any medium, provided the Earthquake‐triggered landslides and their consequences are major hazards in mountains. The destructive original work is properly cited, the use is non‐commercial and no modifica- earthquakes in Taiwan (M 7.6, 1999), China (M 7.9, 2008), Nepal (M 7.8, 2015), and New Zealand (M 7.8, tions or adaptations are made. 2016) caused widespread landsliding and prompted researchers to study with increasing detail how

FAN ET AL. 1 Reviews of Geophysics 10.1029/2018RG000626

earthquake‐triggered landslides erode landscapes. Early work by Simonett (1967) and Pearce and Watson (1983, 1986) recognized the importance of earthquakes in abruptly raising rates of erosion, sediment transport, and deposition. Detailed work on the 1999 Chi‐Chi, Taiwan earthquake (e.g., Dadson et al., 2004), proposed that earthquakes can change mountain landscapes instantaneously by triggering downslope movement of soils, debris, and rock landslides involving 103–1010 m3; more gradual flushing of these loose materials from the affected areas continues for some time after the triggering earthquake. Earthquake magnitude and depth are first‐order controls on the degree of landscape disturbance, modulated by the character of incoming seismic waves, topography, rock‐mass properties, groundwater conditions, and other factors (Fan, Scaringi, et al., 2018; Gorum et al., 2011). Seismic shaking is an instantaneous perturbation, but its effects attenuate, and the cumulative effects of earthquakes can leave a mark in the evolution of mountain landscapes (Hovius et al., 2011; Marc, Hovius, & Meunier, 2016). A cascade of processes actually takes place after a large continental earthquake (Figure 1). These processes bring differing degrees of hazard, which pose a risk when interacting with the human world. Spatial frequency distributions of earthquake‐triggered landslides are influenced by several factors, including source materials (i.e., the soil/regolith cover or the type of underlying intact bedrock), movement mechanisms (sliding, falling, flowing, or their combinations), ground‐motion characteristics, and sizes (in terms of planform area involved, depth, and volume displaced; Figure 1, red background). The shaking also can weaken slopes across the landscape and make them more prone to delayed failure (increased rate of occurrence). Days to years after the earthquake (blue background), landslide dams formed by earthquake‐ triggered landslides can breach and generate floods. Remobilizations and coalescence of landslide debris as a consequence of heavy rainfall can generate destructive debris flows. Years to decades later (yellow background), floods can continue, albeit less frequently. Delayed slope failures also can occur, and slow processes such as river aggradation will become apparent as the earthquake‐generated debris progressively moves downstream. Description and discussion around these processes and their cause‐effect links form the structure of this review, as presented in section 1.1. Many landslide disasters since 1900 have had an earthquake origin (Chen et al., 2012; Froude & Petley, 2018; Gorum et al., 2011; section 2.1). Investigating the patterns of these processes is essential for risk mitigation, emergency response, reconstruction, and increasing resilience (Das et al., 2018; Flentje & Chowdhury, 2018; Petley, 2011). Integrating these processes over longer time scales reveals the way earthquakes shape mountain belts and provides diagnostics for identifying past earthquakes in sediment and landforms. A large body of literature on specific aspects of the earthquake‐induced surface processes has grown over the past five decades. More than a thousand articles were published presenting case studies, models, and discussions on the impacts of the 2008 Wenchuan earthquake alone (Fan, Juang, et al., 2018). However, no comprehensive reviews have covered thus far the complete role of earthquake‐induced landslides and their consequences across a variety of landscapes (and seismic patterns) and in short‐ to long‐term time scales. Only a small number of specialized reviews and meta‐analyses touching upon specific aspects of this topic have been published. The most widely recognized works, which were relevant for the preparation of our comprehensive review, are reported in Table 1.

1.1. Structure of the Review Seismic shaking can trigger landslides of many sizes from small, shallow failures in soil to large, deep rock avalanches (Figure 1, red background). We use the expressions coseismic landslides or earthquake‐triggered landslides (EQTLs) to refer to them. Their distribution depends on the patterns of incoming seismic waves, geology, and topography, and earthquakes can produce hot spots of high landslide density. In section 2, we explain the current knowledge about EQTLs and ways to accurately and completely identify, map, and analyze their distribution (section 2.1); section 2.2 discusses the processes by which seismic shaking initiates slope failure. Section 3 is devoted to post‐seismic processes in rivers loaded by coseismic landslide debris (Figure 1, blue background). Section 3.1 summarizes characteristics of landslide dams and lakes. Section 3.2 describes how the post‐seismic landscape is sensitive to rainstorms so that more landslides than normal can occur in the months to years after the earthquake. Hillslopes weakened and fractured by shaking can experience

FAN ET AL. 2 Reviews of Geophysics 10.1029/2018RG000626

Figure 1. Chains of geologic hazards triggered by a strong continental earthquake and reviewed in this work. Causal relations between hazards are indicated. Red background shows different types of coseismic landslides; blue background indicates the post‐seismic cascade of hazards in days to years later; and yellow background represents the long‐term impact of an earthquake, years to decades later, and perhaps longer.

an accelerated degradation: Weathering proceeds at an increased rate, cracks form or expand, and post‐ seismic landslides initiate (Figure 1). Section 4 discusses improved tools for earthquake‐induced hazard and risk assessments. The cascade of surface processes initiated by an earthquake has the potential to produce damage, disrupt lifelines and services, and cause loss of life; a comprehensive hazard and risk appraisal has to acknowledge both the immediate and protracted consequences of earthquakes. Section 5 deals with the sediment cascade after an earthquake, from the ﬁrst years during which sediment mobility is at its peak to its eventual drop back to ambient rates. An important related question concerns

FAN ET AL. 3 Reviews of Geophysics 10.1029/2018RG000626

Table 1 A Summary of Key Papers, Reviews, and Meta‐Analyses on Earthquake‐Triggered Landslides and Related Topics Focus Key references

Summary and analyses of reports of earthquake‐triggered landslides and snow Keefer (1984, 1994, 2002); Podolskiy et al. (2010a) and ice avalanches Guidelines for the preparation of earthquake‐triggered landslide inventories Harp et al. (2011); Xu (2015) Spatial patterns and size statistics of earthquake‐triggered landslides Malamud et al. (2004); Marc et al. (2017); Marc, Hovius, and Meunier (2016) Physical methods for seismic slope stability analysis Jibson (2011) Physical and statistical landslide hazard assessments accounting for the seismic Godt et al. (2008); Jibson et al. (1998, 2000); Nowicki Jessee et al. (2018); shaking Nowicki et al. (2014) Near‐ﬁeld and far‐ﬁeld patterns of earthquake‐triggered landslides Delgado et al. (2011); Gorum et al. (2011); Meunier et al. (2007, 2008) Post‐seismic landslide rates Marc et al. (2015); Fan et al. (2019) Sediment export from seismically active mountains and the mass balance of Hovius et al. (1997, 2000); Marc, Hovius, and Meunier (2016) earthquakes

how long it takes for coseismic debris to evacuate the affected area versus remaining stored in place. Sediment deposits can be valuable archives for paleoseismic research. Section 6 broadens the picture of earthquake‐induced landscape erosion by considering the volumes and fate of sediment (Figure 1, yellow background) produced over multiple earthquake cycles to estimate their contribution to the geologic evolution of mountain landscapes. In section 7, we summarize our current understanding of how earthquakes change mountain landscapes and highlight some perspectives for future research. We offer four technical supplements that present techniques for analyzing EQTL inventories (supporting information section S1), modeling landslide runout (section S2), mechanisms and modeling of debris ﬂows (section S3), and strategies for risk mitigations (section S4). We also include a Glossary of technical terms and acronyms. We focus on terrestrial settings, particularly mountainous areas. We do not consider submarine landsliding, the generation of tsunamis by coastal or offshore earthquakes and their geomorphological and environmental consequences, or earthquake‐volcano interactions. These are subjects of extensive research (Avouris et al., 2017; Dawson, 1994; Hill et al., 2002; Linde & Sacks, 1998; MacInnes et al., 2009; Manga & Brodsky, 2006; Richmond et al., 2011) beyond the scope of our review, as are studies on EQTL‐generated tsunamis in mountain lakes (e.g., Ichinose et al., 2000; Kremer et al., 2012; Schnellmann et al., 2002) and site effects of seismic shaking due to soil and rock types (Bhattacharya et al., 2011; Evans et al., 2009; Huang & Yu, 2013; Kayen et al., 2013; Wang et al., 2014; Zhang & Wang, 2007).

2. Coseismic Landslides Coseismic landslides, or EQTLs, are mass movements triggered within seconds to minutes of strong ground shaking. These can include small, shallow soil failures and rock falls; large, deep slumps and slides; and rapidly moving, devastating rock avalanches. All downslope mass movements that occur after that are referred to as post‐seismic landslides. The key to understanding coseismic landslide hazard lies both in understanding regional patterns of the entire collection of landslides triggered by a given earthquake and understanding the triggering and runout processes at the local or even laboratory scale. In the following section, we discuss different ways coseismic landslides are studied and what has been learned so far from these studies, starting from the regional perspective and working down to the laboratory scale.

2.1. Mapping and Spatial Patterns 2.1.1. Inventories of EQTLs Detailed inventories are key to study the distribution of coseismic landslides, to assess the main mechanisms of slope failure, and to generate earthquake‐induced landslide hazard maps. In practice, it can be very difﬁ- cult to distinguish coseismic landslides from mass movements that have been generated either before the earthquake or by aftershocks in the days/weeks between the earthquake occurrence and the acquisition of images used for mapping. The challenge is to generate so‐called event‐based inventories (Guzzetti

FAN ET AL. 4 Reviews of Geophysics 10.1029/2018RG000626

et al., 2012) or multiple‐occurrence regional landslide events (Crozier, 2005) that depict the mass movements triggered by a single earthquake.

The generation of EQTL inventories has received much attention over the past decades. Keefer (1984) studied landslides triggered by 40 strong historical earthquakes from the 1811/1812 New Madrid, Missouri sequence (three events M ~ 7.5) to the 1980 Mammoth Lakes, California (M 6.1) earthquake. His study correlated earthquake magnitude with the area over which landslides were triggered, the maximum epicentral distance, and the maximum fault‐rupture distance; some of these data were updated and refined subsequently (Keefer, 2002). Similar efforts were carried out by Rodrıguez et al. (1999), and national databases also have been generated (Hancox et al., 2002; Papadopoulos & Plessa, 2000; Prestininzi & Romeo, 2000). Mapping EQTLs uses at least four methods (Guzzetti et al., 2012; Keefer, 2002; Soeters & van Westen, 1996; van Westen et al., 2008; Xu, 2015): (1) visual image interpretation, (2) (semi)automated classification based on spectral characteristics, (3) (semi)automated classification based on elevation differences, and (4) field investigation (Table 2).

Visual image interpretation is the most common approach for generating EQTL inventories (Schmitt et al., 2017; Tanyas, Allstadt, & van Westen, 2018). Interpreting aerial photos aided by stereoscopic vision allows tracking of features and remains a benchmark in landslide mapping (Guzzetti et al., 2012). Yet obtaining aerial photography of areas affected by earthquakes is expensive, which curtails the completeness of some inventories (Dai et al., 2011; Xu, Xu, & Shyu, 2015). Since the launch of NASA's (National Aeronautics and Space Administration) Landsat‐1 in 1972, satellite images have afforded a more systematic coverage of landslides. The stereoscopic capacity and higher spatial resolution (10–20 m) of SPOT (launched in 1986) images proved very useful in mapping landslides, and subsequently, Landsat TM and ETM+, IRS 1C/1D, and SPOT imagery have been used widely for preparing inventories (Crowley et al., 2003; Gupta & Saha, 2001; Salzmann et al., 2004; Zhou et al., 2002). The entry of Space Imaging Inc. (now DigitalGlobe) into the satellite market with IKONOS‐2 in 1999 was a milestone in landslide mapping. With 1‐m resolution, the image quality rivaled that of aerial photos of 1:10,000 scale (Nichol & Wong, 2005). More recent platforms include commercial satellites such as WorldView (0.31 m), GeoEye (0.41 m), Pleiades (0.5 m), and QuickBird (0.82 m). Today, Planet Constellation offers image resolutions of 0.72 (SkySat) to 3 m (Dove, RapidEye) for a given area daily. The Copernicus Sentinel‐2 mission provides free multispectral images with 10‐m resolution that cover the Earth between 56°S and 84°N, on average once every 3 days (Figure 2). Aerial reconnaissance from fixed‐wing aircraft or helicopter is a rapid, though selective, way to assess the scope of landsliding in an earthquake‐affected area (Rosser et al., 2014; Van Dissen et al., 2013), but it is not suitable for systematic surveys. Unmanned Aerial Vehicles (UAVs) can provide detailed videos, images, 3‐D point clouds, and ortho‐rectified images for parts of the affected area or for individual landslides (C. Tang et al., 2016), but they are too limited in radius to map landslides over large areas. Google Earth Pro stores historical imagery that aids change detections and can be used for EQTL mapping free of cloud cover (Gorum et al., 2013). Specific crowdsourcing tools for landslide mapping are now becoming available, such as NASA Landslide Reporter 2019 (Kirschbaum & Stanley, 2018), although more work needs to be done to make them suitable for collaborative web‐mapping of EQTL inventories due to lack of recent images and specific tasking per mapper, as is the case in Humanitarian OpenStreetMap 2019 (Winsemius et al., 2019).

The interpretation of satellite or aerial imagery should ideally involve stereo images and experts trained in identifying, classifying, and validating the landslides based on image tone, texture, shape, size, and pattern (Soeters & van Westen, 1996). Landslides should be mapped as polygons rather than points to document their areas and volumes. Where possible, landslide polygons should separate initiation areas from runout areas because only source areas are used in statistical landslide hazard assessment. The identification of individual landslides is important but is difficult to pinpoint where they merge. Landslides should be classified by type, depth class, and river‐blocking potential (Figure 3). Semiautomated satellite‐image classification is a rapidly developing tool producing increasingly reliable landslide sizes (Fan, Juang, et al., 2018), though careful expert validation is still needed (hence the term semiautomated). Image classifications based on spectral image characteristics are increasing in accuracy through the use of unsupervised and supervised algorithms, machine‐learning approaches, and object‐based image analysis (Anders et al., 2011; Moosavi et al., 2014; Stumpf & Kerle, 2011). Individual landslide polygons

FAN ET AL. 5 Reviews of Geophysics 10.1029/2018RG000626

Table 2 Overview of Techniques for Creating Earthquake‐Induced Landslide Inventories (After van Westen et al., 2008) Group Advantages Disadvantages Speciﬁc technique Advantages Disadvantages

Visual image Expert interpretation Subjective method can UAV video Detailed mapping of Only applicable over interpretation allows for detailed and give different results or images landslides in small small areas. accurate mapping of depending on areas. Could be Permission to use landslides as polygons, experience and available as overflight UAVs can be with relevant attributes, capability of videos, individual problematic in many such as landslide type, the mapper. images, or 3‐D views areas. Clouds, large and differentiation It is time‐consuming, of point clouds and altitude differences, between erosional, and inventories ortho‐rectified images. and wind can transport, and only become influence the survey. accumulation parts. available after Aerial video or Cover larger areas and Very expensive to hire Several new several weeks/months. images from flying height can be helicopter/airplane. developments can Spatial accuracy can be helicopter/ adjusted to obtain Cloud cover can be a speed up mapping, problematic if mappers airplane more detailed images. major obstacle. such as collaborative use incorrectly Also, video coverage is web mapping, use of georeferenced images. possible during same social media, UAV Large differences in flight. Stereoscopic images, and Google mapping can exist images are ideal for Earth history viewer. between different landslide inventory mapping teams for the mapping. same area. Using history Direct comparison of Geometric distortion can viewer in images from different be a problem, and Google Earth periods. No additional conversion from Pro or costs for acquiring Google Earth KMZ to collaborative images. Specific tools GIS is cumbersome. web mapping. for collaborative Post‐event images web‐mapping of might not be available landslides are now or have cloud available, such as problems. NASA Landslide Reporter. High‐resolution Cover large areas and can Persistent cloud cover satellite images be acquired at different can hinder image times to show acquisition. High costs evolution of landslides. involved. Can be converted into stereoscopic images for optimal interpretation. High‐resolution Shaded relief maps from Acquiring LiDAR data shaded relief LiDAR‐derived over large areas in a maps (e.g., bare‐surface models post‐seismic situation from LiDAR are the best to map is very expensive, and or UAV) landslide forms even processing is time under forest cover. consuming. Stereo‐image interpretation of such images is ideal tool for detailed landslide interpretation. (Semi)automated Rapid assessment of Requires cloud‐free Change Relatively fast method to Requires good quality classification landslide‐affected images, which can be detection analyze changes in and medium‐ to based on spectral areas, over wide problematic to acquire land cover by using high‐resolution characteristics regions, and commonly in certain areas and both pre‐ and spectral images before the best approach for a seasons. Satellite images post‐seismic imagery and after the rapid assessment soon can be costly, although and analyzing the earthquake, which can after the earthquake. decrease in cost and changes in Normalized be problematic to Some advanced lower‐resolution Green Vegetation acquire due to cloud methods also allow alternative are free. Index in case of cover. identification of Misclassification of multispectral images landslide types. Rapid areas, overprediction of or graytones in case of developments using landslide areas. Difficult

FAN ET AL. 6 Reviews of Geophysics 10.1029/2018RG000626

Table 2 (continued)

Group Advantages Disadvantages Speciﬁc technique Advantages Disadvantages

machine learning and to separate Panchromatic (black cloud computing allows amalgamated and white) images. for better results. landslide zones. Pixel‐based Rapid method to detect Cloud coverage during Very limited methods bare areas after an data acquisition. information on earthquake, based on Oversimplification of Normalized Green landslide areas. Vegetation Index. Confusion with other Machine learning bare‐land use areas. approaches and cloud computing provide opportunity for rapid classification in future. Object‐Based Using spectral Scale parameter that Image information in determines the size of Analysis combination with other individual polygons is characteristics (e.g., difficult to determine. shape and slope) to Misclassifications are outline individual common. polygons with similar aspects and remove false positives based on other characteristics. Also, landslide types can be identified. (Semi)automated Comparison of Pre‐earthquake Structure Photogrammetric UAV images can only be classification high‐resolution digital Digital Elevation for motion technique for generated for based on elevation models allows Models are generation of point relatively elevation calculation of small generally of low clouds and Digital small areas. In some differences changes in elevation quality, thus making Elevation Models from areas, it is difficult to caused by landslide it difficult to quantify images derived from obtain permits for depletion or changes. Post‐seismic UAV's. Fast and rapid flying UAVs. accumulation. Can be high‐resolution method. used also as monitoring DEMs are expensive Photogrammetry Photogrammetric Image acquisition costs tool in post‐seismic and time‐consuming techniques for can be high, and situation. to generate over large generating Digital cloud‐free images can areas using LiDAR. Elevations Models from be difficult to acquire. overlapping aerial Specific stereo satellite photographs or satellite images (e.g., Pleiades) images. Cover large are expensive. areas with good accuracy. Laser scanning Terrestrial Laser Acquiring LiDAR data Scanning provides overlarge areas in a rapid point clouds of post‐seismic situation slopes. Airborne Laser is very expensive, and Scanning provides very processing is detailed point clouds time‐consuming. over surface also under forest cover. Interferometric Interferometric Synthetic Cannot be used for rapid Synthetic Aperture Radar landslide events and is Aperture provides information therefore not suitable Radar on slow‐moving for mapping rapid landslides over large earthquake induced areas using relatively landslides. PSInSAR low‐cost images. (Permanent PSInSAR (Permanent Scatterers Scatterers Interferometry) Interferometry) and other techniques requires sufficient provide millimeter

FAN ET AL. 7 Reviews of Geophysics 10.1029/2018RG000626

Table 2 (continued)

Group Advantages Disadvantages Speciﬁc technique Advantages Disadvantages

accuracy of permanent scatterers displacements and is (buildings and rocks). particularly suitable for post‐event monitoring of displacements. Field investigation Field methods are Field methods only allow Geomorphological Conventional method for It is only possible to visit methods essential for obtaining evaluation of small mapping evaluating the small areas. Access to the field evidence of areas or even individual geomorphologic and steep areas is difficult landslides and form the landslides. Cannot be geologic setting of the or impossible. validation of the other used to map landslides earthquake‐triggered Commonly difficult to methods. over large areas. landslides. Essential for get a good overview of understanding the the phenomena. causal mechanism of landslides. Ultimate validation method for the other remote methods. Mobile GIS Using mobile GIS and Battery life of devices can GPS for attribute data be problematic, as will collection allows rapid the readability of mapping and is used in screens in sunny combination with the environments. GPS above remote methods, signal receipt might be from which the problematic in steep landslide inventory is terrain. Data can be obtained, which is then lost when equipment validated and adjusted. breaks. Road surveys Relatively fast method to Only limited to roads and evaluate landslide mapping of landslides events along road away from roads is networks. Can also use limited. Therefore, the dashboard cameras and inventory will be software (e.g., biased and difficult to Mapillary or Google use in area‐wide Street View) to collect studies. Roads could data and map later in be blocked. the office. Interviews Using questionnaires, Local population could workshops, and so be in shock as a result forth, local population of the event and might can provide very useful have moved away to information on how the shelters. They might landslides behaved exaggerate/underplay during the earthquake. information for Essential for specific motives. understanding the Cross‐checking mechanism and impact is needed. of the event. Geophysical Wide range of tools Point‐based surveys available to measure investigations give geometrical, only information at geotechnical, site investigation level. geophysical, and Difficulty in accessing hydrological properties sites with equipment. that are essential for Expensive and analyzing the stability time‐consuming. of earthquake‐triggered landslides. Note. GIS = Geographic Information System; UAV = Unmanned Aerial Vehicle.

FAN ET AL. 8 Reviews of Geophysics 10.1029/2018RG000626

Figure 2. Timeline of development of optical satellite sensors that are used in resource and hazard inventories.

now can be outlined successfully using Object‐Based Image Analysis (K. ‐T. Chang, Hwang, et al., 2011), although problems of separating merged landslide deposits remain. Spectral, topographic, and shape metrics also can characterize EQTLs according to their type (Martha et al., 2010). Semiautomated methods also include comparisons of (mostly LiDAR‐ or UAV‐derived) digital elevation models (DEMs) and their pixel‐wise differences before and after an earthquake. This requires removing artificial objects and vegetation to generate surface models. Several nonoptical sensors overcome the issue of cloud cover (Williams et al., 2017) while tracking surface deformations at millimeter precision via synthetic aperture radar (Casagli et al., 2016; Guzzetti et al., 2012). Although this method is not suitable for detection of rapid EQTLs with long runout, it has potential for monitoring the activity of coseismic landslides in the months and years after an earthquake. Field checking of landslide inventories remains essential for validating these remote sensing techniques (Table 2; Harp et al., 2016; Harp & Jibson, 1995, 1996). Although these methods generally are not used as the primary tool for EQTL inventory mapping due to difficulty in accessing the entire affected area after an earthquake, they are essential for validating the other methods. Mapping landslides in the field provides detailed data on type, internal composition, and failure mechanism of landslides, which commonly are

FAN ET AL. 9 Reviews of Geophysics 10.1029/2018RG000626

Figure 3. Example of mapping earthquake‐induced landslides as polygons with attribute information.

difficult to assess using remote sensing. Field mapping of landslides is time‐consuming and intractable for the thousands of landslides triggered during earthquakes. Despite, or maybe even because of, the variety of methods, EQTL inventory mapping has led to greatly differing results. For example, Parker et al. (2017) used an automatic and visually cross‐checked mapping method for landslides triggered by the 2008 Wenchuan earthquake. Their results differed significantly from those based on visual interpretation alone (Fan, Juang, et al., 2018; Gorum et al., 2011; Xu, Xu, Shen, et al., 2014), mainly because adjacent landslides were generally mapped as a single larger landslide, which overes- timated landslide size and underestimated landslide counts (Figure 4). Poorly mapped landslide locations also can affect the accuracy of spatial queries regarding causative factors such as rock or soil type. Given the large differences in EQTL inventories, standard criteria for assessing the quality of landslide inventories are desirable (Marc & Hovius, 2015; Tanyas et al., 2017; Xu, 2015). Several studies (Guzzetti et al., 2012; Harp et al., 2011; Xu, 2015) proposed criteria, for example, (1) all identifiable landslides should be mapped; (2) landslide boundaries should be mapped as polygons rather than just mapping a landslide point location; (3) regional landslide mapping should span the entire affected area, and the outline of the mapped area should be indicated; (4) complex landslides should be divided into distinct groups; (5) landslides should be mapped as soon as possible after the earthquake; (6) stereo coverage or draping over a DEM to allow oblique views is preferred; (7) landslides must be plotted in a projected coordinate system; and (8) visual interpretation is preferable over automated extraction. Few mapping studies have met all these criteria (e.g., Gorum et al., 2014; Xu et al., 2013; Xu, Xu, & Shyu, 2015), which might explain some of the large differences in size and positional accuracy between inventories (Marc & Hovius, 2015; Xu, Xu, Yao, & Dai, 2014; Figure 4). Many studies summarized the properties of EQTL inventories (Gorum et al., 2014; Harp et al., 2011; Havenith et al., 2016; Keefer, 1984, 2002; Rodrıguez et al., 1999; Tanyas et al., 2017; Xu, 2015). The most complete is that of Tanyas et al. (2017), which compiled data on 363 earthquakes and collated 66 digital coseismic landslide inventories that were classified as (1) comprehensive digital inventories with full coverage of all landslides and their types; (2) noncomprehensive digital inventories with partial coverage, type‐only descriptions, or point data; (3) reports on EQTLs without a digital inventory; (4) reports on EQTLs without any inventory; and (5) earthquakes with potential, though no reported landslides. Tanyas et al. (2017) also presented the associated metadata for digitally available EQTL inventories. The provided metadata contain

FAN ET AL. 10 Reviews of Geophysics 10.1029/2018RG000626

Figure 4. Comparison of earthquake‐triggered landslide mapping of the same area by different groups. Example from the epicentral area of the 2008 Wenchuan earthquake. (a) A post‐seismic satellite image showing the landslides that were triggered and subsequently mapped. (b)–(d) shows a point inventory mapped by Gorum et al. (2013) and polygon inventories (red) mapped by three different groups: (b) Z. Z. Li, Jiao, et al. (2014); (c) Xu, Xu, Yao, and Dai (2014); and (d) Fan, Domènech, et al. (2018).

valuable information for an overall evaluation of inventories. Considering the metadata, the users can decide if an inventory is appropriate for the purpose of their application. This also alerts users about some data limitations. For example, in this database, some inventories include landslides triggered by a sequence of earthquakes rather than a single mainshock (e.g., 1980 Mount Diablo and 1993 Finisterre Mountains). Schmitt et al. (2017) created an open repository to host the digital inventories that they have permission to share through the U.S. Geological Survey ScienceBase platform (Table S1 and Figure 5). New landslide inventories are those from the 2016 M 7.8 Kaikōura, New Zealand (Massey et al., 2018; Sotiris et al., 2016), 2017 M 6.5 Jiuzhaigou, China (Fan, Scaringi, et al., 2018; Xu, Wang, et al., 2018), and 2012 M 5.7 Yiliang, China (JJ. Zhang, Wang, Zhang, et al., 2014) earthquakes. The largest and most detailed of the inventories is that of the 2008 Wenchuan, China earthquake, which now contains about 200,000 individual coseismic landslides covering a total area of 1,160 km2 (Xu, Xu, Yao, & Dai, 2014); a subinventory lists 828 coseismic landslide dams (Fan, van Westen, et al., 2012). As can be seen from Figure 5, EQTL events have occurred in most of the tectonically active mountain areas in the world, with a predominance in warm, temperate, and humid regions. However, earthquake‐triggered snow and rock avalanches also have occurred in cold regions, where they can travel long distances through ice and snow cover. Glazovskaya et al. (1992) determined that the 6.2% of the Earth's surface that has snow‐ cover depth exceeding 30–50 cm, a slope steepness >17°, and a slope height of 20–30 m are prone to snow avalanches. Until recently, very little systematic documentation of earthquake‐triggered snow avalanches has been made (Chernouss et al., 2006; Podolskiy et al., 2010a). The most comprehensive and recent publication by Podolskiy et al. (2010a) reports 22 historical cases of earthquake‐triggered snow avalanches that occurred between 1899 and 2010 in cold Arctic regions and highly elevated mountainous terrains (e.g., the Himalayas). Important cases include rock and ice avalanches in the Chugach Mountains triggered by the 1964 Alaska earthquake (Post, 1967; Tuthill & Laird, 1966); the 1,580 large landslides triggered by the 2002 Denali earthquake over the glaciated terrain of Alaska (Gorum et al., 2014; Jibson et al., 2004); the 1,448 snowmelt‐induced landslides following the 2004 Mid‐Niigata prefecture earthquake (Akiyama et al., 2006), and the long runout landslides triggered over snow during the 2011 Nagano prefecture earthquake in Japan (Has et al., 2012; Yamasaki et al., 2014).

FAN ET AL. 11 Reviews of Geophysics 10.1029/2018RG000626

Figure 5. Distribution of earthquakes having digitally available earthquake‐triggered landslide inventories listed in Table S1 (Supplementary Information, adapted from Tanyas et al., 2017; Schmitt et al., 2017). Data repository can be accessed online (https://www.sciencebase.gov/catalog/item/583f4114e4b04fc80e3c4a1a). The background is from Amante and Eakins (2009; https://doi.org/10.1594/PANGAEA.769615).

2.1.2. Size Statistics Size statistics of landslides provide useful input for hazard assessments (e.g., Guzzetti et al., 2005) and studies of landscape evolution (e.g., Korup et al., 2012) because probabilities of landslide size, total area, and volume can be estimated from their empirical distributions. These size distributions are mostly reported as normalized histograms. 2.1.2.1. The Frequency‐Area Distribution of Landslides Many studies argue that in most EQTL inventories, the distributions of medium and large landslide areas follow an inverse power law:

f ¼ cAβ; (1)

where f is the frequency density, c is a dimensional constant, A is the landslide area, and β is a scaling exponent (Figure 6; e.g., Malamud et al., 2004). This relation also has been reported for landslide triggers such as storms and rapid snowmelt (Malamud et al., 2004), and its seemingly universal applicability has motivated the application of these statistics for hazard assessments (Guzzetti et al., 2005). In most EQTL inventories, small landslides depart from the power law and form a rollover (Figure 6a) below which they are less frequent (e.g., Malamud et al., 2004; Tanyas, van Westen, et al., 2018). The approximate location where the frequency‐size distribution diverges from power law is commonly referred to as the cutoff and is interpreted either as a statistical sampling artifact or as a function of the bulk property of the hillslopes; the scaling exponent describes the relevance of larger landslides (Figure 6). 2.1.2.2. The Magnitude Scale of Landslide Events Frequency‐area distributions can be used to identify a landslide‐event magnitude scale (Malamud et al., 2004) to characterize the geomorphic severity of an earthquake. Keefer (1984) pioneered this idea by considering the log‐transformed total landslide count triggered by an earthquake: An event triggering 102–103 landslides is classified as a 2; 103–104 landslides is classified as a 3, and so forth. Malamud et al. (2004) suggested that the completeness of landslide inventories can be assessed by comparing the frequency‐size distribution of a partial inventory with the proposed empirical curves (section S1). However, using 45 earthquake‐induced landslide inventories, Tanyas, van Westen, et al. (2018) found that the empirical curves proposed by Malamud et al. (2004) did not fit many inventories and that fitting was subjective and prone to

FAN ET AL. 12 Reviews of Geophysics 10.1029/2018RG000626

Figure 6. Some examples of probability‐area distributions derived from digital earthquake‐triggered landslide inventories (Table S1) mapped for different earthquakes showing (a) the curves fit by a double‐Pareto distribution using the code of Rossi et al. (2012) and (b) power law fits. Probability‐density distributions (gray lines) are plotted for 32 earthquake‐triggered landslide events from Tanyas, van Westen, et al. (2018). Characteristic features are labeled in inset plots. A few examples from specific earthquake inventories are highlighted in color to show the range of typical curve shapes (left) and the upper and lower bounds observed for power law exponents (right).

uncertainty. They updated the method allowing for more flexible curve shapes having different β values and proposed a more reproducible and robust way to determine the landslide‐event magnitude. Reliable landslide‐event magnitudes might allow direct comparison of different landslide‐triggering events. Tanyas et al. (2019) proposed a regression equation of globally available morphometric and seismic variables for the near‐real‐time prediction of EQTL‐event magnitude, total landslide area (Malamud et al., 2004; Tanyas, van Westen, et al., 2018), and volume (Malamud et al., 2004). 2.1.2.3. The Relation Between Landslide Volume and Surface Area Compared to landslide area, determining landslide volume is difficult because it requires knowing either pre‐ and post‐landslide surface geometry or the subsurface geometry of the slope failure (Guzzetti et al., 2009). Even techniques such as subtraction of high‐resolution pre‐ and post‐landslide DEMs can only estimate the volumetric change because some of the source area can overlap with deposits (C. Tang et al., 2019). Therefore, empirical relations commonly are used to estimate landslide volume. Several authors argue for a power law relation between the landslide‐affected area (A) and volume (V), that is, γ V ¼ αA ; (2) and that landslide geometry is irrespective of specific mechanical properties (Guzzetti et al., 2009; Klar et al., 2011). For example, Guzzetti et al. (2009) examined 677 slides and estimated a scaling exponent (γ) of 1.45. Klar et al. (2011) used limit‐equilibrium principles to explore the relation between the landslide volume and surface area and argued that γ ranges between 1.32 and 1.38. Larsen et al. (2010) noted considerable variation in the scaling exponent with landslide material: γ values for soil are 1.1–1.3 and for bedrock are 1.3–1.6. They also showed that total landslide volume is very sensitive to γ such that small differences in γ can cause a prediction error of one or more orders of magnitude in total landslide volume. 2.1.2.4. Uncertainties in Landslide Size Statistics Korup et al. (2012) stated that minute numerical errors in model parameters of frequency‐size distribution of landslides can cause uncertainty greater than a factor of 2 in total landslide volumes. They conclude that such minor errors can cause large uncertainties in erosion or mobilization rates inferred from size statistics. To address this issue, frequency‐size distribution analyses need to be conducted using reliable EQTL inventories, and the uncertainties caused by inventory data need to be examined in detail. The reliability of landslide size statistics hinges on the reliability of the inventory. Four major issues can compromise the quality of a landslide inventory and its size statistics: (1) mapping accuracy of landslide

FAN ET AL. 13 Reviews of Geophysics 10.1029/2018RG000626

boundaries, (2) coalescing landslides, (3) overlapping landslide sources and deposits, and (4) geometric distortions in the image and due to topography. Properly georeferenced imagery with adequate spatial resolution is needed for the accurate delineation of landslide boundaries and thus for the accurate assessment of landslide areas. The level of expertise of mappers and the time invested in mapping are important, even when an inventory is created using semiautomated mapping techniques. The out- lines of landslides can be difficult to map if images are poor or contrasts low or if two or more landslides coalesce or gradually do so after an earthquake; if many such cases exist, this can also alter landslide size distributions (e.g., Marc & Hovius, 2015; Tanyas, Allstadt, & van Westen, 2018) Figure 7. Data distribution between earthquake magnitude and power law exponent fit to probability‐density distributions with uncertainties calcu- and bias volumetric estimates (Z. Li, Jiao, et al., 2014). The factors control- lated by using the data from Tanyas, Allstadt, and van Westen (2018) for 45 ling landslide runout differ from those controlling slope failure that create earthquake‐triggered landslide inventories. the landslide source area (Frattini & Crosta, 2013). Thus, landslide source area should be treated separately in landslide frequency‐size analyses because the depositional part of landslides might affect landslide‐size statistics, particularly for long‐runout landslides (Tanyas, van Westen, et al., 2018). However, landslide sources and deposits are rarely mapped separately (e.g., Frattini & Crosta, 2013), particularly in EQTL inventories (Tanyas et al., 2017). To address this issue, Tanyas, van Westen, et al. (2018) exclude debris flows from the examined EQTL inventories because of their larger runout and depositional areas. Fixed ratios between source and deposit areas (Larsen et al., 2010) can be a rough guide to differentiate sources of landslides (e.g., Marc & Hovius, 2015; Marc, Hovius, & Meunier, 2016). 2.1.2.5. Interpretation of Landslide Size Statistics Although no clear physical explanation has been found (Hergarten, 2003), power law distributions of EQTL inventories are typical of other natural hazards such as earthquakes and forest fires (e.g., Hergarten, 2003; White et al., 2008). Topography is a candidate explanation for the power law distribution of landslides (e.g., Frattini & Crosta, 2013; Liucci et al., 2017; ten Brink et al., 2009) because topography is the main factor limiting the boundaries of sliding material. Valagussa et al. (2019) argue that the intensity of ground motion has a stronger influence on landslides size distribution and noted that stronger ground shaking can trigger larger landslides. Yet only three out of six EQTL inventories they examined supported this line argument. Ground‐motion characteristics such as frequency content and duration might also alter the size distribution (Jibson, 2011). For example, Jibson et al. (2004) report that the 2002 Denali earthquake triggered significantly lower concentrations of small rock falls and rock slides compared to earthquakes with magnitudes at least as high. They argued that this was because the earthquake was deficient in high‐frequency energy and attendant high‐amplitude accelerations. This hypothesis has not been tested yet and thus requires further analysis. The power law exponents (β values) of landslide inventories range from 1.4 to 3.7 (Stark & Guzzetti, 2009; Tanyas, van Westen, et al., 2018; Van Den Eeckhaut et al., 2007; Figure 7). In terms of hazard assessment, a large value of β indicates a larger proportion of small to medium landslides. A lower β, on the other hand, signifies higher proportion of large landslides. Differences in β might reflect regional differences in structural geology, morphology, hydrology, and climate (Bennett et al., 2012; Chen, 2009; Densmore et al., 1998; Dussauge‐Peisser et al., 2002; Hergarten, 2012; C. Li, Ma, et al., 2011; Sugai et al., 1995); however, no compelling relationship has been discovered as yet partly because of the uncertainty in estimating β, which can be about 40% in some cases (Tanyas, van Westen, et al., 2018). Similarly, the relation between earthquake magnitudes and power law exponents of landslide size distributions remains noisy (Figure 7). As mentioned above, also the reason for the rollover and the divergence from the power law is controversial: Some have argued that it is controlled by the mechanical properties of the substrate (Bennett et al., 2012; Frattini & Crosta, 2013; Guthrie et al., 2008; Pelletier et al., 1997; Stark & Guzzetti, 2009; Van Den Eeckhaut et al., 2007); others have speculated that it can be in part a mapping artifact caused by lack of spatial (Hungr et al., 1999; Stark & Hovius, 2001) or temporal (Tanyas, van Westen, et al., 2018; Williams et al., 2018) resolution. If the rollover is purely a mapping artifact, then refined models of the size distribution of landslides are needed. Tanyas, van Westen, et al. (2018) found that some inventories lack rollovers

FAN ET AL. 14 Reviews of Geophysics 10.1029/2018RG000626

(Figure 6a), and they noted that both physical explanations and other arguments can work together in forming a rollover. Some studies consider the rollover to be the lowest landslide size at which the inventory can be assumed to be complete (Parker et al., 2015; Van Den Eeckhaut et al., 2007). 2.1.3. Spatial Patterns and Controlling Factors Coseismic landslides are triggered as a result of a complex interaction between several triggering factors, in particular, seismicity and predisposing factors such as soil/rock mass properties, slope geometry, and structural features (Table 3). The triggering factors associated with ground shaking are the most important because they control the regional pattern of EQTL distribution (Nowicki et al., 2014). These controlling factors are highly interrelated. For example, the soil/rock mass properties of a slope are primarily controlled by the physical properties of a lithological unit forming the slope. However, to assess the mass properties of a slope, other factors must be considered, including weathering processes, the effects of previous earthquakes/landslides, anthropogenic influence, land cover, and land use types (Table 3). Some of the controlling factors have several interactions. For example, weathering processes affect both the soil/rock mass and geometry of a slope. Another example is that climate and weathering interact because the moisture content can enhance chemical weathering. Each EQTL event has unique combinations of these controlling factors that control the characteristics of each event (Tanyas, van Westen, et al., 2018), such as the locations, types, sizes, and the total areas affected as well as the density and the total number of EQTLs. For example, the total area of landslides triggered by the 2008 Wenchuan earthquake (M 7.9) was about tenfold that of EQTLs triggered by the 2015 Gorkha earthquake (M 7.8), even though they had similar magnitudes and occurred in similar geomorphic and seismotectonic settings. The difference was caused by specific controlling factors related to the characteristics of the fault rupture and its orientation (Kargel et al., 2016). More efforts are needed to better understand the controlling factors and their interactions to be able to predict the distribution of coseismic landslides in the future. Doing that will require the controlling factors to be spatially represented so that they can be used in a predictive model. Table 3 gives a general indication of whether the various controlling factors can be mapped. Assessment of slope stability is a geotechnical problem and requires the identification of each component of driving and resisting forces along a slip surface. This requires a comprehensive investigation beginning from the physical properties of the slope material to the geometry of the sliding surface, which can be either a discrete surface or a zone of deformation. Full‐fledged summaries of these investigations are given by Wyllie and Mah (2014) and Duncan et al. (2014), among others. In regional studies, detailed geotechnical data required for deterministic evaluation of slope stability are not available in most cases. Therefore, researchers prefer either simplified physical modeling approaches (Jibson et al., 1998, 2000) or statistical methods (Nowicki et al., 2014; Nowicki Jessee et al., 2018) considering proxies for causal factors in regional studies. Whereas the controlling factors for EQTL are to some extent comparable to those for rainfall‐triggered landslides, the common factors related to land use, hydrology, soils, and geology are not described in detail here. Descriptions of their importance can be found in van Westen et al. (2008) and Reichenbach et al. (2018). 2.1.3.1. Earthquake Characteristics Earthquake characteristics, such as magnitude and depth, play an important role in EQTL distribution (Keefer, 1984; Rodrıguez et al., 1999). However, earthquakes having similar magnitudes can cause different levels of shaking and damage (Wald et al., 2003) because of the interaction of other factors such as energy released by faulting, directivity, topographic amplification, and shaking frequency and duration. For example, for earthquakes having unilateral fault rupture (e.g., the 2008 Wenchuan and 2015 Gorkha earthquakes; Roback et al., 2018; Xu, Xu, Yao, & Dai, 2014; Xu et al., 2016), the ground motions can be substantially different along the fault. This commonly results in a stronger concentration of coseismic landslides along the ruptured fault. Another factor controlling the ground motion is the fault‐rupture mechanism. Different types of seismogenic faults commonly generate varying EQTL patterns (Gorum & Carranza, 2015; Meunier et al., 2013; Tatard & Grasso, 2013). Figure 8 gives a schematic illustration of the importance of fault‐rupture mechanism on the distribution of EQTL.

In the case of reverse faults, the density of EQTLs is higher on the hanging wall (Sato et al., 2007; Xu, Xu, Shen, et al., 2014; Xu & Xu, 2012; Figure 8). In case of a parallel fault on the hanging wall of a reverse

FAN ET AL. 15 A TAL. ET FAN Table 3 Overview of Triggering and Predisposing Factors that Control EQTLs With an Indication of Their Relative Importance and the Feasibility of Spatial Representation Can be Commonly used approaches/data Groups/subgroups Controlling factors Importance mapped? Examples references to represent the controlling factors

Triggering factor Seismic effect Magnitude VH VH Keefer (1984) Seismic Arias Depth VH VH Meunier et al. (2007) acceleration, intensity Duration VH VH Nowicki Jessee velocity, and (Jibson, et al. (2018) intensity 1993) Aftershocks H VH Tiwari et al. (2017) parameters Focal mechanism VH VH Gorum and Carranza (Nowicki et al., (2015) 2014) Rupture length VH VH Marc, Hovius, and Meunier (2016) Asperity distribution H M Gorum et al. (2014)

Surface rupture VH VH C. Xu (2014) Geophysics of Reviews Distance from the M VH Kritikos et al. (2015) seismic source Topographic VH VL Meunier et al. (2007) ampliﬁcation Earthquake directivity VH M Roback et al., 2018) Arias intensity H H Jibson (2007) Predisposing Soil/rock mass Topography Elevation L‐M VH Gorum et al. (2011) SRTM, ASTER, ALOS DEM conditions properties, Internal relief M‐H VH Tanyas et al. (2017) slope geometry, and Slope steepness VH VH Zevenbergen and structural features Thorne (1987) Slope direction M‐H VH Lombardo et al. (2018) Planar and proﬁle M‐H VH Heerdegen and Beran curvature (1982) Relative slope position M‐H VH Böhner and Selige (2006) Topographic Wetness M VH K. J. Beven and Kirkby Index (1979) Geology Lithological types VH VH Nadim et al. (2006) Global Lithology Map Rock characteristics VH L‐M Duncan et al. (2014) (Hartmann & (geotechnical) Moosdorf, 2012) Relation discontinuities VH L Ghosh et al. (2010) with topography Discontinuities and faults VH M‐H Chigira and Yagi (2006) Soil Soil types H H Khazai and Sitar SoilGrids (Hengl (2004) et al., 2014) Soil thickness VH M C. Tang et al. (2011)

Soil characteristics VH L‐M Wyllie and Mah (2014) 10.1029/2018RG000626 (geotechnical/ hydrological) Hydrology Climatic zone VH VH Tanyas et al. (2017) Bioclimate Regions (Metzger et al., Antecedent rainfall VH H H. B. Wang et al. (2007) 2013), IMERG (Huffman et al., prior to earthquake 2010), and TMPA (Huffman et al., Rainfall directly after VH H Sassa et al. (2007) 2010; Huffman & Bolvin, 2013) the earthquake Distance to rivers H VH Tanyas et al. (2019) Land use Land cover/land use M‐H VH Yi et al. (2019) GlobCover (Arino et al., 2008) Normalized Difference M‐H VH Song et al. (2012) 16 Vegetation Index Reviews of Geophysics 10.1029/2018RG000626

seismogenic fault, the area between the two faults commonly has a high

ection concentration of EQTLs (Basharat et al., 2016; Kamp et al., 2008; Wang ﬂ et al., 2002; Xu, Xu, & Shyu, 2015; Xu, Xu, Yao, & Dai, 2014). Landslides triggered by strike‐slip faulting tend to have a symmetric distribution around the fault (Gorum et al., 2014; Xu & Xu, 2014) but in a narrower zone (Figure 8). The landslides triggered by a normal fault also are more likely to occur on the hanging wall (Xu, Ma, et al., 2018), but there are too few cases to draw a conclusion. The main reasons for the differences in EQTL distribution due to the difference in faulting style are the following: (1) The highest ground motions tend to occur closer to the surface Commonly used approaches/data to represent the controlling factors fault rupture (Abrahamson & Somerville, 1996), and a dipping fault surface (reverse or normal) has larger areas closer to the ground surface than a vertical fault surface (more typical of strike slip) thus resulting in a higher concentration of EQTLs on hanging walls (Tatard & Grasso, 2013); (2) hanging walls also can trap energy, resulting in stronger ground motions due to the constructive interference of seismic waves (B. Shi et al., Ayala et al. ‐ 1998); and (3) the hanging wall wedge is much smaller than the footwall, thus the same force applied to both sides of the fault will result in greater (2017) (2006)

Satellite Precipitation Analysis. Also, examples are given of research accelerations in the hanging wall (Oglesby et al., 2000). Additionally, ‐ although both normal and reverse faulting events experience this hanging wall effect, reverse faults tend to generate stronger shaking than normal faults (Oglesby et al., 2000). Can be mapped? Examples references In addition to the type of faulting, Gorum et al. (2014) argued that high landslide density zones could be related to asperities that can be observed along the fault plane, which are zones where a high amount of energy is released during rupturing (Ruiz et al., 2011) due to maximum friction (Hall‐Wallace, 1998). Huang and Fan (2013) considered characteristics of the rupturing event to understand the distribution of EQTLs for the 2008 Wenchuan earthquake, and they suggest a control of locking fault junctions, which are rarely failing intersections of fault segments (Shen et al., 2009). Huang and Fan (2013) associated the high landslide density zones with the fault junction zones where a large amount of energy was released by their rupturing. satellitE Retrievals for GPM; TMPA = TRMM Multi ‐ To capture the effect of seismicity considering all the above‐listed control-

Land cover changesDistance to roads H M H VH Alcántara Stanley and Kirschbaum ling factors, the most representative models can be derived via numerical simulations (e.g., Imperatori & Mai, 2015; Lee et al., 2009, 2008). triggered landslide occurrence spatially.

‐ However, such approaches have high computational costs. Given this fact, the most suitable ground‐motion data worldwide available for most earthquakes are from the U.S. Geological Survey ShakeMap database (Worden & Wald, 2016), which provides deterministic estimates of ground‐motion parameters including peak ground acceleration (PGA), peak ground velocity (PGV), Modified Mercalli Intensity, and spectral acceleration. Several studies have shown that landslide probabilities substantially change when Groups/subgroups Controlling factors Importance applying PGA, PGV, and Modified Mercalli Intensity as principal seismic factors in the landslide model because they directly reflect ground‐motion properties (Meunier et al., 2007; Nowicki et al., 2014; Nowicki Jessee et al., 2018). 2.1.3.2. Predisposing Conditions Although the seismicity factors are most important in controlling the distribution of EQTLs, the predisposing conditions and their interaction with seismicity also play an important role in the spatial patterns of landslid- . VL = very low; L = low; M = moderate; H = high; VH = very high; SRTM = Shuttle Radar Topography Mission; ASTER = Advanced Spaceborne Thermal Emission and Re ing. For example, topographic site amplification strongly affects ground Radiometer; DEM = digital elevation model IMERG = Integrated Multi that investigated the role of these factors for earthquake Note Table 3 (continued) motions and thus coseismic landslide occurrence (Meunier et al., 2007,

FAN ET AL. 17 Reviews of Geophysics 10.1029/2018RG000626

Figure 8. Schematic representation of the effect of fault type on the distribution of earthquake‐triggered landslides.

2008). The clustering of landslides near slope crests where topographic amplification is the greatest (Davis & West, 1973; Rizzitano et al., 2014) is an example that the predisposing factors for EQTLs are not necessarily similar to those of rainfall‐triggered ones. L. Shen et al. (2016) showed that slope aspect can affect EQTLs primarily as a result of seismic factors, such as directions of geological block movement, crustal stress, and seismic wave propagation. These were different from the influence of slope aspect on nonseismic landslides, which mainly relate to differences in vegetation cover, sunlight exposure, rainfall, and soil moisture (Kamp et al., 2008; Yalcin, 2008). The coupled effect of seismicity and rainfall increases the magnitude of landslide‐triggering events (Faris & Wang, 2014; Sassa et al., 2007). For instance, the Hokkaido region in Japan was hit by a M 6.6 earthquake on 5 September 2018 immediately after the Typhoon Jebi occurred on 27 August 2018 (Yamagishi & Yamazaki, 2018), which recorded over 100 mm of rain in the area where the earthquake triggered many landslides. There may be an additional effect to the seismic shaking coming from the antecedent rainfall, which may have increased pore pressures and modified the slope equilibrium favoring failure mechanisms during the ground motion. Other predisposing factors such as soil/rock mass properties, slope geometry, and structural features also affect the spatial distribution of landslides. Topographic factors can be easily derived using globally available DEMs such as the Shuttle Radar Topography Mission (NASA Jet Propulsion Laboratory; Abrams & Hook, 2013; Farr et al., 2007), the Advanced Spaceborne Thermal Emission and Reflection Radiometer Global DEM (Tachikawa et al., 2011), the ALOS Global Digital Surface Model (Tadono et al., 2016), or the WorldDEM by Airbus 2019. Using these data sources, DEMs having resolutions as fine as 12 m can be obtained. On the other hand, the required data regarding some other predisposing conditions are generally not globally available in high resolution (e.g., soil type or lithology), although global products are becoming available, such as SoilGrids (Hengl et al., 2014) or OneGeology (Laxton et al., 2010). The association between specific conditions (e.g., land use and soil types) and landslide susceptibility is not straightforward to characterize globally (e.g., Stanley & Kirschbaum, 2017). For example, a global‐scale geologic map is provided by Sayre et al. (2014) with 250‐m resolution. However, although the units might have the same geologic description, they can have a completely different role in landslide initiation because of differences in associated structural and geotechnical features. Unfortunately, neither structural features nor detailed lithological data are globally available at larger mapping scales (larger than 1:100,000). As a result, structurally controlled failures that can create large landslides (Chigira & Yagi, 2006) are not taken into account in regional

FAN ET AL. 18 Reviews of Geophysics 10.1029/2018RG000626

multivariate analysis. Similarly, hydrologic conditions generally are not considered because no high‐ resolution groundwater or precipitation data are globally available, with some rare, but insightful exceptions (e.g., Kogure & Okuda, 2018). As a result, all available statistical models that allow us to understand the factors controlling landsliding (Nowicki Jessee et al., 2018; Robinson et al., 2017) are hampered by some data availability issues.

2.2. Initiation and Failure Mechanisms: Experimental Studies 2.2.1. Landslide Hot Spots and Seismic Site Response Clustering of EQTLs near ridge crests and secondary clusters in colluvial slope toes and above inner gorges was observed by Meunier et al. (2008) for the 1994 Northridge, 1999 Chi‐Chi, and 1993 Finisterre earthquakes. Topography‐controlled landslide hot spots also were identified in the patterns of landslides triggered by the 2001 El Salvador, 2008 Wenchuan, 2010 Haiti, 2010–2011 Canterbury, 2013 Lushan, and 2017 Jiuzhaigou earthquakes (Evans & Bent, 2004; Fan, Scaringi, et al., 2018; Hough et al., 2010; Li et al., 2018; Massey et al., 2018; Xu, Xu, Shen, et al., 2014; Xu, Xu, Shyu, Gao, et al., 2015). The markedly uneven distribution of EQTLs along slopes differs from that of rainfall‐triggered landslides, which tend to be more evenly distributed along different parts of slopes (Meunier et al., 2008) because they are primarily controlled by the patterns of precipitation and by the coupling between slope geometry and the hydromechanical properties of the slope material. The response of a slope to the dynamic stress induced by seismic waves is the result of a complex interaction among the frequency and energy content of the seismic waves, which depend on the source mechanism of the earthquake, the path‐specific attenuation, and the local site conditions (material, layering, topography, etc.) that can amplify or de‐amplify the shaking at specific wavelengths (Massey et al., 2017; Meunier et al., 2008; Sepúlveda et al., 2005). Several studies investigated the relative contributions of the earthquake source, path, and local site effects on the site‐specific amplification phenomena (Ashford & Sitar, 2002; Del Gaudio & Wasowski, 2011; Gischig et al., 2015; Jibson, 2011; Moore et al., 2011; Rizzitano et al., 2014; Strenk & Wartman, 2011). However, no single factor was found to be responsible for clustering of landslides; instead, different combinations of above‐mentioned factors work for different cases (Massey et al., 2017). Site‐specific resonance (i.e., amplification) can occur through constructive interference of seismic waves. Reflected, refracted, and/or diffracted waves, at the free surface or at interfaces between geological structures with strong impedance contrast, can all contribute to resonance phenomena. The shape and orientation of these surfaces can cause amplification at multiple frequencies, depending on topography and the thickness and lateral extent of the geological structures (Del Gaudio et al., 2014). Strong impedance contrasts producing local amplification can be caused by a soil or debris layer overlying a rock formation, by highly fractured zones within more intact materials, or by weathered materials overlying less weathered materials (Bourdeau & Havenith, 2008; Bozzano et al., 2008; Del Gaudio & Wasowski, 2011; Gischig et al., 2015; Moore et al., 2011). Site‐dependent anisotropy also can occur, which consists of a pronounced resonance anisotropy that causes important shaking amplification along a preferential direction (Del Gaudio & Wasowski, 2007). Surface morphology (inclination, convexity, and height) can cause amplification at both larger ridge scales and smaller site scales (Hough et al., 2010; Kaiser et al., 2014; Meunier et al., 2008). Several studies demonstrated amplification phenomena in slopes prone to landsliding (Del Gaudio & Wasowski, 2007, 2011; Hartzell et al., 2017; Moore et al., 2011). These works show that amplification observed at different stations reveals peaks coinciding with potential sliding directions, and this implies greater slope susceptibility to seismic failure with respect to what can be estimated by conventional (pseudo‐static) slope stability analyses (Del Gaudio et al., 2014). Although accurate information on site‐amplification effects can be obtained by recording strong‐motion events for a long period of time (Hartzell et al., 2017), the need for a very dense accelerometer monitoring network (i.e., multiple stations on each slope) in earthquake‐prone areas makes the retrieval of such information unfeasible, and alternative methods have come into use. After observing that ground motions from 3‐D earthquake simulations show that topographic curvature correlates with topographic amplification, Maufroy et al. (2015) proposed a methodology to estimate the amplification by exploiting such correlation. Rai et al. (2016) also suggested an empirical approach calibrated on a set of recordings from small‐ to medium‐magnitude earthquakes in California. Their model relies on a parameter termed relative elevation,

FAN ET AL. 19 Reviews of Geophysics 10.1029/2018RG000626

Figure 9. Summary of experimental approaches (discussed in this section) to investigate the failure mechanism, dynamics, and kinematics of earthquake‐triggered landslides.

which quantifies topography using the elevation of a site relative to its surroundings and improves the ground‐motion attenuation model previously proposed by Chiou et al. (2010). Analysis of ambient noise as a substitute for strong‐motion accelerograms (Figure 9) is helpful in evaluating topographic site amplification, for instance, through the horizontal‐to‐vertical noise spectral‐ratio technique, first proposed by Nogoshi and Igarashi (1971) and Nakamura (1989). Its working principle is that a strong contrast of impedance between a surficial soft layer and a more rigid substratum causes an amplification of the horizontal components of seismic background noise at the same frequencies at which the shear‐ wave amplification reaches its maximum (Del Gaudio et al., 2014). Consequently, a pronounced peak in the H/V (Horizontal/Vertical) spectral ratios at site‐specific frequencies can reveal site‐specific (and directional) resonance conditions, and information on S wave velocities of the surficial material also can be obtained (Burjánek et al., 2010; Danneels et al., 2008; Del Gaudio et al., 2008; Gallipoli & Mucciarelli, 2007; Hancox et al., 2002; Havenith et al., 2002; Jongmans et al., 2009; Louie, 2001; Méric et al., 2007; Mreyen et al., 2017; Nunziata et al., 2009; Ohori et al., 2002; Torgoev et al., 2013). Site‐dependent directivity was first observed by Bonamassa and Vidale (1991) and Spudich et al. (1996) and has been related in some cases to the presence of preexisting landslides that caused strong material contrasts at defined interfaces (Rial, 1996; Xu et al., 1996). Because directional amplification cannot be predicted from the surface topography alone, ambient‐noise analysis constitutes an important tool to obtain insight on the subsurface structure (Del Gaudio et al., 2014), estimate the amplifications that a slope can undergo during a seismic event, and evaluate the likelihood of slope failure in different conditions. 2.2.2. Laboratory Tests Using Shaking Tables Shaking tables are experimental devices that can be used to investigate the response of model slopes, artificial structures, or building components to seismic shaking or other types of vibration (Figure 9). They are particularly useful in studying the initiation mechanisms of EQTLs, for back‐analyzing the behavior of landslides that lacked adequate monitoring (R. Huang et al., 2013) and for measuring mechanical parameters of weak layers conducive to failure (e.g., Podolskiy et al., 2015). In fact, the limited information on the strength properties of soils and rocks under dynamic forcing hampers the applicability of advanced physically based numerical models that account for seismic shaking explicitly. To overcome this limitation, laboratory equipment capable of testing soil and rock samples or assemblies of various sizes under dynamic conditions has become increasingly used (Lin & Wang, 2006; Wang & Sassa, 2009; Wartman et al., 2005; Wasowski et al., 2011). Among others, Lin and Wang (2006), Liu et al. (2013), and G. Fan, Zhang, et al. (2016) performed shaking‐ table tests on rock slopes having different stiffnesses, strengths, and bedding orientations. They reported a positive correlation between the amplification of the seismic acceleration and the height of the model slopes, which can derive from the reduction of stiffness along height (Ambraseys, 2009; Dakoulas & Gazetas, 1986).

FAN ET AL. 20 Reviews of Geophysics 10.1029/2018RG000626

These studies also demonstrated the role of weak layers (dipping in the direction of the slope) and fractures in enhancing seismic displacements and increasing the likelihood of failure.

Through shaking‐table tests on sand slopes, Katz and Aharonov (2006) demonstrated the role of material heterogeneity in deﬁning the size‐distribution pattern of landslides. They observed that small and shallow landslides occur as slumps within rather homogeneous and unconsolidated soil deposits. The size of these landslides is essentially controlled by the depth of the deposit over the underlying rock, which generally is heterogeneous due to the presence of layers, bedding planes, joints, and fractures. This heterogeneity is partly responsible for the power law decay in natural landslide size distributions.

R. Huang et al. (2013) used a shaking table to study the failure mechanism of a specific landslide triggered by the 2008 Wenchuan earthquake. They reproduced the pattern of deformation prior to the general failure and observed that the location of the failure plane in the model was consistent with that of the maximum horizontal acceleration. Based on this, the authors developed a simple empirical model to predict the depth of landslides in rock slopes from the characteristics of the seismic shaking and those of the rock material, with the occurrence of deeper landslides associated with the presence of softer/weaker rocks (i.e., rocks with smaller elastic modulus and less cohesion). Wartman et al. (2005) investigated how permanent displacements in slopes are generated to verify the reliability of slope‐displacement predictions obtained through the sliding‐block approach of Newmark (1965). They demonstrated that such an approach provides good first‐order estimates of deformation, as later con- firmed also by Wang and Lin (2011). However, these estimates predict too much displacement for slopes that experienced small displacements, but they more closely match those cases in which the sliding resistance decays toward the residual strength (attained after large displacements). Y. ‐M. Hsieh, Lee, et al. (2011) demonstrated the need for shaking‐table tests to verify the reliability of sliding thresholds, which can be much higher than the actual ones if obtained through static tilt tests. Although slope conditions are well controlled and different scenarios can be tested accurately, upscaling the results obtained through shaking‐table tests on meter‐size slopes to the size of actual slopes can be problematic because material behavior scales nonlinearly. To overcome this problem, a geotechnical centrifuge can be used (Brennan & Madabhushi, 2009; Liang & Knappett, 2017; Taboada‐Urtuzuastegui et al., 2002; Zhang, 2018; Zhang et al., 2017). By spinning hundreds of times per minute, the centrifuge can impose large accelerations on the test slope, producing stress‐strain patterns that more closely resemble those found in natural slopes. Scaling of metric quantities is then performed in relation to the ratio between the imposed centrifugal acceleration in the test slope and the gravitational acceleration in the natural slope. A recent study conducted in a centrifuge has been reported by Zhang et al. (2017), who investigated the case of a slope made of a loess layer overlying a mudstone layer. They found amplifications of the horizontal and vertical seismic acceleration at the slope crest as much as three times those at the base and demonstrated the role of the contrast in the properties of the two materials in contact in producing tensile cracks at the crest and the shear zone at the contact. 2.2.3. Numerical Approaches to Assess Stability and Permanent Displacements Jibson (2011) summarized the existing methods to assess the seismic stability of slopes and to estimate the amount of permanent inertial displacements. The methods are classified into three categories, each having specific strengths, weaknesses, and applicability to different situations: (1) pseudostatic analysis, (2) stress‐ deformation analysis, and (3) permanent‐displacement analysis (Figure 9).

Pseudostatic analyses were developed early in the history of soil mechanics (Terzaghi, 1950). In such analyses, a permanent body force representing the seismic shaking is introduced in conventional static limit‐ equilibrium calculations. Generally, only the horizontal component of the seismic shaking is modeled because the cumulative effect of the vertical forces is assumed to be negligible. However, simulations and experiments (Huang et al., 2001; Yuan et al., 2015; Zhang et al., 2015) show that vertical shaking can affect the results significantly. In fact, in the presence of vertical acceleration, weak materials may fail under a smaller amount of shear stress due to reduction of normal stress or tensile failure (Podolskiy et al., 2010b). On the other hand, the use of a permanent and unidirectional force to represent the seismic shaking in pseudostatic analyses significantly underestimates the seismic stability of a slope. Consequently, scaling coefficients must be chosen to account for only a fraction of the PGA, which in fact is effective only for a

FAN ET AL. 21 Reviews of Geophysics 10.1029/2018RG000626

very short time (Blake et al., 2002; Jibson, 2011; Jibson & Michael, 2009; Kramer, 1996; Marcuson, 1981; Seed, 1979; Terzaghi, 1950; Wieczorek et al., 1985). The use of stress‐deformation analyses began to spread in the 1960s following the development of the finite‐ element method. Initially devised for general engineering applications, the method was soon used to model the static and dynamic deformations of slopes. Other modeling strategies followed, including the finite‐ difference method (Mitchell & Griffiths, 1980), discontinuous deformation analysis (Shi & Goodman, 1985), discrete‐element analysis (Cundall, 1971), the rigid‐block spring method (Kawai, 1978), and the boundary‐element method (Brebbia & Wrobel, 1980). With time, increasingly complex and computationally intensive models (e.g., nonlinear and inelastic) were developed (Boulanger & Ziotopoulou, 2012; Elgamal et al., 1990; Gerolymos & Gazetas, 2005; Griffiths & Prevost, 1988; Prevost, 1981; Ziotopoulou & Boulanger, 2016). However, the use of complex models is practical only if spatially dense and high‐quality input data are available. Thus, advanced stress‐deformation analyses generally are used only in strategic construction projects or to model individual slopes that affect essential infrastructures or lifelines. The application of such analyses to larger and generic regions remains unfeasible, as the availability of the necessary input data is normally insufficient (Jibson, 2011). Permanent displacement analysis was introduced by Newmark (1965), who conceptualized coseismic landslides as individual rigid blocks that move on an inclined plane when a critical (yield) acceleration is exceeded. This critical acceleration depends on the inclination of the sliding surface from the horizontal and on the factor of safety, the ratio of resisting to driving forces or moments in the slope. The input for the method is an actual or synthetic accelerogram. Velocity and displacement of the block as a function of time can be simply obtained by successively integrating the portions of the accelerogram in which the critical acceleration is exceeded. The modeled landslide is assumed to behave as a rigid‐plastic body with no internal deformation, and several other model assumptions generally are made for simplicity (Jibson, 2011). Newmark's method has inspired more sophisticated methods that are able to model the dynamic response of landslide material more realistically and yield more accurate displacement estimates. These models can be categorized into decoupled and coupled analyses. Decoupled analyses account for the internal deformability of the landslide mass under seismic shaking (Lin & Whitman, 1986; Seed & Martin, 1966). The most commonly used procedure was proposed by Makdisi and Seed (1978) and consists of two steps. First, a dynamic‐response analysis of the slope is performed assuming no failure surface, and an average acceleration for the sliding mass is obtained by averaging accelerations calculated in several points in the slope. Second, the resulting acceleration is taken as the input for a Newmark analysis to evaluate the permanent displacement. This analysis does not account for the effects of sliding displacement on the ground motion and does not consider amplifications in different parts of the slope. Coupled models overcome some of the limitations of decoupled analyses by jointly modeling the dynamic response of the sliding mass and the permanent displacement. They are the most sophisticated form of sliding‐block analysis but also the most computationally intensive. Bray and Travasarou (2007) developed a simplified approach that uses a nonlinear, fully coupled sliding‐block model to produce a semiempirical relationship for predicting displacement. Jibson et al. (2013) published a software package for the implemen- tation of rigid‐block, decoupled, and coupled analyses. Displacements obtained through modeling are not the actual landslide displacements. Rather, they represent an index of the performance of slopes under seismic excitation, which must be interpreted carefully in relation to the probable effect on a potential landslide (Jibson, 2007; Jibson et al., 1998, 2000; Pradel et al., 2005; Rathje & Bray, 2000). In general, threshold displacement can be set depending on the characteristics of the landslide material. For instance, evaluated displacements in the range of 5–10 cm can corre- spond to ground cracking and to previously undeformed soils approaching the residual strength condition. A static stability analysis can thus provide useful information on the post‐seismic stability of slopes (Jibson & Keefer, 1993).

2.3. Main Research Challenges Identifying and predicting the spatial distribution patterns and occurrence probability of EQTLs remains challenging. Regarding the development of empirically or physically based models to predict size statistics and spatial distributions of EQTLs, the following issues should be addressed in future research:

FAN ET AL. 22 Reviews of Geophysics 10.1029/2018RG000626

1. Calibration of both physically and statistically based models relies upon the availability of EQTL inventories. Although more EQTL inventories are available than before, the representativeness and completeness of the inventories are still a concern. For example, Tanyas et al. (2017) showed that 50% of all accessible EQTL inventories are from Asia and only 5% are from South America. Moreover, available inventories differ in quality and completeness (Tanyas et al., 2017), which can directly affect the reliability of hazard assessments (e.g., Pellicani & Spilotro, 2015). 2. For both physically and statistically based models, the accessibility of necessary layers of input parameters, particularly at a global level, and estimates of their uncertainties is a continuing challenge. More efforts are needed to better understand the controlling factors and their interactions in order to be able to predict the distribution of coseismic landslides in future. For that, it is also very important to consider whether the controlling factors can be spatially represented so that they can be used in a predictive model. 3. Ground‐motion characterization is an important component of rapid‐assessment studies, and ShakeMap is the most commonly used tool to derive the required parameters. However, ShakeMap currently does not directly account for topographic ampliﬁcation (Sepúlveda et al., 2005), duration of shaking (Jibson, 2011; Jibson et al., 2004), or rupture direction (e.g., Gallen et al., 2017), all of which can be factors in the landslide initiation process (Jibson, 2011; Jibson et al., 2004). 4. As discussed in section 2.2, experimental studies are helpful for understanding the initiation and failure mechanism of EQTLs, but how to solve the scale effect and to apply the results for predicting EQTLs for a large area still need more efforts. Particularly, the shear strength of soils and rocks strongly inﬂuences slope stability, but none of the current available methods can estimate shear strengths and their relative contribution to landslide susceptibility at a regional scale (Dreyfus et al., 2013).

3. Post‐Seismic Geological Hazards 3.1. Formation and Failure of EQTL Dams 3.1.1. An Overview of Earthquake‐Triggered Landslide Dams One of the immediate post‐seismic hazards are landslides that dam rivers, forming lakes that inundate areas upstream and dams that might fail and catastrophically release stored volumes of water and sediment (Figure 1). These landslide dams (Costa & Schuster, 1988) are common during or shortly after strong earthquakes (Evans et al., 2011; Keefer, 1984; Korup, 2004b; Schuster et al., 1998). Their lifespans range from a few hours to thousands of years. Landslide dams are a well‐known hazard given their potential of releasing outburst floods; they commonly require emergency engineering mitigation and evacuation (Delaney & Evans, 2015; Fan, van Westen, et al., 2012; O'Connor & Costa, 2004; Peng & Zhang, 2012b, 2013b; Shi et al., 2017). Landslide dams can attenuate or accentuate water and sediment fluxes, channel geometry and stability, and valley evolution (Hewitt, 2006; Korup et al., 2010). For example, Fan, van Westen, et al. (2012) estimated that landslide dams induced by the 2008 Wenchuan earthquake temporarily stored about 0.6 km3 of water and sediment. Both gradual and catastrophic breaches of landslide dams release sediment downstream, deposit- ing material on fans, floodplains, and deltas (Hewitt et al., 2008). Channel aggradation of several to several tens of meters due to such increases in sediment yield has been reported after major earthquakes (e.g., Yanites et al., 2010). Landslide‐dammed lakes that persist for hundreds of years can attenuate flood waves, but even breached dams can control river morphology (Burchsted & Daniels, 2014; Hermanns et al., 2011; Hewitt et al., 2008), especially valley‐floor sedimentation by locking local base levels (Korup, 2005; Korup & Tweed, 2007) or by creating boulder‐armored knickpoints that can remain in a river reach for thousands of years (e.g., Korup, 2006a, 2006b). Clusters of long‐lived dams might suggest a seismic origin (Strom, 2015), and the lake sediment can yield clues about past earthquakes (Avşar et al., 2014, 2016; Howarth et al., 2012, 2014, 2016; Moernaut et al., 2014; Waldmann et al., 2011). Further discussion of these sediment‐related effects is provided in section 5. Despite several reviews (Costa & Schuster, 1988; Evans et al., 2011; Hewitt et al., 2008; Korup, 2002), few systematic analyses focused on earthquake‐induced landslide dams; nonetheless, case studies abound (Dai et al., 2005; Dong et al., 2011; Fan, Xu, van Westen, et al., 2017; Schneider, 2009). We suspect that more scientific and practical insights can be gained from studying earthquake‐induced landslide dams to better inform risk assessment and emergency response.

FAN ET AL. 23 Reviews of Geophysics 10.1029/2018RG000626

3.1.2. Controls on the Formation and Stability of Landslide Dams Schuster et al. (1998) listed four controls of the pattern of earthquake‐generated landslide dams: (1) seismic intensity (i.e., high PGA and duration of strong shaking); (2) steep hillslopes and uneven topography; (3) weak and weathered lithologies; and (4) high soil moisture or pore‐water pressures. Apart from these controls that apply to landslides in general, the volume and mobility of landslides, the topography of the valley ﬂoor, and hydrologic factors (e.g., the distance to river network) determine whether landslides form dams (Fan et al., 2014; Tacconi Stefanelli et al., 2016). Predicting where, when, and how a landslide forms a dam remains problematic, though, and more studies were concerned with monitoring or quantifying the stability of landslide dams. The regional patterns and controls of coseismic landslide dams have received less research attention compared to EQTLs. Most inventories of landslide dams deal with historical cases, though generally with unknown triggers and few data on the location, size, type of landslides, or damage caused. For example, the trigger of more than half of 232 landslide dams examined in New Zealand remains unknown (Korup, 2004a). After the Wenchuan earthquake, Fan, van Westen, et al. (2012) compiled the largest event‐based landslide‐dam inventory (n = 828) to date and argued that the propensity for river damming depended on landslide type and initiation volume as well as on relief and mean hillslope steepness (Fan et al., 2014). Further candidate controls include the distance from rivers, their stream power index, and drainage density (Tang et al., 2018). Physically based models such as smoothed‐particle hydrodynamics and discrete element methods can offer insights into how landslide dams form (Braun et al., 2018; Zhao et al., 2017). Much effort concerns understanding and predicting the stability and longevity of landslide dams. Early work explored several geomorphic predictors that separated stable from unstable dams (Tacconi Stefanelli et al., 2016), although the distinction between intact and failed dams might be more appropriate. Some candidate

predictors are the landslide dam volume (Vd), height, width, length, lake volume, and upstream catchment area (Ac; Casagli & Ermini, 1999; Dong et al., 2009, 2011; Ermini & Casagli, 2003). A blockage index,

BI ¼ log V d=Ac: (3)

BI revealed some initial trends of landslide‐dam stability in the Northern Apennines (Casagli & Ermini, 1999); no landslide dams were reported for BI < 3, whereas known landslide dams had remained intact for hundreds of years or longer for BI > 5. Such empirical thresholds are prone to vary, depending on study area and data quality. X. Fan, van Westen, et al. (2012) found that, in the wake of the Wenchuan earthquake, no landslide dams formed where BI < 2, whereas most of dams were still intact about a year after the earthquake where BI >5. A landslide dam's internal structure, material properties, grain‐size distribution, volume and rate of water and sediment inflow, and seepage processes all affect the stability and longevity of landslide dams (Korup & Tweed, 2007). Unfortunately, the internal structure is exposed mostly after dam failure, though geophysical measurements and geotechnical tests have begun to elucidate some internal characteristics of selected dams (Casagli et al., 2003; Dunning & Armitage, 2011; Wang et al., 2013; G. Wang, Furuya, et al., 2016; Weidinger, 2011; Weidinger et al., 2002). Reviews of the sedimentology of landslides, particularly rock slides and rock avalanches, have contributed important details regarding grain‐size distribution, the strength of undercut deposits, or possible pathways for percolating water (e.g., Dufresne et al., 2016). Landslide dams can fail through overtopping, piping, and slope failure (Costa & Schuster, 1988). Overtopping is the most frequent mode (Schuster, 1993; L. Zhang, Peng, et al., 2016), with water spilling over the dam crest and eroding a channel through the dam (Manville, 2001). Piping means internal erosion by seepage, which removes solid particles and produces tubular underground conduits that appear initially as springs or seepage on the downstream dam face (Singh, 1996). Slope failure is initiated when the hydraulic pressure of the impounded water overcomes the shear resistance of the dam material (Walder & O'Connor, 1997) and commonly occurs with piping and overtopping when vertical erosion oversteepens the breach sidewalls. The mechanisms of dam breach are less well understood and commonly elude direct observation. Many dams were breached artificially to avoid uncontrolled outburst flooding (Canuti et al., 1999; Fan, van Westen, et al., 2012; N. Liu, Chen, et al., 2010; Xu et al., 2009); thus, natural mechanisms of dam breach could not be observed. Field‐scale dam‐breach experiments have been conducted to understand the failure mechanism of landslide dams (Yan et al., 2017; Zhou et al., 2013). Broadband seismometers can capture characteristic

FAN ET AL. 24 Reviews of Geophysics 10.1029/2018RG000626

Figure 10. Climate‐normalized landslide rates (solid circles, left axis) and debris‐flow‐triggering rainfall (open circles, right axis) before and after some recent earthquakes. Landslide rates are the ratio between the average pre‐earthquake landslide volumes and the landslide volumes for the given interval of time. The area‐volume scaling proposed by Larsen et al. (2010) for mixed soil and bedrock landslide data sets is used here to estimate volumes from total landslide areas (although other choices are possible). Vertical bars indicate the error (±1 standard deviation) associated with the scaling. The climatic normalization was performed by dividing the volumes by the cumulative rainfall recorded between the date of the previous remote sensing image and the given time. The size of the open circles (debris flow‐triggering rainfall) is proportional to the number of debris flows triggered by that rainfall on that date. Data from Domènech et al. (2018); Fan, Domènech, et al. (2018); Fan et al. (2019); Khattak et al. (2010); Marc et al. (2015); and Saba et al. (2010).

seismic signals generated by floods, debris flow, and bedload flux and thus help to establish the timing and reach of dam failure and subsequent sediment transport (e.g., Chao et al., 2015; Feng, 2012; Yan et al., 2017).

3.2. Landslide Remobilizations, Debris Flows, and Post‐Seismic Landslides Enhanced rates of landsliding are common in earthquake‐hit regions. Rainfall can either trigger new landslides (Marc et al., 2015) from rock material that is weakened and damaged by shaking or remobilize EQTL material (Domènech et al., 2019; Fan et al., 2019; Fan, Domènech, et al., 2018) that can feed debris flows for many years after the earthquake (Figure 1; Lin et al., 2004; Tang et al., 2009). Still, our knowledge of the timing and extent of post‐seismic landslides, debris flows, and sediment transport hinges on observations rather than predictions. This uncertainty makes reconstruction planning difficult, especially for the potentially affected valley floors. 3.2.1. Spatial and Temporal Patterns of Geohazards After Earthquakes Increased rates of geomorphic activity after large earthquakes were first described for the 1897 Assam earthquake. Oldham (1899) collated some of the earliest accounts of post‐seismic landslides and debris flows. Except for a pioneering study of the 1929 Murchison, New Zealand earthquake (Pearce & Watson, 1986), post‐seismic geohazards remained largely unstudied until the 1999 Chi‐Chi, Taiwan earthquake (e.g., Hovius et al., 2000). Compared to prior rates of landsliding and sediment fluxes, the 1999 Chi‐Chi earthquake caused a rapid increase in suspended sediment fluxes in rivers (Dadson et al., 2004), followed by increased landsliding rates during 2001 Typhoon Toraji (Hovius et al., 2011). Subsequent analyses suggested that enhanced rates of landsliding and sediment flux, when normalized for climate records, can be main- tained for as long as a decade (Marc et al., 2015). Enhanced rates of landsliding immediately after earthquakes might not be driven solely by rainfall or aftershocks but also by a lowered strength of hillslope materials due to intensive mainshock shaking. Estimates of post‐seismic landsliding rates come from detailed compilations of inventories collected from satellite imagery (e.g., Fan et al., 2019; Marc et al., 2015), and these studies show that pulses of enhanced landslide frequency and sediment flushing occurred within 1–10 years after the earthquake. Many studies have reported that both rates of landsliding and sediment transport decay toward background rates within a decade (Dadson et al., 2004; Hovius et al., 2011; Khattak et al., 2010; C. ‐W. Lin, Liu, et al., 2006; Marc et al., 2015; Shou et al., 2011; Yang et al., 2017; S. Zhang, Zhang, et al., 2016; Zhang & Zhang, 2017; Figure 10).

FAN ET AL. 25 Reviews of Geophysics 10.1029/2018RG000626

Detailed mapping of 34 time slices between 1994 and 2006, and thus before and after the 1999 Chi‐Chi earthquake, argued for the importance of rock damage relative to climatic controls and seismic forcing (Dadson et al., 2004; Marc et al., 2015; Figure 10). Landslide rates V correlated with the total precipitation P exponentially:

= V ¼ AeP B; (4)

where A and B are empirical constants that reflect the pre‐earthquake susceptibility to slope failure of the studied area and the geomorphic efficacy of rainfall (Marc et al., 2015). In regions affected by the 2008 Wenchuan earthquake, rates of enhanced post‐seismic landsliding were longer than previously observed, though less than a decade (Fan, Domènech, et al., 2018; Tang et al., 2016; Yang et al., 2017; S. Zhang, Zhang, et al., 2016). Yet all inventories document intense and short periods of post‐seismic landslides and debris flows, which has implications for the overall sediment budget of the earthquake (see section 5). For example, the post‐seismic sediment pulse associated with the 2008 Wenchuan earthquake mobilized 0.6–1.5 km3 of material. The volume of active landslides decreased by >70% within 3 years and practically normalized within 7 years after the mainshock, even though most of the sediment remained in place. This suggests that the legacy of sediment generated by earthquakes could last, in different forms, for much longer than the decade after the earthquake (Fan, Domènech, et al., 2018). The reported magnitude and frequency of post‐seismic failures differ between study areas. Post‐seismic landslide size distributions for the 1993 Finisterre, 1999 Chi‐Chi, and 2008 Iwate earthquakes were indistin- guishable from coseismic landslide inventories (Marc et al., 2015), and the authors inferred a uniform ground‐strength decrease at shallow depths during the earthquake. This is followed by a ground‐strength recovery independent of the failure size or depth. Contrasting observations were reported during the post‐ seismic remobilizations of coseismic landslides triggered by the 2008 Wenchuan earthquake (Fan, Domènech, et al., 2018). Although smaller coseismic landslides stabilized quickly, the larger and deeper ones remained prone to reactivation several years after the earthquake. 3.2.2. Proposed Mechanisms for Enhanced and Decaying Landslide Activity in Coseismic Deposits The shape of the frequency distribution of post‐seismic landslides and debris flows, peaking in the aftermath of an earthquake and decaying back to pre‐earthquakes values within years, is consistent across different earthquakes (Figure 10). Some authors argue that we can see the mechanisms that control the pattern of landslides before and after an earthquake from changes in their size distribution. With respect to landslide remobilization, debris flows seem to play a major role. Active landslides tend to shed debris flows along narrow gullies, partly via Hortonian overland flow driving sediment entrainment (e.g., Domènech et al., 2019; van Asch et al., 2014) or collapse of water‐saturated landslide material (Iverson, 1997). Experiments on artificial slopes using debris from coseismic landslides provide insight into how changes to grain‐size distributions of landslide debris affect the initiation and runout of flow‐like landslides (Hu et al., 2017; Hu, Hicher, et al., 2018; Hu, Scaringi, et al., 2018; Wang & Sassa, 2001, 2003). These studies pointed out that hydrologi- cally driven grain coarsening, for example, through surficial and internal erosion and depletion of fine particles, can improve the stability condition of the deposits by increasing their hydraulic conductivity and intergranular contacts, reducing the probability of failure in larger rainstorms (Hu et al., 2017; Hu, Hicher, et al., 2018). Monitoring studies of post‐seismic debris flows reported a trend toward coarser flows that traveled shorter distances and initiated in larger EQTL deposits (e.g., Chen et al., 2014; Zhang et al., 2013). These coarser debris flows indicate either reworking or winnowing of fines from landslide debris or more storage of coarser material in channel beds. Progressive depletion of erodible debris, increasing soil hydraulic conductivity, increasing friction in sediment due to compaction, and increasing root reinforcement in regrowing vegetation cover might explain the decay of debris‐flow activity in the years after the earthquake (e.g., Domènech et al., 2019; Hu, Scaringi, et al., 2018; Z. Li, Jiao, et al., 2014; Saito et al., 2014; Shen et al., 2017; Yang et al., 2018). Exhaustion of the total volume of available sediment (Cruden & Hu, 1993) is unlikely because most of the EQTL sediment (as much as 80–90%) can remain on the hillslope for many years after the initial event. Landslides that remain active after the earthquake tend to occur closer to channels and in parts of the landscape with higher contributing drainage areas (Fan et al., 2019) than dormant landslide deposits, although remobilization might not depend solely on fluvial undercutting (e.g., Croissant, Lague, Davy, et al., 2017;

FAN ET AL. 26 Reviews of Geophysics 10.1029/2018RG000626

Croissant, Lague, Steer, & Davy, 2017) where stream power is low and coarse sediment protects the base of landslide deposits. Several hypotheses concern the reduction of post‐seismic landslide activity. First, changes in the connectivity of landslide debris within the fluvial or hillslope hydrological system could change debris‐flow triggering thresholds (Z. Li, Jiao, et al., 2014, see section 5). Incision by debris flows into large landslide masses can dis- turb groundwater flow, potentially draining the shallow subsurface and reducing the potential for high pore pressures required to trigger debris flows. Second, extensive mitigation works can dampen the triggering of debris flows (e.g., Liu et al., 2017), and check dams and other retaining systems limit sediment export from low‐order catchments. Third, vegetation can quickly reclaim slopes denuded by earthquakes, particularly in humid tropical and temperate regions, thus reducing overland flow and stabilizing hillslopes by root anchor- ing (e.g., O'Loughlin, 1985; Schmidt et al., 2003; Watson et al., 1999) and eventually shutting down potential sediment sources for debris flows (Y. Liu, Liu, & Ge, 2010). Fourth, grain coarsening resulting from progressive winnowing of fines or recurring landsliding (Hu, Hicher, et al., 2018; Hu, Scaringi, et al., 2018; S. Zhang, Zhang, & Chen, 2014) increases hydraulic conductivity, making surface runoff and excess pore pressure less likely. A measured spike in suspended sediment concentrations (grain size < 0.25–0.5 mm) in rivers soon after the Wenchuan earthquake, followed by a decay in subsequent years (Li et al., 2017; J. Wang, Jin, et al., 2015; Wang et al., 2017), might be some evidence of this coarsening process or simply reflect artifacts of debris flow (S. Zhang et al., 2013). 3.2.3. Controls on Enhanced and Decaying Post‐Seismic Landsliding Post‐seismic landslides can form at higher rates during heavy rainfall, possibly driven by clearing of oversteepened slopes in the crowns of EQTLs, vegetation and groundwater changes, and altered mechanical properties of bedrock and regolith (Marc et al., 2015). Oversteepened slopes in the crowns of coseismic landslides contributed less than half of the total post‐seismic landslide volume in documented cases (Schöpa et al., 2018). Vegetation and groundwater level changes have time scales that differ from those of the post‐ seismic landslide transient (<10 years). Earthquake‐induced strength reduction might occur via different mechanisms in colluvium, regolith, and bedrock. In the colluvium and regolith, mechanical damage by decompaction and the extension and dilation of shallow cracks could affect the landslide triggering rate. Bedrock cracks, joints, and bedding planes can form or widen during shaking. The occurrence of landslides having a range of depths suggests that these different mechanisms produce a similar geomorphic response. These changes in subsurface material properties seem to be transient and can only be exploited during extreme precipitation events, as landslide rates remain low in intervals without extreme precipitation (Marc et al., 2015). It is difficult to validate these proposed changes in the shear strength of hillslope material because direct geotechnical testing relies on small ele- mental volumes, and obtaining a statistically significant amount of data is too expensive and time‐ consuming. A possible workaround is to use the characteristic velocities of seismic waves, which depend on the elastic modulus of the medium in which they propagate. Laboratory experiments found a reduction of elastic moduli caused by moderate strains in rock samples and a subsequent logarithmic recovery or healing (Johnson et al., 1996; Rivière et al., 2015). Similar results were obtained during field experiments of ground shaking (Lawrence et al., 2009; Renaud et al., 2014). Ambient noise interferometry and techniques that analyze seismic waves via deconvolution revealed changes of subsurface seismic velocities at depths of 100 m to several kilometers after medium to large earthquakes (e.g., Brenguier et al., 2008; Hobiger et al., 2014). In those cases, the velocity drop due to the earthquake is consis- tently followed by a progressive, nonlinear recovery over months to years. Widespread substrate damage has been attributed to strong ground motion and faulting, and the post‐seismic recovery has been attributed to healing of cracks and grain‐to‐grain contacts (Sawazaki & Snieder, 2013). These mechanisms also could affect the strength of hillslope material. Therefore, studying the coevolution of epicentral landslide rates and shallow seismic velocities might go some way toward explaining observed transients in post‐seismic mass wasting, at least that detaching bedrock, because soils and debris mantles remain unaffected by changes in bulk rock‐mass strength. 3.2.4. Triggering Rainfalls for Debris Flows Rainfall is a common trigger of landslides (e.g., Crosta & Frattini, 2001; Froude & Petley, 2018; Segoni et al., 2018), and the processes relating rainfall to the occurrence and size of slope failure, any remobilization, acceleration, and fluidization have been widely studied (e.g., Crozier & Glade, 1999; Jamei et al., 2017;

FAN ET AL. 27 Reviews of Geophysics 10.1029/2018RG000626

Figure 11. Changes in rainfall thresholds for debris‐ﬂow occurrence after the 1999 Chi‐Chi and 2008 Wenchuan earthquakes. The peak hourly rainfall intensity during the rainfall event that triggered debris ﬂows is plotted. Data for the Chi‐Chi earthquake are from Lin et al. (2004) and refer to the Chenyulan River watershed; data for the Wenchuan earthquake are from S. Zhang and Zhang (2017) and refer to the epicentral area from Yingxiu to Gengda.

Montgomery & Dietrich, 1994; Olivares & Damiano, 2007; Wieczorek et al., 2000). Yet we can still learn much about the variability and scaling of meteorological and hydrological factors and processes controlling slope failure (Almeida et al., 2017; Beven & Germann, 2013; Corominas et al., 2013; Crosta & Frattini, 2001; von Ruette et al., 2014). Technical advances now permit very detailed measurements of hydromechanical changes within a landslide body (Kogure & Okuda, 2018). Shallow landslides and debris flows are more directly related to patterns of rainfall intensity, duration, and antecedent rainfall (e.g., Aleotti, 2004; Guzzetti et al., 2008). Rainfall thresholds represent a condition that, when exceeded, are likely to trigger landslides (Guzzetti et al., 2008). Thresholds can be lower bounds of known hydrological conditions (e.g., rainfall, infiltration, and soil moisture) that triggered landslides or upper bounds of these same conditions that did not trigger landslides (Reichenbach et al., 1998; Staley et al., 2013). The spread between these thresholds is typically large or poorly delimited. Simple empirical relations might be inaccurate because they neglect important controlling factors (such as the variability in topography and soil types) and should best be reported probabilistically if used in early warning and risk mitigation systems (e.g., Basher et al., 2015; Piciullo et al., 2017; Rosi et al., 2012; Vennari et al., 2014). Practical thresholds already have been incorporated in regional early warning systems (e.g., Bogaard & Greco, 2018; Segoni et al., 2018, 2017), see also section S4.3 for an overview of the applications in different countries. Rainfall thresholds can represent changes in hydrological and geotechnical conditions of slopes, including effects of earthquake‐induced slope weakening, weathering, wildfires, climate, land use changes, slope‐ strengthening processes, and soil‐cover depletion (e.g., Brooks et al., 2002; Cannon et al., 2008; Fan, Juang, et al., 2018; Gariano et al., 2017; Kean et al., 2011; C. ‐W. Lin, Liu, et al., 2006; Shieh et al., 2009). The ubiquitous landslide debris following the 1999 Chi‐Chi and 2008 Wenchuan earthquakes (see sections 3.2.1–3.2.3) was considered an important factor for decreased rainfall thresholds and increased frequency of landslide remobilization and debris flows (e.g., H. Chen & Hawkins, 2009; Huang & Li, 2014; Lin et al., 2004; Shieh et al., 2009; Tang et al., 2016, 2009; Yu et al., 2014; S. Zhang & Zhang, 2017). In an area affected by the 1999 Chi‐Chi earthquake, Lin et al. (2004) and C.‐ W. Chang, Lin, and Tsai (2011) found that the debris flow‐triggering rainfall intensity and total precipitation dropped significantly, whereas large debris flows became more frequent in smaller catchments. After the 2008 Wenchuan earthquake, more frequent and destructive debris flows occurred in the earthquake‐stricken region as well, and rainfall thresholds for debris flow triggering were lowered significantly (e.g., Yu et al., 2014; Zhou & Tang, 2014), as much as 75% (Zhang & Zhang, 2017; Figure 11). They also reported similar reductions of 60% after the Chi‐Chi earthquake and of 50% after the 2005 Kashmir earthquake (M 7.6). These critical rainfall thresholds gradually recover to pre‐earthquake values (Figures 10 and 11). How long it takes for this recovery is still an open

FAN ET AL. 28 Reviews of Geophysics 10.1029/2018RG000626

question, and estimates range from less than a decade to as much as a century (Cui et al., 2009; Huang & Fan, 2013; Koi et al., 2008; Yu et al., 2014; Zhang & Zhang, 2017). 3.2.5. Delayed Failures Predisposed by Earthquakes Earthquakes can initiate, propagate, widen, or interconnect cracks on rock exposures to some depth, particularly near ridge crests and high on the slopes where the shaking is enhanced by topographic effects (see section 2.2.1). Cracks facilitate weathering by thermal stress and freeze‐thaw cycles, oxidation and hydrolysis, moisture, and infiltration of electrolyte solutions (e.g., Ayonghe et al., 1999; Price, 1995; Scott et al., 2016; Whalley, 1984). Thus, slope failures conditioned by such weathering might still happen several decades after the earthquake (e.g., Towhata, 2013). Post‐seismic landslides driven by long‐term rock degradation have been reported from Japan (Okamoto et al., 2013), and precursory earthquakes might play a role in enhancing landsliding triggered by the next earthquake (Parker et al., 2015; Wei et al., 2014). A recent example of a destructive landslide on a slope possibly preconditioned by an earthquake is that of the 2017 Xinmo landslide, Sichuan, China, which mobilized several million cubic meters of rock from a mountain ridge and destroyed an entire village, killing many residents (Fan, Xu, Scaringi, et al., 2017; Fan, Xu, & Scaringi, 2018; Scaringi et al., 2018). The slope was intensely shaken by the 1933 Diexi earthquake (M 7.5) that also triggered large landslides nearby. Some of them produced landslide‐ dammed lakes, such as the still‐in‐place Diexi lake (Cheng et al., 2008; Ling, 2015). Although the predisposing action of the 1933 earthquake on the Xinmo landslide has not been proven beyond doubt, the possibility exists. Signs of the incipient landslide were unnoticed, but the subsequent analysis of satellite optical and radar images revealed several interconnected cracks near the mountain ridge in place for more than a decade, and slow deformations occurred in the months prior to the event (Fan, Xu, Scaringi, et al., 2017). Five to ten thousand people die because of aseismic landslides each year in the world (Froude & Petley, 2018; Petley, 2012), with a trend that is increasing over time possibly because of human activity and increasing global population. An unknown number of these occurred in seismic regions and might have been delayed post‐seismic landslides. Fan, Xu, and Scaringi (2018) argued that high‐elevation mountain ridges in inhab- ited seismic regions should be systematically scanned for structural features (e.g., extensive cracks and slow deformations) that can foretell rock damage or imminent failure, so that real‐time monitoring, warning systems, and other countermeasures can be deployed to reduce the risk from such failures. However, delayed post‐seismic landslides remain a poorly explored research direction, primarily because they are difficult to recognize. Few studies have attempted to quantify the potential for subsurface damage as a function of earthquakes size nor have studies quantified how earthquake damage contributes to regolith production and transport from hillslopes. Seismic reflection or acoustic emission data could help to estimate rock damage (Clarke & Burbank, 2011; Von Voigtlander et al., 2018) and thus to evaluate post‐seismic responses of rock‐mass properties. 3.3. Main Research Challenges Despite a growing number of inventories of coseismic landslides (see section 2.1), detailed data and time series of the geomorphic reworking of thus produced material are few (X. Fan et al., 2019). Future research might wish to consider the following: 1. The causes for the spiked landslide rates after earthquakes remain insufficiently understood. Although rainfall‐triggered remobilization of coseismic deposits and their transformation into flow‐like landslides are now tangible at reduced complexity and in simplified laboratory experiments, more effort is needed to upscale this understanding to hillslopes or catchments in a way that allows predictions and overcome the limitations of empirical threshold‐based approaches. 2. Decadal to centennial effects of earthquakes on slope stability and landslide rates need more enquiry. More systematic studies of time‐dependent effects of earthquakes on soil and rock strength and degradation and slope monitoring in seismic areas might help to identify less obvious post‐seismic slope failures. Observations from regions hit by earthquakes in the past decades or centuries can also be beneficial to understand what to expect on mountain slopes affected by recent earthquakes. 3. Forecasting the potential of landslide dams has been aided by growing inventories. These observations have led to developments in empirical models of landslide‐dam formation. Yet physically based insights on the propensity to, and longevity of, landslide dams still are in their infancy.

FAN ET AL. 29 Reviews of Geophysics 10.1029/2018RG000626

4. Though many studies have examined the sedimentology of rock avalanches and rock slides (Dufresne et al., 2016; Dufresne & Dunning, 2017; Dunning & Armitage, 2011; Weidinger et al., 2014), hydrologic and meteorologic measurements are limited, in situ monitoring has been sparse, and geotechnical features of dam materials (i.e., particle size distribution, porosity, hydraulic conductivity, and mechanical properties) need better study.

4. Hazard and Risk Assessment, Management, and Mitigation 4.1. Susceptibility, Hazard, and Risk Landslide susceptibility maps classify terrain into zones that have different vulnerability to slope failure due to different combinations of causal factors. Landslide hazard maps also quantify probabilities (spatial, temporal, and intensity probabilities) for these zones (Corominas et al., 2013). Susceptibility maps are the most ubiquitous and are generated in cases where probabilistic appraisals are not feasible, for example, where landslide data have no time stamps. Landslide hazard maps can show different types of information, depending on the method with which they have been generated and the available historical data. Landslide frequency is generally derived from the frequency of the triggering factor (e.g., earthquake or precipitation). Landslide hazard maps that were generated from statistical or heuristic methods can display the expected landslide density as spatial‐temporal probability (densities for specific frequencies). Mass‐movement maps generated through physically based initiation and runout modeling can show the landslide intensity (as safety factor, failure probability, impact pressure, deposit depth, or flow velocity) for specified scenarios or return periods (e.g., Jibson & Michael, 2009). Risk is defined as the probability of harmful consequences and, hence, expected losses, for example, death, injury, property, livelihoods, disrupted economic activity, or environment damaged. This damage results from the interaction of hazard and vulnerable conditions (UNISDR, 2017). For the estimation of risk, several important components are required: hazard (temporal and spatial probability and intensity), quantification and characteristics of the exposed elements‐at‐risk (e.g., buildings, people, and roads), and the physical vulnerability of these elements (relating the hazard intensity to the degree of damage for specific types of elements‐at‐risk). If the various components of the risk equation can be spatially quantified for a given set of hazard scenarios (HSs) and elements‐at‐risk, the risk can be analyzed using the following equation (van Westen et al., 2017):

hi P7¼1 Risk ¼ ∑ ∫ PðÞj ×∑ PðÞj × AðÞj × V ðÞj ; (5) All Hazards PT ¼0 THS All EaR SHS ER HS ER HS

where

P(THS) is the temporal probability of a certain HS. A HS is a hazard event of a certain type (e.g., a landslide with a speciﬁed magnitude and recurrence interval);

P(SHS) is the spatial probability that a particular location is affected given a certain HS (e.g., the landslide density for a given susceptibility class). In the case of hazard intensity maps derived through physically based modeling, the spatial probability is considered to be 1;

A(ERHS) is the quantiﬁcation of the number of exposed elements at risk given a certain HS (e.g., number of people, number of buildings, monetary values, and hectares of land); and

V(ERHS) is the vulnerability (as a value between 0 and 1) of elements at risk given the hazard intensity under the speciﬁc HS. Estimating coseismic and post‐seismic landslide susceptibility and hazard, let alone risk, is thus challenging, as illustrated in Figure 12, which shows the expected changes in active landslides, exposed elements‐at‐risk, and methods used for generating landslide inventories, susceptibility, hazard, and risk, after a major landslide‐triggering earthquake.

FAN ET AL. 30 Reviews of Geophysics 10.1029/2018RG000626

Figure 12. Summary of expected changes in active landslides and exposed elements‐at‐risk, and methods used for generating landslide inventories, and assessing susceptibility, hazard and risk, after a major landslide‐triggering earthquake. (a) Schematic representation of the number of landslides before and after the earthquake. The arrows above indicate triggering events (earthquakes in red and rainfall events in blue), where the size of the arrows indicates different magnitudes. The earthquake triggers coseismic landslides, some of which could dam rivers. After the earthquake, the landslide dams decay rapidly as they are breached or stabilize, and also the number of active earthquake‐triggered landslides decrease gradually, but extreme rainfall events might reactivate some of them (light red) and also trigger new landslides (blue); (b) due to earthquake damage, the number of elements‐at‐risk (e.g., buildings) decrease drastically and start to increase after the event, depending on the speed of recovery. Extreme rainfall events can lead to debris ﬂows and landslides, which can damage the reconstructed infrastructure, leading to a sawtooth pattern; (c) methods for landslide inventory mapping, consisting of Object‐Based Image Analysis and visual interpretation, depending on the availability of cloud‐free high‐ resolution imagery (red arrows represent for earthquake‐triggered landslide and blue arrows represent for post‐seismic debris ﬂows); (d) methods for assessing landslide susceptibility. This must be updated constantly after the occurrence of the earthquake and subsequent rainfall events. Immediately after the earthquake, a rapid assessment of landslide density is required for rapid response planning before landslide inventories can be generated; (e) methods for the assessment of hazards before, immediately after, and in the years following the earthquake; (f) landslide risk assessment, before, immediately after, and in the years following the earthquake, leading to a constant decrease in risk due to recovery of the natural and built environment. QRA represents the Quantitative Risk Assessment; PAGER represents the Prompt Assessment of Global Earthquakes for Response (Earle et al., 2009). Both PAGER and Shakemap are products of the U.S. Geological Survey Earthquake Hazards Program.

FAN ET AL. 31 Reviews of Geophysics 10.1029/2018RG000626

Figure 13. Six episodes of a hazard chain that affected the town of Beichuan, China, following the Wenchuan earthquake.

4.2. Chains of Hazards and the Related Risk We illustrate the scope of a seismic hazard cascade by the chain of events that struck the town of Beichuan in six major episodes during and after the Wenchuan earthquake (Figure 13; L. M. Zhang, Zhang, & Huang, 2014). In episode I, strong seismic ground shaking caused the collapse of about 80% of the buildings in the Beichuan town. In episode II, the large Wangjiayan landslide and the Jingjiashan rock avalanche triggered by the shaking buried parts of the town, killing more than 2,300 people. A large landslide at Tangjiashan, 4.5 km upstream of the town, blocked the Jianjiang River, forming a landslide dam with a lake volume of 3.16 × 108 m3; two smaller landslide dams also formed just upstream of the town. In episode III, on 10 June 2008, the Tangjiashan landslide dam was overtopped, and the resulting flood wave overtopped the two smaller dams (Fan, Tang, et al., 2012), increasing the discharge and causing widespread damage in Beichuan; fortu- nately, the residents had been evacuated earlier (Peng & Zhang, 2013a, 2013b). In episode IV, on 23–25 September 2008, the landslide debris around the town fed severe rainfall‐triggered debris flows that buried much of the old town that remained unaffected by episodes II and III. The Jianjiang River that runs through town also changed its course after the September 2008 debris flows. In episode V, the riverbed near the town rose by 15–38 m due to sedimentation over the 5 years from 2008 to 2013. Finally, in episode VI, some 80% of town was flooded in July 2013; due to the significant rising of the riverbed, a flood that was much smaller than the 10 June 2008 dam‐break flood caused more severe inundations. The deserted town of Beichuan now serves as a memorial for the tragic earthquake and the cascade of adverse impacts that it triggered. Many landslides and debris flows following the Wenchuan earthquake remained lethal in the area for many years following the earthquake, which indicates that the hazard appraisals need to be extended beyond

FAN ET AL. 32 Reviews of Geophysics 10.1029/2018RG000626

Table 4 Recommended Procedure for Assessing the Risks From Hazard Chains Task Key steps

What can go wrong? 1. Select likely extreme hazards beyond the design levels as reference initiating hazards. 2. Fully consider hazard interactions and cascading hazards. 3. Evaluate the development of an initial hazard over its life cycle. The hazard events might not occur at the same time. What is the likelihood that it will 1. Express possible hazard interactions over space and time using event trees. go wrong? 2. Evaluate the occurrence probabilities of the hazard events using physically or statistically based methods (Keefer, 1984; Liu et al., 2015; L. L. Zhang, Zhang, et al., 2011). What are the consequences if it 1. Evaluate the space within which each individual hazard evolves over its life cycle. goes wrong? 2. Consider possible cascading effects. 3. Identify the elements at risk in the space affected by the hazards during their life cycle periods. 4. Investigate possible interactions among the vulnerabilities to the individual hazards and the effect of early warnings.

traditional practice. Only the filling of landslide‐dammed lakes allowed for sufficient warning times to allow evacuation and installation of engineering measures to minimize impact. Risk management involves at least three questions: (1) What can go wrong? (2) What is the likelihood that it will go wrong? (3) What are the consequences if it goes wrong? Assessing risks from multiple hazards generally relies on independent analyses that adopt disparate procedures and resolutions, so that several problems arise with this strategy: 1. Separate investigations of individual processes can misestimate the general risk profile (Bell & Glade, 2012); 2. Comparing risks of different origins is difficult (Marzocchi et al., 2012); 3. The weighting of different hazards and vulnerability factors are snapshots requiring regular updating; 4. The assumption of independent risk sources neglects possible interactions among the hazards (Kappes et al., 2012; Marzocchi et al., 2012); 5. Assuming that vulnerability is independent of risk neglects changes in the resilience against subsequent hazards; and 6. The aggregation of individual, independent risks might misestimate the total risk. Quantifying the risk from hazard chains thus remains challenging. Landslides evolve as a disaster chain with unknown outcomes, and hazards can be correlated. Uncertainties and feedbacks are part of the entire hazard chain and warrant more research (Kappes et al., 2012; Lacasse et al., 2012). Based on the experience of Beichuan (Figure 13) and other studies, L. M. Zhang, Zhang, and Huang (2014) and Zhang and Zhang (2017) propose a preliminary protocol to assess risks from a hazard cascade, empha- sizing hazard feedbacks and the concept of a life cycle: 1. Life cycle of hazards. Each process in a hazard cascade has a certain period of pronounced activity (Figure 13). Hence, each hazard has a life cycle; for example, from the formation of a landslide dam to its failure or the time needed for a sediment pulse to pass a given river reach. The first step is to identify these life cycles. 2. Duration. Starting from the earthquake, each individual hazard should be evaluated over its entire life cycle. 3. Extent. Each individual hazard should be investigated over its potential affected area during its life cycle. For landslide dams, consider the maximum possible inundation zone upstream and the potential flooding downstream. 4. Elements at risk. Elements at risk should be identified in areas potentially covered by hazards over their life cycles (Table 4; Chen et al., 2016; L. M. Zhang, Zhang, & Huang, 2014). Table 5 summarizes the feasibility of quantifying the risk of the earthquake hazard cascade before, immediately after, several years after, and decades after an earthquake. 4.3. Pre‐earthquake Hazard and Risk Assessment Estimating EQTL susceptibility, hazard, and risk requires knowledge about possible locations of future earthquakes and their characteristics, including magnitude, depth, rupture length, presence of surface

FAN ET AL. 33 A TAL. ET FAN Table 5 Multihazard Relations and the Feasibility of Quantifying the Risk Components Risk components

Before an Multihazard Temporal Spatial probability Intensity Physical vulnerability Elements‐at‐risk earthquake phenomena probability Rainfall‐ induced Depends on Physically‐based, statistical/ If runout Often no suitable intensity is Can be obtained from classiﬁcation of satellite landslides landslide and heuristic modeling modeling is available. Possible for or drone images, voluntary rainfall included, runout. mapping through records otherwise only OpenStreetMap, density detailed survey, or through existing databases. Earthquake‐ Many scenario Density for scenarios only No intensity Only consider destruction of Speciﬁc attributes need to be collected induced earthquakes measure exposed elements (type, use, value, and people) for landslides possible available risk assessment.

OpenStreetMap or survey Geophysics of Reviews Rainfall‐induced If discharge/ Flood modeling if sufficiently detailed data on Linking with stage‐damage flood rainfall elevation land cover and river cross sections are curves records available. available Earthquake‐ Difficult to Prediction of locations of Flood models Linking with stage‐damage induced‐flood quantify many damming landslides is not limited by lack curves chains possible of parameters Shortly after Coseismic Not relevant: Rapid modeling of landslide Use landslide Rapid damage mapping using voluntary geographic information (VGI, e.g., landslide Event has density and object‐oriented inventory to Humanitarian OpenStreepMap), satellite image classification, and drones. activity already image analysis and characterize happened interpretation locations Failure of Rapid Modeling dam break flood scenarios is time‐ Known curves but unknown Rapid mapping of new situation, using VGI. coseismic assessment of consuming. Uncertainty dam break mechanism, buildingconditions landslide dams dam failure downstream flood modeling requires high‐ hazard resolution DEM. During Landslide dam Monitoring lake Modeling dam break flood scenarios, based on Linking building type with Rapidly changing, multiple mapping stages several breakout floods level and monitoring dam, downstream flood modeling known curves needed, using classification of satellite or years after inflow requires high‐resolution DEM drone images, voluntary mapping through Rainfall‐induced Changing rainfall Monitoring through satellite Quantification of Often no intensity measure OpenStreetMap, and detailed survey. Specific landslide thresholds images or drones expected possible. Only consider attributes need to be collected (type, use, reactivation volume and destruction of exposed value, and people) for risk assessment. runout is elements OpenStreetMap or survey difficult Rainfall‐induced Changing rainfall Many watersheds with debris Unknown volume Develop new curves or use V debris flows thresholds flow probability in modeling = 1 approach.

Rainfall‐induced Depend on many Location and volume of blocking is very difficult to Linking with stage‐damage 10.1029/2018RG000626 floods due to chains predict. Flood modeling for potential scenarios is curves mass possible but depends on damming scenarios. movement activity Decades after Rainfall‐induced Return periods Flood modeling using new DEM with new Linking with stage‐damage Updated data using classification of satellite or flood in based on past sediments curves drone images, voluntary mapping through sediment‐filled records OpenStreetMap, and detailed survey. Specific valleys cannot be used attributes need to be collected (type, use, Depends on Physically based, statistical/ If runout Often no intensity measure value, and people) for risk assessment. 34 landslide and heuristic modeling modeling is OpenStreetMap or survey Reviews of Geophysics 10.1029/2018RG000626

rupture, and source mechanism (Marc, Hovius, & Meunier, 2016). This information is rarely known beforehand, though a set of plausible earthquake scenarios can be identified that might produce different landslide patterns in terms of failure types, size distributions, and density. Nadim et al. (2006) used a PGA with a 475‐year return period derived from the Global Seismic Hazard Program as one of the triggering factors in a weighted multifactor approach to estimate EQTL susceptibility world- ements are considered to wide. Such maps highlight possible regions of concern but cannot be used for spatial planning (Cepeda et al., 2013). Detailed quantitative earthquake‐induced risk assessment methods, based on event‐tree analysis, have been developed for specific high‐potential‐loss facilities, such as nuclear power stations or hydropower dams (Kwag & Hahm, 2018), but cannot be applied over large areas. 4.3.1. Physically Based Models The basis of many studies on EQTL susceptibility and hazard assessment is Newmark's (1965) model, which assumes a rigid block sliding on an inclined plane under the influence of seismic shaking (section 2.2.3). The cumulative downslope displacement induced by an increment of shaking is an index of the seismic stability of the slope (Jibson, 2007). Wilson and Keefer (1983) used strong‐motion records and field observations of a landslide triggered by the 1979 Coyote Creek (California) earth- exposed elements quake to show that Newmark's model fairly accurately predicts the Only consider destruction of dynamic behavior of landslides on natural slopes. Wilson and Keefer

Risk components (1985) proposed a framework for using Newmark's model to produce regional‐scale seismic slope stability maps, and Wieczorek et al. (1985) applied this concept for an experimental seismic slope stability map. Jibson (1993) developed a simpliﬁed Newmark method for regional analy- included, otherwise only density measure

No intensity sis by predicting Newmark displacement as a function of earthquake shaking intensity and critical acceleration, that is, the minimum ground acceleration required to initiate slope failure; Jibson (2007) updated this methodology using a much larger data set. Jibson et al. (1998, 2000) compared Newark modeled displacements to landslide data and developed a simple mathematical relation demonstrating that the predicted Newmark displacement correlates with landslide frequency, offering a simple physically based approach to estimate seismic landslide hazards at regional scale. Newmark‐displacement models have been reﬁned with last earthquake data on local slope steepness; geotechnical properties such as cohesion, Learn from the pattern of the friction angle, and density linked to geological units; and groundwater levels (Del Gaudio et al., 2003; Hsieh & Lee, 2011; Jibson et al., 1998, 2000; Liu et al., 2018; Miles & Keefer, 2001; Wang & Lin, 2011; Y. Wang, Song, et al., 2016). Newmark displacement analysis can be based on deterministic or probabilistic approaches. Strong‐motion data are the main rainfall records earthquakes

Many scenario input for deterministic analyses (Biondi et al., 2004; Gallen et al., 2017; Rodriguez‐Peces et al., 2011; Romeo, 2000), whereas probabilistic approaches use ground‐motion parameters such as Arias intensity or PGA for different return periods speciﬁed by seismic hazard studies. induced

‐ Among the many proposed reﬁnements of the method (Gallen et al., 2015; Grant et al., 2016; Kaynia et al., 2011; Saade et al., 2016), the most landslides in recovered terrain may be triggered byearthquake next

Rainfall Landslides that challenging aspect remains the availability of reliable data on the strengths of soils and rocks (Dreyfus et al., 2013) and ground motion. The latter is particularly important because predicting the topographic amplification in the steep areas in which most landslides occur is most dif- . Red = low; Orange = moderate; Green = High; VGI = Volunteered Geographic Information; DEM = digital elevation model. V = 1 indicates that the exposed el ficult (Meunier et al., 2008; Sepúlveda et al., 2005). Yet the simplified Table 5 (continued) Note experience total destruction (vulnerability = 1). See text for explanation. Newmark approach has the advantage of capturing the basic physical

FAN ET AL. 35 Reviews of Geophysics 10.1029/2018RG000626

processes (Allstadt et al., 2017). However, to predict EQTLs, this approach requires assessment of stresses and mechanical properties as a function of topography. 4.3.2. Statistical Models Statistical models predict the probability of landslide occurrence for a given area. Most approaches integrate seismic metrics, such as PGA, with terrain conditions, such as slope steepness, elevation, distance to drainage, or lithology (Jibson et al., 2000; Lee et al., 2008; Miles & Keefer, 2000, 2001, 2007, 2009). Marc, Hovius, and Meunier (2016) and Marc et al. (2017) introduced a model that considers landsliding as a function of reduced substrate strength caused by ground shaking. The model builds on a set of scaling relations between spatial landslide density, landslide area and volume, seismic ground acceleration, fault length, earthquake source depth, and seismic moment. The model draws on the widely observed relation between landsliding and a minimum level of ground shaking (Harp & Jibson, 2002; Khazai & Sitar, 2004). The ground shaking depends on the source mechanism, the wave propagation, and attenuation along the path to the surface, as well as site and directivity effects (Boore & Atkinson, 2008). For the sake of simplicity, attenuation is assumed to be linear, directivity is neglected, and site response is taken as a constant average response over the entire landscape. This model, therefore, does not retain any spatial information and only predicts a bulk landslide volume or area over the affected landscape. The model considers that multiple seismic sources can shake the landscape and that the total landslide volume reﬂects the sum of all these point sources. This modeling approach ignores static variability in material strength and the possibility that substrate strength changes progressively during prolonged shaking. The model uses a calibrated relation between earthquake rupture length and moment (Leonard, 2010), adjusted for earthquake mechanism (strike‐slip or dip‐slip). For a magnitude greater than 6.75, the acceleration saturates, and larger earthquakes do not cause higher landslide densities. The model by Marc, Hovius, and Meunier (2016) allows anticipating bulk landslide volumes given basic information about earthquake magnitude, source location, mechanism, and topographic steepness. These parameters also can be assessed quickly after an earthquake to obtain a ﬁrst con- straint on coseismic landsliding or combined with a landslide size‐frequency distribution (Hovius et al., 1997; Malamud et al., 2004). Logistic regression (Nowicki et al., 2014; Nowicki Jessee et al., 2018; Parker et al., 2015) and fuzzy logic (Kritikos et al., 2015) are among the most widely used data‐driven methods. Statistical methods also helped to generate susceptibility maps based on the coseismic landslide distribution of the 2005 Kashmir (Kamp et al., 2008), 2008 Wenchuan (Y. Wang, Song, et al., 2016), and 2015 Gorkha (Shrestha et al., 2018) earthquakes. Yet maps cannot readily be used to assess landslide susceptibility in future earthquakes with likely different characteristics and different landslide patterns. Historical earthquakes can provide useful information if coseismic landslide inventories are available, but there are very few cases where such inventories are available for earthquakes having different magnitudes in the same area.

Little is known about whether and how patterns of historical earthquakes and earthquake‐induced landslides (Cauzzi et al., 2014; Nowicki Jessee et al., 2018; Tanyas, Allstadt, & van Westen, 2018) relate to earthquake scenarios in a given area. This would require generating a ground‐motion prediction map for each scenario earthquake. Modeling of ground motion for potential earthquakes and integrating the results with statistical models trained on historical EQTL inventories could be an item of future research, perhaps also as input in probabilistic seismic hazard models. 4.3.3. Hazard Assessment for Large Landslides Earthquakes in mountainous regions also can trigger a small number of very large landslides (>107 m3). Examples of very large EQTLs since 1900 include the following: Usoi, Tajikistan, 2 × 109 m3 in 1907 (Evans et al., 2011); Bairaman, Papua New Guinea, 1.8 × 108 m3 in 1985 (King et al., 1987, 1989); Tsaoling, Taiwan, 1.3 × 108 m3 in 1999 (Chigira et al., 2003; Tang et al., 2009); Hattian Bala, Pakistan, 8.5 ×107 m3 in 2005 (Dunning et al., 2007; Owen et al., 2008); and Daguangbao landslide, China, 109 m3 in 2008 (Huang & Fan, 2013). Some massive prehistoric landslides are believed to have been triggered by earthquakes, such as the Seymareh rock slide, Zagros Mountains, Iran, one of the largest subaerial landslides on Earth with an estimated volume of about 4 × 1010 m3 (Roberts & Evans, 2013).

The susceptibility to the generation of very large EQTLs varies according to the geologic and geomorphic setting. Sediment‐storage landscapes such as loess terrain and sensitive‐clay basins are particularly susceptible. Many large landslides have been triggered in the Loess Plateau of China (Sun et al., 2017; Zhuang et al.,

FAN ET AL. 36 Reviews of Geophysics 10.1029/2018RG000626

2018) and in other loess‐mantled parts of Central Asia (Evans et al., 2009; Ishihara et al., 1990) in historical time. The 1920 M 8.5 Haiyuan earthquake (Zhang & Wang, 1995; Zhang & Wang, 2007), for example, caused the widespread collapse of loess slopes in China's Gansu Province and the occurrence of many catastrophic flowslides, some of which dammed rivers forming significant landslide‐dammed lakes. In the Champlain Sea basin of eastern Canada, the February 1663 earthquake triggered widespread collapse of subaerial and submarine slopes including massive (>107 m3) sensitive‐clay landslides (e.g., Filion et al., 1991; Syvitski & Schafer, 1996). Large EQTLs are of particular interest from geomorphic and risk‐management perspectives because of (1) the contribution they make to coseismic denudation and sediment transfer in affected landscapes, (2) the direct coseismic hazard to human activities they represent, (3) the secondary post‐seismic hazard they generate through the formation of landslide dams (Evans et al., 2011; Fan, Xu, van Westen, et al., 2017) and the immediate and longer‐term geomorphic impact of landslide damming and landslide‐dam failure, and (4) their tendency to persist in the landscape, as both scars and deposits. Predicting the possible locations and volumes of very large EQTL is not possible with the current state of knowledge. No studies exist that successfully predicted the location and magnitude of very large landslides as first‐time failures. However, forensic studies on the factors that control their occurrence (Evans et al., 2009) could lead to better understanding of their causal mechanisms. These, in combination with improved methods for collecting global data sets on (structural) geology, geomorphology, and topography and improved modeling capabilities of ground shaking and topographic amplification, could lead to prediction of potential large landslide sites in future (Tropeano et al., 2017). 4.3.4. Landslide Dam Hazard and Risk Somewhat like very large landslides, few studies have addressed the assessment of landslide‐dam risk resulting from earthquake‐induced landslides. Because landslide dams are consequential hazards of earthquakes, any hazard and risk appraisal should investigate each component of the hazard chain: earthquake → coseismic/post‐seismic landslides → damming → dam failure → dam‐break flood. Event trees are a standard tool for capturing such hazard chains (Gill & Malamud, 2014) by quantifying and multiplying the probabilities of the possible mutually exclusive outcomes that occur in sequence (Marzocchi et al., 2012; see Figure 14; Fan, van Westen, et al., 2012). Using event trees requires knowledge of possible interactions between all process outcomes.

4.4. Rapid EQTL Susceptibility Assessment Swiftly estimating the extent and distribution of landslide‐affected areas immediately after an earthquake is important for coordinating response efforts; highlighting coordinates of landslide‐blocked roads, rivers, and critical lifelines; and contributing to preliminary loss estimates (Wald et al., 2003; Wegscheider et al., 2013). This assessment is relevant for the days (and possibly weeks) until the first landslide inventory maps become available. Obtaining such data at sufficient reliability is time‐consuming despite many advanced mapping techniques (Martha et al., 2017). For example, the EQTL inventory for the 2015 Gorkha earthquake took six analyst teams about a month to create in one of the fastest rapid‐response campaigns ever mounted (Kargel et al., 2016; Robinson et al., 2017). For the Wenchuan earthquake, the landslide inventory mapping took more than 1 year to complete (Gorum et al., 2011; Xu, Xu, Yao, & Dai, 2014). Progress in post‐event susceptibility and hazard modeling can be gaged by the amount of systematic data collection (Figure 15). Before the 1994 Northridge earthquake, general statistical relations were made between earthquake magnitude and the area affected by landslides or the maximum landslide distance from either the epicenter or the rupture zone (Keefer, 1984). Jibson and Harp (2012) found that these distances varied by tectonic settings and thus magnitude was an unsuitable parameter to characterize landslide‐affected areas. PGA contours offer more conclusive correlations with landslide extent and density (Jibson & Harp, 2016; Meunier et al., 2007). Wilson and Keefer (1985) originally proposed a minimum threshold of 0.05 g for coseismic landsliding, based on data from 40 earthquakes collected by Keefer (1984). Jibson and Harp (2016) analyzed six earthquakes with well‐mapped landslide data and suggested a PGA range of 0.02–0.08 g as the minimum ground acceleration capable of triggering landslides; for snow avalanches, this limit is 0.03 g (Podolskiy et al., 2010b). The 1994 Northridge earthquake was the first for which extensive data on engineering properties of geologic units, ground shaking, slope, and triggered landslides were available to permit a detailed regional analysis (Jibson et al., 1998, 2000). After the earthquake, real‐time digital

FAN ET AL. 37 Reviews of Geophysics 10.1029/2018RG000626

Figure 14. Probabilistic event tree for earthquake‐triggered landslide dam‐break flood hazard. The six steps (nodes) pro- ceed from regional to local scale. A branch that terminates with “Clone” is identical to the central branch. For example, at node 1, the other magnitude bins are identical to the central 6 ≤ M < 7 branch. At node 2, Si represents the spatial probability (susceptibility) of landslide type i, where av = avalanche; sl = slide; flw=flow; fal = fall, given specified seismic source parameters (i.e., magnitude, fault type and length, and peak ground acceleration). At node 3, Fi stands for the size‐frequency distribution of landslide type i, which can be analyzed using the methods presented in section 2.1.2.

seismic networks were expanded in many areas (Wald et al., 2003), and the U.S. Geological Survey ShakeMap system was developed (Wald et al., 1999) to provide estimates of ground‐motion parameters in near‐real time for signiﬁcant earthquakes worldwide. These ground‐motion predictions are now commonly used in many EQTL susceptibility and hazard assessments (Allstadt et al., 2017). The 2008 Wenchuan earthquake (Figure 15) triggered the largest number of coseismic landslides ever recorded ~200,000; (Fan, Domènech, et al., 2018; Tanyas et al., 2017; Xu, Xu, Yao, & Dai, 2014). Using data from this as well as the 1994 Northridge and 1976 Guatemala earthquakes, Godt et al. (2008) developed a global model for the rapid assessment of EQTLs that uses ground‐motion parameters from ShakeMap and globally available data to estimate the critical acceleration of slopes at a regional scale. Without any formal quantitative validation, the authors stressed that their model offered qualitatively successful results. Parker et al. (2015) trained a statistical model using two New Zealand earthquakes (1929 Buller, Murchison, and 1968 Inangahua). Tested on the 1994 Northridge, 1999 Chi‐Chi, and 2008 Wenchuan earthquakes, however, the authors concluded that their model overpredicted landslide occurrence. The 2015 Gorkha earthquake provided the opportunity to test multiple models in real time. Immediately following the earthquake, three different landslide hazard maps, based on the methods developed by Kritikos et al. (2015), Parker et al. (2015), and Gallen et al. (2015), were posted online. Gallen et al. (2017) made a detailed evaluation of the statistical landslide prediction models of Kritikos et al. (2015) and Parker et al. (2015) including a quantitative validation using the landslide inventory created by Roback et al. (2018) and concluded that the models signiﬁcantly overpredicted the landslide‐affected area due to inaccurate shaking estimates provided by ShakeMap. As time evolved following the 2015 Gorkha earthquake, the focus of EQTL studies shifted from modeling methods back to inventory data because researchers realized that to improve their models they needed better data sets. The number of EQTL inventories have been increasing rapidly. Tanyas et al.

FAN ET AL. 38 Reviews of Geophysics 10.1029/2018RG000626

Figure 15. Progress in earthquake‐triggered landslide modeling studies through time. The earthquake inventories used to compute cumulative numbers of inventories are those reported in Tanyas et al. (2017). Vertical lines indicate the time of earthquakes signiﬁcant to earthquake‐triggered landslide research (red) and products useful for developing models (blue dashed line). Sciencebase refers to the release of the openly available inventory repository (Schmitt et al., 2017).

(2017) gathered 66 EQTL inventories from around the world and worked with their authors to make them publicly accessible through the U.S. Geological Survey ScienceBase platform developed by Schmitt et al. (2017). This compilation provides a resource for training models. Using 23 of these inventories, Nowicki Jessee et al. (2018) updated the statistical model proposed by Nowicki et al. (2014) for the near‐real‐time prediction of EQTL probability, by increasing the grid resolution to 250 m and using PGV rather than PGA as the primary ground‐motion input. Nowicki Jessee et al. (2018) based the model on 23 inventories to represent a wider range of earthquake magnitudes, climatic, and tectonic settings. Although the model still overpredicts landslide probabilities (Allstadt et al., 2018), it was rigorously tested and validated. Tanyas et al. (2019) carried out a similar statistical analysis for worldwide earthquake‐induced landslide susceptibility assessment using a slope unit approach, which are terrain parti- tions attributed to similar hydrological and geomorphological conditions and to processes that shape natural landscapes. They used a set of 25 earthquake‐induced landslide events that were categorized based on the similarity between causal factors to determine the most relevant training set to make a prediction for a given landslide event. The results show that categorizing landslide events introduces a remarkable improvement in the modeling performance of many events. The categorization of existing inventories can be applied within any statistical, global approach to earthquake‐induced landslide events. The U.S. Geological Survey's Prompt Assessment of Global Earthquakes for Response (PAGER) system (Earle et al., 2009) provides rapid estimates of earthquake‐speciﬁc economic losses and fatalities, but it does not explicitly account for losses due to landslides. This long‐recognized problem has initiated research toward the goal of explicitly including ground failure in PAGER's loss estimates (Wald, 2013). In a step toward that goal, the U.S. Geological Survey has recently released a new near‐real‐time earthquake product, Ground Failure, which considers both landslides and liquefaction and provides an overall assessment of hazard and population exposure as well as geospatial maps of hazard. The system considers global earthquakes, and any models implemented currently must be applicable worldwide. The input data for such models must also be available globally, and the models must produce geospatial maps of probability. Three models are currently implemented for landslides—those presented by Godt et al. (2008), Nowicki Jessee

FAN ET AL. 39 Reviews of Geophysics 10.1029/2018RG000626

Figure 16. A proposed hazard assessment model and mitigation workﬂow for landslide dams modiﬁed from Fan, Zhan, et al. (2018).

et al. (2018), and Nowicki et al. (2014) with the last serving as the default model for alert‐level determination and for display on interactive web maps. As of October 2018, this system provides qualitative descriptors of hazard and loss estimates. Full integration with PAGER necessitates quantitative estimates, but further research and development is required to reach that point. Allstadt et al. (2018) used remote sensing and ﬁeld observations from the 2016 Kaikōura (New Zealand) earthquake to evaluate the three models that are currently implemented in the U.S. Geological Survey Ground Failure earthquake product. They examined the model performances and how prediction maps changed as the ShakeMap ground motion estimates evolved through time. For this test case, the authors concluded that (1) any of the three models were useful for rough prediction of EQTL patterns, (2) all models overpredicted the hazard, and (3) the temporal change of the ShakeMap models that occurs as more data on fault geometry, felt reports, strong ground‐motion data are incorporated in the hours to days after the earthquake has a strong positive effect on model output.

4.5. Rapid EQTL Dam Hazard Assessment Landslide dams are another important process in the earthquake hazard cascade that requires rapid assessment. This includes appraisals of dam stability, potential dam‐failure mechanisms, and modeling of possible outburst ﬂoods (Figure 16). The longevity of landslide dams ranges from minutes to thousands of years, depending on many factors, including dam volume, size, shape, and sorting of blockage material, rates of seepage through the blockage, and rates of sediment and water ﬂow into the newly formed lake (Costa & Schuster, 1988). Studies of the longevity of landslide dams (Costa & Schuster, 1988, 1991; Ermini & Casagli, 2003; Fan, Tang, et al., 2012; Peng & Zhang, 2012a, 2012b) can be summarized as follows: 85–

FAN ET AL. 40 Reviews of Geophysics 10.1029/2018RG000626

87% of dams that reportedly failed lasted less than 1 year, 40–71% less than 1 month, and 27–34% less than 1 day. The stability of landslide dams depends heavily on their internal composition, which can be problematic to uncover if the time between formation and failure is short (Costa & Schuster, 1988; Korup, 2005; Korup & Tweed, 2007). Rapid predictions of landslide‐dam stability thus must remain prone to high uncertainty. Casagli et al. (2003) proposed to use geomorphic parameters for a rapid assessment of dam stability as a prequel to more accurate and time‐consuming geotechnical stability analyses (Dunning et al., 2005). After the Wenchuan earthquake, Xu et al. (2009) proposed a rapid hazard assessment matrix by considering dam height, composition material, and maximum capacity of the landslide‐dammed lakes, which was applied extensively in the month after the earthquake. Thirty‐two of the dams were evaluated to have high dam‐ breach hazard and were breached artificially by the Chinese government. Landslide dam stability and breaching is a well‐studied topic (e.g., Bonelli, 2013, 2018). Studying the likely dam‐failure mechanisms can involve scenarios of overtopping, piping, erosion, suffusion, entrainment, and gravitational collapse (Walder & O'Connor, 1997). For breach mechanisms such as piping, major numerical advancements have been made (e.g., Rotunno et al., 2017) including discrete numerical modeling with hydromechanical coupling (e.g., Tran et al., 2017). Numerous empirical equations express peak discharge at the dam site with metrics such as impounded lake volume, depth, and area (Costa & Schuster, 1988; Evans, 1986; see also Peng & Zhang, 2012a, 2012b, for overview). The optimal approach is to apply a physically based model to simulate the dam‐failure process, predict the resulting outburst flood hydrograph, and map the downstream routing of the flood wave. One commonly used physical model is BREACH, developed by the U.S. National Weather Service (Fread, 1991). Fan, Tang, et al. (2012) applied BREACH to predict the flood hydrographs for the Tangjiashan landslide dam and used the output hydrographs in the 1‐D‐2‐D SOBEK hydrodynamic model to simulate spatial variations in flood parameters such as arrival time and flood depth. Proxies of the erodibility of dam materials and breach process also have been applied (Chang & Zhang, 2010; Chen et al., 2015; Davies et al., 2007; Shi et al., 2015; G. Wang, Furuya, et al., 2016), as well as a Geographic Information System‐based hydrodynamic model (Croissant, Lague, Steer, & Davy, 2017; M. Li, Sung, et al., 2011).

4.6. Post‐Seismic Landslide and Debris‐Flow Hazard and Risk Assessments Disaster management in earthquake‐affected mountainous areas focuses on the response phase for immediate aid. This is followed by the recovery phase for rehabilitation and reconstruction (UNISDR, 2017). Depending on the scale of the earthquake and resilience of the affected society, recovery can be highly variable (Figure 12) or even entail abandoning of the disaster area (Zobel & Khansa, 2014). In any case, recovery efforts should be informed by assessments of hazards and risks in the earthquake cascade, which need to be updated regularly (van Westen & Greiving, 2017). EQTL inventories are important data in this regard. Yet such inventories commonly are not detailed enough to provide time stamps of rainstorms that reactivate coseismic landslides, unless images have high repeat rates. Acquiring such data from earthquake disaster areas requires collaboration between different organiza- tions and ideally also local communities, in reporting landslide events, for instance, through web‐based tools (Frigerio et al., 2014). Vegetation monitoring is an important component to analyze the changes in post‐ seismic landslide hazard and risk because vegetation regrowth can stabilize EQTL areas (Yang et al., 2018). Forest cover and agriculture can be both a cause of post‐seismic landslides and elements at risk (W. ‐T. Lin, Lin, & Chou, 2006; Yang et al., 2018). Monitoring the changes of landslide material in the post‐seismic landscape informs estimates of sediment loads and ﬂooding in rivers. Drilling and geophysical surveys (Jongmans & Garambois, 2007), geometric estimates, physically based models, empirical relations, and DEMs‐of‐difference all allow quantifying landslide volumes. The most widely used method for landslide volume estimation is to rely on regression models between landslide area and volume (Guzzetti et al., 2009). Although generally applicable, this method can result in large errors for both individual landslides and total inventories (Larsen et al., 2010). More accurate results, albeit for fewer landslides, come from repeat LiDAR surveys corrected for disturbing effects of vegetation or other surface objects (Chen et al., 2006; Chen et al., 2013; Tseng et al., 2013). In practice, pre‐ earthquake LiDAR DEMs are rarely available, so that other, less detailed alternatives, for example, contour lines or photogrammetry, must be considered (Fan, Xu, & Scaringi, 2018; Fan, Xu, Scaringi, et al., 2017).

FAN ET AL. 41 Reviews of Geophysics 10.1029/2018RG000626

4.6.1. Hazard Assessment Predicting the dynamics of coseismic landslide scars and deposits yields insight on how quickly effects of widespread mass wasting attenuate. In the area affected by the Wenchuan earthquake, Fan et al. (2019) created a multitemporal post‐seismic debris‐flow database that includes 527 debris flows in 244 watersheds. They also collected rainfall data from 91 rain gages, which they correlated to analyze the relation between rainfall and debris‐flow characteristics (data set release: Domènech et al., 2018). The study showed that the total number of active landslides decreased rapidly within 7 years (from 9,189 in 2008 to 221 in 2015), and the post‐seismic reactivations occurred randomly within the studied watershed, thus making them very difficult to predict. They also concluded that even with such a large data set, it remains difficult to establish rainfall‐triggering thresholds empirically, due to varying lithological, soil, and land‐cover conditions in the watersheds and the scarcity of reliably reported events. Critical thresholds might work for warning purposes in individual watersheds but may be inapplicable to others (Huang et al., 2015). Physically based models and geotechnical investigations are more suitable for analyzing post‐seismic changes in landslide‐triggering rainfall thresholds. Many studies use regional versions of the infinite‐slope model, ideally coupled with hydrological models to incorporate changing groundwater levels or changing rainfall infiltration. Examples of these models are SHALSTAB (Montgomery & Dietrich, 1994), SINMAP (Pack et al., 1998), and STARWARS + PROBSTAB (van Beek, 2002). By design, infinite‐slope models apply to shallow, translational landslides. S. Zhang et al. (2018) used a Monte Carlo variant of this model to express uncertainty of soil mechanical parameters in a model for rainfall‐induced shallow landslides in the Wenchuan earthquake area. More locally, Chen and Zhang (2015) applied a physically based distributed model to predict rainfall‐induced shallow slope failures in two‐layer soils under various rainfall conditions. Whereas 2‐D and 3‐D slope stability models were initially only applicable at the site scale, modern 3‐D models such as Scoops3D (Reid et al., 2015) and r.slope.stability (Mergili et al., 2014) allow simulations over large areas and different geometries for the slip surface. Parameterizing these models needs to account for the spatial variation of geotechnical and hydrological parameters (Reid et al., 2015). Empirical runout models such as Flow‐R combine flow patterns and failure probability to create runout susceptibility maps (Horton et al., 2013) and are useful at a regional scale. Runout models such Flow‐2D, RAMMS, MassMove2D (Beguería et al., 2009), Debris Mobility Model (Kwan & Sun, 2006), and AschFlow (Luna et al., 2016) have seen much use in debris‐flow hazard and risk assessment (e.g., M. Chang et al., 2017), but they require input on discharge, sediment concentration, initiation volumes, and entrainment. In any case, such models can offer insights into the relationships between inputs (e.g., rainfall) and consequence (e.g., runout; van Asch et al., 2014). Few of the available physically based models combine slope stability, erosion, entrainment, and runout. Thus, recent attention has turned to integrated models for hydrologic and sedimentary processes. Fan, Lehmann, McArdell, and Or (2017) introduced an innovative way to model catchment‐scale landslide behavior by linking rainfall‐induced landslide initiation with debris‐flow runout models. N. Zhang and Matsushima (2016) developed a soil‐water mixing model coupled with a depth‐integrated particle method for debris flows by considering sediment entrainment and rainfall characteristics in the Wenchuan epicentral area. Bout et al. (2018) developed an integrated model of slope hydrology, slope failure, mass movements, and runout (OPENLISEM), which allowed for an improved modeling of the hazard interactions that can occur during an extreme rainfall event, such as landslide initiation, runoff, landslide damming, flooding, and debris flow. Integrated simulation of hydrology, entrainment, and debris‐flow initiation also has been attempted by Chen and Zhang (2015) in the EDDA model, by Hu et al. (2016), and by Lehmann and von Rütte (2017) in the STEP TRAMM model. These models consider various interacting processes including rainfall runoff, erosion, landslide initiation, debris flows, and flooding, in different scenarios, for example, for changing sediment availability or vegetation stands. Many of these models underestimate landslide volumes due to the use of a simplified landslide‐initiation mechanism. 4.6.2. Exposure, Vulnerability, and Risk Assessment The elements‐at‐risk can be classified in various ways depending on the intended risk‐management application (van Westen et al., 2008), for example, for the design of early warning systems, the distribution of population in space and time is most relevant. In the planning of risk‐reduction measures, cost‐benefit analysis might be required to compare different alternatives, and it is then important to quantify physical objects such as buildings in monetary terms. The interaction of elements at risk and hazard footprints (the area

FAN ET AL. 42 Reviews of Geophysics 10.1029/2018RG000626

affected by a hazard of a given intensity) defines the exposure and the vulnerability of the elements‐at‐risk. The risk components (hazard, exposure, and vulnerability) all change after an earthquake (Figure 12). Severely damaged buildings and other infrastructure might need to be demolished. Temporary shelters are required for the homeless, and choosing suitable sites is an immediate task (Fang et al., 2011). After the response phase, elements‐at‐risk change rapidly during the rehabilitation and reconstruction phases (Dalen et al., 2012; Y. Wang, Zou, & Li, 2015). For mapping and characterizing these changing elements‐ at‐risks, remote sensing combined with Volunteered Geographic Information (Schelhorn et al., 2014) can be useful. Volunteered Geographic Information has increasingly assisted the reassessment of elements‐at‐ risk after disaster through initiatives such as Humanitarian OpenStreet Map (Geiß & Taubenböck, 2017). Vulnerability is another key component in the estimation of losses and risk from natural hazards. It remains a multifaceted component, and numerous perspectives of how to best characterize vulnerability have been proposed (Birkmann, 2006; Fuchs, 2009; Shirley et al., 2012). The engineering and natural sciences communities focus on physical vulnerability as “the degree of loss to a given element, or set of elements, within the area affected by a hazard, expressed on a scale of 0 (no loss) to 1 (total loss)” (UNDRO, 1984). Vulnerability also can be expressed as fragility, reflecting the probability of reaching a certain damage state for a given level of intensity. Factors that confound vulnerability estimates in the case of EQTLs include differing landslide types and processes; difficulty to express landslide intensity; limited information on precursory, triggering, and controlling conditions; few accurate or limited observational data for process‐intensity estimation (Papathoma‐Köhle et al., 2011); and sparse damage data, especially on buildings or lives lost (Fuchs, 2009). Vulnerability curves express the relation between hazard intensity (e.g., impact pressure) and the fraction of damage directly mapped after disasters, simulated numerically or obtained from expert opinion (Fuchs et al., 2007; Jakob et al., 2012; Zeng et al., 2012). Dynamic runout models for debris flows offer out- puts on the depths, velocities, and impact pressures of landslides and debris flows and can thus aid vulnerability assessment. Care should be taken to use vulnerability curves derived for different building types and environments. Physical vulnerability curves for post‐seismic debris flows have been derived after the Chi‐ Chi and Wenchuan earthquakes (Lo et al., 2012; Zeng et al., 2014). When these are compared to European curves, they show large differences, primarily because the buildings in Wenchuan were newly constructed after the earthquake and were much more resistant than those used for developing the European curves. In some cases, the vulnerability of buildings or roads can be assessed using Differential Interferometric Synthetic Aperture Radar measurements combined with damage observations (Bianchini et al., 2015; Peduto et al., 2017). 4.7. Main Research Challenges In conclusion (see also the summary in Table 5), few studies offer a comprehensive quantitative risk assessment for post‐seismic hydrometeorological hazards because of several challenges: 1. Quantifying hazard chains spatially is difficult. Risk assessment of coseismic and post‐seismic landslides is hampered by the complications in downscaling probabilistic seismic hazard assessment results to the local level, where topographic and soil amplification can modulate landslide size (volume) distributions and hence the pattern of physical impact. 2. Reactivation of coseismic landslides controls the flushing of sediment into and out of the local drainage network, but few conceptual and physical models currently deal with sustained slope‐failure dynamics, which makes this component difficult to predict. More effort also should be given to the modeling of failure surfaces so that actual volumes can be calculated. 3. Monitoring of changing elements‐at‐risk such as buildings, population, infrastructure, land use, and land cover is essential for post‐seismic risk assessment. New tools such as drones and high‐resolution satellite imagery; voluntary geographic information; and crowdsourcing could turn out to be valuable ingredients in multihazard risk assessment. Vulnerability assessment also requires more attention, for example, in the form of comparable physical vulnerability and fragility curves.

5. The Post‐Seismic Sediment Cascade EQTLs can weaken and erode large quantities of material by landsliding. This sediment is initially deposited on hillslopes or valley bottoms, where it is prone to subsequent reworking in the sedimentary cascade. Local aggradation ﬁlls channels, reduces accommodation space, and changes the probability and severity of

FAN ET AL. 43 Reviews of Geophysics 10.1029/2018RG000626

Figure 17. Illustration of how sediment from earthquake‐triggered landslides that is delivered to the fluvial system can lead to river‐channel aggradation. This results in a decrease in accommodation space and thus increases the probability and severity of overbank flooding. The sequence of events described represents a cascade of sedimentary hazards that can hamper reconstruction following earthquakes. (a) River channels prior to delivery of sediment from landslides. (b) Aggraded channels that can no longer accommodate storm flow, leading to flooding (after Dingle et al., 2017). (c) Imagery and photographs of the coseismic Wenjia landslide and the Mianyuan River in Sichuan, China, immediately after and then years after the earthquake. The landslide deposits were remobilized and developed into a catastrophic debris flow during the 13 August 2010 rainstorm. The debris flow partially dammed the Mianyuan River, causing flooding of the newly reconstructed Qingping town. Red outlined buildings were partially buried by the post‐seismic debris flow.

flooding (Figure 17), this affects post‐seismic resilience and recovery. The longer‐term relevance of such sediment flushing, for example, for landscape evolution, is discussed in section 6. This chapter considers how and over what time scales the sediment transport occurs and what some of the consequences can be. A growing literature seeks to understand the sequence of sedimentary impacts of landslides on channel processes (Chen & Petley, 2005; Croissant, Lague, Steer, & Davy, 2017; Davies, 2017; Hancox et al., 2005; Korup, 2004b), especially where buildings, bridges, and other critical infrastructure are in the path of landslide‐derived sediment pulses. Work has focused on documenting the transfer of sediment from hillslopes to channels (e.g., Fan et al., 2019; Fan, Domènech, et al., 2018; Tang et al., 2016), the resulting fluvial sediment pulse and river‐channel aggradation (Dadson et al., 2004; Pearce & Watson, 1983; J.

FAN ET AL. 44 Reviews of Geophysics 10.1029/2018RG000626

Figure 18. Multiyear suspended‐sediment and event‐based landslide records from the Chi‐Chi earthquake in Taiwan (a) and the Wenchuan earthquake, China (b). (a) is modified after Hovius et al. (2011), where K is defined as the suspended‐ sediment concentration normalized for river discharge. (b) shows the volume of landslides before and after the Wenchuan earthquake from the multitemporal landslide inventory generated by Fan et al. (2019) and downstream suspended‐sediment gain data from J. Wang, Jin, et al. (2015), reflecting the increase in sediment along the Fujiang River reach affected by EQTL. In both cases, the notable seismic enhancement of sediment fluxes gradually decays to baseline levels over ~5 years, similar to the time scale over which post‐seismic landslide activity also wanes. The relation between these remains to be better understood, but both may have a similar underlying control. The horizontal dashed lines represent the average of normalized landslide areas (in a) and volume of landslides in (b) before the Chi‐Chi and Wenchuan earthquake. Vertical dashed blue lines show major precipitation events.

Wang, Jin, et al., 2016; J. Wang, Jin, et al., 2015; Yanites et al., 2010), and the long‐term record preserved in ﬂoodplain and fan sediment (Davies & Korup, 2007; Schwanghart et al., 2016).

5.1. Observations of the Sedimentary Response to Earthquakes 5.1.1. Post‐Seismic Suspended Sediment Fluxes: Insights From Chi‐Chi and Wenchuan The sedimentary response to large earthquakes has been recognized for many decades (Keefer, 1994; Pain & Bowler, 1973; Pearce & Watson, 1986) and earlier accounts compiled by Ambraseys (2009), but the 1999 Chi‐ Chi and 2008 Wenchuan earthquakes spurred the quantitative understanding of post‐seismic sediment dynamics. Records of suspended sediment concentration and water discharge collected routinely at river gaging stations in Taiwan and Sichuan helped to constrain pre‐earthquake conditions and quantify the magnitude and timing of perturbation (Dadson et al., 2004; Hovius et al., 2011; Huang & Montgomery, 2012; J. Wang, Jin, et al., 2016). Suspended sediment fluxes increased following both earthquakes as a result of landslide inputs (Figure 18), mostly driven by intensive rainfall and high runoff in subsequent years; this process highlights the role of fluvial transport capacity in regulating the pace of landslide‐sediment evacuation. Though transport capacity is clearly important for sediment transport, across different river catchments affected by the Wenchuan earthquake, sediment fluxes are correlated with the proportion of catchment area affected by landsliding more clearly than with hydrological parameters, such as runoff and stream power (Li et al., 2017). Also, in both earthquake areas, sediment fluxes relaxed to pre‐earthquake background levels

FAN ET AL. 45 Reviews of Geophysics 10.1029/2018RG000626

within a decade (Domènech et al., 2018, 2019; Fan et al., 2019; Fan, Domènech, et al., 2018; Hovius et al., 2011; Huang & Montgomery, 2012) despite landslide material remaining in the landscape (Figure 18). These observations suggest that there is a complicated relationship between sediment supply and the transport capacity of rivers following large earthquakes; at least in the rivers studied after Chi‐Chi and Wenchuan, fluvial transport capacity alone did not seem to control sediment flux, and instead, suspended‐sediment transport appeared to have been at least partly determined by supply. The interplay of transport and supply limitation (Hovius et al., 2000) was influenced by different factors following the two earthquakes. In Taiwan, post‐seismic typhoons triggered many new landslides, providing new sediment at the same time as raising fluvial transport capacity. Typhoon‐triggered landslides shifted increasingly closer to river channels over time, accentuating their potential role as fluvial sediment sources (Hovius et al., 2011). In Wenchuan, fewer new post‐seismic landslides formed, but debris flows triggered during post‐seismic storms delivered much coseismic landslide sediment into rivers (Domènech et al., 2019; Fan, Domènech, et al., 2018. Fan et al., 2019; Tang et al., 2016; S. Zhang, Zhang, et al., 2016). Few detailed gaging‐station data are available for other landslide‐triggering earthquakes for quantifying changes in suspended sediment fluxes, though some other data are becoming available (e.g., the 2010 Maule earthquake, Chile; Tolorza et al., 2016). Even without detailed time series, analyses of long‐term sediment‐flux records with respect to patterns of active seismicity support the notion that earthquakes are important erosional drivers (Vanmaercke et al., 2014, 2017). Anticipatory planning could open possibilities for immediate post‐seismic measurements where baseline data exist, but monitoring has stopped in recent decades, as is the case in southern California (Warrick & Mertes, 2009). Proactively initiating baseline monitoring would be prudent in locations highly prone to large earthquakes in the near future, such as in the Southern Alps of New Zealand or the Transverse Ranges of southern California. 5.1.2. Observations of Bedload Sediment Response Most gaging data capture only suspended river loads. Once entrained by rivers, a fraction of this suspended material is removed to distal sedimentary basins, or offshore, where its sequestration can have geodynamic implications. Coarser bed sediment accumulates in channels and alluvial fans. For example, in the Matiri River in New Zealand, Pearce and Watson (1986) estimated the amount of sediment persisting from landslides triggered by the 1929 Murchison earthquake. They suggest that much of the landslide material remained in low‐order streams even after 50 years. A study of small catchments in Japan by Koi et al. (2008) showed that contemporary sediment yields in steep rivers were still largely influenced by the 1923 Kanto earthquake. Along these lines, Yanites et al. (2010, 2011) concluded that segments of the Peikang River in Taiwan would likely take hundreds of years to evacuate sediment derived from landslides triggered by the Chi‐Chi earthquake. Stolle et al. (2019) report that rivers around Pokhara, Nepal's second largest city, are still adjusting today to massive aggradation most likely triggered during several strong medieval earthquakes. Most of the active channel network is currently incised in thick valley fills that are less than a thousand years old. Repeated surveys in the Wenchuan earthquake area showed progressive movement of landslide materials from hillslopes into channels between 2008 and 2013, amounting to <10% of the total landslide volume in 5 years (S. Zhang, Zhang, et al., 2016). In some reaches, rivers incised by tens of meters in the Sichuan Basin downstream from the landslide source regions (N. Fan, Nie, et al., 2016); this observation is similar to what Yanites et al. (2010) reported for the Peikang River after the Chi‐Chi earthquake. Systematic observations would allow more general inferences of the pattern of upstream aggradation and downstream incision. Concentrations of the cosmogenic nuclide 10Be in fluvial quartz grains offer additional insight into bedload dynamics. This cosmogenic isotope is produced by interaction of cosmic rays with minerals in the upper few meters of the Earth's crust. Landslides unearth material from depth with low or negligible 10Be concentrations (Brown et al., 1995). As this material mixes with pre‐earthquake sediment in rivers, the 10Be concentration in quartz, a mineral resistant to weathering, can track the addition and eventual removal of landslide material (Wang et al., 2017; West et al., 2014). This approach requires 10Be data from before a major earthquake and is limited to grain sizes that can be sampled and analyzed (and for which pre‐ earthquake data exist), but its application in tandem with other studies of post‐landslide sediment transport is promising. Another promising technical development is the application of seismic techniques to monitoring active bedload transport (e.g., Chao et al., 2015; Schmandt et al., 2013); such techniques could find compelling application in understanding the transport of landslide‐derived sediment.

FAN ET AL. 46 Reviews of Geophysics 10.1029/2018RG000626

Figure 19. Schematic diagram illustrating the components needed to build a comprehensive model of post‐seismic sediment transport, including key interconnec- tions (red arrows) that lead to dynamic changes in connectivity, grain size, and ﬂow conditions during the post‐seismic evolution of the sedimentary system. Components are ranked based on evaluation of the current relative state of knowledge for each component (ranked out of 5, with 5/5 being most well‐known). For example, landslide mapping is now relatively well developed and sophisticated, and this information can be relatively quickly acquired from satellite images, whereas grain‐size information is scarce. Green shapes relate to material supply; light purple shapes relate to material transport.

Future efforts to quantify bedload sediment response to earthquakes might be able to use more high‐ resolution satellite imagery and possibly survey large areas with UAVs. For example, Yanites et al. (2011) used satellite imagery and field surveys to provide a regional assessment of aggradation dynamics over multiple years in central Taiwan to test the influence of intermittent sedimentation on long‐term incision rates. Yet the removal of landslide‐derived bed sediment could take many decades and thus extend beyond most current monitoring programs. Several possible avenues are open to extend this observational window to tar- get sedimentary sinks, including reservoirs, debris basins, and dammed lakes as records of post‐seismic sediment fluxes (e.g., Foster, 2010; Lavé & Burbank, 2004). 5.1.3. Sedimentary Records of Past Earthquakes Prehistoric sedimentary archives offer further information about the sedimentary legacy of past earthquakes. These records commonly lack the chronological resolution to quantify dynamics of sediment transport in detail, but they can provide insight into the sedimentary response integrated over decades to millennia (Howarth et al., 2016). Lake sediment can be a useful archive of post‐seismic sediment input (Howarth et al., 2016), although lacustrine turbidites and deformed layers (Karlin & Abella, 1992; Moernaut et al., 2014; Sims, 1973) might not always represent landslide inputs from surrounding catchments. Along the Anatolian Fault in Turkey, Avşar et al. (2014, 2016) used the geochemical fingerprint of ultramafic rocks associated with EQTLs to pin down the timing and duration of sediment pulses, which they found lasted 5–10 years. Alluvial deposits also can record sediment perturbations after earthquakes, with a notable example being the alluvial fill around Pokhara, now recognized as the legacy of catastrophic sedimentation following medieval earthquakes (Schwanghart et al., 2016). Thus far, work in these alluvial settings has primarily focused on identifying characteristic signatures that can be pinned to earthquake events, although Stolle et al. (2019) reconstructed former geomorphic surfaces from these valley fills to infer elevated erosion rates and sediment yields averaged over more than half a millennium. Finally, studies of channels and floodplains that have formed in response to large, nonseismic landslide inputs offer a basis for comparison and thus for estimating the seismic legacy of massive valley aggradation (Korup, 2004b).

5.2. Modeling Post‐Seismic Sediment Transport and Its Role in the Hazard Cascade The handful of detailed post‐seismic sediment budgets highlights that research still misses a predictive framework for estimating the effects of single earthquakes on the fluvial system, much less a sequence of earthquakes. Predicting the fluvial response to earthquake‐triggered landsliding depends on anticipating the rate at which sediment makes its way from hillslope deposits into and through the fluvial system over years to millennia (Figure 19).

FAN ET AL. 47 Reviews of Geophysics 10.1029/2018RG000626

Several studies modeled landslide‐ or debris‐flow derived sediment pulses in the fluvial system as stochastic processes (e.g., Benda & Dunne, 1997; Cui, Parker, Lisle, et al., 2003; Cui, Parker, Pizzuto, & Lisle, 2003), which introduces a probabilistic element to models of sediment transport following earthquakes (Croissant, Lague, Davy, et al., 2017; Croissant, Lague, Steer, & Davy, 2017). However, modeling sediment dynamics associated with the input from several thousand or even tens of thousands of discrete landslide sources highlight at least two important questions: 1. What determines how much landslide sediment is delivered to high‐order river channels, as opposed to being isolated on hillslopes? 2. What determines the ability of landslide sediment to move through a fluvial network?

Addressing these questions with more ﬁeld data will help to quantify the links in the sedimentary chain and inform predictive modeling. Efforts to address these questions are considered in the next two sections.

5.3. Landslide Connectivity Connectivity refers to both the fraction and rate of coseismic landslide material being delivered to channels as well as the underlying processes. Landslide deposits not connected to the fluvial system can remain mar- ooned for long periods of time; such deposits are removed gradually only by slow, diffusive hillslope transport such as soil creep. Several approaches can be used to measure connectivity in landslide inventories. Slope‐area relationships derived from digital elevation data can be used to define a threshold catchment area for fluvial channels (Montgomery, 2001), and landslides can then be defined as connected to channels if their lowest elevation point touches or intersects the channel (Dadson et al., 2004; Li et al., 2016). However, there are several uncertainties in applying this method. One is that it is unclear whether rivers in a landslide‐affected region share a common initiation threshold. Another is that the threshold area that defines fluvial channels is rarely distinct in slope‐area space (Montgomery, 2001), partly due to methodolo- gical uncertainties (Clubb et al., 2014; Passalacqua et al., 2010) and partly due to the destruction of channel heads by earthquake‐induced landslides. For the Gorkha earthquake landslide inventory, uncertainty in the channel‐head area threshold led to variations of 33–43% of landslide area and 49–60% of volume calculated as being fluvially connected (Roback et al., 2018). Another challenge is that data resolution can hamper accurate estimates; G. Li et al. (2016) argued that the resolution of digital elevation data can cause errors of ~10% in estimated connectivity. Finally, physical contact of a landslide deposit with a river channel might not reveal much about the rates of sediment transfer; processes become more important in changing connectivity in low‐order channels and hillslopes. An alternative metric of connectivity is the normalized distance from each landslide to ridges and valleys (Huang & Montgomery, 2012; Meunier et al., 2008), so that low values up to an arbitrary threshold indicate a higher likelihood of landslide material entering river channels. This metric can yield results similar to those derived using area thresholds, but it is also uncertain. Yet another alternative is to assess connectivity from the contrast between vegetation, landslides, and river channels in high‐resolution imagery (West et al., 2011). Regardless of method, most studies have found that between 5% and 20% of coseismic landslides inter- sected with rivers (Dadson et al., 2004, for Chi‐Chi; Li et al., 2016, for Wenchuan; Roback et al., 2018, for Gorkha). This percentage increases to 40–60% if considering landslide volume, at least for the Wenchuan and Gorkha cases, as larger landslides with longer runout are more likely to reach channels (Li et al., 2016). Still, this leaves about half the landslide volume apparently stranded from the drainage network. One reason is that seismically triggered landslides are most likely to initiate near ridges (Densmore & Hovius, 2000; Harp et al., 1981; Meunier et al., 2008); they might be less likely to reach channels than rainfall‐triggered landslides. Studies comparing different river catchments affected by the Wenchuan earthquake found that mapped connectivity does not significantly improve correlation between catchment‐wide landsliding and post‐seismic suspended sediment fluxes (Li et al., 2016), let alone 10Be concentrations in river‐bed sediment (Wang et al., 2017). It seems that connectivity might not fully reflect the ability of landslide debris to reach channels (Figure 20), perhaps because of one or more of the following:

1. Loose and bare material in landslide deposits on hillslopes might be highly mobile during rainstorms, whereas coarse clasts might armor landslide debris.

FAN ET AL. 48 Reviews of Geophysics 10.1029/2018RG000626

Figure 20. Connectivity between landslide deposits and channels modulates the supply of landslide sediment to river systems. (a) Only the toe of large landslide deposits connected to channels is immediately accessible for entrainment; (b) coarse clasts armor landslide debris against erosion; (c) the gradual removal of fine material, leaving coarse material behind that is more difficult to remove; (d and f) landslide material apparently isolated from rivers can become mobile via hillslope transport processes, enhanced by rilling and gullying on landslide deposits; and (e) debris flows can provide a dynamic mechanism for transporting material from colluvial parts of the landscape to rivers.

2. Low‐order channels might lack the transport capacity to move landslide material, even when it is apparently connected to the fluvial system. 3. Debris flows could entrain large fractions of landslide material on hillslopes and deliver it to rivers. 4. Only high and commensurately rare flood stages might entrain landslide material impinging on channels. Coarse material in headwaters can be transported to higher‐order rivers by debris flows, or it can remain in place where it weathers and abrades, gradually reducing in grain size until it can be transported by floods. Meanwhile, landslide deposits deposited in larger rivers downstream can gradually coarsen as fines are win- nowed, leaving behind boulders lags. Thus, both grain size and connectivity are dynamic properties that change as the sedimentary system responds to landslide inputs. These factors might need more consideration in work quantifying connectivity from landslide maps, and a more dynamic view of connectivity might help to reconcile the seemingly juxtaposed transport and supply limitations of post‐seismic sediment fluxes. Further progress might be made by detailed analyses of remote‐sensing and field surveys of debris flows and armoring of landslide deposits. For example, monitoring studies of erosional activity such as rilling and gullying of landslide surfaces indicate that over 6 years after the Wenchuan earthquake, the rate of this erosion decreased with time without any clear pattern (Domènech et al., 2019; Fan, Domènech, et al., 2018; Tang et al., 2016).

FAN ET AL. 49 Reviews of Geophysics 10.1029/2018RG000626

5.4. Grain Size and Landslide‐Sediment Mobility Grain size determines whether and how frequently landslide sediment delivered to river channels can be transported. Whether flow conditions exceed the critical bedload shear stress (Shields stress) and initiate particle motion depends on local flow depth and runoff. Smaller grains have a lower critical Shields stress and require less flow depth for transport, whereas larger clasts move only in large and rare flows. Few studies have quantified the grain‐size distributions of landslide deposits, and fewer still link these to fluvial sediment dynamics. Landslide inputs to rivers can measurably increase bedload grain size (Attal et al., 2015; Attal & Lavé, 2006). Several studies tried to physically model links between grain size of coseismic landslide deposits (e.g., Qi et al., 2012; Wang et al., 2013; L. M. Zhang, Xu, et al., 2011; S. Zhang, Zhang, & Chen, 2014) and the volumes of sediment delivered to rivers, but accurately measuring grain‐size distributions of landslide deposits is challenging (Casagli et al., 2003). Hence, a more systematic understanding of the effects of EQTLs on fluvial grain‐size distributions is important for fully understanding post‐seismic sediment dynamics.

5.5. Time Needed for Sediment Evacuation Understanding of the catastrophic aggradation of rivers fed by coseismic landslide debris has become more diversified because of detailed water‐ and sediment‐discharge records before and after the 1999 Chi‐Chi and 2008 Wenchuan earthquakes (as discussed in detail in above). Yet the response of alluvial systems to individual earthquakes or sequences of earthquakes still rarely features in longer‐term studies, especially those focused on sedimentary deposits. How long do terraces and fan deposits that were formed in the years after strong earthquakes persist on valley floors? Studies generally devoted to such landforms rarely invoke earthquakes and commonly resort to long‐term tectonic shifts in base level, climatic oscillations, or human disturbance. For example, 100‐ to 200‐m‐thick fan terraces line many bedrock rivers of the Central Range of Taiwan (Hsieh & Chyi, 2010). Radiocarbon ages of these fluvial and debris‐flow deposits scatter widely throughout the Holocene without any clear link to major climatic shifts, such that episodic disturbances by earthquakes and tropical cyclones seem more plausible generators of such thick, dissected valley fills. Some large alluvial fans could have formed or at least changed catastrophically during or shortly after strong earthquakes and could thus contain valuable information about the timing and magnitude of ensuing sediment pulses (M. ‐L. Hsieh, Liew, & Chen, 2011; Keefer, 1999; Langridge et al., 2012). One example is Pokhara, which appears to be built entirely on a 120‐km2 fan‐shaped deposit debouching from the Higher Himalayan Annapurna Massif. Detailed topographic analysis, radiocarbon dating, and sediment‐ provenance work indicate that 4–5km3 of Higher Himalayan sediment was deposited catastrophically up to 70 km from a small source area above 3,500 m a.s.l. (above sea level) in at least three major pulses coin- cident with documented dates of strong earthquakes in medieval times (Schwanghart et al., 2016). These episodes of rapid valley aggradation infilled nearly a dozen tributary mouths for upstream distances as great as 8 km and ponded several lakes that persist. Such settings offer much needed insights into the response and recovery of rivers from EQTLs over decades to centuries, if not millennia. In coastal settings, earthquake‐induced sediment pulses can leave long‐lived signatures in the form of large dune ridges that form at river mouths in response to massive amounts of fluvial sediment transferred by longshore drift. Sets of parallel dune ridges in particular can contain sedimentary signals of past earthquakes (Goff et al., 2008) and represent the coastal end member of a suite of landforms within the seismic staircase of cascading processes. This conceptual model invokes EQTLs that increase river sediment loads, which in turn promote coastal dune building, given short enough transport distances. The model also predicts that multiple dune ridges along tectonically active coasts would not only record sediment flux driven by river discharge, sea‐level, and climatic changes but would also bear an imprint of past earthquakes. From these observations, we infer that one overarching question concerning post‐seismic sediment dynamics is how long landslide material remains in the landscape. The time needed for evacuating landslide material also determines the duration of cascading hazards. The residence and flushing time of landslide debris is highly variable. Reported post‐seismic suspended sediment pulses are mostly short‐lived (<10 years), but coarse sediment aggradation in river channels lasts longer, and many of the largest landslide deposits remain visible for millennia, at least. General understanding of what controls these time scales depends on building a more complete picture of the sediment cascade from hillslope to channels (Figure 19).

FAN ET AL. 50 Reviews of Geophysics 10.1029/2018RG000626

5.6. Research Challenges and Directions With this goal in mind of developing a quantitative model for the sediment dynamics following large earthquakes, we identify several directions for future research: 1. Gather more data from future earthquakes through proactive baseline monitoring of sediment ﬂuxes in seismically active regions and preparing for events where baseline data already exist. 2. Develop further methods of high‐resolution imagery for tracking post‐seismic sediment pulses and channel changes, for example, by employing high‐resolution DEMs. Use lakes, reservoirs, and debris basins as recorders of post‐seismic sediment pulses. 3. Quantify connectivity as a dynamic property and build process‐based models of the full sediment cascade from hillslopes to channels. 4. Develop techniques to efﬁciently measure the grain size of landslide material and its changes over time.

6. The Lasting Legacy of EQTLs Landslides that scar hillslopes and deposits that clog river valleys are the most obvious geomorphic signatures of earthquakes in mountains. Yet the impact of EQTLs can persist long after scars have been revegetated, deposits eroded, and river channels evacuated. Growing data on the magnitude and distribution of EQTLs have helped to advance our understanding of the longer‐term effects of landslides as erosional agents. One question in this regard is that of a seismic mass balance: Is the net effect of large earthquakes to add material that builds topography or to remove material via landslides? Quantitative appraisals of the legacy of EQTLs have relied largely on statistical models, and more recent work has added numerical simulations. The statistical approach uses size distributions of EQTLs to infer average regional erosion rates (Keefer, 1994). These density‐adjusted rates can be compared with rates of tectonic uplift (G. Li, West, et al., 2014; Marc, Hovius, & Meunier, 2016). This strategy hinges on appropriate statistical models, sufﬁcient sample size, negligible mapping bias, and robust error propagation when combining earthquake return times with landslide counts. Long‐term projections need to consider uncertainties in the frequency and magnitude of landslide‐triggering earthquakes, how landslides respond to a given shaking intensity, the effect of precursory triggers and slope instability, and the gradual removal of landslide material. Numerical simulations rely on some of these assumptions but also include physical constraints on slope stability and transport laws for river processes (Croissant, Lague, Davy, et al., 2017).

6.1. Quantifying Seismic Erosion Rates The contribution of individual large earthquakes to the uplift and erosion of mountain belts was addressed with growing detail after the 1999 Chi‐Chi and 2008 Wenchuan earthquakes (Hovius et al., 2011; Parker et al., 2011), following Keefer (1984, 1994, 2002) and Malamud et al. (2004). These studies have stimulated greater interest in the concept of a seismic erosion rate referring to the rate of coseismic landslide erosion averaged through time (G. Li et al., 2017). Earthquakes also contribute to weathering via physical break- down of rock masses into mobile regolith (Clarke & Burbank, 2010, 2011) and comminuting detached rock materials during coseismic landslide runout. Quantifying the contribution of these processes is open to research, though this research is challenged by accurately measuring soil and regolith depths in steep topography and also by developing a theory of coseismic regolith generation. Landslide volumes provide a ﬁrst‐order estimate of erosion rates for individual earthquakes (Keefer, 1994; Malamud et al., 2004; Marc, Hovius, & Meunier, 2016; Parker et al., 2011). Directly measuring the volume of thousands of landslides generated during large earthquakes is intractable, and most volumetric estimates rely on area‐volume scaling (Larsen et al., 2010). Long‐term landslide rates can be estimated by combining the return periods of earthquakes and associated landslide volumes (Keefer, 1994; Malamud et al., 2004), but generally, these relationships have order‐of‐magnitude uncertainties because much of the variability of coseismic landslide volumes arises from contributions of local steepness, earthquake magnitude, seismic moment, hydrology, and rock‐mass strength. This uncertainty highlights the need for models of landslide initiation that incorporate a wider range of factors and that can be scaled up to the orogen scale.

FAN ET AL. 51 Reviews of Geophysics 10.1029/2018RG000626

In quantifying seismic erosion rates based on landslide volumes, a key assumption is that landslide material is removed from mountain belts and that this removal does not influence crustal isostasy. Landslide‐derived sediment export in the decade following the Chi‐Chi earthquake reduced surface uplift by as much as 0.25 m on average, more than 30% of the mass added by coseismic uplift (Hovius et al., 2011). In the Longmenshan along the eastern margin of the Tibetan Plateau, landslide‐ driven seismic erosion rates projected for multiple earthquake cycles could be of similar magnitude to denudation rates derived from sediment yields, cosmogenic radionuclides, and thermochronology (Li et al., 2017). But is all the coseismic landslide mass removed before the next earthquake? In the Nepal Himalaya, large earthquakes have return periods of hundreds of years or shorter (Avouac, 2007), and not all coseismic landslide material might be removed before the next earthquake. In the Longmenshan return, periods of large earthquakes are longer, and Parker et al. (2011) argue for relatively little relict landslide debris predating the Wenchuan earthquake. Numerical simulations suggest rivers have the capacity to remove material relatively quickly compared to the return time of such large events (Croissant, Lague, Steer, & Davy, 2017). Yet earthquake‐induced landslides had 20–80% of their total volume contained in the 10 largest landslides (Marc, Hovius, Meunier, Gorum, & Uchida, 2016; see Figure 21 for illustration of importance of 10 largest landslides in erosional volume for the Wenchuan earthquake). These largest landslides commonly exceed 0.1 km3, and landslides of this size have persisted for thousands of years in mountain belts throughout the world (Korup et al., 2007). The rate at which sediment generated by coseismic landslides exits mountain ranges could thus depend largely on the erosion rate of large landslides. A compilation of the fate of large landslide deposits in mountains worldwide documents residence times of up to tens of millennia; some potentially EQTL deposits in mountain forelands are several million years old. Specific sediment yields from – – eroding large landslide bodies can exceed 105 t·km 2·year 1 in the first year after emplacement and remain – – elevated at >103 t·km 2·year 1 thousands of years later (Korup, 2012; Stolle et al., 2019). The persistence of these large deposits presents challenges in estimating long‐term seismic erosion rates from landslide volumes alone.

6.2. Earthquakes and Long‐Term Landscape Evolution Most studies agree that the total volume of coseismic landslides produced during a strong earthquake is substantial compared to available data on total erosional fluxes (Dadson et al., 2004; Hovius et al., 2011; Li et al., 2016). This observation has led to the question of whether earthquakes in compressional settings mainly generate topography via crustal shortening and thickening or whether they mainly destroy topography by erosion (Figure 21a). Some studies have balanced the relative magnitude of uplift and erosion over multiple earthquakes by using elastic half‐space models to simulate uplift (Figure 21b) and suggested that some earthquake magnitudes (M > 8) might be more important than others for net topographic growth of mountain belts (G. Li, West, et al., 2014; Marc, Hovius, Meunier, Gorum, & Uchida, 2016). But reconciling the combined effects of uplift, rock weakening, and erosion by earthquakes relative to aseismic uplift, chemical weathering, and fluvial incision might merit more consideration in the future. Thrusting and reverse faulting accommodate a given fraction of the shortening associated with compressional mountain belts. Yet estimates for the total amount of permanent vertical deformation caused by these earthquakes differ broadly (Avouac, 2007; Meade, 2010). Traditional models hold that the role of earthquakes in generating topography is minor because the associated movement acts in opposition to the accu- mulated interseismic strain (Simpson, 2015). For example, geodetic data indicate that the High Himalaya subsided by 0.6 m above the failure plane of the 2015 Gorkha earthquake (Elliott et al., 2016). However, permanent deformation associated with earthquakes is recorded through the uplift of geomorphic surfaces (Kelsey, 1990). This permanent deformation reflects a combination of inelastic strain that accumulates in the upper crust (Baker et al., 2013), redistribution of mass by gravitational and viscous deformation (King et al., 1988; Simpson, 2015), and mass redistribution by erosion and deposition (Densmore et al., 2012; King et al., 1988; Molnar, 2012), as well as other aseismic processes in the lithosphere such as crustal flow (Royden et al., 1997). The relative contributions of aseismic and coseismic processes to surface uplift remain controversial. For mountains fringing the Tibetan Plateau, plausible end‐member models argue for either an entirely aseismic origin (Royden et al., 1997) or an entirely coseismic origin (Hubbard & Shaw, 2009; Wang et al., 2012). What is the role of earthquakes in landform evolution if aseismic processes dominate rock

FAN ET AL. 52 Reviews of Geophysics 10.1029/2018RG000626

Figure 21. Summary of the concept of earthquake mass balance using the Wenchuan earthquake as an example. (a) Cartoon illustrating the mass balance between earthquake‐induced rock uplift and landslide‐facilitated fluvial export of sediment; (b) diagram showing the calculation of volume balance for compressional earthquakes in the context of an elastic half space model (modified after Marc, Hovius, Meunier, Gorum, & Uchida, 2016); (c–f) for the Wenchuan earthquake, swath profiles of coseismic displacement, landslide volume, and net volume changes calculated in 1‐km‐wide strips (difference between coseismic displacement and landslide volume, considering all landslide volume and landslide volume excluding 10 largest landslides), projected to directions perpendicular and parallel to the fault trace (coseismic displacement data from Fielding et al., 2013; landslide data from G. Li, West, et al., 2014); (g) map view of the Wenchuan earthquake, landslides, fault trace, and the swath profile in the context of the Tibetan Plateau and the Longmenshan Mountains. Yellow dots are coseismic landslides.

uplift? This open question could be important to understand the geodynamic consequences of seismically triggered landsliding. The question of whether and to what extent EQTLs contribute to the mass or volume balance of seismic deformation remains open. Even if seismic erosion does not balance seismic uplift, landslides can have

FAN ET AL. 53 Reviews of Geophysics 10.1029/2018RG000626

Table 6 Summary of the Topographic Legacies of Earthquake‐Triggered Landslides Possible effects of earthquake‐triggered landslides on large‐scale topography • Driving erosional losses from seismically active mountains; for example, determining the “earthquake mass or volume balance” (Figure 21; Hovius et al., 2011; G. Li, West, et al., 2014; Marc, Hovius, Meunier, Gorum, & Uchida, 2016; Parker et al., 2011) • Focusing denudation along seismically active mountain fronts (Li et al., 2017) reorganizing drainage basins (Dahlquist et al., 2018; Hovius et al., 1998) Potential morphological ﬁngerprints of landslides and some example studies • Planar hillslopes associated with earthquake‐induced landslides in landscapes with fewer inner gorges due to preferential initiation near ridges (Densmore & Hovius, 2000) • Clusters of giant landslides, ≫106 m3 volume (Crozier et al., 1995), with mechanistic back analysis to distinguish seismic triggering (Jibson, 2009; Jibson & Keefer, 1993) • Relicts of landslide‐dammed lakes (Korup & Wang, 2015) • Bowl‐shaped source areas (Turnbull & Davies, 2006) • Alluvial deposits from landslide‐driven aggradation (Schwanghart et al., 2016) • Parallel dune ridges along coastlines (Goff et al., 2008)

important effects on the long‐term evolution of orogenic systems. Landslide‐driven seismic erosion can contribute to focusing denudation near active faults (Li et al., 2017). The resulting unloading could in turn affect seismicity (Steer et al., 2014) and generate possible feedbacks between earthquakes and erosion. EQTLs can also change hillslope morphology, as is evident from pre‐ and post‐seismic elevation models. For example, Ren et al. (2017) reported a decrease in mean hillslope angle of the Longmenshan after the Wenchuan earthquake. Landslides also can affect topographic evolution via drainage‐basin reorganization. Very large landslides can establish the spatial pattern of fluvial incision in uplifting landscapes such as Papua New Guinea (Hovius et al., 1998). At smaller scales, Dahlquist et al. (2018) found that a subset of landslides triggered during the Wenchuan and Gorkha earthquakes, as well as during Typhoon Morakot in Taiwan, caused divide migration by cutting off ridges. By changing the distribution of fluvial erosive power, these migrations could generate feedbacks between seismicity and landsliding. Thus, including the spatial distribution of landslides might be important for developing landscape‐ evolution models. In summary, earthquakes are important for the erosion of steep mountains, yet we do not yet have a systematic understanding of how they contribute to the overall erosion budget. Currently coseismic uplift and coseismic erosion are treated independently of each other and without the context of interseismic erosional processes; this separation limits our ability to assess the role of large earthquakes in the evolution of mountain belts. Coupling models of landsliding, hillslope erosion, and fault dynamics could show a route toward filling this gap.

6.3. Topographic Legacy of EQTLs How do the geomorphic impacts of EQTLs play out over decades, centuries, and millennia? Which evidence remains in the landscape that could help diagnose the timing and intensity of past earthquakes? Few studies cover the topographic fingerprints of EQTLs beyond time spans of several years or decades, though the studies that do indicate that human memory is short in this regard. For example, much of the extensive deposits dumped by the catastrophic rock/ice avalanche and debris flow from Nevados Huascaran, Peru, mobilized during a strong earthquake in 1970 (Evans et al., 2009) now sustain new settlements and agricultural fields. In deciphering the longer‐term relevance of EQTLs, we seek to identify clear geomorphic legacies (Table 6). Despite detailed insights on the patterns and characteristics of EQTLs, we still face substantial shortcomings when it comes to inverting this knowledge and determining a seismic trigger from given landslide evidence (Jibson, 1996, 2009). Within steep‐relief mountains, EQTLs that form close to ridge crests might be distin- guishable from rainfall‐triggered landslides that tend to initiate in areas of topographic convergence close to valley bottoms (Densmore & Hovius, 2000; Hovius et al., 2011; Meunier et al., 2008). In these cases, landscapes dominated by EQTLs might have more planar hillslopes with fewer inner gorges (Densmore & Hovius, 2000), although these features have been challenging to identify in the field (Clarke & Burbank, 2010). Researchers have speculated that single, or especially clustered, deposits of giant (≫106 m3) landslides

FAN ET AL. 54 Reviews of Geophysics 10.1029/2018RG000626

are possible and potentially also long‐lived proxies of past strong ground shaking (Crozier et al., 1995). However, without a discernible statistical difference between rainfall and earthquake‐induced landslide distributions (Malamud et al., 2004), separating the effects of extreme earthquakes from those of extreme storms will be difficult. For a given region and denudation rate, the largest landslide deposits should be those most prone to remain for the longest (Korup & Clague, 2009), and some of the largest still recognizable landslide deposits have survived in semiarid to arid landscapes for several millions of years, such as in the active subduction zone of northern Chile (Mather et al., 2014). Giant landslides lower catchment divides, drive drainage capture, perturb or dam rivers, create long‐lived channel knickpoints, or even cause drainage reversals; the largest reported rock‐slope failures have buried hundreds of square kilometers below thick hummocky debris sheets (Shoaei, 2014). Even where landslide‐ dammed lakes do not survive, their former reservoirs commonly retain fringes of lake sediment and delta deposits. Clusters of giant landslide deposits are not limited to high topographic relief but can also affect indistinct, low‐gradient, semidesert landscapes such as the Caspian Sea basin, possibly offering evidence of rare but strong intraplate earthquakes (Pánek et al., 2016). The spatial clustering of landslide‐dammed lakes, even if equal in age, is an ambiguous proxy for inferring isoseismals of earthquake intensity because strong rainstorms can also form numerous lakes more or less coevally (Korup & Wang, 2015). Yet even a clear geomorphic footprint of large‐scale slope instability might be difficult to interpret. Turnbull and Davies (2006) proposed that the distinctly bowl‐shaped source areas of many deep, earthquake‐triggered rock slides closely resemble glacial cirques. The implication is that seismic imprints on rock slopes in glacier- ized terrain might be clearly visible and persistent but difficult to distinguish from true glacial cirques. Mechanistic back analyses using force or moment limit equilibria can indicate to first order whether aseismic causes and triggers alone were sufficient for mobilizing a given large landslide, thus excluding a possible earthquake origin (Jibson, 2009; Jibson & Keefer, 1993). However, constraining the necessary input parameters generally overlooks the previous history of disturbances for the hillslope in question (Parker et al., 2015) and how close it was to failure prior to the triggering earthquake. Systematic work on large subaqu- eous slope failures, including turbidites, mass‐transport complexes, and debrites in seismically active regions, relies heavily on the interpretation of earthquake triggers, with great potential for documenting longer‐term geomorphic impact especially when combined with independent terrestrial records (Beck, 2009; Leithold et al., 2018; Van Daele et al., 2013). 6.4. Landslides, the Carbon Cycle, and Possible Climate‐Seismicity Links Widespread landsliding can alter biogeochemical fluxes with regional to global consequences. Most notably, landsliding can modify fluxes of carbon associated both with chemical weathering and organic material eroded from hillslopes (Emberson et al., 2016, 2017; Frith et al., 2018; Hilton, Galy, et al., 2011; Hilton , Meunier, et al., 2011). Although the modifications of these fluxes are small relative to the annual fluxes associated with photosynthesis, respiration, and fossil‐fuel consumption, when integrated over 105–106 years of geologic time, even small perturbations to rates of weathering and organic carbon burial or oxidation can

have dramatic effects on the concentration of CO2 in the atmosphere (e.g., Caldeira et al., 1999). Evaluating these possible links requires quantifying the effects of individual events, and this quantiﬁcation

is critical to establishing whether EQTLs act as a source or sink of CO2 from the atmosphere. Growing data sets, though mostly associated with storms, have documented the effect of landslides in eroding standing vegetation and delivering organic material to river systems where it can be readily eroded, primarily as particulate organic carbon (POC; Clark et al., 2015; Hilton, Galy, et al., 2011; Hilton, Meunier, et al., 2011; Ramos Scharrón et al., 2012). Rivers draining the Longmenshan also had POC ﬂuxes twice as high following the Wenchuan earthquake than before (J. Wang, Jin, et al., 2016). Estimating the carbon‐ cycle implications of these changes hinges on knowing both the source of eroded carbon and its fate (Figure 22). Organic carbon in landslide material can be derived both from carbon in bedrock (petrogenic

carbon [POCpetro]), which has been sequestered from the atmosphere for millions of years, and carbon derived from recently photosynthesized biomass (biospheric carbon [POCbio]), which comprises soil and vegetal carbon that resided on hillslope soils and can range in age from years to thousands of years (Blair et al., 2004; Hilton et al., 2008; Hilton, Galy, et al., 2011; Hilton, Meunier, et al., 2011). Once mobilized during

landsliding, both POCpetro and POCbio can be retained in landslide deposits, removed by rivers and buried in downstream sediment, or oxidized during ﬂuvial transport, releasing the carbon back to the atmosphere as

FAN ET AL. 55 Reviews of Geophysics 10.1029/2018RG000626

Figure 22. Summary of possible effects of earthquake‐triggered landslides (and landsliding in general) on the carbon cycle. Carbon‐cycle effects are related to modifying ﬂuxes of both organic carbon and inorganic carbon, the latter via chemical weathering reactions. Dashed arrows reﬂect connections that are not well known; for example, there is clear evidence that biospheric organic carbon is stripped from hillsides and eroded from landslide deposits, but its ultimate fate (burial vs. oxidation during transport) is not well understood; this leaves open questions about the direction in which landslides shift atmospheric CO2.

CO2. If POCbio is retained in sediment while new vegetation regrows on old landslide scars, it will serve as a CO2 sink. If POCbio is oxidized, it will have little effect on long‐term carbon fluxes. Preservation of POCpetro also will have little effect, though oxidation of POCpetro will act as a CO2 source by releasing old carbon to the atmosphere. The net effect of landsliding on the organic carbon cycle thus depends on several factors that are challenging to measure (Figure 22). Landslide deposits also can act as weathering reactors, reflected in enhanced solute fluxes from drainage waters (Emberson et al., 2016). But much of this solute load can be derived from sulfide oxidation and carbonate weathering (Emberson et al., 2017, 2018), at least where these operate. Silicate mineral weathering

produces alkalinity that consumes CO2 through precipitation of carbonates such as CaCO3 (Berner & Raiswell, 1983; Revelle & Suess, 1957). In contrast, sulfide oxidation can release CO2 to the atmosphere, with effects that can last for millions of years (Torres et al., 2014). Thus, the net effect of chemical weathering in EQTLs on the carbon cycle could depend on the balance of silicate versus sulfide dissolution, which, in turn, depends on lithology. Quantifying the carbon‐cycle imprint of earthquakes also needs to account for the hydrologic response to seismic shaking (Manga & Wang, 2007). After the Wenchuan earthquake, solute fluxes increased notably in the Min Jiang, the largest river draining the epicentral area; yet these increases did not match the spatial pattern of landsliding, suggesting a hydrologic control (Jin et al., 2016). Systematic flowpath changes could have equally important carbon‐cycle implications, especially if they sustain long‐term increases in solute fluxes. But these changes might be difficult to distinguish conclusively from those associated with landslide weathering. Intriguingly, the hydrologic role in weathering might relate to

FAN ET AL. 56 Reviews of Geophysics 10.1029/2018RG000626

landsliding in interesting ways that operate in more than one direction. The role of groundwater in determining site response to seismic shaking is recognized increasingly (e.g., Cox et al., 2012; Wang et al., 2001). Hydrologic conditions can be a key control on coseismic landsliding, but these conditions commonly remain unresolved. Relationships between hydrology, coseismic landsliding, and weathering merit further investigation. Altogether, earthquakes and associated landslides could play a meaningful role in linking tectonic activity and global climate, but understanding this role remains challenging given the multiple possible effects at work (Figure 22).

6.5. Research Challenges and Directions Work over the last decade has stimulated interest in the long‐term consequences of EQTLs, with several pro- vocative hypotheses about earthquake mass balance, the topographic fingerprint of landslides, and the carbon cycle and climate effects. Many open questions and major challenges remain, with several exciting directions for future work including: 1. Developing a more complete understanding of how EQTLs relate to long‐term erosion, both in terms of whether and how much landslide sediment is removed from mountain belts, and to what extent landsliding plays a role in regolith production on hillslopes. Scalable models for these processes will be important to answering questions about the long‐term geomorphic role of EQTLs. 2. Incorporating landsliding and episodic erosion in geodynamic and landscape‐evolution models, so that the concept of earthquake volume balance can be placed into the larger context of seismic and interseismic deformation. Such an effort might consider how landslides affect spatial distribution of erosion, including by drainage reorganization. 3. Evaluating whether distinct characteristics of EQTLs can be conclusively distinguished from rainfall‐ triggered landslides in a way that could be applied confidently for paleo‐analysis. 4. Quantifying the effects of landsliding on organic carbon and chemical weathering fluxes in diverse settings and developing models that can provide general insight into the role of earthquakes on the global carbon cycle.

7. Summary and Perspectives Widespread landsliding in the epicentral areas of large earthquakes is now widely recognized as a pivotal geological process, one that raises basic questions about the production of topography by coseismic uplift, the destruction of topography by coseismic erosion, and the routing of excess material from mountains to sedimentary sinks. Steep, active mountain ranges, such as those in New Zealand, Japan, Taiwan, China, and California, have invited many studies of landsliding and sediment ﬂushing driven by earthquakes. A few destructive earthquakes of the past two decades have allowed more systematic work on the mechanisms, patterns, and rates of earthquake‐induced landsliding. Advances in technology now allow partly or fully automated mapping of tens to thousands of landslides in short periods of time. Such inventories offer a knowledge base for studying how hillslopes and rivers respond to a given seismic‐shaking and ground‐ deformation pattern, including effects such as spreading and attenuation of seismic waves away from an earthquake source and geologic or topographic site effects. These inventories also have reinforced research at the interface between seismology, geomorphology, and natural hazards. With now more than 60 inventories available, more general models describing the expected landsliding for a given earthquake are possible, aided by the physically based interpretation and modeling.

The growing data sets provide the foundation for emerging understanding of EQTLs. Controls on the rate at which earthquake‐generated material is produced include how earthquake energy translates into local ground acceleration, as well as effects of soil and rock‐mass strength, vegetation disturbance, and hydrology. These and other controls are likely responsible for the broad spectrum of rates with which mountain landscapes respond to comparable seismic disturbance. Long‐term observation of mountain areas affected by earthquakes shows that rates of landsliding remain high for months to years after an earthquake and that landslide debris moves through a cascade from hillslopes to river channels and into sedimentary sinks. This chain of potentially adverse events has been known to geomorphologists for some time, but it still awaits a better integration into natural hazard and risk appraisals, which would be further enabled by

FAN ET AL. 57 Reviews of Geophysics 10.1029/2018RG000626

more quantitative understanding. DEMs comparing pre‐ to post‐seismic land surfaces hold promise to quantify the processes and rates in the earthquake hazard cascade with accuracy not previously possible. Our review of a large body of literature identifies several research challenges: 1. Any useful prediction of earthquake‐triggered landsliding requires, apart from more data on case studies, details of the earthquake source mechanism and the propagation of seismic waves and data on the topographic, hydrologic, and biologic controls on slope stability. In this context, studying the cumulative effects of earthquakes in sequence can be instructive. 2. The full cycle of earthquake perturbation and relaxation needs attention and also a definition of the reference state of the landscape. Establishing long‐running observatories where all major landscape parameters that can be affected by an earthquake are measured systematically and repeatedly might be a viable option, though this requires a standardized strategy for optimal investment. 3. Rapid and coordinated scientific response after an earthquake is essential to separate coseismic and post‐ seismic dynamics and their impacts across the entire earthquake‐affected area. The deployment of strategic equipment from comprehensive surveys is desirable including repeated, regional acquisition of high‐resolution topographic data and collection of seismic data to track sediment in the landscape. 4. The removal of coseismic debris remains insufficiently documented, and sediment residence times differ between individual earthquakes. Systematic documentation of post‐seismic landslides and fluvial sediment transport is important to understand sediment cascades that highlight how sedimentary signals move into settled lowlands and geologic records. The removal of debris is grain‐size‐specific, so that ideally future models can anticipate mass fluxes from the granulometry produced by EQTLs in a given lithology. 5. Documented transient increases of landslide rates after earthquakes still need mechanistic explanations, considering, for example, the damage in surface near‐surface hillslope materials caused by seismic shaking. 6. The controls on the occurrence of very large landslides remain poorly understood. Because they can be exceptionally damaging, we need criteria by which we can predict where such landslides are most likely to occur. We need criteria for distinguishing between very large landslides caused by earthquakes and nonseismic slides. This would facilitate the use of large landslide deposits for aiding paleoseismic records where other methods cannot be applied. 7. Landscape response to earthquakes is so far considered mainly in terms of dry mass wasting, rarely considering antecedent weather conditions, groundwater hydrology, vegetation, or weathering. To constrain the importance and operation of these factors, systematic observations are required, in parallel with data on seismic shaking, mass wasting, and sediment transport. 8. The notion of an earthquake hazard cascade with protracted sedimentary response and associated hazards must be communicated to the professional communities dealing with mitigation, resilience, and disaster management. Together with these communities, we must work toward the prediction of dynamic disaster scenarios. Scientifically, the community might wish to aim at a comprehensive, generic understanding of earthquake‐ induced landsliding, which can be used to robustly determine its role in landscape dynamics through models and observations. Such a general understanding might enable fuller exploration of the impact of earthquakes on the all Earth systems. Beyond science, the prize is safer societies in seismically active mountain regions, with quantitative knowledge of the patterns and likelihood of hazardous processes and of the likely economic and personal risks due to earthquake‐induced landslide crises. Glossary Cascade of geohazards – See Hazard chain. Catchment – A hydrological zone denoted by forming lines separating the area between two adjacent drainage basins. Chain of geohazards – See Hazard chain. Continuum model – A family of numerical methods that are suitable for simulating landslide runouts. They simulate the landslide mass as the sum of one or more continuous domains and generally treat the movement as that of a fluid using equations from fluid mechanics to characterize it.

FAN ET AL. 58 Reviews of Geophysics 10.1029/2018RG000626

Controlling factor – A physical characteristic, property, or condition that exerts an influence over a particular process. For example, if this process is the initiation of a landslide, different factors (topographic, seismic, hydrologic, geologic, geotechnical, climatic, and anthropic) can promote or hinder landslide initiation. The angle of inclination of a slope, the permeability of a soil layer, or the ground shaking caused by an earthquake are all factors that can control landslide initiation. Coseismic landslide – A landslide that is a direct consequence of earthquake‐induced ground motion. Critical Shields stress – A constant nondimensional value at which it is assumed that sediment transport begins, which is also known as bed‐shear stress. It was proposed by Shields (1936). Dammed lake – See Landslide dam. DDA – Discontinuous deformation analysis is a type of discrete element method that accounts for the interaction of particles along discontinuities in fractured and jointed rock. Debris flow – A surging flow of debris (a mixture of soil and rock fragments) and water that can travel at high speed along steep slopes and within channels. Debris flows can increase in volume as they move by incorporating additional material from the slope or channel surface or channel sides. Discrete element method – A family of numerical methods that can be used to simulate landslides. In these methods, the soil or rock masses are decomposed into distinct elements (e.g., spheres and blocks) that mutually interact. Both 2‐D and 3‐D models are available. Denudation – All of the natural processes that result in the wearing away or progressive lowering of the Earth's surface, including weathering, erosion, mass wasting, and transportation (adapted from AGI, 2005). Digital elevation model (DEM) – A three‐dimensional representation of a part of the Earth's surface. It generally is composed of ground‐elevation values arranged in a raster grid having uniform spacing. Data from a digital elevation can be processed to create data layers and maps portraying topographic characteristics such as slope angle, slope curvature, and drainage divides. Early‐warning system – See Landslide early‐warning system. Element at risk – Elements at risk are the population, properties, economic activities, including public services, or any other defined values exposed to hazards in a given area. EQTL – Earthquake‐triggered landslide. See Coseismic landslide. Erosion – Natural processes by which Earth materials are loosened, dissolved, or worn away and transported from one place to another. Erosive processes include running water (including rainfall), waves and currents, moving ice, and wind (adapted from AGI, 2005). Escarpment ‐ A long cliff or a steep slope formed as an effect of faulting or erosion. Escarpment is also used interchangeably with scarp. Escarpment refers to the margin between two landforms, while scarp is synonymous with a cliff or steep slope. Exposed value – The accounting of assets or inhabitants of a region prone to natural disasters. Exposure – See Exposed value.

FAN ET AL. 59 Reviews of Geophysics 10.1029/2018RG000626

Fault – A discontinuity in a volume of rock, across which there has been significant displacement in opposite directions as a result of rock‐mass movement, mostly formed from the action of plate‐tectonic forces. An inclined fault has a foot wall (below the fault surface) and a hanging wall (above the fault surface). Fault rupture – Displacement along a fault induced by an earthquake. Fault rupture initiates at the focus of an earthquake and then radiates outward along the fault surface. FEM – Finite element method. A family of numerical methods that can be used to simulate some behaviors and processes in soil and rock, such as rainwater infiltration and groundwater seepage, contaminant transport, heat transfer, stress distribution, and deformations. In these methods, the soil or rock masses are represented as continuous idealizations that are divided into elements, linked explicitly only at their joints. Geohazard – A physical process or phenomenon, such as an earthquake or landslide, that can occur with or without warning or triggering and that has the potential to substantially affect the natural and (or) built environment. Geologic time scale – A system of chronological dating that relates geological strata (stratigraphy) to time. Geotechnical centrifuge – A centrifuge that uses physical and scaled prototype slope models to recreate in situ behavior within a laboratory scale. Geological hazard – See Geohazard. GIS – Geographic Information System. A computer software package designed to analyze, handle, store, present, or capture geographic or spatial data. Grain‐size distribution – Also known as particle‐size distribution. A list of mathematical values containing information of diameters of individual granular matter and their contributory percentage of mass compared with the total mass of all grains. Gully – V‐shaped channel that is an erosional landform created by running water on a hillside. Hazard – The probability that, in a specified area, a geologic process (such as a landslide or an earthquake) will occur over a specific time period or under the influence of a specific triggering mechanism. For example, a measure of seismic hazard is the probability of exceeding a particular level of ground motion in a year. Hazard chain – A sequence of hazards that have a cause‐and‐effect relationship with one another. An example would be an earthquake whose shaking triggers a landslide that then dams a river; this creates a lake that can overtop the landslide dam and cause downstream flooding. Hydraulic conductivity – Capacity of any solid or semisolid matter to transport fluid (e.g., water with dissolved minerals and suspended particles) through its pore spaces. Hydraulic head – The height above a datum plane (usually the groundwater table) supported by the hydraulic pressure at a given point in a groundwater system. Hydraulic gradient – The rate of change in the total hydraulic head per unit distance of flow in a given direction. Hydrological forcing – The yearly cycles of rainfall, snowfall, snowmelt, groundwater circulation, and surface water flows.

FAN ET AL. 60 Reviews of Geophysics 10.1029/2018RG000626

Intensity – The measured value of a physical parameter directly relevant to the potential for damage or loss of life from an eventual catastrophic natural phenomenon. For example, PGA is a measure of seismic intensity; cumulative rainfall is a measure of extreme weather intensity. Landslide – A general term referring to a wide variety of mass‐movement landforms and processes involving downslope transport, under gravitational influence, of soil and rock material. This general term thus includes slope movements that do not involve actual sliding, such as debris flows, rock falls, and rock slides (adapted from AGI, 2005; Cruden & Varnes, 1996; Varnes, 1984). Landslide dam – A landslide that blocks a stream or river drainage and impounds an upstream lake (Landslide‐dammed lake). Landslide dams can be transient or permanent and have widely differing physical characteristics such as size, shape, composition, and permeability. Landslide‐dammed lake – See Landslide dam. Landslide early‐warning system – A system having both technical and social components that is designed to issue warnings of impending damaging landslides before they occur. These components and the aims of the alert issued can differ as a function of the scale at which the system is employed. The technical component entails physical or empirical models to assess the probability of landslide occurrence over defined warning zones, typically through forecasting and monitoring of meteorological variables, in order to give generalized warnings to communities and institutions working with hazard‐mitigation measures (at regional scale) or to provide timely alarms for the temporary evacuation of people from areas where, at specific times, the risk level to which they are exposed can become intolerably high (at local scale; Segoni et al., 2017). Landslide inventory – A database that records the location and other characteristics of landslides that are identified through field investigation, remote sensing, historical records, or other types of research. Landslide inventories can be subdivided by the type of trigger that caused the landslides (e.g., an earthquake and a rainstorm). Mass wasting – A general term for the downslope transport of soil and rock material under the influence of gravity (AGI, 2005). Normal fault – A fault that is formed by downward movement of the hanging wall against the footwall. Paleoclimate – Past climate during the longer time scale of geological history studied as a specialized field (paleoclimatology). Paleoseismicity – Earthquakes recorded in the geologic record, most of them unknown in terms of human descriptions or seismograms. Geologic records of past earthquakes can include faulted layers of sediment and rock, injections of liquefied sand, landslides, abruptly raised or lowered shorelines, and tsunami deposits. PGA – Peak Ground Acceleration. The maximum ground acceleration observed in a strong‐motion record induced by an earthquake. PGD – Peak Ground Displacement. The maximum ground displacement observed in a strong‐motion record induced by an earthquake. PGV – Peak Ground Velocity. The maximum ground velocity observed in a strong‐motion record induced by an earthquake. Porosity – A measure of the relative volume of pores in a solid material. Post‐seismic landslide – A landslide that occurs after an earthquake, for which the earthquake acted as a predisposing factor but not as the direct trigger. Predisposing factor – See Controlling factor.

FAN ET AL. 61 Reviews of Geophysics 10.1029/2018RG000626

Rainfall threshold – The smallest rainfall event that can trigger a debris flow or other type of landslide in a given area at a given time. Ravine – A fluvial landform of relatively steep cross‐sectional sides and steep gradients. Regolith – The layer of fragmented and unconsolidated rock material, whether formed in place or deposited after being transported, that nearly everywhere forms the surface of the land and overlies or covers intact bedrock (adapted from AGI, 2005). Resilience – The capacity of a community to recover the living standard they enjoyed prior the occurrence of a natural disaster. Reverse fault – A fault structure resulting from the upward movement of the hanging wall against the footwall. Risk – The exposure of humans and their built environment to geologic hazard. For example, a landslide is a type of geologic hazard; the lives and property that could be damaged or destroyed by that landslide encompass the risk from that hazard. Risk assessment – A quantitative examination of the oncoming adverse effects of an anticipated (geological) hazard. Risk mitigation – A set of strategies and countermeasures aimed at reducing the risk for people, infrastructure, ecosystem, and so forth that derives from a hazard or a set of hazards. According to the strategy, this is achieved by reducing the hazard directly, and/or by reducing the exposed value, and/or its vulnerability. Runout – The distance a landslide moves. It can be measured along the movement path, in straight line, or in horizontal projection, from the landslide scarp to the farthermost point of the toe or as displacement of the center of mass. Together with the height drop of the landslide mass, it can be used to estimate the mobility. Shaking table – An experimental device used to study the response of model slopes and man‐made structures or building components to the seismic shaking or other types of vibration by subjecting them to base ground motion, usually induced by mechanical actuators. Shields stress – A nondimensional number used to calculate the initiation of motion of sediment in a fluid flow. It is a nondimensionalization of shear stress. Site effect – The amplification or de‐amplification of seismic motion (acceleration, velocity, and displacements) at specific frequencies and in specific directions that results from the interaction between the seismic waves, which reach the site after a path‐specific attenuation, and the local site conditions (material, layering, topography, etc.). Soil grading – A classification of soils based on the relative content of different particle sizes. See also Grain‐size distribution. Strike‐slip fault – A fault structure resulting from horizontal movement in opposing directions across a vertical or steeply dipping fault. Susceptibility – The temporally stable properties of a natural or built system that render it more or less likely to experience a given hazard in a specific triggering scenario. For example, landslide susceptibility is a function of slope properties such as topography (slope angle, etc.), material strength, and geologic discontinuities that make a slope more or less susceptible to failure during an earthquake or heavy rainfall. Trigger – A process or event that initiates another process or event. For instance, rainfall or earthquakes can trigger landslides.

FAN ET AL. 62 Reviews of Geophysics 10.1029/2018RG000626

UAV ‐ Unmanned Aerial Vehicles. A small airplane or helicopter (quadcopter, hexacopter, etc.) operated without an onboard pilot and remotely controlled from the ground. Smaller UAVs are also termed drones. Vulnerability – The probability of observing a particular damage state or the probability of observing a death if a given intensity measure is exceeded. One example is the probability of observing collapse of a building if a PGA larger than 5 m/s2. Watershed – see Catchment.

Author Contributions X. F. and R. H. were responsible for coordinating with all coauthors and generating most of the ﬁgures, with input from all authors. O. K. and G. S. contributed to improving and ﬁnalizing of the paper. N. H. contributed to sections 1 and 7. C. J. W., H. T., G. S., R. W. J., K. E. A., and C. X. contributed to section 2. T. C. H., G. S., and X. F. contributed to section 3. C. J. W., L. Z., S. G. E., and X. F. contributed to section 4. A. J. W., T. C. H., O. K., and L. G. contributed sections 5 and 6. X. P., Q. X., and R. H. provided helpful discussions and funding for the work. R. W. J. and K. E. A. polished the language. All authors reviewed the paper and collected relevant data and literature.

Acknowledgments The authors wish to thank Yunus Ali, References Srikrishnan Siva Subramanian, and Abrahamson, N. A., & Somerville, P. G. (1996). Effects of the hanging wall and footwall on ground motions recorded during the Northridge Lanxin Dai for their help in processing earthquake. Bulletin of the Seismological Society of America, 86(1B), S93–S99. some data and figures. The authors are Abrams, M. J., & Hook, S. J. (2013). NASA's Hyperspectral Infrared Imager (HyspIRI). In C. Kuenzer, & S. Dech (Eds.), Thermal infrared grateful to the Editor‐in‐chief (Fabio remote sensing: Sensors, methods, applications, (pp. 117–130). Dordrecht: Springer Netherlands. https://doi.org/10.1007/978‐94‐007‐ Florindo), the Associate Editor (Kate 6639‐6_6 Huihsuan Chen), the Editorial Board of AGI (2005). In K. K. E. Neuendorf, J. P. Mehl, Jr., & J. A. Jackson (Eds.), American Geological Institute, Glossary of Geology, (5th ed.). Reviews of Geophysics, the two Alexandria, Virginia, USA: Springer Science & Business Media. reviewers (Evgeny Podolskiy and Akiyama, K., Kurihara, J., & Ogawa, K. (2006). Characteristics of slope failure induced by the Mid Niigata prefecture earthquake and the Wei‐An Chao), and Alex Grant for snow melting after the snow term. In Disaster Mitigation of Debris Flows, Slope Failures and Landslides, (pp. 749–758). Tokyo: Universal providing constructive comments Academy Press Inc. during the preparation and revision of Alcántara‐Ayala, I., Esteban‐Chávez, O., & Parrot, J. F. (2006). Landsliding related to land‐cover change: A diachronic analysis of hillslope the manuscript. Part of this research instability distribution in the Sierra Norte, Puebla, Mexico. CATENA, 65(2), 152–165. https://doi.org/10.1016/j.catena.2005.11.006 has been funded by the Funds for Aleotti, P. (2004). A warning system for rainfall‐induced shallow failures. Engineering Geology, 73(3–4), 247–265. https://doi.org/10.1016/j. Creative Research Groups of China enggeo.2004.01.007 (Grant No. 41521002), the National Allstadt, K. E., Jibson, R. W., Thompson, E. M., Massey, C. I., Wald, D. J., Godt, J. W., & Rengers, F. K. (2018). Improving near‐real‐time Science Fund for Outstanding Young coseismic landslide models: Lessons learned from the 2016 Kaikōura, New Zealand, earthquake. Bulletin of the Seismological Society of Scholars of China (Grant No. America, 108(3B), 1649–1664. https://doi.org/10.1785/0120170297 41622206), and the International Allstadt, K. E., Thompson, E. M., Hearne, M., Novicki Jessee, M. A., Zhu, J., Wald, D. J., & Tanyaş, H. (2017). Integrating landslide and Cooperation Grant (NSFC‐ liquefaction hazard and loss estimates with existing USGS real‐time earthquake information products. In 16th World Conference on RCUK_NERC) Resilience to Earthquake Engineering. U.S. Geological Survey. Earthquake‐induced landslide risk in Almeida, S., Holcombe, E. A., Pianosi, F., & Wagener, T. (2017). Dealing with deep uncertainties in landslide modeling for disaster risk China (41661134010). Any use of trade, reduction under climate change. Natural Hazards and Earth System Sciences, 17(2), 225–241. https://doi.org/10.5194/nhess‐17‐225‐2017 firm, or product names is for descriptive Amante, C., & Eakins, B. W. (2009). ETOPO1 Global Relief Model converted to PanMap layer format. NOAA‐National Geophysical Data purposes only and does not imply Center, PANGAEA. https://doi.org/10.1594/PANGAEA.769615 endorsement by the U.S. Government. Ambraseys, N. (2009). Earthquakes in the Mediterranean and Middle East: A multidisciplinary study of seismicity up to 1900. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139195430 Anders, N. S., Seijmonsbergen, A. C., & Bouten, W. (2011). Segmentation optimization and stratified object‐based analysis for semi‐ automated geomorphological mapping. Remote Sensing of Environment, 115(12), 2976–2985. https://doi.org/10.1016/j.rse.2011.05.007 Arino, O., Bicheron, P., Achard, F., Latham, J., Witt, R., & Weber, J.‐L. (2008). The most detailed portrait of Earth. European Space Agency, 136,25–31. Ashford, S. A., & Sitar, N. (2002). Simplified method for evaluating seismic stability of steep slopes. Journal of Geotechnical and Geoenvironmental Engineering, 128(2), 119–128. https://doi.org/10.1061/(ASCE)1090‐0241(2002)128:2(119) Attal, M., & Lavé, J. (2006). Changes of bedload characteristics along the Marsyandi River (central Nepal): Implications for understanding hillslope sediment supply, sediment load evolution along fluvial networks, and denudation in active orogenic belts. In Special Paper 398: Tectonics, Climate, and Landscape Evolution, (Vol. 398, pp. 143–171). Boulder, CO: Geological Society of America. https://doi.org/ 10.1130/2006.2398(09) Attal, M., Mudd, S. M., Hurst, M. D., Weinman, B., Yoo, K., & Naylor, M. (2015). Impact of change in erosion rate and landscape steepness on hillslope and fluvial sediments grain size in the Feather River basin (Sierra Nevada, California). Earth Surface Dynamics, 3(1), 201–222. https://doi.org/10.5194/esurf‐3‐201‐2015 Avouac, J.‐P. (2007). Dynamic processes in extensional and compressional settings—Mountain building: From earthquakes to geological deformation. In Treatise on geophysics, (pp. 377–439). Amsterdam: Elsevier. https://doi.org/10.1016/B978‐044452748‐6.00112‐7 Avouris, D. M., Carn, S. A., & Waite, G. P. (2017). Triggering of volcanic degassing by large earthquakes. Geology, 45(8), 715–718.

FAN ET AL. 63 Reviews of Geophysics 10.1029/2018RG000626

Avşar, U., Hubert‐Ferrari, A., Batist, M. D., & Fagel, N. (2014). A 3400 year lacustrine paleoseismic record from the North Anatolian Fault, Turkey: Implications for bimodal recurrence behavior. Geophysical Research Letters, 41, 377–384. https://doi.org/10.1002/2013GL058221 Avşar, U., Jónsson, S., Avşar, Ö., & Schmidt, S. (2016). Earthquake‐induced soft‐sediment deformations and seismically amplified erosion rates recorded in varved sediments of Köyceğiz Lake (SW Turkey): Earthquake records in lacustrine varves. Journal of Geophysical Research: Solid Earth, 121, 4767–4779. https://doi.org/10.1002/2016JB012820 Ayonghe, S., Mafany, G., Ntasin, E., & Samalang, P. (1999). Seismically activated swarm of landslides, tension cracks, and a rockfall after heavy rainfall in Bafaka, Cameroon. Natural Hazards, 19(1), 13–27. https://doi.org/10.1023/A:1008041205256 Baker, A., Allmendinger, R. W., Owen, L. A., & Rech, J. A. (2013). Permanent deformation caused by subduction earthquakes in northern Chile. Nature Geoscience, 6(6), 492–496. https://doi.org/10.1038/ngeo1789 Basharat, M., Ali, A., Jadoon, I. A. K., & Rohn, J. (2016). Using PCA in evaluating event‐controlling attributes of landsliding in the 2005 Kashmir earthquake region, NW Himalayas, Pakistan. Natural Hazards, 81(3), 1999–2017. https://doi.org/10.1007/s11069‐016‐2172‐9 Basher, L., Harrison, D., Phillips, C., & Marden, M. (2015). What do we need for a risk management approach to steepland plantation forests in erodible terrain. New Zealand Journal of Forestry, 60(2), 7–10. Beck, C. (2009). Late Quaternary lacustrine paleo‐seismic archives in north‐western Alps: Examples of earthquake‐origin assessment of sedimentary disturbances. Earth‐Science Reviews, 96(4), 327–344. https://doi.org/10.1016/j.earscirev.2009.07.005 Beguería, S., Van Asch, T. W. J., Malet, J.‐P., & Gröndahl, S. (2009). A GIS‐based numerical model for simulating the kinematics of mud and debris flows over complex terrain. Natural Hazards and Earth System Science, 9(6), 1897–1909. https://doi.org/10.5194/nhess‐9‐ 1897‐2009 Bell, R., & Glade, T. (2012). Multi‐hazard analysis in natural risk assessments. Landslides. Safety & Security Engineering Series, 1–10. Benda, L., & Dunne, T. (1997). Stochastic forcing of sediment routing and storage in channel networks. Water Resources Research, 33(12), 2865–2880. https://doi.org/10.1029/97WR02387 Bennett, D. P., Cuss, R. J., Vardon, P. J., Harrington, J. F., Philp, R. N., & Thomas, H. R. (2012). Data analysis toolkit for long‐term, large‐ scale experiments. Mineralogical Magazine, 76(8), 3355–3364. https://doi.org/10.1180/minmag.2012.076.8.48 Berner, R. A., & Raiswell, R. (1983). Burial of organic carbon and pyrite sulfur in sediments over phanerozoic time: A new theory. Geochimica et Cosmochimica Acta, 47(5), 855–862. https://doi.org/10.1016/0016‐7037(83)90151‐5 Beven, K., & Germann, P. (2013). Macropores and water flow in soils revisited: REVIEW. Water Resources Research, 49, 3071–3092. https:// doi.org/10.1002/wrcr.20156 Beven, K. J., & Kirkby, M. J. (1979). A physically based, variable contributing area model of basin hydrology/Un modèle à base physique de zone d'appel variable de l'hydrologie du bassin versant. Hydrological Sciences Journal, 24(1), 43–69. https://doi.org/10.1080/ 02626667909491834 Bhattacharya, S., Hyodo, M., Goda, K., Tazoh, T., & Taylor, C. A. (2011). Liquefaction of soil in the Tokyo Bay area from the 2011 Tohoku (Japan) earthquake. Soil Dynamics and Earthquake Engineering, 31(11), 1618–1628. https://doi.org/10.1016/j. soildyn.2011.06.006 Bianchini, S., Pratesi, F., Nolesini, T., & Casagli, N. (2015). Building deformation assessment by means of persistent scatterer interferometry analysis on a landslide‐affected area: The Volterra (Italy) case study. Remote Sensing, 7(4), 4678–4701. https://doi.org/10.3390/ rs70404678 Biondi, G., Condorelli, A., Maugeri, M., & Mussumeci, G. (2004). Earthquake‐triggered landslide hazards in the Catania area. In WIT Transactions on Ecology and the Environment (Vol. 77, pp. 115–130). Ashurst, Southampton: WIT Press. Birkmann, J. (2006). Measuring vulnerability to promote disaster‐resilient societies: Conceptual frameworks and definitions. Measuring Vulnerability to Natural Hazards: Towards Disaster Resilient Societies, 1,9–54. Blair, N. E., Leithold, E. L., & Aller, R. C. (2004). From bedrock to burial: The evolution of particulate organic carbon across coupled watershed‐continental margin systems. Marine Chemistry, 92(1–4), 141–156. https://doi.org/10.1016/j.marchem.2004.06.023 Blake, T. F., Hollingsworth, R. A., Stewart, J. P., D'Antonio, R., Earnest, J., Gharib, F., et al. (2002). Recommended procedures for imple- mentation of DMG special publication 117 guidelines for analyzing and mitigating landslide hazards in California. Los Angeles Section Geotechnical Group, Document Published by the Southern California Earthquake Center. Bogaard, T., & Greco, R. (2018). Invited perspectives: Hydrological perspectives on precipitation intensity‐duration thresholds for landslide initiation: Proposing hydro‐meteorological thresholds. Natural Hazards and Earth System Sciences, 18(1), 31–39. https://doi.org/ 10.5194/nhess‐18‐31‐2018 Böhner, J., & Selige, T. (2006). Spatial prediction of soil attributes using terrain analysis and climate regionalisation. Gottinger Geographische Abhandlungen, 115,13–28. Bonamassa, O., & Vidale, J. E. (1991). Directional site resonances observed from aftershocks of the 18 October 1989 Loma Prieta earthquake. Bulletin of the Seismological Society of America, 81(5), 1945–1957. Bonelli, S. (2013). Erosion in geomechanics applied to dams and levees. Hoboken, USA: John Wiley & Sons. https://doi.org/10.1002/ 9781118577165 Bonelli, S. (2018). Internal erosion in earthdams, dikes and levees. New York, NY: Springer Berlin Heidelberg. Boore, D. M., & Atkinson, G. M. (2008). Ground‐motion prediction equations for the average horizontal component of PGA, PGV, and 5%‐damped PSA at spectral periods between 0.01 s and 10.0 s. Earthquake Spectra, 24(1), 99–138. https://doi.org/10.1193/ 1.2830434 Boulanger, R. K., & Ziotopoulou, K. (2012). PM4Sand (Version 2): A sand plasticity model for earthquake engineering applications”‐ Department of Civil & Environmental Engineering, UC Davis, Center for Geotechnical Modeling, Report No. UCD/CGM‐12‐01. Bourdeau, C., & Havenith, H.‐B. (2008). Site effects modelling applied to the slope affected by the Suusamyr earthquake (Kyrgyzstan, 1992). Engineering Geology, 97(3–4), 126–145. https://doi.org/10.1016/j.enggeo.2007.12.009 Bout, B., Lombardo, L., van Westen, C. J., & Jetten, V. G. (2018). Integration of two‐phase solid fluid equations in a catchment model for flashfloods, debris flows and shallow slope failures. Environmental Modelling & Software, 105,1–16. https://doi.org/10.1016/j. envsoft.2018.03.017 Bozzano, F., Lenti, L., Martino, S., Paciello, A., & Scarascia‐Mugnozza, G. (2008). Self‐excitation process due to local seismic amplification responsible for the reactivation of the Salcito landslide (Italy) on 31 October 2002. Journal of Geophysical Research, 113, B10312. https:// doi.org/10.1029/2007JB005309 Braun, A., Cuomo, S., Petrosino, S., Wang, X., & Zhang, L. (2018). Numerical SPH analysis of debris flow run‐out and related river damming scenarios for a local case study in SW China. Landslides, 15(3), 535–550. https://doi.org/10.1007/s10346‐017‐0885‐9 Bray, J. D., & Travasarou, T. (2007). Simplified procedure for estimating earthquake‐induced deviatoric slope displacements. Journal of Geotechnical and Geoenvironmental Engineering, 133(4), 381–392. https://doi.org/10.1061/(ASCE)1090‐0241(2007)133:4(381)

FAN ET AL. 64 Reviews of Geophysics 10.1029/2018RG000626

Brebbia, C. A., & Wrobel, L. C. (1980). The boundary element method. In Computer Methods in Fluids.(A81‐28303 11‐34, (pp. 26–48). London: Pentech Press, Ltd. Brenguier, F., Campillo, M., Hadziioannou, C., Shapiro, N. M., Nadeau, R. M., & Larose, E. (2008). Postseismic relaxation along the San Andreas Fault at Parkfield from continuous seismological observations. Science, 321(5895), 1478–1481. https://doi.org/10.1126/ science.1160943 Brennan, A. J., & Madabhushi, S. P. G. (2009). Amplification of seismic accelerations at slope crests. Canadian Geotechnical Journal, 46(5), 585–594. https://doi.org/10.1139/T09‐006 Brooks, S., Crozier, M., Preston, N., & Anderson, M. (2002). Regolith stripping and the control of shallow translational hillslope failure: Application of a two‐dimensional coupled soil hydrology‐slope stability model, Hawke's Bay, New Zealand. Geomorphology, 45(3–4), 165–179. https://doi.org/10.1016/S0169‐555X(01)00153‐2 Brown, E. T., Stallard, R. F., Larsen, M. C., Raisbeck, G. M., & Yiou, F. (1995). Denudation rates determined from the accumulation of in situ‐produced 10Be in the Luquillo Experimental Forest, Puerto Rico. Earth and Planetary Science Letters, 129(1–4), 193–202. https://doi. org/10.1016/0012‐821X(94)00249‐X Burchsted, D., & Daniels, M. D. (2014). Classification of the alterations of beaver dams to headwater streams in northeastern Connecticut, USA. Geomorphology, 205,36–50. https://doi.org/10.1016/j.geomorph.2012.12.029 Burjánek, J., Gassner‐Stamm, G., Poggi, V., Moore, J. R., & Fäh, D. (2010). Ambient vibration analysis of an unstable mountain slope. Geophysical Journal International, 180(2), 820–828. https://doi.org/10.1111/j.1365‐246X.2009.04451.x Caldeira, K., Berner, R., Sundquist, E. T., Pearson, P. N., & Palmer, M. R. (1999). Seawater pH and atmospheric carbon dioxide. Science, 286(5447), 2043a–22043a. https://doi.org/10.1126/science.286.5447.2043a Cannon, S. H., Gartner, J. E., Wilson, R. C., Bowers, J. C., & Laber, J. L. (2008). Storm rainfall conditions for floods and debris flows from recently burned areas in southwestern Colorado and southern California. Geomorphology, 96(3–4), 250–269. https://doi.org/10.1016/j. geomorph.2007.03.019 Canuti, P., Casagli, N., & Catani, F. (1999). Successes and failures in fighting landslides: Some experiences from Italy and elsewhere. In Sub‐ Forum on Science and Technology in support of Natural Disaster Reduction (p. 89). Casagli, N., Cigna, F., Bianchini, S., Hölbling, D., Füreder, P., Righini, G., et al. (2016). Landslide mapping and monitoring by using radar and optical remote sensing: Examples from the EC‐FP7 project SAFER. Remote Sensing Applications: Society and Environment, 4, 92–108. https://doi.org/10.1016/j.rsase.2016.07.001 Casagli, N., & Ermini, L. (1999). Geomorphic analysis of landslide dams in the Northern Apennine. Transactions of the Japanese Geomorphological Union, 20(3), 219–249. Casagli, N., Ermini, L., & Rosati, G. (2003). Determining grain size distribution of the material composing landslide dams in the Northern Apennines: Sampling and processing methods. Engineering Geology, 69(1–2), 83–97. https://doi.org/10.1016/S0013‐ 7952(02)00249‐1 Cauzzi, C., Edwards, B., Fäh, D., Clinton, J., Wiemer, S., Kästli, P., et al. (2014). New predictive equations and site amplification estimates for the next‐generation Swiss ShakeMaps. Geophysical Journal International, 200(1), 421–438. https://doi.org/10.1093/gji/ ggu404 Cepeda, J., Schwendtner, B., Quan Luna, B., Diaz, M., & Molina, G. (2013). Landslide hazard and risk assessment in El Salvador. UNISDR global assessment Report 2013—GAR13. Chang, C.‐W., Lin, P.‐S., & Tsai, C.‐L. (2011). Estimation of sediment volume of debris flow caused by extreme rainfall in Taiwan. Engineering Geology, 123(1–2), 83–90. https://doi.org/10.1016/j.enggeo.2011.07.004 Chang, D. S., & Zhang, L. M. (2010). Simulation of the erosion process of landslide dams due to overtopping considering variations in soil erodibility along depth. Natural Hazards and Earth System Sciences, 10(4), 933–946. https://doi.org/10.5194/nhess‐10‐933‐2010 Chang, K.‐T., Hwang, J.‐T., Liu, J.‐K., Wang, E.‐H., & Wang, C.‐I. (2011). Apply two hybrid methods on the rainfall‐induced landslides interpretation (pp. 1–5). Presented at the 2011 19th International Conference on Geoinformatics, IEEE Chang, M., Tang, C., Van Asch, T. W., & Cai, F. (2017). Hazard assessment of debris flows in the Wenchuan earthquake‐stricken area, South West China. Landslides, 14(5), 1783–1792. https://doi.org/10.1007/s10346‐017‐0824‐9 Chao, W.‐A., Wu, Y.‐M., Zhao, L., Tsai, V. C., & Chen, C.‐H. (2015). Seismologically determined bedload flux during the typhoon season. Scientific Reports, 5(1), 8261. https://doi.org/10.1038/srep08261 Chen, C.‐Y. (2009). Sedimentary impacts from landslides in the Tachia River Basin, Taiwan. Geomorphology, 105(3–4), 355–365. https:// doi.org/10.1016/j.geomorph.2008.10.009 Chen, H., & Hawkins, A. B. (2009). Relationship between earthquake disturbance, tropical rainstorms and debris movement: An overview from Taiwan. Bulletin of Engineering Geology and the Environment, 68(2), 161–186. https://doi.org/10.1007/s10064‐009‐0209‐y Chen, H., & Petley, D. N. (2005). The impact of landslides and debris flows triggered by Typhoon Mindulle in Taiwan. Quarterly Journal of Engineering Geology and Hydrogeology, 38(3), 301–304. https://doi.org/10.1144/1470‐9236/04‐077 Chen, H. X., & Zhang, L. M. (2015). EDDA 1.0: Integrated simulation of debris flow erosion, deposition and property changes. Geoscientific Model Development, 8(3), 829–844. https://doi.org/10.5194/gmd‐8‐829‐2015 Chen, H. X., Zhang, L. M., & Zhang, S. (2014). Evolution of debris flow properties and physical interactions in debris‐flow mixtures in the Wenchuan earthquake zone. Engineering Geology, 182, 136–147. https://doi.org/10.1016/j.enggeo.2014.08.004 Chen, H. X., Zhang, S., Peng, M., & Zhang, L. M. (2016). A physically‐based multi‐hazard risk assessment platform for regional rainfall‐ induced slope failures and debris flows. Engineering Geology, 203,15–29. https://doi.org/10.1016/j.enggeo.2015.12.009 Chen, R.‐F., Chang, K.‐J., Angelier, J., Chan, Y.‐C., Deffontaines, B., Lee, C.‐T., & Lin, M.‐L. (2006). Topographical changes revealed by high‐resolution airborne LiDAR data: The 1999 Tsaoling landslide induced by the Chi–Chi earthquake. Engineering Geology, 88(3–4), 160–172. https://doi.org/10.1016/j.enggeo.2006.09.008 Chen, S.‐C., Lin, T.‐W., & Chen, C.‐Y. (2015). Modeling of natural dam failure modes and downstream riverbed morphological changes with different dam materials in a flume test. Engineering Geology, 188, 148–158. https://doi.org/10.1016/j.enggeo.2015.01.016 Chen, X. L., Zhou, Q., Ran, H., & Dong, R. (2012). Earthquake‐triggered landslides in southwest China. Natural Hazards and Earth System Sciences, 12(2), 351–363. https://doi.org/10.5194/nhess‐12‐351‐2012 Chen, Z., Lei, T., Yan, Q., Hu, H., Xiong, M., & Li, Z. (2013). Measuring and calculation methods for landslide volume with 3‐D laser scanner in Wenchuan earthquake area. Transactions of the Chinese Society of Agricultural Engineering, 29(8), 135–144. Cheng, G., Wang, X., He, X., Fan, J., & Fan, J. (2008). Outburst risk of barrier lakes in Sichuan, China. Journal of Mountain Science, 5(3), 189–193. https://doi.org/10.1007/s11629‐008‐0208‐7 Chernouss, P., Fedorenko, Y., Mokrov, E., & Barashev, N. (2006). Studies of seismic effects on snow stability on mountain slopes. Polar Meteorology and Glaciology, 20,62–73.

FAN ET AL. 65 Reviews of Geophysics 10.1029/2018RG000626

Chigira, M., Wang, W.‐N., Furuya, T., & Kamai, T. (2003). Geological causes and geomorphological precursors of the Tsaoling landslide triggered by the 1999 Chi‐Chi earthquake, Taiwan. Engineering Geology, 68(3–4), 259–273. https://doi.org/10.1016/S0013‐ 7952(02)00232‐6 Chigira, M., & Yagi, H. (2006). Geological and geomorphological characteristics of landslides triggered by the 2004 Mid Niigta prefecture earthquake in Japan. Engineering Geology, 82(4), 202–221. https://doi.org/10.1016/j.enggeo.2005.10.006 Chiou, B., Youngs, R., Abrahamson, N., & Addo, K. (2010). Ground‐motion attenuation model for small‐to‐moderate shallow crustal earthquakes in California and its implications on regionalization of ground‐motion prediction models. Earthquake Spectra, 26(4), 907–926. https://doi.org/10.1193/1.3479930 Clark, K. E., West, A. J., Hilton, R. G., Asner, G. P., Quesada, C. A., Silman, M. R., et al. (2015). Storm‐triggered landslides in the Peruvian Andes and implications for topography, carbon cycles, and biodiversity. Earth Surface Dynamics Discussions, 3(3), 631–688. https://doi. org/10.5194/esurfd‐3‐631‐2015 Clarke, B. A., & Burbank, D. W. (2010). Bedrock fracturing, threshold hillslopes, and limits to the magnitude of bedrock landslides. Earth and Planetary Science Letters, 297(3–4), 577–586. https://doi.org/10.1016/j.epsl.2010.07.011 Clarke, B. A., & Burbank, D. W. (2011). Quantifying bedrock‐fracture patterns within the shallow subsurface: Implications for rock mass strength, bedrock landslides, and erodibility. Journal of Geophysical Research, 116, F04009. https://doi.org/10.1029/ 2011JF001987 Clubb, F. J., Mudd, S. M., Milodowski, D. T., Hurst, M. D., & Slater, L. J. (2014). Objective extraction of channel heads from high‐resolution topographic data. Water Resources Research, 50, 4283–4304. https://doi.org/10.1002/2013WR015167 Corominas, J., van Westen, C., Frattini, P., Cascini, L., Malet, J.‐P., Fotopoulou, S., et al. (2013). Recommendations for the quantitative analysis of landslide risk. Bulletin of Engineering Geology and the Environment., 73(2), 209–263. https://doi.org/10.1007/s10064‐013‐ 0538‐8 Costa, J. E., & Schuster, R. L. (1988). The formation and failure of natural dams. Geological Society of America Bulletin, 100(7), 1054–1068. https://doi.org/10.1130/0016‐7606(1988)100<1054:TFAFON>2.3.CO;2 Costa, J. E., & Schuster, R. L. (1991). Documented historical landslide dams from around the world. U.S. Geological Survey. (Report No. 91– 239) (p. 494). Vancouver, WA. https://doi.org/10.3133/ofr91239 Cox, S. C., Rutter, H., Sims, A., Manga, M., Weir, J., Ezzy, T., et al. (2012). Hydrological effects of the M W 7.1 Darfield (Canterbury) earthquake, 4 September 2010, New Zealand. New Zealand Journal of Geology and Geophysics, 55(3), 231–247. https://doi.org/10.1080/ 00288306.2012.680474 Croissant, T., Lague, D., Davy, P., Davies, T. R. H., & Steer, P. (2017). A precipiton‐based approach to model hydro‐sedimentary hazards induced by large sediment supplies in alluvial fans: Morphodynamic modeling of alluvial fan hydro‐sedimentary hazards. Earth Surface Processes and Landforms, 42(13), 2054–2067. https://doi.org/10.1002/esp.4171 Croissant, T., Lague, D., Steer, P., & Davy, P. (2017). Rapid post‐seismic landslide evacuation boosted by dynamic river width. Nature Geoscience, 10(9), 680–684. https://doi.org/10.1038/ngeo3005 Crosta, G. B., & Frattini, P. (2001). Rainfall thresholds for triggering soil slips and debris flow. In: Proc. of the 2nd EGS Plinius Conference on Mediterranean Storms: Publication CNR GNDCI, 2547, pp. 463–487. Crowley, J. K., Hubbard, B. E., & Mars, J. C. (2003). Analysis of potential debris flow source areas on Mount Shasta, California, by using airborne and satellite remote sensing data. Remote Sensing of Environment, 87(2–3), 345–358. https://doi.org/10.1016/j. rse.2003.08.003 Crozier, M. J. (2005). Multiple‐occurrence regional landslide events in New Zealand: hazard management issues. Landslides, 2(4), 247–256. https://doi.org/10.1007/s10346‐005‐0019‐7 Crozier, M. J., Deimel, M. S., & Simon, J. S. (1995). Investigation of earthquake triggering for deep‐seated landslides, Taranaki, New Zealand. Quaternary International, 25,65–73. https://doi.org/10.1016/1040‐6182(94)00036‐5 Crozier, M. J., & Glade, T. (1999). Frequency and magnitude of landsliding: Fundamental research issues. Zeitschrift Für Geomorphologie Supplement Volumes, 115, 141–155. https://doi.org/10.1127/zfgsuppl/115/1999/141 Cruden, D. M., & Hu, X. Q. (1993). Exhaustion and steady state models for predicting landslide hazards in the Canadian Rocky Mountains. Geomorphology, 8(4), 279–285. https://doi.org/10.1016/0169‐555X(93)90024‐V Cruden, D. M., & Varnes, D. J. (1996). Landslide types and processes, Special Report. Transportation Research Board, National Academy of Sciences, 247,36–75. Cui, Y., Parker, G., Lisle, T. E., Gott, J., Hansler‐Ball, M. E., Pizzuto, J. E., et al. (2003). Sediment pulses in mountain rivers: 1. Experiments: Sediment pulse experiments. Water Resources Research, 39(9), 1239. https://doi.org/10.1029/2002WR001803 Cui, Y., Parker, G., Pizzuto, J., & Lisle, T. E. (2003). Sediment pulses in mountain rivers: 2. Comparison between experiments and numerical predictions: Sediment pulse numerical solution. Water Resources Research, 39(9), 1240. https://doi.org/10.1029/ 2002WR001805 Cui, Y.‐J., Le, T. T., Tang, A. M., Delage, P., & Li, X. L. (2009). Investigating the time‐dependent behaviour of Boom clay under thermo- mechanical loading. Géotechnique, 59(4), 319–329. https://doi.org/10.1680/geot.2009.59.4.319 Cundall, P. A. (1971). A computer model for simulating progressive, large‐scale movement in blocky rock system. In Proceedings of the International Symposium on Rock Mechanics, 1971. Dadson, S. J., Hovius, N., Chen, H., Dade, W. B., Lin, J.‐C., Hsu, M.‐L., et al. (2004). Earthquake‐triggered increase in sediment delivery from an active mountain belt. Geology, 32(8), 733. https://doi.org/10.1130/G20639.1 Dahlquist, M. P., West, A. J., & Li, G. (2018). Landslide‐driven drainage divide migration. Geology, 46(5), 403–406. https://doi.org/10.1130/ G39916.1 Dai, F. C., Lee, C. F., Deng, J. H., & Tham, L. G. (2005). The 1786 earthquake‐triggered landslide dam and subsequent dam‐break flood on the Dadu River, southwestern China. Geomorphology, 65(3–4), 205–221. https://doi.org/10.1016/j.geomorph.2004.08.011 Dai, F. C., Xu, C., Yao, X., Xu, L., Tu, X. B., & Gong, Q. M. (2011). Spatial distribution of landslides triggered by the 2008 Ms 8.0 Wenchuan earthquake, China. Journal of Asian Earth Sciences, 40(4), 883–895. https://doi.org/10.1016/j.jseaes.2010.04.010 Dakoulas, P., & Gazetas, G. (1986). Seismic shear strains and seismic coefficients in dams and embankments. Soil Dynamics and Earthquake Engineering, 5(2), 75–83. https://doi.org/10.1016/0267‐7261(86)90001‐1 Dalen, K., Flatø, H., Liu, J., & Zhang, H. (2012). Recovering from the Wenchuan earthquake: Living conditions and development in disaster areas 2008–2011. Oslo: Fafo. Danneels, G., Bourdeau, C., Torgoev, I., & Havenith, H.‐B. (2008). Geophysical investigation and dynamic modelling of unstable slopes: Case‐study of Kainama (Kyrgyzstan). Geophysical Journal International, 175(1), 17–34. https://doi.org/10.1111/j.1365‐ 246X.2008.03873.x

FAN ET AL. 66 Reviews of Geophysics 10.1029/2018RG000626

Das, J. T., Puppala, A. J., Bheemasetti, T. V., Walshire, L. A., & Corcoran, M. K. (2018). Sustainability and resilience analyses in slope stabilisation. Proceedings of the Institution of Civil Engineers ‐ Engineering Sustainability, 171(1), 25–36. https://doi.org/10.1680/ jensu.16.00054 Davies, T. R. H. (2017). Mountain process geomorphology: Conceptual progress in the Southern Alps. In J. Shulmeister (Ed.), Landscape and quaternary environmental change in New Zealand, (pp. 205–233). Paris: Atlantis Press. https://doi.org/10.2991/978‐94‐6239‐237‐3_6 Davies, T. R. H., & Korup, O. (2007). Persistent alluvial fanhead trenching resulting from large, infrequent sediment inputs. Earth Surface Processes and Landforms, 32(5), 725–742. https://doi.org/10.1002/esp.1410 Davies, T. R. H., Manville, V., Kunz, M., & Donadini, L. (2007). Modeling landslide dambreak flood magnitudes: case study. Journal of Hydraulic Engineering, 133(7), 713–720. https://doi.org/10.1061/(ASCE)0733‐9429(2007)133:7(713) Davis, L. L., & West, L. R. (1973). Observed effects of topography on ground motion. Bulletin of the Seismological Society of America, 63(1), 283–298. Dawson, A. G. (1994). Geomorphological effects of tsunami run‐up and backwash. In Geomorphology and natural hazards, (pp. 83–94). Amsterdam, Netherlands: Elsevier. https://doi.org/10.1016/B978‐0‐444‐82012‐9.50010‐4 Del Gaudio, V., Coccia, S., Wasowski, J., Gallipoli, M. R., & Mucciarelli, M. (2008). Detection of directivity in seismic site response from microtremor spectral analysis. Natural Hazards and Earth System Sciences, 8(4), 751–762. https://doi.org/10.5194/nhess‐8‐751‐ 2008 Del Gaudio, V., Muscillo, S., & Wasowski, J. (2014). What we can learn about slope response to earthquakes from ambient noise analysis: An overview. Engineering Geology, 182, 182–200. https://doi.org/10.1016/j.enggeo.2014.05.010 Del Gaudio, V., Pierri, P., & Wasowski, J. (2003). An approach to time‐probabilistic evaluation of seismically induced landslide hazard. Bulletin of the Seismological Society of America, 93(2), 557–569. https://doi.org/10.1785/0120020016 Del Gaudio, V., & Wasowski, J. (2007). Directivity of slope dynamic response to seismic shaking. Geophysical Research Letters, 34, L12301. https://doi.org/10.1029/2007GL029842 Del Gaudio, V., & Wasowski, J. (2011). Advances and problems in understanding the seismic response of potentially unstable slopes. Engineering Geology, 122(1–2), 73–83. https://doi.org/10.1016/j.enggeo.2010.09.007 Delaney, K. B., & Evans, S. G. (2015). The 2000 Yigong landslide (Tibetan Plateau), rockslide‐dammed lake and outburst flood: Review, remote sensing analysis, and process modelling. Geomorphology, 246, 377–393. https://doi.org/10.1016/j.geomorph.2015.06.020 Delgado, J., Garrido, J., López‐Casado, C., Martino, S., & Peláez, J. A. (2011). On far field occurrence of seismically induced landslides. Engineering Geology, 123(3), 204–213. https://doi.org/10.1016/j.enggeo.2011.08.002 Densmore, A. L., Ellis, M. A., & Anderson, R. S. (1998). Landsliding and the evolution of normal‐fault‐bounded mountains. Journal of Geophysical Research, 103(B7), 15203–15219. https://doi.org/10.1029/98JB00510 Densmore, A. L., & Hovius, N. (2000). Topographic fingerprints of bedrock landslides. Geology, 28(4), 371–374. https://doi.org/10.1130/ 0091‐7613(2000)28<371:TFOBL>2.0.CO;2 Densmore, A. L., Parker, R. N., Rosser, N. J., de Michele, M., Yong, L., Runqiu, H., et al. (2012). Reply to “Isostasy can't be ignored”. Nature Geoscience, 5(2), 83–84. https://doi.org/10.1038/ngeo1385 Dingle, E. H., Attal, M., & Sinclair, H. D. (2017). Abrasion‐set limits on Himalayan gravel flux. Nature, 544(7651), 471–474. https://doi.org/ 10.1038/nature22039 Domènech, G., Fan, X., Scaringi, G., van Asch, T. W. J., Xu, Q., Huang, R., & Hales, T. C. (2019). Modelling the role of material depletion, grain coarsening and revegetation in debris flow occurrences after the 2008 Wenchuan earthquake. Engineering Geology, 250,34–44. https://doi.org/10.1016/j.enggeo.2019.01.010 Domènech, G., Yang, F., Guo, X., Fan, X., Scaringi, G., Dai, L., et al. (2018). Two multi‐temporal datasets to track the enhanced landsliding after the 2008 Wenchuan Earthquake [Data set]. Zenodo [data set]. https://doi.org/10.5281/zenodo.1405489 Dong, J.‐J., Tung, Y.‐H., Chen, C.‐C., Liao, J.‐J., & Pan, Y.‐W. (2009). Discriminant analysis of the geomorphic characteristics and stability of landslide dams. Geomorphology, 110(3–4), 162–171. https://doi.org/10.1016/j.geomorph.2009.04.004 Dong, J.‐J., Tung, Y.‐H., Chen, C.‐C., Liao, J.‐J., & Pan, Y.‐W. (2011). Logistic regression model for predicting the failure probability of a landslide dam. Engineering Geology, 117(1–2), 52–61. https://doi.org/10.1016/j.enggeo.2010.10.004 Dreyfus, D., Rathje, E. M., & Jibson, R. W. (2013). The influence of different simplified sliding‐block models and input parameters on regional predictions of seismic landslides triggered by the Northridge earthquake. Engineering Geology, 163,41–54. https://doi.org/ 10.1016/j.enggeo.2013.05.015 Dufresne, A., Bösmeier, A., & Prager, C. (2016). Sedimentology of rock avalanche deposits—Case study and review. Earth‐Science Reviews, 163, 234–259. https://doi.org/10.1016/j.earscirev.2016.10.002 Dufresne, A., & Dunning, S. A. (2017). Process dependence of grain size distributions in rock avalanche deposits. Landslides, 14(5), 1555–1563. https://doi.org/10.1007/s10346‐017‐0806‐y Duncan, J. M., Wright, S. G., & Brandon, T. L. (2014). Soil strength and slope stability. Hoboken, NJ: John Wiley & Sons. Dunning, S. A., & Armitage, P. J. (2011). The grain‐size distribution of rock‐avalanche deposits: Implications for natural dam stability. In Natural and artificial rockslide dams, (pp. 479–498). Berlin, Heidelberg: Springer. Dunning, S. A., Mitchell, W. A., Rosser, N. J., & Petley, D. N. (2007). The Hattian Bala rock avalanche and associated landslides triggered by the Kashmir Earthquake of 8 October 2005. Engineering Geology, 93(3–4), 130–144. https://doi.org/10.1016/j.enggeo.2007.07.003 Dunning, S. A., Petley, D. N., Rosser, N. J., & Strom, A. L. (2005). The morphology and sedimentology of valley confined rock‐avalanche deposits and their effect on potential dam hazard. In O. Hungr, R. Fell, R. Couture, & E. Eberhardt (Eds.), Proceedings of the International Conference on Landslide Risk Management, Edited by, (pp. 691–701). London: Taylor & Francis, Balkema. Dussauge‐Peisser, C., Helmstetter, A., Grasso, J.‐R., Hantz, D., Desvarreux, P., Jeannin, M., & Giraud, A. (2002). Probabilistic approach to rock fall hazard assessment: Potential of historical data analysis. Natural Hazards and Earth System Science, 2(1/2), 15–26. https://doi. org/10.5194/nhess‐2‐15‐2002 Earle, P. S., Wald, D. J., Jaiswal, K. S., Allen, T. I., Hearne, M. G., Marano, K. D., et al. (2009). Prompt Assessment of Global Earthquakes for Response (PAGER): A system for rapidly determining the impact of earthquakes worldwide. U.S. Geological Survey Open‐File Report, 1131, 15. Elgamal, A.‐W. M., Scott, R. F., Succarieh, M. F., & Yan, L. (1990). La Villita dam response during five earthquakes including permanent deformation. Journal of Geotechnical Engineering, 116(10), 1443–1462. https://doi.org/10.1061/(ASCE)0733‐ 9410(1990)116:10(1443) Elliott, J. R., Jolivet, R., González, P. J., Avouac, J.‐P., Hollingsworth, J., Searle, M. P., & Stevens, V. L. (2016). Himalayan megathrust geometry and relation to topography revealed by the Gorkha earthquake. Nature Geoscience, 9(2), 174–180. https://doi.org/10.1038/ ngeo2623

FAN ET AL. 67 Reviews of Geophysics 10.1029/2018RG000626

Emberson, R., Galy, A., & Hovius, N. (2017). Combined effect of carbonate and biotite dissolution in landslides biases silicate weathering proxies. Geochimica et Cosmochimica Acta, 213, 418–434. https://doi.org/10.1016/j.gca.2017.07.014 Emberson, R., Hovius, N., & Galy, A. (2018). Weathering of reactive mineral phases in landslides acts as a source of carbon dioxide in mountain belts. Journal of Geophysical Research: Earth Surface, 123, 2695–2713. https://doi.org/10.1029/2018JF004672 Emberson, R., Hovius, N., Galy, A., & Marc, O. (2016). Chemical weathering in active mountain belts controlled by stochastic bedrock landsliding. Nature Geoscience, 9(1), 42–45. https://doi.org/10.1038/ngeo2600 Ermini, L., & Casagli, N. (2003). Prediction of the behaviour of landslide dams using a geomorphological dimensionless index. Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group, 28(1), 31–47. https://doi.org/10.1002/ esp.424 Evans, S. G. (1986). The maximum discharge of outburst floods caused by the breaching of man‐made and natural dams. Canadian Geotechnical Journal, 23(3), 385–387. https://doi.org/10.1139/t86‐053 Evans, S. G., & Bent, A. L. (2004). The Las Colinas landslide, Santa Tecla: A highly destructive flowslide triggered by the January 13, 2001, El Salvador earthquake. SPECIAL PAPERS‐GEOLOGICAL SOCIETY OF AMERICA,25–38. Evans, S. G., Delaney, K. B., Hermanns, R. L., Strom, A., & Scarascia‐Mugnozza, G. (2011). The formation and behaviour of natural and artificial rockslide dams; implications for engineering performance and hazard management. In Natural and artificial rockslide dams, (pp. 1–75). Berlin, Heidelberg: Springer. Evans, S. G., Roberts, N. J., Ischuk, A., Delaney, K. B., Morozova, G. S., & Tutubalina, O. (2009). Landslides triggered by the 1949 Khait earthquake, Tajikistan, and associated loss of life. Engineering Geology, 109(3–4), 195–212. https://doi.org/10.1016/j. enggeo.2009.08.007 Fan, G., Zhang, J., Wu, J., & Yan, K. (2016). Dynamic response and dynamic failure mode of a weak intercalated rock slope using a shaking table. Rock Mechanics and Rock Engineering, 49(8), 3243–3256. https://doi.org/10.1007/s00603‐016‐0971‐7 Fan, L., Lehmann, P., McArdell, B., & Or, D. (2017). Linking rainfall‐induced landslides with debris flows runout patterns towards catchment scale hazard assessment. Geomorphology, 280,1–15. https://doi.org/10.1016/j.geomorph.2016.10.007 Fan, N., Nie, R., Wang, Q., & Liu, X. (2016). Dramatic undercutting of piedmont rivers after the 2008 Wenchuan Ms 8.0 Earthquake. Scientific Reports, 6(1), 37108. https://doi.org/10.1038/srep37108 Fan, X., Domènech, G., Scaringi, G., Huang, R., Xu, Q., Hales, T. C., et al. (2018). Spatio‐temporal evolution of mass wasting after the 2008 Mw 7.9 Wenchuan Earthquake revealed by a detailed multi‐temporal inventory. Landslides, 15(12), 2325–2341. https://doi.org/10.1007/ s10346‐018‐1054‐5 Fan, X., Juang, C. H., Wasowski, J., Huang, R., Xu, Q., Scaringi, G., et al. (2018). What we have learned from the 2008 Wenchuan Earthquake and its aftermath: A decade of research and challenges. Engineering Geology, 241,25–32. https://doi.org/10.1016/j. enggeo.2018.05.004 Fan, X., Rossiter, D. G., van Westen, C. J., Xu, Q., & Görüm, T. (2014). Empirical prediction of coseismic landslide dam formation: Coseismic landslide dam formation. Earth Surface Processes and Landforms, 39(14), 1913–1926. https://doi.org/10.1002/esp.3585 Fan, X., Scaringi, G., Domènech, G., Yang, F., Guo, X., Dai, L., et al. (2019). Two multi‐temporal datasets that track the enhanced landsliding after the 2008 Wenchuan earthquake. Earth System Science Data, 11(1), 35–55. https://doi.org/10.5194/essd‐11‐35‐2019 Fan, X., Scaringi, G., Xu, Q., Zhan, W., Dai, L., Li, Y., et al. (2018). Coseismic landslides triggered by the 8th August 2017 M s 7.0 Jiuzhaigou earthquake (Sichuan, China): Factors controlling their spatial distribution and implications for the seismogenic blind fault identification. In Landslides, (pp. 1–17). Germany: Springer. Fan, X., Tang, C. X., van Westen, C. J., & Alkema, D. (2012). Simulating dam‐breach flood scenarios of the Tangjiashan landslide dam induced by the Wenchuan Earthquake. Natural Hazards and Earth System Sciences, 12(10), 3031–3044. https://doi.org/10.5194/nhess‐ 12‐3031‐2012 Fan, X., van Westen, C. J., Xu, Q., Gorum, T., & Dai, F. (2012). Analysis of landslide dams induced by the 2008 Wenchuan earthquake. Journal of Asian Earth Sciences, 57,25–37. https://doi.org/10.1016/j.jseaes.2012.06.002 Fan, X., Xu, Q., & Scaringi, G. (2018). Brief communication: Post‐seismic landslides, the tough lesson of a catastrophe. Natural Hazards and Earth System Sciences, 18(1), 397–403. https://doi.org/10.5194/nhess‐18‐397‐2018 Fan, X., Xu, Q., Scaringi, G., Dai, L., Li, W., Dong, X., et al. (2017). Failure mechanism and kinematics of the deadly June 24th 2017 Xinmo landslide, Maoxian, Sichuan, China. Landslides, 14(6), 2129–2146. https://doi.org/10.1007/s10346‐017‐0907‐7 Fan, X., Xu, Q., van Westen, C. J., Huang, R., & Tang, R. (2017). Characteristics and classification of landslide dams associated with the 2008 Wenchuan earthquake. Geoenvironmental Disasters, 4(1). https://doi.org/10.1186/s40677‐017‐0079‐8 Fan, X., Zhan, W., Dong, X., van Westen, C., Xu, Q., Dai, L., et al. (2018). Analyzing successive landslide dam formation by different triggering mechanisms: The case of the Tangjiawan landslide, Sichuan, China. Engineering Geology, 243, 128–144. https://doi.org/ 10.1016/j.enggeo.2018.06.016 Fang, C., Wu, F., & Li, M. (2011). Suitability evaluation of population and settlements spatial layout after Wenchuan Earthquake. Journal of Geographical Sciences, 21(3), 539–548. https://doi.org/10.1007/s11442‐011‐0862‐5 Faris, F., & Wang, F. (2014). Stochastic analysis of rainfall effect on earthquake induced shallow landslide of Tandikat, West Sumatra, Indonesia. Geoenvironmental Disasters, 1(1), 12. https://doi.org/10.1186/s40677‐014‐0012‐3 Farr, T. G., Rosen, P. A., Caro, E., Crippen, R., Duren, R., Hensley, S., et al. (2007). The Shuttle Radar Topography Mission. Reviews of Geophysics, 45, RG2004. https://doi.org/10.1029/2005RG000183 Feng, Z. (2012). The seismic signatures of the surge wave from the 2009 Xiaolin landslide‐dam breach in Taiwan. Hydrological Processes, 26(9), 1342–1351. https://doi.org/10.1002/hyp.8239 Fielding, E. J., Sladen, A., Li, Z., Avouac, J.‐P., Bürgmann, R., & 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. Geophysical Journal International, 194(2), 1138–1166. https://doi.org/10.1093/gji/ggt155 Filion, L., Quinty, F., & Bégin, C. (1991). A chronology of landslide activity in the valley of Rivière du Gouffre, Charlevoix, Quebec. Canadian Journal of Earth Sciences, 28(2), 250–256. https://doi.org/10.1139/e91‐024 Flentje, P., & Chowdhury, R. (2018). Resilience and sustainability in the management of landslides. Proceedings of the Institution of Civil Engineers ‐ Engineering Sustainability, 171(1), 3–14. https://doi.org/10.1680/jensu.16.00045 Foster, I. D. (2010). Lakes and reservoirs in the sediment cascade. In Sediment Cascades: An Integrated Approach, (pp. 345–376). Chichester, UK: John Wiley & Sons Ltd. https://doi.org/10.1002/9780470682876.ch12 Frattini, P., & Crosta, G. B. (2013). The role of material properties and landscape morphology on landslide size distributions. Earth and Planetary Science Letters, 361, 310–319. https://doi.org/10.1016/j.epsl.2012.10.029 Fread, D. L. (1991). BREACH: An erosion model for earth dam failures. Hydrologic Research Laboratory, U.S. National Weather Service.

FAN ET AL. 68 Reviews of Geophysics 10.1029/2018RG000626

Frigerio, S., Schenato, L., Bossi, G., Cavalli, M., Mantovani, M., Marcato, G., & Pasuto, A. (2014). A web‐based platform for automatic and continuous landslide monitoring: The Rotolon (Eastern Italian Alps) case study. Computers & Geosciences, 63,96–105. https://doi.org/ 10.1016/j.cageo.2013.10.015 Frith, N. V., Hilton, R. G., Howarth, J. D., Gröcke, D. R., Fitzsimons, S. J., Croissant, T., et al. (2018). Carbon export from mountain forests enhanced by earthquake‐triggered landslides over millennia. Nature Geoscience, 11(10), 772–776. https://doi.org/10.1038/s41561‐018‐ 0216‐3 Froude, M. J., & Petley, D. N. (2018). Global fatal landslide occurrence 2004 to 2016. Natural Hazards and Earth System Sciences, 18(8), 2161–2181. https://doi.org/10.5194/nhess‐18‐2161‐2018 Fuchs, S. (2009). Susceptibility versus resilience to mountain hazards in Austria‐paradigms of vulnerability revisited. Natural Hazards & Earth System Sciences, 9(2), 337–352. https://doi.org/10.5194/nhess‐9‐337‐2009 Fuchs, S., Heiss, K., & Hübl, J. (2007). Towards an empirical vulnerability function for use in debris flow risk assessment. Natural Hazards and Earth System Science, 7(5), 495–506. https://doi.org/10.5194/nhess‐7‐495‐2007 Gallen, S. F., Clark, M. K., Godt, J. K., Roback, K., & Niemi, N. A. (2017). Application and evaluation of a rapid response earthquake‐ triggered landslide model to the 25 April 2015 Mw 7.8 Gorkha earthquake, Nepal. Tectonophysics, 714,173–187. https://doi.org/10.1016/ j.tecto.2016.10.031 Gallen, S. F., Clark, M. K., & Godt, J. W. (2015). Coseismic landslides reveal near‐surface rock strength in a high‐relief, tectonically active setting. Geology, 43(1), 11–14. https://doi.org/10.1130/G36080.1 Gallipoli, M. R., & Mucciarelli, M. (2007). Effetti direzionali in registrazioni sismometriche in aree in frana e bordi di bacino. XXVI GNGTS Conference, Rome, 13–15 November 2007. Gariano, S. L., Rianna, G., Petrucci, O., & Guzzetti, F. (2017). Assessing future changes in the occurrence of rainfall‐induced landslides at a regional scale. Science of The Total Environment, 596‐597, 417–426. https://doi.org/10.1016/j.scitotenv.2017.03.103 Geiß, C., & Taubenböck, H. (2017). One step back for a leap forward: Toward operational measurements of elements at risk. Netherlands: Springer. Gerolymos, N., & Gazetas, G. (2005). Constitutive model for 1‐D cyclic soil behaviour applied to seismic analysis of layered deposits. Soils and Foundations, 45(3), 147–159. https://doi.org/10.3208/sandf.45.3_147 Ghosh, S., Günther, A., Carranza, E. J. M., van Westen, C. J., & Jetten, V. G. (2010). Rock slope instability assessment using spatially distributed structural orientation data in Darjeeling Himalaya (India). Earth Surface Processes and Landforms, 35(15), 1773–1792. https:// doi.org/10.1002/esp.2017 Gill, J. C., & Malamud, B. D. (2014). Reviewing and visualizing the interactions of natural hazards: Interactions of Natural Hazards. Reviews of Geophysics, 52, 680–722. https://doi.org/10.1002/2013RG000445 Gischig, V. S., Eberhardt, E., Moore, J. R., & Hungr, O. (2015). On the seismic response of deep‐seated rock slope instabilities—Insights from numerical modeling. Engineering Geology, 193,1–18. https://doi.org/10.1016/j.enggeo.2015.04.003 Glazovskaya, T. G., Myagkov, S. M., Troshkina, E. S., Akif'eva, K. V., Kondakova, N. L., & Kravtcova, V. I. (1992). Rasprostranenie irejim lavin [Avalanches spreading and regime]. In S. M. Myagkov, & L. A. Kanaev (Eds.), Geografiya Lavin [Geography of Avalanches], (pp. 43–111). Moscow: Moscow State University Press. Godt, J. W., Sener, B., Verdin, K., Wald, D. J., Earle, P. S., Harp, E. L., & Jibson, R. W. (2008). Rapid assessment of earthquake‐induced landsliding. In Proceedings of the First World Landslide Forum, (Vol. 4, pp. 219–222). Tokyo: United Nations University. Goff, J., McFadgen, B., Wells, A., & Hicks, M. (2008). Seismic signals in coastal dune systems. Earth‐Science Reviews, 89(1‐2), 73–77. https:// doi.org/10.1016/j.earscirev.2008.03.002 Gorum, T., & Carranza, E. J. M. (2015). Control of style‐of‐faulting on spatial pattern of earthquake‐triggered landslides. International Journal of Environmental Science and Technology, 12(10), 3189–3212. https://doi.org/10.1007/s13762‐015‐0752‐y Gorum, T., Fan, X., van Westen, C. J., Huang, R., Xu, Q., Tang, C., & Wang, G. (2011). Distribution pattern of earthquake‐induced landslides triggered by the 12 May 2008 Wenchuan earthquake. Geomorphology, 133(3–4), 152–167. https://doi.org/10.1016/j. geomorph.2010.12.030 Gorum, T., Korup, O., van Westen, C. J., van der Meijde, M., Xu, C., & van der Meer, F. D. (2014). Why so few? Landslides triggered by the 2002 Denali earthquake, Alaska. Quaternary Science Reviews, 95,80–94. https://doi.org/10.1016/j.quascirev.2014. 04.032 Gorum, T., van Westen, C. J., Korup, O., van der Meijde, M., Fan, X., & van der Meer, F. D. (2013). Complex rupture mechanism and topography control symmetry of mass‐wasting pattern, 2010 Haiti earthquake. Geomorphology, 184, 127–138. https://doi.org/10.1016/j. geomorph.2012.11.027 Grant, A., Wartman, J., & Abou‐Jaoude, G. (2016). Multimodal method for coseismic landslide hazard assessment. Engineering Geology, 212, 146–160. https://doi.org/10.1016/j.enggeo.2016.08.005 Griffiths, D. V., & Prevost, J. H. (1988). Two‐and three‐dimensional dynamic finite element analyses of the Long Valley Dam. National Center for Earthquake Engineering Research. Gupta, R., & Saha, A. (2001). Mapping debris flows in the Himalayas, natural resource management. GISdevelopment. Net, 4. Guthrie, R. H., Deadman, P. J., Cabrera, A. R., & Evans, S. G. (2008). Exploring the magnitude–frequency distribution: A cellular automata model for landslides. Landslides, 5(1), 151–159. https://doi.org/10.1007/s10346‐007‐0104‐1 Guzzetti, F., Ardizzone, F., Cardinali, M., Rossi, M., & Valigi, D. (2009). Landslide volumes and landslide mobilization rates in Umbria, central Italy. Earth and Planetary Science Letters, 279(3–4), 222–229. https://doi.org/10.1016/j.epsl.2009.01.005 Guzzetti, F., Mondini, A. C., Cardinali, M., Fiorucci, F., Santangelo, M., & Chang, K.‐T. (2012). Landslide inventory maps: New tools for an old problem. Earth‐Science Reviews, 112(1–2), 42–66. https://doi.org/10.1016/j.earscirev.2012.02.001 Guzzetti, F., Peruccacci, S., Rossi, M., & Stark, C. P. (2008). The rainfall intensity–duration control of shallow landslides and debris flows: An update. Landslides, 5(1), 3–17. https://doi.org/10.1007/s10346‐007‐0112‐1 Guzzetti, F., Reichenbach, P., Cardinali, M., Galli, M., & Ardizzone, F. (2005). Probabilistic landslide hazard assessment at the basin scale. Geomorphology, 72(1–4), 272–299. https://doi.org/10.1016/j.geomorph.2005.06.002 Hall‐Wallace, M. K. (1998). Can earthquakes be predicted? Journal of Geoscience Education, 46(5), 439–449. https://doi.org/10.5408/1089‐ 9995‐46.5.439 Hancox, G. T., McSaveney, M. J., Manville, V. R., & Davies, T. R. H. (2005). The October 1999 Mt Adams rock avalanche and subsequent landslide dam‐break flood and effects in Poerua river, Westland, New Zealand. New Zealand Journal of Geology and Geophysics, 48(4), 683–705. https://doi.org/10.1080/00288306.2005.9515141 Hancox, G. T., Perrin, N. D., & Dellow, G. D. (2002). Recent studies of historical earthquake‐induced landsliding, ground damage, and MM intensity in New Zealand. Bulletin‐New Zealand Society for Earthquake Engineering, 35(2), 59–95.

FAN ET AL. 69 Reviews of Geophysics 10.1029/2018RG000626

Harp, E. L., & Jibson, R. W. (1995). Inventory of landslides triggered by the 1994 Northridge, California earthquake. U.S. Geological Survey Open‐File Report. 95‐213. Harp, E. L., & Jibson, R. W. (1996). Landslides triggered by the 1994 Northridge, California, earthquake. Bulletin of the Seismological Society of America, 86(1 SUPPL. B), S319–S332. Retrieved from. http://pubs.er.usgs.gov/publication/70019319 Harp, E. L., & Jibson, R. W. (2002). Anomalous concentrations of seismically triggered rock falls in Pacoima Canyon: Are they caused by highly susceptible slopes or local amplification of seismic shaking? Bulletin of the Seismological Society of America, 92(8), 3180–3189. https://doi.org/10.1785/0120010171 Harp, E. L., Jibson, R. W., & Schmitt, R. G. (2016). Map of landslides triggered by the January 12, 2010, Haiti earthquake. https://pubs.er. usgs.gov/publication/sim3353. Harp, E. L., Keefer, D. K., Sato, H. P., & Yagi, H. (2011). Landslide inventories: The essential part of seismic landslide hazard analyses. Engineering Geology, 122(1–2), 9–21. https://doi.org/10.1016/j.enggeo.2010.06.013 Harp, E. L., Wilson, R. C., & Wieczorek, G. F. (1981). Landslides from the February 4, 1976, Guatemala earthquake. Washington, DC: U.S. Government Printing Office. Hartmann, J., & Moosdorf, N. (2012, September 12). Global Lithological Map Database v1.0 (gridded to 0.5° spatial resolution). PANGAEA. https://doi.org/10.1594/PANGAEA.788537 Hartzell, S., Leeds, A. L., & Jibson, R. W. (2017). Seismic response of soft deposits due to landslide: The Mission Peak, California, Landslide. Bulletin of the Seismological Society of America, 107(5), 2008–2020. https://doi.org/10.1785/0120170033 Has, B., Noro, T., Maruyama, K., Nakamura, A., Ogawa, K., & Onoda, S. (2012). Characteristics of earthquake‐induced landslides in a heavy snowfall region—Landslides triggered by the northern Nagano prefecture earthquake, March 12, 2011, Japan. Landslides, 9(4), 539–546. https://doi.org/10.1007/s10346‐012‐0344‐6 Havenith, H.‐B., Jongmans, D., Faccioli, E., Abdrakhmatov, K., & Bard, P. Y. (2002). Site effect analysis around the seismically induced Ananevo Rockslide, Kyrgyzstan. Bulletin of the Seismological Society of America, 92(8), 3190–3209. https://doi.org/10.1785/0120010206 Havenith, H.‐B., Torgoev, A., Braun, A., Schlögel, R., & Micu, M. (2016). A new classification of earthquake‐induced landslide event sizes based on seismotectonic, topographic, climatic and geologic factors. Geoenvironmental Disasters, 3(1). https://doi.org/10.1186/s40677‐016‐0041‐1 Heerdegen, R. G., & Beran, M. A. (1982). Quantifying source areas through land surface curvature and shape. Journal of Hydrology, 57(3–4), 359–373. https://doi.org/10.1016/0022‐1694(82)90155‐X Hengl, T., de Jesus, J. M., MacMillan, R. A., Batjes, N. H., Heuvelink, G. B. M., Ribeiro, E., et al. (2014). SoilGrids1km—Global Soil Information Based on Automated Mapping. PLOS ONE, 9(8), e105992. https://doi.org/10.1371/journal.pone.0105992 Hergarten, S. (2003). Landslides, sandpiles, and self‐organized criticality. Natural Hazards and Earth System Science, 3(6), 505–514. https:// doi.org/10.5194/nhess‐3‐505‐2003 Hergarten, S. (2012). Topography‐based modeling of large rockfalls and application to hazard assessment. Geophysical Research Letters, 39, L13402. https://doi.org/10.1029/2012GL052090 Hermanns, R. L., Hewitt, K., Strom, A., Evans, S. G., Dunning, S. A., & Scarascia‐Mugnozza, G. (2011). The classification of rockslide dams. In Natural and artificial rockslide dams, (pp. 581–593). Berlin, Heidelberg: Springer. Hewitt, K. (2006). Disturbance regime landscapes: Mountain drainage systems interrupted by large rockslides. Progress in Physical Geography, 30(3), 365–393. https://doi.org/10.1191/0309133306pp486ra Hewitt, K., Clague, J. J., & Orwin, J. F. (2008). Legacies of catastrophic rock slope failures in mountain landscapes. Earth‐Science Reviews, 87(1–2), 1–38. https://doi.org/10.1016/j.earscirev.2007.10.002 Hill, D. P., Pollitz, F., & Newhall, C. (2002). Earthquake‐volcano interactions. Physics Today, 55(11), 41–47. https://doi.org/10.1063/ 1.1535006 Hilton, R. G., Galy, A., & Hovius, N. (2008). Riverine particulate organic carbon from an active mountain belt: Importance of landslides: LANDSLIDE MOBILIZED POC. Global Biogeochemical Cycles, 22, GB1017. https://doi.org/10.1029/2006GB002905 Hilton, R. G., Galy, A., Hovius, N., Horng, M.‐J., & Chen, H. (2011). Efficient transport of fossil organic carbon to the ocean by steep mountain rivers: An orogenic carbon sequestration mechanism. Geology, 39(1), 71–74. https://doi.org/10.1130/G31352.1 Hilton, R. G., Meunier, P., Hovius, N., Bellingham, P. J., & Galy, A. (2011). Landslide impact on organic carbon cycling in a temperate montane forest. Earth Surface Processes and Landforms, 36(12), 1670–1679. https://doi.org/10.1002/esp.2191 Hobiger, M., Wegler, U., Shiomi, K., & Nakahara, H. (2014). Single‐station cross‐correlation analysis of ambient seismic noise: Application to stations in the surroundings of the 2008 Iwate‐Miyagi Nairiku earthquake. Geophysical Journal International, 198(1), 90–109. https:// doi.org/10.1093/gji/ggu115 Horton, P., Jaboyedoff, M., & Rudaz, B. et, & Zimmermann, M. (2013). Flow‐R, a model for susceptibility mapping of debris flows and other gravitational hazards at a regional scale. Natural Hazards and Earth System Sciences, 13(4), 869–885. https://doi.org/10.5194/nhess‐13‐ 869‐2013 Hough, S. E., Altidor, J. R., Anglade, D., Given, D., Janvier, M. G., Maharrey, J. Z., et al. (2010). Localized damage caused by topographic amplification during the 2010 M 7.0 Haiti earthquake. Nature Geoscience, 3(11), 778–782. https://doi.org/10.1038/ngeo988 Hovius, N., Meunier, P., Lin, C.‐W., Chen, H., Chen, Y.‐G., Dadson, S., et al. (2011). Prolonged seismically induced erosion and the mass balance of a large earthquake. Earth and Planetary Science Letters, 304(3–4), 347–355. https://doi.org/10.1016/j.epsl.2011.02.005 Hovius, N., Stark, C. P., & Allen, P. A. (1997). Sediment flux from a mountain belt derived by landslide mapping. Geology, 25(3), 231–234. https://doi.org/10.1130/0091‐7613(1997)025<0231:SFFAMB>2.3.CO;2 Hovius, N., Stark, C. P., Hao‐Tsu, C., & Jiun‐Chuan, L. (2000). Supply and removal of sediment in a landslide‐dominated mountain belt: Central Range, Taiwan. The Journal of Geology, 108(1), 73–89. https://doi.org/10.1086/314387 Hovius, N., Stark, C. P., Tutton, M. A., & Abbott, L. D. (1998). Landslide‐driven drainage network evolution in a pre‐steady‐state mountain belt: Finisterre Mountains, Papua New Guinea. Geology, 26(12), 1071. https://doi.org/10.1130/0091‐7613(1998)026<1071:LDDNEI>2.3. CO;2 Howarth, J. D., Fitzsimons, S. J., Norris, R. J., & Jacobsen, G. E. (2012). Lake sediments record cycles of sediment flux driven by large earthquakes on the Alpine fault, New Zealand. Geology, 40(12), 1091–1094. https://doi.org/10.1130/G33486.1 Howarth, J. D., Fitzsimons, S. J., Norris, R. J., & Jacobsen, G. E. (2014). Lake sediments record high intensity shaking that provides insight into the location and rupture length of large earthquakes on the Alpine Fault, New Zealand. Earth and Planetary Science Letters, 403, 340–351. https://doi.org/10.1016/j.epsl.2014.07.008 Howarth, J. D., Fitzsimons, S. J., Norris, R. J., Langridge, R., & Vandergoes, M. J. (2016). A 2000 yr rupture history for the Alpine fault derived from Lake Ellery, South Island, New Zealand. Geological Society of America Bulletin, 128(3–4), 627–643. https://doi.org/10.1130/ B31300.1

FAN ET AL. 70 Reviews of Geophysics 10.1029/2018RG000626

Hsieh, M.‐L., & Chyi, S.‐J. (2010). Late Quaternary mass‐wasting records and formation of fan terraces in the Chen‐yeo‐lan and Lao‐nung catchments, central‐southern Taiwan. Quaternary Science Reviews, 29(11–12), 1399–1418. https://doi.org/10.1016/j. quascirev.2009.10.002 Hsieh, M.‐L., Liew, P.‐M., & Chen, H.‐W. (2011). Early Holocene catastrophic mass‐wasting event and fan‐delta development on the Hua‐ tung coast, eastern Taiwan. Geomorphology, 134(3–4), 378–393. https://doi.org/10.1016/j.geomorph.2011.07.012 Hsieh, S.‐Y., & Lee, C.‐T. (2011). Empirical estimation of the Newmark displacement from the Arias intensity and critical acceleration. Engineering Geology, 122(1–2), 34–42. https://doi.org/10.1016/j.enggeo.2010.12.006 Hsieh, Y.‐M., Lee, K.‐C., Jeng, F.‐S., & Huang, T.‐H. (2011). Can tilt tests provide correct insight regarding frictional behavior of sliding rock block under seismic excitation? Engineering Geology, 122(1–2), 84–92. https://doi.org/10.1016/j.enggeo.2010.11.008 Hu, W., Dong, X. J., Xu, Q., Wang, G. H., van Asch, T. W. J., & Hicher, P. Y. (2016). Initiation processes for run‐off generated debris flows in the Wenchuan earthquake area of China. Geomorphology, 253, 468–477. https://doi.org/10.1016/j.geomorph.2015.10.024 Hu, W., Hicher, P.‐Y., Scaringi, G., Xu, Q., Van Asch, T. W. J., & Wang, G. (2018). Seismic precursor to instability induced by internal erosion in loose granular slopes. Géotechnique, 68(11), 989–1001. https://doi.org/10.1680/jgeot.17.P.079 Hu, W., Scaringi, G., Xu, Q., & Huang, R. (2018). Internal erosion controls failure and runout of loose granular deposits: Evidence from flume tests and implications for postseismic slope healing. Geophysical Research Letters, 45, 5518–5527. https://doi.org/10.1029/ 2018GL078030 Hu, W., Scaringi, G., Xu, Q., Pei, Z., Van Asch, T. W. J., & Hicher, P.‐Y. (2017). Sensitivity of the initiation and runout of flowslides in loose granular deposits to the content of small particles: An insight from flume tests. Engineering Geology, 231,34–44. https://doi.org/10.1016/ j.enggeo.2017.10.001 Huang, C. C., Lee, Y. H., Liu, H. P., Keefer, D. K., & Jibson, R. W. (2001). Influence of surface‐normal ground acceleration on the initiation of the Jih‐Feng‐Erh‐Shan Landslide during the 1999 Chi‐Chi, Taiwan, Earthquake. Bulletin of the Seismological Society of America, 91(5), 953–958. https://doi.org/10.1785/0120000719 Huang, J., Huang, R., Ju, N., Xu, Q., & He, C. (2015). 3D WebGIS‐based platform for debris flow early warning: A case study. Engineering Geology, 197,57–66. https://doi.org/10.1016/j.enggeo.2015.08.013 Huang, M. Y. F., & Montgomery, D. R. (2012). Fluvial response to rapid episodic erosion by earthquake and typhoons, Tachia River, central Taiwan. Geomorphology, 175‐176, 126–138. https://doi.org/10.1016/j.geomorph.2012.07.004 Huang, R., & Fan, X. (2013). The landslide story. Nature Geoscience, 6(5), 325–326. https://doi.org/10.1038/ngeo1806 Huang, R., & Li, W. (2014). Post‐earthquake landsliding and long‐term impacts in the Wenchuan earthquake area, China. Engineering Geology, 182, 111–120. https://doi.org/10.1016/j.enggeo.2014.07.008 Huang, R., Zhao, J., Ju, N., Li, G., Lee, M. L., & Li, Y. (2013). Analysis of an anti‐dip landslide triggered by the 2008 Wenchuan earthquake in China. Natural Hazards, 68(2), 1021–1039. https://doi.org/10.1007/s11069‐013‐0671‐5 Huang, Y., & Yu, M. (2013). Review of soil liquefaction characteristics during major earthquakes of the twenty‐first century. Natural Hazards, 65(3), 2375–2384. https://doi.org/10.1007/s11069‐012‐0433‐9 Hubbard, J., & Shaw, J. H. (2009). Uplift of the Longmen Shan and Tibetan plateau and the 2008 Wenchuan (M = 7.9) earthquake. Nature, 458(7235), 194–197. https://doi.org/10.1038/nature07837 Huffman, G. J., Adler, R. F., Bolvin, D. T., & Nelkin, E. J. (2010). The TRMM multi‐satellite precipitation analysis (TMPA). In Satellite rainfall applications for surface hydrology, (pp. 3–22). Dordrecht: Springer. Huffman, G. J., & Bolvin, D. T. (2013). TRMM and other data precipitation data set documentation. NASA, Greenbelt, USA, 28(2.3), 1. Hungr, O., Evans, S. G., & Hazzard, J. (1999). Magnitude and frequency of rock falls and rock slides along the main transportation corridors of southwestern British Columbia. Canadian Geotechnical Journal, 36(2), 224–238. https://doi.org/10.1139/t98‐106 Ichinose, G. A., Anderson, J. G., Satake, K., Schweickert, R. A., & Lahren, M. M. (2000). The potential hazard from tsunami and seiche waves generated by large earthquakes within Lake Tahoe, California‐Nevada. Geophysical Research Letters, 27(8), 1203–1206. https:// doi.org/10.1029/1999GL011119 Imperatori, W., & Mai, P. M. (2015). The role of topography and lateral velocity heterogeneities on near‐source scattering and ground‐ motion variability. Geophysical Journal International, 202(3), 2163–2181. https://doi.org/10.1093/gji/ggv281 Ishihara, K., Okusa, S., Oyagi, N., & Ischuk, A. (1990). Liquefaction‐induced flow slide in the collapsible loess deposit in Soviet Tajik. Soils and Foundations, 30(4), 73–89. https://doi.org/10.3208/sandf1972.30.4_73 Iverson, R. M. (1997). The physics of debris flows. Reviews of Geophysics, 35(3), 245–296. https://doi.org/10.1029/97RG00426 Jakob, M., Stein, D., & Ulmi, M. (2012). Vulnerability of buildings to debris flow impact. Natural Hazards, 60(2), 241–261. https://doi.org/ 10.1007/s11069‐011‐0007‐2 Jamei, M., Hamrouni, F., & Trabelsi, H. (2017). Application of the hydraulic gradient method for physical modeling of rainfall induced landslide: The optimal design for a physical laboratory model. In JTC1 workshop on advances in landslide understanding, Barcelona. Jibson, R. W. (1993). Predicting earthquake‐induced landslide displacements using Newmark's sliding block analysis. Transportation Research Record, 1411,9–17. Jibson, R. W. (1996). Use of landslides for paleoseismic analysis. Engineering Geology, 43(4), 291–323. https://doi.org/10.1016/S0013‐ 7952(96)00039‐7 Jibson, R. W. (2007). Regression models for estimating coseismic landslide displacement. Engineering Geology, 91(2–4), 209–218. https:// doi.org/10.1016/j.enggeo.2007.01.013 Jibson, R. W. (2009). Using landslides for paleoseismic analysis. International Geophysics, 95, 565–601. https://doi.org/10.1016/S0074‐ 6142(09)95008‐2 Jibson, R. W. (2011). Methods for assessing the stability of slopes during earthquakes—A retrospective. Engineering Geology, 122(1–2), 43–50. https://doi.org/10.1016/j.enggeo.2010.09.017 Jibson, R. W., & Harp, E. L. (2012). Extraordinary distance limits of landslides triggered by the 2011 Mineral, Virginia, Earthquake. Bulletin of the Seismological Society of America, 102(6), 2368–2377. https://doi.org/10.1785/0120120055 Jibson, R. W., & Harp, E. L. (2016). Ground motions at the outermost limits of seismically triggered landslides. Bulletin of the Seismological Society of America, 106(2), 708–719. https://doi.org/10.1785/0120150141 Jibson, R. W., Harp, E. L., & Michael, J. A. (1998). A method for producing digital probabilistic seismic landslide hazard maps: An example from the Los Angeles, California, area. U.S. Geological Survey. (Report No. 98–113). https://doi.org/10.3133/ofr98113 Jibson, R. W., Harp, E. L., & Michael, J. A. (2000). A method for producing digital probabilistic seismic landslide hazard maps. Engineering Geology, 58(3–4), 271–289. https://doi.org/10.1016/S0013‐7952(00)00039‐9

FAN ET AL. 71 Reviews of Geophysics 10.1029/2018RG000626

Jibson, R. W., Harp, E. L., Schulz, W., & Keefer, D. K. (2004). Landslides triggered by the 2002 Denali Fault, Alaska, earthquake and the inferred nature of the strong shaking. Earthquake Spectra, 20(3), 669–691. https://doi.org/10.1193/1.1778173 Jibson, R. W., & Keefer, D. K. (1993). Analysis of the seismic origin of landslides: Examples from the New Madrid seismic zone. Geological Society of America Bulletin, 105(4), 521–536. https://doi.org/10.1130/0016‐7606(1993)105<0521:AOTSOO>2.3.CO;2 Jibson, R. W., & Michael, J. A. (2009). Maps showing seismic landslide hazards in Anchorage, Alaska. U.S. Geological Survey Scientific Investigations Map 3077, 11 p., 2 sheets, scale 1:25,000. (Report No. 3077). https://doi.org/10.3133/sim3077 Jibson, R. W., Rathje, E. M., Jibson, M. W., & Lee, Y. W. (2013). SLAMMER: Seismic LAndslide Movement Modeled using Earthquake Records (ver.1.1, November 2014). U.S. Geological Survey Techniques and Methods, book 12, chapter B1, unpaged. Jin, Z., West, A. J., Zhang, F., An, Z., Hilton, R. G., Yu, J., et al. (2016). Seismically enhanced solute fluxes in the Yangtze River headwaters following the A.D. 2008 Wenchuan earthquake. Geology, 44(1), 47–50. https://doi.org/10.1130/G37246.1 Johnson, P. A., Zinszner, B., & Rasolofosaon, P. N. (1996). Resonance and elastic nonlinear phenomena in rock. Journal of Geophysical Research, 101(B5), 11,553–11,564. https://doi.org/10.1029/96JB00647 Jongmans, D., Bièvre, G., Renalier, F., Schwartz, S., Beaurez, N., & Orengo, Y. (2009). Geophysical investigation of a large landslide in glaciolacustrine clays in the Trièves area (French Alps). Engineering Geology, 109(1–2), 45–56. https://doi.org/10.1016/j. enggeo.2008.10.005 Jongmans, D., & Garambois, S. (2007). Geophysical investigation of landslides: A review. Bulletin de La Société Géologique de France, 178(2), 101–112. https://doi.org/10.2113/gssgfbull.178.2.101 Kaiser, A. E., Holden, C., & Massey, C. I. (2014). Site amplification, polarity and topographic effects in the Port Hills during the Canterbury earthquake sequence. In: GNS Science Consultancy Report 2014/121. Kamp, U., Growley, B. J., Khattak, G. A., & Owen, L. A. (2008). GIS‐based landslide susceptibility mapping for the 2005 Kashmir earthquake region. Geomorphology, 101(4), 631–642. https://doi.org/10.1016/j.geomorph.2008.03.003 Kappes, M. S., Keiler, M., von Elverfeldt, K., & Glade, T. (2012). Challenges of analyzing multi‐hazard risk: A review. Natural Hazards, 64(2), 1925–1958. https://doi.org/10.1007/s11069‐012‐0294‐2 Kargel, J. S., Leonard, G. J., Shugar, D. H., Haritashya, U. K., Bevington, A., Fielding, E. J., et al. (2016). Geomorphic and geologic controls of geohazards induced by Nepals 2015 Gorkha earthquake. Science, 351(6269), aac8353. https://doi.org/10.1126/science. aac8353 Karlin, R. E., & Abella, S. E. (1992). Paleoearthquakes in the Puget Sound region recorded in sediments from Lake Washington, USA. Science, 258(5088), 1617–1620. https://doi.org/10.1126/science.258.5088.1617 Katz, O., & Aharonov, E. (2006). Landslides in vibrating sand box: What controls types of slope failure and frequency magnitude relations? Earth and Planetary Science Letters, 247(3–4), 280–294. https://doi.org/10.1016/j.epsl.2006.05.009 Kawai, T. (1978). New discrete models and their application to seismic response analysis of structures. Nuclear Engineering and Design, 48(1), 207–229. https://doi.org/10.1016/0029‐5493(78)90217‐0 Kayen, R., Moss, R. E. S., Thompson, E. M., Seed, R. B., Cetin, K. O., Kiureghian, A. D., et al. (2013). Shear‐wave velocity‐based probabilistic and deterministic assessment of seismic soil liquefaction potential. Journal of Geotechnical and Geoenvironmental Engineering, 139(3), 407–419. https://doi.org/10.1061/(ASCE)GT.1943‐5606.0000743 Kaynia, A. M., Skurtveit, E., & Saygili, G. (2011). Real‐time mapping of earthquake‐induced landslides. Bulletin of Earthquake Engineering, 9(4), 955–973. https://doi.org/10.1007/s10518‐010‐9234‐2 Kean, J. W., Staley, D. M., & Cannon, S. H. (2011). In situ measurements of post‐fire debris flows in southern California: Comparisons of the timing and magnitude of 24 debris‐flow events with rainfall and soil moisture conditions. Journal of Geophysical Research, 116, F04019. https://doi.org/10.1029/2011JF002005 Keefer, D. K. (1984). Landslides caused by earthquakes. Geological Society of America Bulletin, 95(4), 406. https://doi.org/10.1130/0016‐ 7606(1984)95<406:LCBE>2.0.CO;2 Keefer, D. K. (1994). The importance of earthquake‐induced landslides to long‐term slope erosion and slope‐failure hazards in seismically active regions. In Geomorphology and Natural Hazards (pp. 265–284). Amsterdam, Netherlands: Elsevier. https://doi.org/ 10.1016/B978‐0‐444‐82012‐9.50022‐0 Keefer, D. K. (1999). Earthquake‐induced landslides and their effects on alluvial fans. Journal of Sedimentary Research, 69(1), 84–104. https://doi.org/10.2110/jsr.69.84 Keefer, D. K. (2002). Investigating landslides caused by earthquakes—A historical review. Surveys in Geophysics, 23(6), 473–510. https:// doi.org/10.1023/A:1021274710840 Kelsey, H. M. (1990). Late Quaternary deformation of marine terraces on the Cascadia Subduction Zone near Cape Blanco, Oregon. Tectonics, 9(5), 983–1014. https://doi.org/10.1029/TC009i005p00983 Khattak, G. A., Owen, L. A., Kamp, U., & Harp, E. L. (2010). Evolution of earthquake‐triggered landslides in the Kashmir Himalaya, northern Pakistan. Geomorphology, 115(1‐2), 102–108. https://doi.org/10.1016/j.geomorph.2009.09.035 Khazai, B., & Sitar, N. (2004). Evaluation of factors controlling earthquake‐induced landslides caused by Chi‐Chi earthquake and comparison with the Northridge and Loma Prieta events. Engineering Geology, 71(1–2), 79–95. https://doi.org/10.1016/S0013‐ 7952(03)00127‐3 King, G. C. P., Stein, R. S., & Rundle, J. B. (1988). The growth of geological structures by repeated earthquakes 1. Conceptual framework. Journal of Geophysical Research, 93, 13,307–13,318. https://doi.org/10.1029/JB093iB11p13307 King, J. P., Loveday, I., & Schuster, R. L. (1989). The 1985 Bairaman landslide dam and resulting debris flow, Papua New Guinea. Quarterly Journal of Engineering Geology and Hydrogeology, 22(4), 257–270. https://doi.org/10.1144/GSL.QJEG.1989.022.04.02 King, J. P., Loveday, I. C., & Schuster, R. L. (1987). Failure of a massive earthquake‐induced landslide dam in Papua New Guinea. Earthquakes & Volcanoes (USGS), 19(2), 40–47. Kirschbaum, D., & Stanley, T. (2018). Satellite‐based assessment of rainfall‐triggered landslide hazard for situational awareness. Earth's Future, 6(3), 505–523. https://doi.org/10.1002/2017EF000715 Klar, A., Aharonov, E., Kalderon‐Asael, B., & Katz, O. (2011). Analytical and observational relations between landslide volume and surface area. Journal of Geophysical Research, 116, F02001. https://doi.org/10.1029/2009JF001604 Kogure, T., & Okuda, Y. (2018). Monitoring the vertical distribution of rainfall‐induced strain changes in a landslide measured by distributed fiber optic sensing with Rayleigh backscattering. Geophysical Research Letters, 45,4033–4040. https://doi.org/10.1029/ 2018GL077607 Koi, T., Hotta, N., Ishigaki, I., Matuzaki, N., Uchiyama, Y., & Suzuki, M. (2008). Prolonged impact of earthquake‐induced landslides on sediment yield in a mountain watershed: The Tanzawa region, Japan. Geomorphology, 101(4), 692–702. https://doi.org/10.1016/j. geomorph.2008.03.007

FAN ET AL. 72 Reviews of Geophysics 10.1029/2018RG000626

Korup, O. (2002). Recent research on landslide dams—A literature review with special attention to New Zealand. Progress in Physical Geography, 26(2), 206–235. https://doi.org/10.1191/0309133302pp333ra Korup, O. (2004a). Geomorphometric characteristics of New Zealand landslide dams. Engineering Geology, 73(1–2), 13–35. https://doi.org/ 10.1016/j.enggeo.2003.11.003 Korup, O. (2004b). Landslide‐induced river channel avulsions in mountain catchments of southwest New Zealand. Geomorphology, 63(1–2), 57–80. https://doi.org/10.1016/j.geomorph.2004.03.005 Korup, O. (2005). Geomorphic hazard assessment of landslide dams in South Westland, New Zealand: Fundamental problems and approaches. Geomorphology, 66(1–4), 167–188. https://doi.org/10.1016/j.geomorph.2004.09.013 Korup, O. (2006a). Effects of large deep‐seated landslides on hillslope morphology, western Southern Alps, New Zealand. Journal of Geophysical Research, 111, F01018. https://doi.org/10.1029/2004JF000242 Korup, O. (2006b). Rock‐slope failure and the river long profile. Geology, 34(1), 45–48. https://doi.org/10.1130/G21959.1 Korup, O. (2012). Landslides and sediment flux: A quaternary perspective. Quaternary International, 279‐280, 253. https://doi.org/10.1016/ j.quaint.2012.08.608 Korup, O., & Clague, J. J. (2009). Natural hazards, extreme events, and mountain topography. Quaternary Science Reviews, 28(11–12), 977–990. https://doi.org/10.1016/j.quascirev.2009.02.021 Korup, O., Clague, J. J., Hermanns, R. L., Hewitt, K., Strom, A., & Weidinger, J. T. (2007). Giant landslides, topography, and erosion. Earth and Planetary Science Letters, 261(3–4), 578–589. https://doi.org/10.1016/j.epsl.2007.07.025 Korup, O., Densmore, A. L., & Schlunegger, F. (2010). The role of landslides in mountain range evolution. Geomorphology, 120(1–2), 77–90. https://doi.org/10.1016/j.geomorph.2009.09.017 Korup, O., Görüm, T., & Hayakawa, Y. (2012). Without power? Landslide inventories in the face of climate change. Earth Surface Processes and Landforms, 37(1), 92–99. https://doi.org/10.1002/esp.2248 Korup, O., & Tweed, F. (2007). Ice, moraine, and landslide dams in mountainous terrain. Quaternary Science Reviews, 26(25–28), 3406–3422 . https://doi.org/10.1016/j.quascirev.2007.10.012 Korup, O., & Wang, G. (2015). Multiple landslide‐damming episodes. In Landslide hazards, risks and disasters, (pp. 241–261). Amsterdam, Netherlands: Elsevier. https://doi.org/10.1016/B978‐0‐12‐396452‐6.00008‐2 Kramer, S. L. (1996). Geotechnical earthquake engineering. In Prentice–Hall international series in civil engineering and engineering mechanics (Chapter 6, pp. 226–228). New Jersey: Prentice‐Hall. Kremer, K., Simpson, G., & Girardclos, S. (2012). Giant Lake Geneva tsunami in AD 563. Nature Geoscience, 5(11), 756–757. https://doi.org/ 10.1038/ngeo1618 Kritikos, T., Robinson, T. R., & Davies, T. R. H. (2015). Regional coseismic landslide hazard assessment without historical landslide inventories: A new approach: Coseismic landslide hazard assessment. Journal of Geophysical Research: Earth Surface, 120,711–729. https://doi.org/10.1002/2014JF003224 Kwag, S., & Hahm, D. (2018). Development of an earthquake‐induced landslide risk assessment approach for nuclear power plants. Nuclear Engineering and Technology, 50(8), 1372–1386. https://doi.org/10.1016/j.net.2018.07.016 Kwan, J. S., & Sun, H. W. (2006). An improved landslide mobility model. Canadian Geotechnical Journal, 43(5), 531–539. https://doi.org/ 10.1139/t06‐010 Lacasse, S., Nadim, F., & HoΦeg, K. (2012). Risk assessment and mitigation in geo‐practice. In Geotechnical engineering state of the art and practice: Keynote lectures from GeoCongress 2012 (pp. 729–764). Langridge, R. M., Basili, R., Basher, L., & Wells, A. P. (2012). Late Holocene landscape change history related to the Alpine Fault determined from drowned forests in Lake Poerua, Westland, New Zealand. Natural Hazards and Earth System Sciences, 12(6), 2051–2064. https://doi.org/10.5194/nhess‐12‐2051‐2012 Larsen, I. J., Montgomery, D. R., & Korup, O. (2010). Landslide erosion controlled by hillslope material. Nature Geoscience, 3(4), 247–251. https://doi.org/10.1038/ngeo776 Lavé, J., & Burbank, D. (2004). Denudation processes and rates in the Transverse Ranges, southern California: Erosional response of a transitional landscape to external and anthropogenic forcing: DENUDATION IN THE TRANSVERSE RANGES. Journal of Geophysical Research, 109, F01006. https://doi.org/10.1029/2003JF000023 Lawrence, Z., Bodin, P., & Langston, C. A. (2009). In situ measurements of nonlinear and nonequilibrium dynamics in shallow, unconsolidated sediments. Bulletin of the Seismological Society of America, 99(3), 1650–1670. https://doi.org/10.1785/0120080177 Laxton, J., Serrano, J.‐J., & Tellez‐Arenas, A. (2010). Geological applications using geospatial standards—An example from OneGeology‐Europe and GeoSciML. International Journal of Digital Earth, 3(sup1), 31–49. https://doi.org/10.1080/ 17538941003636909 Lee, S.‐J., Chan, Y.‐C., Komatitsch, D., Huang, B.‐S., & Tromp, J. (2009). Effects of realistic surface topography on seismic ground motion in the Yangminshan region of Taiwan based upon the spectral‐element method and LiDAR DTM. Bulletin of the Seismological Society of America, 99(2A), 681–693. https://doi.org/10.1785/0120080264 Lee, S.‐J., Chen, H.‐W., Liu, Q., Komatitsch, D., Huang, B.‐S., & Tromp, J. (2008). Three‐dimensional simulations of seismic‐wave propagation in the Taipei basin with realistic topography based upon the spectral‐element method. Bulletin of the Seismological Society of America, 98(1), 253–264. https://doi.org/10.1785/0120070033 Lehmann, P. J., & von Rütte, D. (2017). STEP TRAMM: Landslide Triggering Model (version 2017.8). Software available at http://www. step.ethz.ch/step‐tramm.html. Leithold, E. L., Wegmann, K. W., Bohnenstiehl, D. R., Smith, S. G., Noren, A., & O'Grady, R. (2018). Slope failures within and upstream of Lake Quinault, Washington, as uneven responses to Holocene earthquakes along the Cascadia subduction zone. Quaternary Research, 89(1), 178–200. https://doi.org/10.1017/qua.2017.96 Leonard, M. (2010). Earthquake fault scaling: Self‐consistent relating of rupture length, width, average displacement, and moment release. Bulletin of the Seismological Society of America, 100(5A), 1971–1988. https://doi.org/10.1785/0120090189 Li, C., Ma, T., Zhu, X., & Li, W. (2011). The power–law relationship between landslide occurrence and rainfall level. Geomorphology, 130(3–4), 221–229. https://doi.org/10.1016/j.geomorph.2011.03.018 Li, G., West, A. J., Densmore, A. L., Hammond, D. E., Jin, Z., Zhang, F., et al. (2016). Connectivity of earthquake‐triggered landslides with the fluvial network: Implications for landslide sediment transport after the 2008 Wenchuan earthquake. Journal of Geophysical Research: Earth Surface, 121, 703–724. https://doi.org/10.1002/2015JF003718 Li, G., West, A. J., Densmore, A. L., Jin, Z., Parker, R. N., & Hilton, R. G. (2014). Seismic mountain building: Landslides associated with the 2008 Wenchuan earthquake in the context of a generalized model for earthquake volume balance. Geochemistry, Geophysics, Geosystems, 15, 833–844. https://doi.org/10.1002/2013GC005067

FAN ET AL. 73 Reviews of Geophysics 10.1029/2018RG000626

Li, G., West, A. J., Densmore, A. L., Jin, Z., Zhang, F., Wang, J., et al. (2017). Earthquakes drive focused denudation along a tectonically active mountain front. Earth and Planetary Science Letters, 472, 253–265. https://doi.org/10.1016/j.epsl.2017.04.040 Li, G., West, A. J., Densmore, A. L., Jin, Z., Zhang, F., Wang, J., & Hilton, R. G. (2018). Distribution of earthquake‐triggered landslides across landscapes: Towards understanding erosional agency and cascading hazards. In Y. G. Li (Ed.), Fault‐zone guided wave, ground motion, landslide and earthquake forecast, (pp. 160–190). Berlin, Boston: De Gruyter. https://doi.org/10.1515/ 9783110543551‐170 Li, M., Sung, R., Dong, J., Lee, C., & Chen, C. (2011). The formation and breaching of a short‐lived landslide dam at Hsiaolin Village, Taiwan—Part II: Simulation of debris flow with landslide dam breach. Engineering Geology, 123(1–2), 60–71. https://doi.org/10.1016/j. enggeo.2011.05.002 Li, Z., Jiao, Q., Liu, L., Tang, H., & Liu, T. (2014). Monitoring geologic hazards and vegetation recovery in the Wenchuan earthquake region using aerial photography. ISPRS International Journal of Geo‐Information, 3(1), 368–390. https://doi.org/10.3390/ijgi3010368 Liang, T., & Knappett, J. A. (2017). Centrifuge modelling of the influence of slope height on the seismic performance of rooted slopes. Géotechnique, 67(10), 1–15. https://doi.org/10.1680/jgeot.16.P.072 Lin, C.‐W., Liu, S.‐H., Lee, S.‐Y., & Liu, C.‐C. (2006). Impacts of the Chi‐Chi earthquake on subsequent rainfall‐induced landslides in central Taiwan. Engineering Geology, 86(2–3), 87–101. https://doi.org/10.1016/j.enggeo.2006.02.010 Lin, C.‐W., Shieh, C.‐L., Yuan, B.‐D., Shieh, Y.‐C., Liu, S.‐H., & Lee, S.‐Y. (2004). Impact of Chi‐Chi earthquake on the occurrence of landslides and debris flows: Example from the Chenyulan River watershed, Nantou, Taiwan. Engineering Geology, 71(1–2), 49–61. https://doi.org/10.1016/S0013‐7952(03)00125‐X Lin, J., & Whitman, R. V. (1986). Earthquake induced displacements of sliding blocks. Journal of Geotechnical Engineering, 112(1), 44–59. https://doi.org/10.1061/(ASCE)0733‐9410(1986)112:1(44) Lin, M.‐L., & Wang, K.‐L. (2006). Seismic slope behavior in a large‐scale shaking table model test. Engineering Geology, 86(2–3), 118–133. https://doi.org/10.1016/j.enggeo.2006.02.011 Lin, W.‐T., Lin, C.‐Y., & Chou, W.‐C. (2006). Assessment of vegetation recovery and soil erosion at landslides caused by a catastrophic earthquake: A case study in Central Taiwan. Ecological Engineering, 28(1), 79–89. https://doi.org/10.1016/j.ecoleng.2006.04.005 Linde, A. T., & Sacks, I. S. (1998). Triggering of volcanic eruptions. Nature, 395(6705), 888–890. https://doi.org/10.1038/27650 Ling, S. (2015). Landslide damming in Western Sichuan Province, China, with special reference to the 1786 Dadu River and 1933 Diexi events (Master's thesis, University of Waterloo). Liu, F. Z., Xu, Q., Dong, X. J., Yu, B., Frost, J. D., & Li, H. J. (2017). Design and performance of a novel multi‐function debris flow mitigation system in Wenjia Gully, Sichuan. Landslides, 14(6), 2089–2104. https://doi.org/10.1007/s10346‐017‐0849‐0 Liu, H., Xu, Q., Li, Y., & Fan, X. (2013). Response of high‐strength rock slope to seismic waves in a shaking table test. Bulletin of the Seismological Society of America, 103(6), 3012–3025. https://doi.org/10.1785/0120130055 Liu, J., Shi, J., Wang, T., & Wu, S. (2018). Seismic landslide hazard assessment in the Tianshui area, China, based on scenario earthquakes. Bulletin of Engineering Geology and the Environment, 77(3), 1263–1272. https://doi.org/10.1007/s10064‐016‐0998‐8 Liu, N., Chen, Z., Zhang, J., Lin, W., Chen, W., & Xu, W. (2010). Draining the Tangjiashan Barrier Lake. Journal of Hydraulic Engineering, 136(11), 914–923. https://doi.org/10.1061/(ASCE)HY.1943‐7900.0000241 Liu, Y., Liu, R., & Ge, Q. (2010). Evaluating the vegetation destruction and recovery of Wenchuan earthquake using MODIS data. Natural Hazards, 54(3), 851–862. https://doi.org/10.1007/s11069‐010‐9511‐z Liu, Z., Nadim, F., Garcia‐Aristizabal, A., Mignan, A., Fleming, K., & Luna, B. Q. (2015). A three‐level framework for multi‐risk assessment. Georisk: Assessment and management of risk for engineered systems and geohazards, 9(2), 59–74. Liucci, L., Melelli, L., Suteanu, C., & Ponziani, F. (2017). The role of topography in the scaling distribution of landslide areas: A cellular automata modeling approach. Geomorphology, 290, 236–249. https://doi.org/10.1016/j.geomorph.2017.04.017 Lo, W.‐C., Tsao, T.‐C., & Hsu, C.‐H. (2012). Building vulnerability to debris flows in Taiwan: A preliminary study. Natural Hazards, 64(3), 2107–2128. https://doi.org/10.1007/s11069‐012‐0124‐6 Lombardo, L., Bakka, H., Tanyas, H., van Westen, C., Mai, P. M., & Huser, R. (2018). Geostatistical modeling to capture seismic‐shaking patterns from earthquake‐induced landslides. ArXiv Preprint ArXiv:1807.08513. Louie, J. N. (2001). Faster, better: Shear‐wave velocity to 100 meters depth from refraction microtremor arrays. Bulletin of the Seismological Society of America, 91(2), 347–364. https://doi.org/10.1785/0120000098 Luna, B. Q., Blahut, J., van Asch, T., van Westen, C., & Kappes, M. (2016). ASCHFLOW—A dynamic landslide run‐out model for medium scale hazard analysis. Geoenvironmental Disasters, 3(1), 29. https://doi.org/10.1186/s40677‐016‐0064‐7 MacInnes, B. T., Bourgeois, J., Pinegina, T. K., & Kravchunovskaya, E. A. (2009). Tsunami geomorphology: Erosion and deposition from the 15 November 2006 Kuril Island tsunami. Geology, 37(11), 995–998. https://doi.org/10.1130/G30172A.1 Makdisi, F. L., & Seed, H. B. (1978). Simplified procedure for estimating dam and embankment earthquake‐induced deformations. ASCE Journal of the Geotechnical Engineering Division, 104, 849–867. Malamud, B. D., Turcotte, D. L., Guzzetti, F., & Reichenbach, P. (2004). Landslides, earthquakes, and erosion. Earth and Planetary Science Letters, 229(1–2), 45–59. https://doi.org/10.1016/j.epsl.2004.10.018 Manga, M., & Brodsky, E. (2006). Seismic triggering of eruptions in the far field: Volcanoes and geysers. Annual Review of Earth and Planetary Sciences, 34(1), 263–291. https://doi.org/10.1146/annurev.earth.34.031405.125125 Manga, M., & Wang, C.‐Y. (2007). Earthquake hydrology. In Treatise on geophysics, (pp. 293–320). Amsterdam: Elsevier. https://doi.org/ 10.1016/B978‐044452748‐6.00074‐2 Manville, V. (2001). Techniques for evaluating the size of potential dam‐break floods from natural dams. Institute of Geological & Nuclear Sciences. Marc, O., & Hovius, N. (2015). Amalgamation in landslide maps: Effects and automatic detection. Natural Hazards and Earth System Sciences, 15(4), 723–733. https://doi.org/10.5194/nhess‐15‐723‐2015 Marc, O., Hovius, N., & Meunier, P. (2016). The mass balance of earthquakes and earthquake sequences. Geophysical Research Letters, 43, 3708–3716. https://doi.org/10.1002/2016GL068333 Marc, O., Hovius, N., Meunier, P., Gorum, T., & Uchida, T. (2016). A seismologically consistent expression for the total area and volume of earthquake‐triggered landsliding. Journal of Geophysical Research: Earth Surface, 121, 640–663. https://doi.org/10.1002/ 2015JF003732 Marc, O., Hovius, N., Meunier, P., Uchida, T., & Hayashi, S. (2015). Transient changes of landslide rates after earthquakes. Geology, 43(10), 883–886. https://doi.org/10.1130/G36961.1 Marc, O., Meunier, P., & Hovius, N. (2017). Prediction of the area affected by earthquake‐induced landsliding based on seismological parameters. Natural Hazards and Earth System Sciences, 17(7), 1159–1175. https://doi.org/10.5194/nhess‐17‐1159‐2017

FAN ET AL. 74 Reviews of Geophysics 10.1029/2018RG000626

Marcuson, W. F. (1981). Moderator's report for session on earth dams and stability of slopes under dynamic loads. In Proc., Int. Conf. on recent adv. In Geotech. Earthquake Engrg. And Soil Dyn., Univ. of Missori (Vol. 3). Martha, T. R., Kerle, N., Jetten, V., van Westen, C. J., & Kumar, K. V. (2010). Characterising spectral, spatial and morphometric properties of landslides for semi‐automatic detection using object‐oriented methods. Geomorphology, 116(1–2), 24–36. https://doi.org/10.1016/j. geomorph.2009.10.004 Martha, T. R., Roy, P., Mazumdar, R., Govindharaj, K. B., & Kumar, K. V. (2017). Spatial characteristics of landslides triggered by the 2015 Mw 7.8 (Gorkha) and Mw 7.3 (Dolakha) earthquakes in Nepal. Landslides, 14(2), 697–704. https://doi.org/10.1007/s10346‐016‐0763‐x Marzocchi, W., Garcia‐Aristizabal, A., Gasparini, P., Mastellone, M. L., & Di Ruocco, A. (2012). Basic principles of multi‐risk assessment: A case study in Italy. Natural Hazards, 62(2), 551–573. https://doi.org/10.1007/s11069‐012‐0092‐x Massey, C., Della Pasqua, F., Holden, C., Kaiser, A., Richards, L., Wartman, J., et al. (2017). Rock slope response to strong earthquake shaking. Landslides, 14(1), 249–268. https://doi.org/10.1007/s10346‐016‐0684‐8 Massey, C., Townsend, D., Rathje, E., Allstadt, K. E., Lukovic, B., Kaneko, Y., et al. (2018). Landslides triggered by the 14 November 2016 M w 7.8 Kaikōura earthquake, New Zealand. Bulletin of the Seismological Society of America, 108(3B), 1630–1648. https://doi.org/10.1785/ 0120170305 Mather, A. E., Hartley, A. J., & Griffiths, J. S. (2014). The giant coastal landslides of Northern Chile: Tectonic and climate interactions on a classic convergent plate margin. Earth and Planetary Science Letters, 388, 249–256. https://doi.org/10.1016/j.epsl.2013.10.019 Maufroy, E., Cruz‐Atienza, V. M., Cotton, F., & Gaffet, S. (2015). Frequency‐scaled curvature as a proxy for topographic site‐effect amplification and ground‐motion variability. Bulletin of the Seismological Society of America, 105(1), 354–367. https://doi.org/10.1785/ 0120140089 Meade, B. J. (2010). The signature of an unbalanced earthquake cycle in Himalayan topography? Geology, 38(11), 987–990. https://doi.org/ 10.1130/G31439.1 Mergili, M., Marchesini, I., Rossi, M., Guzzetti, F., & Fellin, W. (2014). Spatially distributed three‐dimensional slope stability modelling in a raster GIS. Geomorphology, 206, 178–195. https://doi.org/10.1016/j.geomorph.2013.10.008 Méric, O., Garambois, S., Malet, J.‐P., Cadet, H., Guéguen, P., & Jongmans, D. (2007). Seismic noise‐based methods for soft‐rock landslide characterization. Bulletin de La Societe Geologique de France, 178(2), 137–148. https://doi.org/10.2113/gssgfbull.178.2.137 Metzger, M. J., Bunce, R. G. H., Jongman, R. H. G., Sayre, R., Trabucco, A., & Zomer, R. (2013). A high‐resolution bioclimate map of the world: A unifying framework for global biodiversity research and monitoring. Global Ecology and Biogeography, 22(5), 630–638. https:// doi.org/10.1111/geb.12022 Meunier, P., Hovius, N., & Haines, A. J. (2007). Regional patterns of earthquake‐triggered landslides and their relation to ground motion. Geophysical Research Letters, 34, L20408. https://doi.org/10.1029/2007GL031337 Meunier, P., Hovius, N., & Haines, J. A. (2008). Topographic site effects and the location of earthquake induced landslides. Earth and Planetary Science Letters, 275(3–4), 221–232. https://doi.org/10.1016/j.epsl.2008.07.020 Meunier, P., Uchida, T., & Hovius, N. (2013). Landslide patterns reveal the sources of large earthquakes. Earth and Planetary Science Letters, 363,27–33. https://doi.org/10.1016/j.epsl.2012.12.018 Miles, S. B., & Keefer, D. K. (2000). Evaluation of seismic slope‐performance models using a regional case study. Environmental & Engineering Geoscience, 6(1), 25–39. https://doi.org/10.2113/gseegeosci.6.1.25 Miles, S. B., & Keefer, D. K. (2001). Seismic landslide hazard for the city of Berkeley. California, U.S. Geological Survey. (Report No. 2378). https://doi.org/10.3133/mf2378 Miles, S. B., & Keefer, D. K. (2007). Comprehensive areal model of earthquake‐induced landslides: technical specification and user guide. U.S. Geological Survey. (Report No. 2007–1072). https://doi.org/10.3133/ofr20071072 Miles, S. B., & Keefer, D. K. (2009). Evaluation of CAMEL—Comprehensive areal model of earthquake‐induced landslides. Engineering Geology, 104(1–2), 1–15. https://doi.org/10.1016/j.enggeo.2008.08.004 Mitchell, A. R., & Griffiths, D. F. (1980). The finite difference method in partial differential equations. New York: John Wiley & Sons Led. Moernaut, J., Daele, M. V., Heirman, K., Fontijn, K., Strasser, M., Pino, M., et al. (2014). Lacustrine turbidites as a tool for quantitative earthquake reconstruction: New evidence for a variable rupture mode in south central Chile. Journal of Geophysical Research: Solid Earth, 119, 1607–1633. https://doi.org/10.1002/2013JB010738 Molnar, P. (2012). Isostasy can't be ignored. Nature Geoscience, 5(2), 83–83. https://doi.org/10.1038/ngeo1383 Montgomery, D. R. (2001). Slope distributions, threshold hillslopes, and steady‐state topography. American Journal of Science, 301(4–5), 432–454. https://doi.org/10.2475/ajs.301.4‐5.432 Montgomery, D. R., & Dietrich, W. E. (1994). A physically based model for the topographic control on shallow landsliding. Water Resources Research, 30(4), 1153–1171. https://doi.org/10.1029/93WR02979 Moore, J. R., Gischig, V., Burjanek, J., Loew, S., & Fah, D. (2011). Site Effects in Unstable Rock Slopes: Dynamic Behavior of the Randa Instability (Switzerland). Bulletin of the Seismological Society of America, 101(6), 3110–3116. https://doi.org/10.1785/ 0120110127 Moosavi, V., Talebi, A., & Shirmohammadi, B. (2014). Producing a landslide inventory map using pixel‐based and object‐oriented approaches optimized by Taguchi method. Geomorphology, 204, 646–656. https://doi.org/10.1016/j.geomorph.2013.09.012 Mreyen, A.‐S., Micu, M., Onaca, A., Cerfontaine, P., & Havenith, H.‐B. (2017). Integrated geological‐geophysical models of unstable slopes in seismic areas. In M. Mikos, B. Tiwari, Y. Yin, & K. Sassa (Eds.), Advancing culture of living with landslides, (pp. 269–279). Cham: Springer International Publishing. https://doi.org/10.1007/978‐3‐319‐53498‐5_31 Nadim, F., Kjekstad, O., Peduzzi, P., Herold, C., & Jaedicke, C. (2006). Global landslide and avalanche hotspots. Landslides, 3(2), 159–173. https://doi.org/10.1007/s10346‐006‐0036‐1 Nakamura, Y. (1989). A method for dynamic characteristics estimation of subsurface using microtremor on the ground surface. Railway Technical Research Institute, Quarterly Reports, 30(1). Newmark, N. M. (1965). Effects of earthquakes on dams and embankments. Geotechnique, 15(2), 139–160. https://doi.org/10.1680/ geot.1965.15.2.139 Nichol, J., & Wong, M. S. (2005). Satellite remote sensing for detailed landslide inventories using change detection and image fusion. International Journal of Remote Sensing, 26(9), 1913–1926. https://doi.org/10.1080/01431160512331314047 Nogoshi, M., & Igarashi, T. (1971). On the amplitude characteristics of microtremor (Part 2). Zisin (Journal of the Seismological Society of Japan. 2nd Ser.), 24(1), 26–40. https://doi.org/10.4294/zisin1948.24.1_26 Nowicki Jessee, M. A., Hamburger, M. W., Allstadt, K., Wald, D. J., Robeson, S. M., Tanyas, H., et al. (2018). A global empirical model for near real‐time assessment of seismically induced landslides. Journal of Geophysical Research: Earth Surface, 123, 1835–1859. https://doi. org/10.1029/2017JF004494

FAN ET AL. 75 Reviews of Geophysics 10.1029/2018RG000626

Nowicki, M. A., Wald, D. J., Hamburger, M. W., Hearne, M., & Thompson, E. M. (2014). Development of a globally applicable model for near real‐time prediction of seismically induced landslides. Engineering Geology, 173,54–65. https://doi.org/10.1016/j. enggeo.2014.02.002 Nunziata, C., De Nisco, G., & Panza, G. F. (2009). S‐waves profiles from noise cross correlation at small scale. Engineering Geology, 105(3–4), 161–170. https://doi.org/10.1016/j.enggeo.2009.01.005 O'Connor, J. E., & Costa, J. E. (2004). The world's largest floods, past and present: Their causes and magnitudes. U.S. Geological Survey. Circular 1254, 13. https://pubs.usgs.gov/circ/2004/circ1254/ Oglesby, D. D., Archuleta, R. J., & Nielsen, S. B. (2000). The three‐dimensional dynamics of dipping faults. Bulletin of the Seismological Society of America, 90(3), 616–628. https://doi.org/10.1785/0119990113 Ohori, M., Nobata, A., & Wakamatsu, K. (2002). A comparison of ESAC and FK methods of estimating phase velocity using arbitrarily shaped microtremor arrays. Bulletin of the Seismological Society of America, 92(6), 2323–2332. https://doi.org/10.1785/0119980109 Okamoto, T., Sakurai, M., Tsuchiya, S., Yoshimatsu, H., Ogawa, K., & Wang, G. (2013). Secondary hazards associated with coseismic landslide. In K. Ugai, H. Yagi, & A. Wakai (Eds.), Earthquake‐induced landslides, (pp. 77–82). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978‐3‐642‐32238‐9_8 Oldham, R. D. (1899). Report of the great earthquake of 12th June, 1897. Office of the Geological survey. Olivares, L., & Damiano, E. (2007). Postfailure mechanics of landslides: Laboratory investigation of flowslides in pyroclastic soils. Journal of Geotechnical and Geoenvironmental Engineering, 133(1), 51–62. https://doi.org/10.1061/(ASCE)1090‐0241(2007)133:1(51) O'Loughlin, C. L. (1985). Effects of forest land use on erosion and slope stability: Report of a seminar. Owen, L. A., Kamp, U., Khattak, G. A., Harp, E. L., Keefer, D. K., & Bauer, M. A. (2008). Landslides triggered by the 8 October 2005 Kashmir earthquake. Geomorphology, 94(1–2), 1–9. https://doi.org/10.1016/j.geomorph.2007.04.007 Pack, R. T., Tarboton, D. G., Goodwin, C. N., & Prasad, A. (1998). SINMAP, a stability index approach to terrain stability hazard mapping. SINMAP User's Manual, Terratech Consulting Ltd. Pain, C. F., & Bowler, J. M. (1973). Denudation following the November 1970 earthquake at Madang, Papua New Guinea. Zeitschrift Fur Geomorphologie, Supp. Bd, 18,91–104. Pánek, T., Korup, O., Minár, J., & Hradecký, J. (2016). Giant landslides and highstands of the Caspian Sea. Geology, 44(11), 939–942. https:// doi.org/10.1130/G38259.1 Papadopoulos, G. A., & Plessa, A. (2000). Magnitude–distance relations for earthquake‐induced landslides in Greece. Engineering Geology, 58(3–4), 377–386. https://doi.org/10.1016/S0013‐7952(00)00043‐0 Papathoma‐Köhle, M., Kappes, M., Keiler, M., & Glade, T. (2011). Physical vulnerability assessment for alpine hazards: State of the art and future needs. Natural Hazards, 58(2), 645–680. https://doi.org/10.1007/s11069‐010‐9632‐4 Parker, R. N., Densmore, A. L., Rosser, N. J., de Michele, M., Li, Y., Huang, R., et al. (2011). Mass wasting triggered by the 2008 Wenchuan earthquake is greater than orogenic growth. Nature Geoscience, 4(7), 449–452. https://doi.org/10.1038/ngeo1154 Parker, R. N., Hancox, G. T., Petley, D. N., Massey, C. I., Densmore, A. L., & Rosser, N. J. (2015). Spatial distributions of earthquake‐induced landslides and hillslope preconditioning in the northwest South Island, New Zealand. Earth Surface Dynamics, 3(4), 501–525. https:// doi.org/10.5194/esurf‐3‐501‐2015 Parker, R. N., Rosser, N. J., & Hales, T. C. (2017). Spatial prediction of earthquake‐induced landslide probability. Natural Hazards and Earth System Sciences Discussions, 1–29. https://doi.org/10.5194/nhess‐2017‐193 Passalacqua, P., Do Trung, T., Foufoula‐Georgiou, E., Sapiro, G., & Dietrich, W. E. (2010). A geometric framework for channel network extraction from lidar: Nonlinear diffusion and geodesic paths. Journal of Geophysical Research, 115, F01002. https://doi.org/10.1029/ 2009JF001254 Pearce, A. J., & Watson, A. J. (1983). Medium‐term effects of two landsliding episodes on channel storage of sediment. Earth Surface Processes and Landforms, 8(1), 29–39. https://doi.org/10.1002/esp.3290080104 Pearce, A. J., & Watson, A. J. (1986). Effects of earthquake‐induced landslides on sediment budget and transport over a 50‐yr period. Geology, 14(1), 52. https://doi.org/10.1130/0091‐7613(1986)14<52:EOELOS>2.0.CO;2 Peduto, D., Ferlisi, S., Nicodemo, G., Reale, D., Pisciotta, G., & Gullà, G. (2017). Empirical fragility and vulnerability curves for buildings exposed to slow‐moving landslides at medium and large scales. Landslides, 14(6), 1993–2007. https://doi.org/10.1007/s10346‐017‐0826‐7 Pelletier, J. D., Malamud, B. D., Blodgett, T., & Turcotte, D. L. (1997). Scale‐invariance of soil moisture variability and its implications for the frequency‐size distribution of landslides. Engineering Geology, 48(3–4), 255–268. https://doi.org/10.1016/S0013‐7952(97)00041‐0 Pellicani, R., & Spilotro, G. (2015). Evaluating the quality of landslide inventory maps: Comparison between archive and surveyed inventories for the Daunia region (Apulia, Southern Italy). Bulletin of Engineering Geology and the Environment, 74(2), 357–367. https:// doi.org/10.1007/s10064‐014‐0639‐z Peng, M., & Zhang, L. M. (2012a). Analysis of human risks due to dam break floods—Part 2: Application to Tangjiashan landslide dam failure. Natural Hazards, 64(2), 1899–1923. https://doi.org/10.1007/s11069‐012‐0336‐9 Peng, M., & Zhang, L. M. (2012b). Analysis of human risks due to dam‐break floods—Part 1: A new model based on Bayesian networks. Natural Hazards, 64(1), 903–933. https://doi.org/10.1007/s11069‐012‐0275‐5 Peng, M., & Zhang, L. M. (2013a). Dynamic decision making for dam‐break emergency management—Part 1: Theoretical framework. Natural Hazards and Earth System Sciences, 13(2), 425–437. https://doi.org/10.5194/nhess‐13‐425‐2013 Peng, M., & Zhang, L. M. (2013b). Dynamic decision making for dam‐break emergency management—Part 2: Application to Tangjiashan landslide dam failure. Natural Hazards and Earth System Sciences, 13(2), 439–454. https://doi.org/10.5194/nhess‐13‐439‐2013 Petley, D. N. (2011). Building resilience to landslides in mountain communities. https://blogs.agu.org/landslideblog/2011/10/12/building‐ resilience‐to‐landslides‐in‐mountain‐communities/. Petley, D. N. (2012). Global patterns of loss of life from landslides. Geology, 40(10), 927–930. https://doi.org/10.1130/G33217.1 Piciullo, L., Gariano, S. L., Melillo, M., Brunetti, M. T., Peruccacci, S., Guzzetti, F., & Calvello, M. (2017). Definition and performance of a threshold‐based regional early warning model for rainfall‐induced landslides. Landslides, 14(3), 995–1008. https://doi.org/10.1007/ s10346‐016‐0750‐2 Podolskiy, E. A., Chambon, G., Naaim, M., & Gaume, J. (2015). Evaluating snow weak‐layer failure parameters through inverse finite element modelling of shaking‐platform experiments. Natural Hazards and Earth System Sciences, 15(1), 119–134. https://doi.org/ 10.5194/nhess‐15‐119‐2015 Podolskiy, E. A., Nishimura, K., Abe, O., & Chernous, P. A. (2010a). Earthquake‐induced snow avalanches: I. Historical case studies. Journal of Glaciology, 56(197), 431–446. https://doi.org/10.3189/002214310792447815 Podolskiy, E. A., Nishimura, K., Abe, O., & Chernous, P. A. (2010b). Earthquake‐induced snow avalanches: II. Experimental study. Journal of Glaciology, 56(197), 447–458. https://doi.org/10.3189/002214310792447833

FAN ET AL. 76 Reviews of Geophysics 10.1029/2018RG000626

Post, A. S. (1967). Effects of the March 1964 Alaska earthquake on glaciers. U.S. Geol. Survey Prof. Paper 544‐D, p. D1‐D42. Pradel, D., Smith, P. M., Stewart, J. P., & Raad, G. (2005). Case history of landslide movement during the Northridge Earthquake. Journal of Geotechnical and Geoenvironmental Engineering, 131(11), 1360–1369. https://doi.org/10.1061/(ASCE)1090‐0241(2005)131:11(1360) Prestininzi, A., & Romeo, R. (2000). Earthquake‐induced ground failures in Italy. Engineering Geology, 58(3–4), 387–397. https://doi.org/ 10.1016/S0013‐7952(00)00044‐2 Prevost, J. H. (1981). DYNAFLOW: A nonlinear transient finite element analysis program. Department of Civil and Environmental Engineering. Princeton, NJ: Princeton University. Price, D. G. (1995). Weathering and weathering processes. Quarterly Journal of Engineering Geology and Hydrogeology, 28(3), 243–252. https://doi.org/10.1144/GSL.QJEGH.1995.028.P3.03 Qi, L., Wang, Z., & Wang, X. (2012). Erosion and sediment budget of the 2008 Wenchuan earthquake: A case study on Mianyuan River basin. Natural Science, 04(08), 640–647. https://doi.org/10.4236/ns.2012.428084 Rai, M., Rodriguez‐Marek, A., & Yong, A. (2016). An empirical model to predict topographic effects in strong ground motion using California small‐to medium‐magnitude earthquake database. Earthquake Spectra, 32(2), 1033–1054. https://doi.org/10.1193/ 113014EQS202M Ramos Scharrón, C. E., Castellanos, E. J., & Restrepo, C. (2012). The transfer of modern organic carbon by landslide activity in tropical montane ecosystems. Journal of Geophysical Research, 117, G03016. https://doi.org/10.1029/2011JG001838 Rathje, E. M., & Bray, J. D. (2000). Nonlinear coupled seismic sliding analysis of Earth structures. Journal of Geotechnical and Geoenvironmental Engineering, 126(11), 1002–1014. https://doi.org/10.1061/(ASCE)1090‐0241(2000)126:11(1002) Reichenbach, P., Cardinali, M., De Vita, P., & Guzzetti, F. (1998). Regional hydrological thresholds for landslides and floods in the Tiber River Basin (central Italy). Environmental Geology, 35(2–3), 146–159. https://doi.org/10.1007/s002540050301 Reichenbach, P., Rossi, M., Malamud, B. D., Mihir, M., & Guzzetti, F. (2018). A review of statistically‐based landslide susceptibility models. Earth‐Science Reviews, 180,60–91. https://doi.org/10.1016/j.earscirev.2018.03.001 Reid, M. E., Christian, S. B., Brien, D. L., & Henderson, S. (2015). Scoops3D—Software to analyze three‐dimensional slope stability throughout a digital landscape. U.S. Geological Survey Techniques and Methods 14‐A1, 218. https://doi.org/10.3133/tm14A1 Ren, Z., Zhang, Z., & Yin, J. (2017). Erosion associated with seismically‐induced landslides in the Middle Longmen Shan Region, Eastern Tibetan Plateau, China. Remote Sensing, 9(8), 864. https://doi.org/10.3390/rs9080864 Renaud, G., Rivière, J., Larmat, C., Rutledge, J. T., Lee, R. C., Guyer, R. A., et al. (2014). In situ characterization of shallow elastic nonlinear parameters with dynamic acoustoelastic testing. Journal of Geophysical Research: Solid Earth, 119, 6907–6923. https://doi.org/10.1002/ 2013JB010625 Revelle, R., & Suess, H. E. (1957). Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric CO2 during the past decades. Tellus, 9(1), 18–27. https://doi.org/10.3402/tellusa.v9i1.9075 Rial, J. A. (1996). The anomalous seismic response of the ground at the Tarzana Hill site during the Northridge 1994 Southern California earthquake: a resonant, sliding block? Bulletin of the Seismological Society of America, 86, 1714–1723. Richmond, B. M., Buckley, M., Etienne, S., Chagué‐Goff, C., Clark, K., Goff, J., et al. (2011). Deposits, flow characteristics, and landscape change resulting from the September 2009 South Pacific tsunami in the Samoan islands. Earth‐Science Reviews, 107(1–2), 38–51. https:// doi.org/10.1016/j.earscirev.2011.03.008 Rivière, J., Shokouhi, P., Guyer, R. A., & Johnson, P. A. (2015). A set of measures for the systematic classification of the nonlinear elastic behavior of disparate rocks. Journal of Geophysical Research: Solid Earth, 120, 1587–1604. https://doi.org/10.1002/ 2014JB011718 Rizzitano, S., Cascone, E., & Biondi, G. (2014). Coupling of topographic and stratigraphic effects on seismic response of slopes through 2D linear and equivalent linear analyses. Soil Dynamics and Earthquake Engineering, 67,66–84. https://doi.org/10.1016/j. soildyn.2014.09.003 Roback, K., Clark, M. K., West, A. J., Zekkos, D., Li, G., Gallen, S. F., et al. (2018). The size, distribution, and mobility of landslides caused by the 2015 M w 7.8 Gorkha earthquake, Nepal. Geomorphology, 301, 121–138. https://doi.org/10.1016/j.geomorph.2017.01.030 Roberts, N. J., & Evans, S. G. (2013). The gigantic Seymareh (Saidmarreh) rock avalanche, Zagros Fold–Thrust Belt, Iran. Journal of the Geological Society, 170(4), 685–700. https://doi.org/10.1144/jgs2012‐090 Robinson, T. R., Rosser, N. J., Densmore, A. L., Williams, J. G., Kincey, M. E., Benjamin, J., & Bell, H. J. A. (2017). Rapid post‐earthquake modelling of coseismic landslide intensity and distribution for emergency response decision support. Natural Hazards and Earth System Sciences, 17(9), 1521–1540. https://doi.org/10.5194/nhess‐17‐1521‐2017 Rodrıguez, C. E., Bommer, J. J., & Chandler, R. J. (1999). Earthquake‐induced landslides: 1980–1997. Soil Dynamics and Earthquake Engineering, 18(5), 325–346. https://doi.org/10.1016/S0267‐7261(99)00012‐3 Rodriguez‐Peces, M. J., García, J., Azañón, J. M., & Jabaloy, A. (2011). Regional hazard assessment of earthquake‐triggered slope instabilities considering site effects and seismic scenarios in Lorca Basin (Spain). Environmental and Engineering Geoscience, 17(2), 183–196. https://doi.org/10.2113/gseegeosci.17.2.183 Romeo, R. (2000). Seismically induced landslide displacements: A predictive model. Engineering Geology, 58(3–4), 337–351. https://doi.org/ 10.1016/S0013‐7952(00)00042‐9 Rosi, A., Segoni, S., Catani, F., & Casagli, N. (2012). Statistical and environmental analyses for the definition of a regional rainfall threshold system for landslide triggering in Tuscany (Italy). Journal of Geographical Sciences, 22(4), 617–629. https://doi.org/10.1007/s11442‐012‐ 0951‐0 Rosser, B. J., Townsend, D., McSaveney, M., & Ries, W. (2014). Landslides and ground damage associated with the M 6.2 Eketahuna earthquake, 20 January 2014, GNS Science Report. Rossi, M., Cardinali, M., Fiorucci, F., Marchesini, I., Mondini, A., Santangelo, M., et al. (2012). A tool for the estimation of the distribution of landslide area in R (Vol. 14, p. 9438). Presented at the EGU General Assembly Conference Abstracts. Rotunno, A. F., Callari, C., & Froiio, F. (2017). Computational modeling of backward erosion piping. In Models, simulation, and experimental issues in structural mechanics, (pp. 225–234). Cham: Springer. https://doi.org/10.1007/978‐3‐319‐48884‐4_12 Royden, L. H., Burchfiel, B. C., King, R. W., Wang, E., Chen, Z., Shen, F., & Liu, Y. (1997). Surface deformation and lower crustal flow in Eastern Tibet. Science, 276(5313), 788–790. https://doi.org/10.1126/science.276.5313.788 Ruiz, S., Kausel, E., Campos, J., Saragoni, G. R., & Madariaga, R. (2011). Identification of high frequency pulses from earthquake asperities along Chilean subduction zone using strong motion. Pure and Applied Geophysics, 168(1–2), 125–139. https://doi.org/10.1007/s00024‐ 010‐0117‐x Saade, A., Abou‐Jaoude, G., & Wartman, J. (2016). Regional‐scale co‐seismic landslide assessment using limit equilibrium analysis. Engineering Geology, 204,53–64. https://doi.org/10.1016/j.enggeo.2016.02.004

FAN ET AL. 77 Reviews of Geophysics 10.1029/2018RG000626

Saba, S. B., van der Meijde, M., & van der Werff, H. (2010). Spatiotemporal landslide detection for the 2005 Kashmir earthquake region. Geomorphology, 124(1–2), 17–25. https://doi.org/10.1016/j.geomorph.2010.07.026 Saito, H., Korup, O., Uchida, T., Hayashi, S., & Oguchi, T. (2014). Rainfall conditions, typhoon frequency, and contemporary landslide erosion in Japan. Geology, 42(11), 999–1002. https://doi.org/10.1130/G35680.1 Salzmann, N., Kääb, A., Huggel, C., Allgöwer, B., & Haeberli, W. (2004). Assessment of the hazard potential of ice avalanches using remote sensing and GIS‐modelling. Norsk Geografisk Tidsskrift‐Norwegian Journal of Geography, 58(2), 74–84. https://doi.org/10.1080/ 00291950410006805 Sassa, K., Fukuoka, H., Wang, F., & Wang, G. (2007). Landslides induced by a combined effect of earthquake and rainfall. In Progress in landslide science, (pp. 193–207). Berlin, Heidelberg: Springer. https://doi.org/10.1007/978‐3‐540‐70965‐7_14 Sato, H. P., Hasegawa, H., Fujiwara, S., Tobita, M., Koarai, M., Une, H., & Iwahashi, J. (2007). Interpretation of landslide distribution triggered by the 2005 Northern Pakistan earthquake using SPOT 5 imagery. Landslides, 4(2), 113–122. https://doi.org/10.1007/s10346‐ 006‐0069‐5 Sawazaki, K., & Snieder, R. (2013). Time‐lapse changes of P‐and S‐wave velocities and shear wave splitting in the first year after the 2011 Tohoku earthquake, Japan: Shallow subsurface. Geophysical Journal International, 193(1), 238–251. https://doi.org/10.1093/gji/ ggs080 Sayre, R., Dangermond, J., Frye, C., Vaughan, R., Aniello, P., Breyer, S., et al. (2014). A new map of global ecological land units—An eco- physiographic stratification approach. Washington, DC: Association of American Geographers. Scaringi, G., Fan, X., Xu, Q., Liu, C., Ouyang, C., Domènech, G., et al. (2018). Some considerations on the use of numerical methods to simulate past landslides and possible new failures: The case of the recent Xinmo landslide (Sichuan, China). Landslides, 15(7), 1359–1375. https://doi.org/10.1007/s10346‐018‐0953‐9 Schelhorn, S.‐J., Herfort, B., Leiner, R., Zipf, A., & de Albuquerque, J. P. (2014). Identifying elements at risk from OpenStreetMap: The case of flooding. In International Conference on Information Systems for Crisis Response and Management (Vol. 11, pp. 508–512). USA: The Pennsylvania State University. Schmandt, B., Aster, R. C., Scherler, D., Tsai, V. C., & Karlstrom, K. (2013). Multiple fluvial processes detected by riverside seismic and infrasound monitoring of a controlled flood in the Grand Canyon. Geophysical Research Letters, 40, 4858–4863. https://doi.org/10.1002/ grl.50953 Schmidt, J., Evans, I. S., & Brinkmann, J. (2003). Comparison of polynomial models for land surface curvature calculation. International Journal of Geographical Information Science, 17(8), 797–814. https://doi.org/10.1080/13658810310001596058 Schmitt, R. G., Tanyas, H., Jessee, M. A. N., Zhu, J., Biegel, K. M., Allstadt, K. E., et al. (2017). An open repository of earthquake‐triggered ground‐failure inventories (ver 2.0, December 2018), U.S. Geological Survey data release, [accessed January 20, 2019], at https://doi.org/ 10.5066/F7H70DB4. Schneider, J. F. (2009). Seismically reactivated Hattian slide in Kashmir, Northern Pakistan. Journal of Seismology, 13(3), 387–398. https:// doi.org/10.1007/s10950‐008‐9103‐5 Schnellmann, M., Anselmetti, F. S., Giardini, D., McKenzie, J. A., & Ward, S. N. (2002). Prehistoric earthquake history revealed by lacustrine slump deposits. Geology, 30(12), 1131–1134. https://doi.org/10.1130/0091‐7613(2002)030<1131:PEHRBL>2.0.CO;2 Schöpa, A., Wei‐An, C., Lipovsky, B. P., Hovius, N., White, R. S., Green, R. G., & Turowski, J. M. (2018). Dynamics of the Askja caldera July 2014 landslide, Iceland, from seismic signalanalysis: precursor, motion and aftermath. Earth Surface Dynamics, 6(2), 467–485. Schuster, R. L. (1993). Landslide dams: Hazards and mitigation. Geografia Fisica e Dinamica Quaternaria, 16,17–19. Schuster, R. L., Wieczorek, G. F., & Hope, D. G. II (1998). Landslide dams in Santa Cruz County, California, resulting from the earthquake. U.S. Geological Survey Professional Paper, 1551,51–70. Schwanghart, W., Bernhardt, A., Stolle, A., Hoelzmann, P., Adhikari, B. R., Andermann, C., et al. (2016). Repeated catastrophic valley infill following medieval earthquakes in the Nepal Himalaya. Science, 351(6269), 147–150. https://doi.org/10.1126/science. aac9865 Scott, C. P., Allmendinger, R. W., González, G., & Loveless, J. P. (2016). Coseismic extension from surface cracks reopened by the 2014 Pisagua, northern Chile, earthquake sequence. Geology, 44(5), 387–390. https://doi.org/10.1130/G37662.1 Seed, H. B. (1979). Considerations in the earthquake‐resistant design of earth and rockfill dams. Géotechnique, 29(3), 215–263. https://doi. org/10.1680/geot.1979.29.3.215 Seed, H. B., & Martin, G. R. (1966). The seismic coefficient in earth dam design. ASCE Journal of the Soil Mechanics and Foundations Division, 92,25–58. Segoni, S., Piciullo, L., & Gariano, S. L. (2018). A review of the recent literature on rainfall thresholds for landslide occurrence. Landslides, 15(8), 1483–1501. https://doi.org/10.1007/s10346‐018‐0966‐4 Segoni, S., Piciullo, L., Gariano, S. L., Bogaard, T. A., & Catani, F. (2017). Special issue: Landslide early warning systems: monitoring systems, rainfall thresholds, warning models, performance evaluation and risk perception. Natural Hazards and Earth System Sciences. Retrieved from https://www.nat‐hazards‐earth‐syst‐sci.net/special_issue896.html Sepúlveda, S. A., Murphy, W., Jibson, R. W., & Petley, D. N. (2005). Seismically induced rock slope failures resulting from topographic amplification of strong ground motions: The case of Pacoima Canyon, California. Engineering Geology, 80(3–4), 336–348. https://doi.org/ 10.1016/j.enggeo.2005.07.004 Shen, L., Xu, C., & Liu, L. (2016). Interaction among controlling factors for landslides triggered by the 2008 Wenchuan, China Mw 7.9 earthquake. Frontiers of Earth Science, 10(2), 264–273. https://doi.org/10.1007/s11707‐015‐0517‐4 Shen, P., Zhang, L. M., Chen, H. X., & Gao, L. (2017). Role of vegetation restoration in mitigating hillslope erosion and debris flows. Engineering Geology, 216, 122–133. https://doi.org/10.1016/j.enggeo.2016.11.019 Shen, Z.‐K., Sun, J., Zhang, P., Wan, Y., Wang, M., Bürgmann, R., et al. (2009). Slip maxima at fault junctions and rupturing of barriers during the 2008 Wenchuan earthquake. Nature Geoscience, 2(10), 718–724. https://doi.org/10.1038/ngeo636 Shi, B., Anooshehpoor, A., Brune, J. N., & Zeng, Y. (1998). Dynamics of thrust faulting: 2D lattice model. Bulletin of the Seismological Society of America, 88(6), 1484–1494. Shi, G. H., & Goodman, R. E. (1985). Two dimensional discontinuous deformation analysis. International Journal for Numerical and Analytical Methods in Geomechanics, 9(6), 541–556. https://doi.org/10.1002/nag.1610090604 Shi, Z. M., Guan, S. G., Peng, M., Zhang, L. M., Zhu, Y., & Cai, Q. P. (2015). Cascading breaching of the Tangjiashan landslide dam and two smaller downstream landslide dams. Engineering Geology, 193, 445–458. https://doi.org/10.1016/j.enggeo.2015.05.021 Shi, Z. M., Xiong, X., Peng, M., Zhang, L. M., Xiong, Y. F., Chen, H. X., & Zhu, Y. (2017). Risk assessment and mitigation for the Hongshiyan landslide dam triggered by the 2014 Ludian earthquake in Yunnan, China. Landslides, 14(1), 269–285. https://doi.org/10.1007/s10346‐ 016‐0699‐1

FAN ET AL. 78 Reviews of Geophysics 10.1029/2018RG000626

Shieh, C.‐L., Chen, Y. S., Tsai, Y. J., & Wu, J. H. (2009). Variability in rainfall threshold for debris flow after the Chi‐Chi earthquake in central Taiwan, China. International Journal of Sediment Research, 24(2), 177–188. https://doi.org/10.1016/S1001‐6279(09)60025‐1 Shields, A. (1936). Anwendung der Aehnlichkeitsmechanik und der Turbulenzforschung auf die Geschiebebewegung. PhD Thesis Technical University Berlin. Shirley, W. L., Boruff, B. J., & Cutter, S. L. (2012). Social vulnerability to environmental hazards. In Hazards vulnerability and environmental justice (pp. 143–160). Abingdon, UK: Routledge. Shoaei, Z. (2014). Mechanism of the giant Seimareh Landslide, Iran, and the longevity of its landslide dams. Environmental Earth Sciences, 72(7), 2411–2422. https://doi.org/10.1007/s12665‐014‐3150‐8 Shou, K. J., Hong, C. Y., Wu, C. C., Hsu, H. Y., Fei, L. Y., Lee, J. F., & Wei, C. Y. (2011). Spatial and temporal analysis of landslides in Central Taiwan after 1999 Chi‐Chi earthquake. Engineering Geology, 123(1‐2), 122–128. https://doi.org/10.1016/j.enggeo.2011.03.014 Shrestha, S., Kang, T.‐S., & Choi, J. C. (2018). Assessment of co‐seismic landslide susceptibility using LR and ANCOVA in Barpak region, Nepal. Journal of Earth System Science, 127(3), 38. https://doi.org/10.1007/s12040‐018‐0936‐1 Simonett, D. S. (1967). Landslide distribution and earthquakes in the Bavani and Torricelli Mountains, New Guinea. Landform Studies from Australia and New Guinea, 64–84. Simpson, G. (2015). Accumulation of permanent deformation during earthquake cycles on reverse faults. Journal of Geophysical Research: Solid Earth, 120, 1958–1974. https://doi.org/10.1002/2014JB011442 Sims, J. D. (1973). Earthquake‐induced structures in sediments of Van Norman Lake, San Fernando, California. Science, 182(4108), 161–163. https://doi.org/10.1126/science.182.4108.161 Singh, V. P. (1996). Dam breach modelling technology, (Vol. 17). Dordrecht, Boston, London: Kluwer Academic Publishers. https://doi.org/ 10.1007/978‐94‐015‐8747‐1 Soeters, R., & van Westen, C. (1996). Slope instability recognition, analysis and zonation. Landslides: Investigation and Mitigation, 247, 129–177. Song, Y., Gong, J., Gao, S., Wang, D., Cui, T., Li, Y., & Wei, B. (2012). Susceptibility assessment of earthquake‐induced landslides using Bayesian network: A case study in Beichuan, China. Computers & Geosciences, 42,189–199. https://doi.org/10.1016/j. cageo.2011.09.011 Sotiris, V., George, P., & Spyros, P. (2016). Map of co‐seismic landslides and surface ruptures for the M 7.8 Kaikoura, New Zealand Earthquake, http://eqgeogr.weebly.com/blog/map‐of‐co‐seismic‐landslides‐surface‐ruptures‐for‐the‐m‐78‐kaikoura‐new‐zealand‐ earthquake‐version‐2. Retrieved from http://eqgeogr.weebly.com/blog/map‐of‐co‐seismic‐landslides‐surface‐ruptures‐for‐the‐m‐78‐ kaikoura‐new‐zealand‐earthquake‐version‐2 Spudich, P., Hellweg, M., & Lee, W. H. K. (1996). Directional topographic site response at Tarzana observed in aftershocks of the 1994 Northridge, California, earthquake: Implications for mainshock motions. Bulletin of the Seismological Society of America, 86(1B), S193–S208. Staley, D. M., Kean, J. W., Cannon, S. H., Schmidt, K. M., & Laber, J. L. (2013). Objective definition of rainfall intensity–duration thresholds for the initiation of post‐fire debris flows in southern California. Landslides, 10(5), 547–562. https://doi.org/10.1007/ s10346‐012‐0341‐9 Stanley, T., & Kirschbaum, D. B. (2017). A heuristic approach to global landslide susceptibility mapping. Natural Hazards, 87(1), 145–164. https://doi.org/10.1007/s11069‐017‐2757‐y Stark, C. P., & Guzzetti, F. (2009). Landslide rupture and the probability distribution of mobilized debris volumes. Journal of Geophysical Research, 114, F00A02. https://doi.org/10.1029/2008JF001008 Stark, C. P., & Hovius, N. (2001). The characterization of landslide size distributions. Geophysical Research Letters, 28(6), 1091–1094. https://doi.org/10.1029/2000GL008527 Steer, P., Simoes, M., Cattin, R., & Shyu, J. B. H. (2014). Erosion influences the seismicity of active thrust faults. Nature Communications, 5(1), 5564. https://doi.org/10.1038/ncomms6564 Stolle, A., Schwanghart, W., Andermann, C., Bernhardt, A., Fort, M., Jansen, J. D., et al. (2019). Protracted river response to medieval earthquakes. Earth Surface Processes and Landforms, 44(1), 331–341. https://doi.org/10.1002/esp.4517 Strenk, P. M., & Wartman, J. (2011). Uncertainty in seismic slope deformation model predictions. Engineering Geology, 122(1–2), 61–72. https://doi.org/10.1016/j.enggeo.2011.03.003 Strom, A. (2015). Natural river damming: Climate‐driven or seismically induced phenomena: Basics for landslide and seismic hazard assessment. In Engineering geology for society and territory‐volume, (Vol. 2, pp. 33–41). Cham: Springer. https://doi.org/10.1007/978‐3‐ 319‐09057‐3_3 Stumpf, A., & Kerle, N. (2011). Object‐oriented mapping of landslides using Random Forests. Remote Sensing of Environment, 115(10), 2564–2577. https://doi.org/10.1016/j.rse.2011.05.013 Sugai, T., Ohmori, H., & Hirano, M. (1995). Rock control on magnitude‐frequency distribution of landslide. In International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts (Vol. 3, p. 116A). Sun, W., Tian, Y., Mu, X., Zhai, J., Gao, P., & Zhao, G. (2017). Loess landslide inventory map based on GF‐1 satellite imagery. Remote Sensing, 9(4), 314. https://doi.org/10.3390/rs9040314 Syvitski, J. P., & Schafer, C. T. (1996). Evidence for an earthquake‐triggered basin collapse in Saguenay Fjord, Canada. Sedimentary Geology, 104(1–4), 127–153. https://doi.org/10.1016/0037‐0738(95)00125‐5 Taboada‐Urtuzuastegui, V., Martinez‐Ramirez, G., & Abdoun, T. (2002). Centrifuge modeling of seismic behavior of a slope in liquefiable soil. Soil Dynamics and Earthquake Engineering, 22(9–12), 1043–1049. https://doi.org/10.1016/S0267‐7261(02)00129‐X Tacconi Stefanelli, C., Segoni, S., Casagli, N., & Catani, F. (2016). Geomorphic indexing of landslide dams evolution. Engineering Geology, 208,1–10. https://doi.org/10.1016/j.enggeo.2016.04.024 Tachikawa, T., Kaku, M., Iwasaki, A., Gesch, D. B., Oimoen, M. J., Zhang, Z., et al. (2011). ASTER global digital elevation model version 2‐ summary of validation results. NASA. Tadono, T., Nagai, H., Ishida, H., Oda, F., Naito, S., Minakawa, K., & Iwamoto, H. (2016). Generation of the 30 m‐mesh global digital surface model by ALOS Prism. International Archives of the Photogrammetry, Remote Sensing & Spatial Information Sciences, 41. Tang, C., Tanyas, H., van Westen, C. J., Tang, C., Fan, X., & Jetten, V. G. (2019). Analysing post‐earthquake mass movement volume dynamics with multi‐source DEMs. Engineering Geology, 248,89–101. https://doi.org/10.1016/j.enggeo.2018.11.010 Tang, C., van Westen, C. J., Tanyas, H., & Jetten, V. G. (2016). Analysing post‐earthquake landslide activity using multi‐temporal landslide inventories near the epicentral area of the 2008 Wenchuan earthquake. Natural Hazards and Earth System Sciences, 16(12), 2641–2655. https://doi.org/10.5194/nhess‐16‐2641‐2016

FAN ET AL. 79 Reviews of Geophysics 10.1029/2018RG000626

Tang, C., Zhu, J., Ding, J., Cui, X. F., Chen, L., & Zhang, J. S. (2011). Catastrophic debris flows triggered by a 14 August 2010 rainfall at the epicenter of the Wenchuan earthquake. Landslides, 8(4), 485–497. https://doi.org/10.1007/s10346‐011‐0269‐5 Tang, C., Zhu, J., Li, W. L., & Liang, J. T. (2009). Rainfall‐triggered debris flows following the Wenchuan earthquake. Bulletin of Engineering Geology and the Environment, 68(2), 187–194. https://doi.org/10.1007/s10064‐009‐0201‐6 Tang, R., Fan, X., Scaringi, G., Xu, Q., van Westen, C. J., Ren, J., & Havenith, H.‐B. (2018). Distinctive controls on the distribution of river‐ damming and non‐damming landslides induced by the 2008 Wenchuan earthquake. Bulletin of Engineering Geology and the Environment,1–19. https://doi.org/10.1007/s10064‐018‐1381‐8 Tanyas, H., Allstadt, K. E., & van Westen, C. J. (2018). An updated method for estimating landslide‐event magnitude. Earth Surface Processes and Landforms, 43(9), 1836–1847. https://doi.org/10.1002/esp.4359 Tanyas, H., Rossi, M., Alvioli, M., van Westen, C. J., & Marchesini, I. (2019). A global slope unit‐based method for the near real‐time prediction of earthquake‐induced landslides. Geomorphology, 327, 126–146. https://doi.org/10.1016/j.geomorph.2018.10.022 Tanyas, H., van Westen, C. J., Allstadt, K. E., Anna Nowicki Jessee, M., Görüm, T., Jibson, R. W., et al. (2017). Presentation and analysis of a worldwide database of earthquake‐induced landslide inventories: Earthquake‐induced landslide inventories. Journal of Geophysical Research: Earth Surface, 122, 1991–2015. https://doi.org/10.1002/2017JF004236 Tanyas, H., van Westen, C. J., Allstadt, K. E., & Jibson, R. W. (2018). Factors controlling landslide frequency–area distributions. Earth Surface Processes and Landforms, 44(4), 900–917. Tatard, L., & Grasso, J. R. (2013). Controls of earthquake faulting style on near field landslide triggering: The role of coseismic slip: Role of coseismic slip on landsliding. Journal of Geophysical Research: Solid Earth, 118, 2953–2964. https://doi.org/10.1002/jgrb.50215 ten Brink, U., Barkan, R., Andrews, B. D., & Chaytor, J. (2009). Size distributions and failure initiation of submarine and subaerial landslides. Earth and Planetary Science Letters, 287(1–2), 31–42. https://doi.org/10.1016/j.epsl.2009.07.031 Terzaghi, K. (1950). Mechanism of landslides. In S. Paige (Ed.), Application of geology to engineering practice, (pp. 83–123). New York, N. Y: Geological Society of America. https://doi.org/10.1130/Berkey.1950.83 Tiwari, B., Ajmera, B., & Dhital, S. (2017). Characteristics of moderate‐ to large‐scale landslides triggered by the M w 7.8 2015 Gorkha earthquake and its aftershocks. Landslides, 14(4), 1297–1318. https://doi.org/10.1007/s10346‐016‐0789‐0 Tolorza, V., Carretier, S., Sepúlveda, S. A., Serey, A., Moya, S., Escobar, P., & Pinto, L. (2016). Limited to null erosive response to the Mw 8.8 Maule Earthquake derived from variations in the sediment concentration of the rivers. In EGU General Assembly Conference Abstracts (Vol. 18, p. 906). Torgoev, A., Lamair, L., Torgoev, I., & Havenith, H.‐B. (2013). A review of recent case studies of landslides investigated in the Tien Shan using microseismic and other geophysical methods. In K. Ugai, H. Yagi, & A. Wakai (Eds.), Earthquake‐induced landslides, (pp. 285–294 ). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978‐3‐642‐32238‐9_29 Torres, M. A., West, A. J., & Li, G. (2014). Sulphide oxidation and carbonate dissolution as a source of CO2 over geological timescales. Nature, 507(7492), 346–349. https://doi.org/10.1038/nature13030 Towhata, I. (2013). Long‐term effects of earthquake‐induced slope failures. Proceedings of the 7th International Conference on Case Histories in Geotechnical Engineering, Chicago 10, http://scholarsmine.mst.edu/icchge/7icchge/session16/10. Tran, D. K., Prime, N., Froiio, F., Callari, C., & Vincens, E. (2017). Numerical modelling of backward front propagation in piping erosion by DEM‐LBM coupling. European Journal of Environmental and Civil Engineering, 21(7–8), 960–987. https://doi.org/10.1080/ 19648189.2016.1248794 Tropeano, G., Silvestri, F., & Ausilio, E. (2017). An uncoupled procedure for performance assessment of slopes in seismic conditions. Bulletin of Earthquake Engineering, 15(9), 3611–3637. https://doi.org/10.1007/s10518‐017‐0113‐y Tseng, C.‐M., Lin, C.‐W., Stark, C. P., Liu, J.‐K., Fei, L.‐Y., & Hsieh, Y.‐C. (2013). Application of a multi‐temporal, LiDAR‐derived, digital terrain model in a landslide‐volume estimation. Earth Surface Processes and Landforms, 38(13), 1587–1601. Turnbull, J. M., & Davies, T. R. H. (2006). A mass movement origin for cirques. Earth Surface Processes and Landforms, 31(9), 1129–1148. https://doi.org/10.1002/esp.1324 Tuthill, S. J., & Laird, W. M. (1966). Geomorphic effects of the earthquake of March 27, 1964, in the Martin‐Bering rivers area, Alaska. US Government Printing Office. UNDRO. (1984). Disaster prevention and mitigation—A compendium of current knowledge. Preparedness Aspects, 11. UNISDR. (2017). Terminology on disaster risk reduction. Report of the open‐ended intergovernmental expert working group on indicators and terminology relating to disaster risk reduction. https://www.preventionweb.net/files/50683_oiewgreportenglish.pdf 41 p. Valagussa, A., Marc, O., Frattini, P., & Crosta, G. (2019). Seismic and geological controls on earthquake‐induced landslide size. Earth and Planetary Science Letters, 506, 268–281. https://doi.org/10.1016/j.epsl.2018.11.005 van Asch, T. W. J., Tang, C., Alkema, D., Zhu, J., & Zhou, W. (2014). An integrated model to assess critical rainfall thresholds for run‐out distances of debris flows. Natural Hazards, 70(1), 299–311. https://doi.org/10.1007/s11069‐013‐0810‐z van Beek, L. P. H. (2002). Assessment of the influence of changes in land use and climate on landslide activity in a Mediterranean environment (Doctoral dissertation). Retrieved from Utrecht University Repository. (https://dspace.library.uu.nl/handle/1874/560). Van Daele, M., Versteeg, W., Pino, M., Urrutia, R., & De Batist, M. (2013). Widespread deformation of basin‐plain sediments in Aysén fjord (Chile) due to impact by earthquake‐triggered, onshore‐generated mass movements. Marine Geology, 337,67–79. https://doi.org/ 10.1016/j.margeo.2013.01.006 Van Den Eeckhaut, M., Poesen, J., Govers, G., Verstraeten, G., & Demoulin, A. (2007). Characteristics of the size distribution of recent and historical landslides in a populated hilly region. Earth and Planetary Science Letters, 256(3–4), 588–603. https://doi.org/10.1016/j. epsl.2007.01.040 Van Dissen, R., McSaveney, M., Townsend, D., Hancox, G., Little, T., Ries, W., et al. (2013). Landslides and liquefaction generated by the Cook Strait and Lake Grassmere earthquakes: A reconnaissance report. Bulletin of the New Zealand Society for Earthquake Engineering, 46(4), 196–200. van Westen, C. J., Castellanos, E., & Kuriakose, S. L. (2008). Spatial data for landslide susceptibility, hazard, and vulnerability assessment: An overview. Engineering Geology, 102(3–4), 112–131. https://doi.org/10.1016/j.enggeo.2008.03.010 van Westen, C. J., & Greiving, S. (2017). Multi‐hazard risk assessment and decision making. In Environmental Hazards Methodologies for Risk Assessment and Management, (pp. 31–94). London: IWA Publishing. https://doi.org/10.2166/9781780407135_0031 van Westen, C. J., Greiving, S., & Dalezios, N. R. (2017). Environmental hazards methodologies for risk assessment and management. In Environmental Hazards Methodologies for Risk Assessment and Management, (pp. 31–94). IWA Publishing, London. https://doi.org/ 10.2166/9781780407135_0031 Vanmaercke, M., Ardizzone, F., Rossi, M., & Guzzetti, F. (2017). Exploring the effects of seismicity on landslides and catchment sediment yield: An Italian case study. Geomorphology, 278, 171–183. https://doi.org/10.1016/j.geomorph.2016.11.010

FAN ET AL. 80 Reviews of Geophysics 10.1029/2018RG000626

Vanmaercke, M., Obreja, F., & Poesen, J. (2014). Seismic controls on contemporary sediment export in the Siret river catchment, Romania. Geomorphology, 216, 247–262. https://doi.org/10.1016/j.geomorph.2014.04.008 Varnes, D. J. (1984). Landslide hazard zonation: A review of principles and practice. Vennari, C., Gariano, S. L., Antronico, L., Brunetti, M. T., Iovine, G., Peruccacci, S., et al. (2014). Rainfall thresholds for shallow landslide occurrence in Calabria, southern Italy. Natural Hazards and Earth System Sciences, 14(2), 317–330. https://doi.org/10.5194/nhess‐14‐ 317‐2014 von Ruette, J., Lehmann, P., & Or, D. (2014). Effects of rainfall spatial variability and intermittency on shallow landslide triggering patterns at a catchment scale. Water Resources Research, 50, 7780–7799. https://doi.org/10.1002/2013WR015122 Von Voigtlander, J., Clark, M. K., Zekkos, D., Greenwood, W. W., Anderson, S. P., Anderson, R. S., & Godt, J. W. (2018). Strong variation in weathering of layered rock maintains hillslope‐scale strength under high precipitation: Influence of weathering on seismic velocities. Earth Surface Processes and Landforms, 43(6), 1183–1194. https://doi.org/10.1002/esp.4290 Wald, D. J. (2013). Adding secondary hazard and ground‐truth observations to PAGER's loss modeling. In ISCRAM. Wald, D. J., Quitoriano, V., Heaton, T. H., Kanamori, H., Scrivner, C. W., & Worden, C. B. (1999). TriNet “ShakeMaps”: Rapid generation of peak ground motion and intensity maps for earthquakes in southern California. Earthquake Spectra, 15(3), 537–555. https://doi.org/ 10.1193/1.1586057 Wald, D. J., Wald, L., Worden, B., & Goltz, J. (2003). ShakeMap, a tool for earthquake response. Walder, J. S., & O'Connor, J. E. (1997). Methods for predicting peak discharge of floods caused by failure of natural and constructed earthen dams. Water Resources Research, 33(10), 2337–2348. https://doi.org/10.1029/97WR01616 Waldmann, N., Anselmetti, F. S., Ariztegui, D., Austin, J. A. Jr., Pirouz, M., Moy, C. M., & Dunbar, R. (2011). Holocene mass‐wasting events in Lago Fagnano, Tierra del Fuego (54 S): Implications for paleoseismicity of the Magallanes‐Fagnano transform fault. Basin Research, 23(2), 171–190. https://doi.org/10.1111/j.1365‐2117.2010.00489.x Wang, C., Chin, C.‐V., Cheng, L.‐H., & Yu, S.‐B. (2001). Coseismic hydrologic response of an alluvial fan to the 1999 Chi‐Chi earthquake, Taiwan. Geology, 29(9), 831–834. https://doi.org/10.1130/0091‐7613(2001)029<0831:CHROAA>2.0.CO;2 Wang, E., Kirby, E., Furlong, K. P., van Soest, M., Xu, G., Shi, X., et al. (2012). Two‐phase growth of high topography in eastern Tibet during the Cenozoic. Nature Geoscience, 5(9), 640–645. https://doi.org/10.1038/ngeo1538 Wang, G., Furuya, G., Zhang, F., Doi, I., Watanabe, N., Wakai, A., & Marui, H. (2016). Layered internal structure and breaching risk assessment of the Higashi‐Takezawa landslide dam in Niigata, Japan. Geomorphology, 267,48–58. https://doi.org/10.1016/j. geomorph.2016.05.021 Wang, G., Huang, R., Kamai, T., & Zhang, F. (2013). The internal structure of a rockslide dam induced by the 2008 Wenchuan (Mw7.9) earthquake, China. Engineering Geology, 156,28–36. https://doi.org/10.1016/j.enggeo.2013.01.004 Wang, G., & Sassa, K. (2001). Factors affecting rainfall‐induced flowslides in laboratory flume tests. Geotechnique, 51(7), 587–599. https:// doi.org/10.1680/geot.2001.51.7.587 Wang, G., & Sassa, K. (2003). Pore‐pressure generation and movement of rainfall‐induced landslides: Effects of grain size and fine‐particle content. Engineering Geology, 69(1–2), 109–125. https://doi.org/10.1016/S0013‐7952(02)00268‐5 Wang, G., & Sassa, K. (2009). Seismic loading impacts on excess pore‐water pressure maintain landslide triggered flowslides. Earth Surface Processes and Landforms, 34(2), 232–241. https://doi.org/10.1002/esp.1708 Wang, H. B., Sassa, K., & Xu, W. Y. (2007). Analysis of a spatial distribution of landslides triggered by the 2004 Chuetsu earthquakes of Niigata Prefecture, Japan. Natural Hazards, 41(1), 43–60. https://doi.org/10.1007/s11069‐006‐9009‐x Wang, J., Jin, Z., Hilton, R. G., Zhang, F., Densmore, A. L., Li, G., & West, A. J. (2015). Controls on fluvial evacuation of sediment from earthquake‐triggered landslides. Geology, 43(2), 115–118. https://doi.org/10.1130/G36157.1 Wang, J., Jin, Z., Hilton, R. G., Zhang, F., Li, G., Densmore, A. L., et al. (2016). Earthquake‐triggered increase in biospheric carbon export from a mountain belt. Geology, 44(6), 471–474. https://doi.org/10.1130/G37533.1 Wang, J., Xiang, W., & Lu, N. (2014). Landsliding triggered by reservoir operation: A general conceptual model with a case study at Three Gorges Reservoir. Acta Geotechnica, 9(5), 771–788. https://doi.org/10.1007/s11440‐014‐0315‐2 Wang, K., & Lin, M. (2011). Initiation and displacement of landslide induced by earthquake—A study of shaking table model slope test. Engineering Geology, 122(1–2), 106–114. https://doi.org/10.1016/j.enggeo.2011.04.008 Wang, W., Godard, V., Liu‐Zeng, J., Scherler, D., Xu, C., Zhang, J., et al., & ASTER Team (2017). Perturbation of fluvial sediment fluxes following the 2008 Wenchuan earthquake: Fluvial sediment fluxes following large‐magnitude earthquake. Earth Surface Processes and Landforms, 42(15), 2611–2622. https://doi.org/10.1002/esp.4210 Wang, W., Nakamura, H., Tsuchiya, S., & Chen, C. (2002). Distributions of landslides triggered by the Chi‐chi Earthquake in Central Taiwan on September 21, 1999. Landslides ‐ Journal of the Japan Landslide Society, 38(4), 318–326. Wang, Y., Song, C., Lin, Q., & Li, J. (2016). Occurrence probability assessment of earthquake‐triggered landslides with Newmark displacement values and logistic regression: The Wenchuan earthquake, China. Geomorphology, 258, 108–119. https://doi.org/10.1016/j. geomorph.2016.01.004 Wang, Y., Zou, Z., & Li, J. (2015). Influencing factors of households disadvantaged in post‐earthquake life recovery: A case study of the Wenchuan earthquake in China. Natural Hazards, 75(2), 1853–1869. https://doi.org/10.1007/s11069‐014‐1400‐4 Warrick, J. A., & Mertes, L. A. K. (2009). Sediment yield from the tectonically active semiarid Western Transverse Ranges of California. GSA Bulletin, 121(7–8), 1054–1070. https://doi.org/10.1130/B26452.1 Wartman, J., Seed, R. B., & Bray, J. D. (2005). Shaking table modeling of seismically induced deformations in slopes. Journal of Geotechnical and Geoenvironmental Engineering, 131(5), 610–622. https://doi.org/10.1061/(ASCE)1090‐0241(2005)131:5(610) Wasowski, J., Keefer, D. K., & Lee, C.‐T. (2011). Toward the next generation of research on earthquake‐induced landslides: Current issues and future challenges. Engineering Geology, 122(1‐2), 1–8. https://doi.org/10.1016/j.enggeo.2011.06.001 Watson, F. G., Vertessy, R. A., & Grayson, R. B. (1999). Large‐scale modelling of forest hydrological processes and their long‐term effect on water yield. Hydrological Processes, 13(5), 689–700. https://doi.org/10.1002/(SICI)1099‐1085(19990415)13:5<689::AID‐HYP773>3.0. CO;2‐D Wegscheider, S., Schneiderhan, T., Mager, A., Zwenzner, H., Post, J., & Strunz, G. (2013). Rapid mapping in support of emergency response after earthquake events. Natural Hazards, 68(1), 181–195. https://doi.org/10.1007/s11069‐013‐0589‐y Wei, X., Chen, N., Cheng, Q., He, N., Deng, M., & Tanoli, J. I. (2014). Long‐term activity of earthquake‐induced landslides: A case study from Qionghai Lake Basin, Southwest of China. Journal of Mountain Science, 11(3), 607–624. https://doi.org/10.1007/s11629‐013‐ 2970‐4 Weidinger, J. T. (2011). Stability and life span of landslide dams in the Himalayas (India, Nepal) and the Qin Ling Mountains (China). In Natural and Artificial Rockslide Dams, (pp. 243–277). Berlin, Heidelberg: Springer. https://doi.org/10.1007/978‐3‐642‐04764‐0_8

FAN ET AL. 81 Reviews of Geophysics 10.1029/2018RG000626

Weidinger, J. T., Korup, O., Munack, H., Altenberger, U., Dunning, S. A., Tippelt, G., & Lottermoser, W. (2014). Giant rockslides from the inside. Earth and Planetary Science Letters, 389,62–73. https://doi.org/10.1016/j.epsl.2013.12.017 Weidinger, J. T., Wang, J., & Ma, N. (2002). The earthquake‐triggered rock avalanche of Cui Hua, Qin Ling Mountains, PR of China—The benefits of a lake‐damming prehistoric natural disaster. Quaternary International, 93, 207–214. West, A. J., Hetzel, R., Li, G., Jin, Z., Zhang, F., Hilton, R. G., & Densmore, A. L. (2014). Dilution of 10Be in detrital quartz by earthquake‐ induced landslides: Implications for determining denudation rates and potential to provide insights into landslide sediment dynamics. Earth and Planetary Science Letters, 396, 143–153. https://doi.org/10.1016/j.epsl.2014.03.058 West, A. J., Lin, C.‐W., Lin, T.‐C., Hilton, R. G., Liu, S.‐H., Chang, C.‐T., et al. (2011). Mobilization and transport of coarse woody debris to the oceans triggered by an extreme tropical storm. Limnology and Oceanography, 56(1), 77–85. https://doi.org/10.4319/lo.2011.56.1.0077 Whalley, W. (1984). High altitude rock weathering processes. The International Karakoram Project, 1, 365–373. White, E. P., Enquist, B. J., & Green, J. L. (2008). On estimating the exponent of power‐law frequency distributions. Ecology, 89(4), 905–912. https://doi.org/10.1890/07‐1288.1 Wieczorek, G. F., Morgan, B. A., & Campbell, R. H. (2000). Debris‐flow hazards in the Blue Ridge of central Virginia. Environmental & Engineering Geoscience, 6(1), 3–23. https://doi.org/10.2113/gseegeosci.6.1.3 Wieczorek, G. F., Wilson, R. C., & Harp, E. L. (1985). Map showing slope stability during earthquakes in San Mateo County, California. Williams, J. G., Rosser, N. J., Hardy, R. J., Brain, M. J., & Afana, A. A. (2018). Optimising 4‐D surface change detection: An approach for capturing rockfall magnitude–frequency. Earth Surface Dynamics, 6(1), 101–119. https://doi.org/10.5194/esurf‐6‐101‐2018 Williams, J. G., Rosser, N. J., Kincey, M. E., Benjamin, J., Oven, K. J., Densmore, A. L., et al. (2017). Satellite‐based emergency mapping: Landslides triggered by the 2015 Nepal earthquake. Natural Hazards and Earth System Sciences Discussions, 18,1–29. https://doi.org/ 10.5194/nhess‐2017‐273 Wilson, R. C., & Keefer, D. K. (1983). Dynamic analysis of a slope failure from the 6 August 1979 Coyote Lake, California, earthquake. Bulletin of the Seismological Society of America, 73(3), 863–877. Wilson, R. C., & Keefer, D. K. (1985). Predicting areal limit of earthquake‐induced landsliding, evaluating earthquake hazards in the Los Angeles Region—An earth‐science perspective. U.S. Geological Survery Professional Paper, 1360, 317–345. Winsemius, H. C., Ward, P. J., Gayton, I., Ten Veldhuis, M., Meijer, D., & Iliffe, M. (2019). Commentary:“The need for a high‐accuracy, open‐access global DEM”. Frontiers in Earth Science, 7, 33. https://doi.org/10.3389/feart.2019.00033 Worden, C. B., & Wald, D. J. (2016). ShakeMap Manual. U.S. Geological Survey. https://doi.org/10.5066/F7D21VPQ Wyllie, D. C., & Mah, C. (2014). Rock slope engineering. New York: CRC Press. Xu, C. (2014). Do buried‐rupture earthquakes trigger less landslides than surface‐rupture earthquakes for reverse faults? Geomorphology, 216,53–57. https://doi.org/10.1016/j.geomorph.2014.03.029 Xu, C. (2015). Preparation of earthquake‐triggered landslide inventory maps using remote sensing and GIS technologies: Principles and case studies. Geoscience Frontiers, 6(6), 825–836. https://doi.org/10.1016/j.gsf.2014.03.004 Xu, C., Ma, S., Tan, Z., Xie, C., Toda, S., & Huang, X. (2018). Landslides triggered by the 2016 Mj 7.3 Kumamoto, Japan, earthquake. Landslides, 15(3), 551–564. https://doi.org/10.1007/s10346‐017‐0929‐1 Xu, C., Wang, S., Xu, X., Zhang, H., Tian, Y., & Ma, S. (2018). A panorama of landslides triggered by the 8 August 2017 Jiuzhaigou, Sichuan Ms7. 0 earthquake. Seismology and Geology, 40(1), 232–260. Xu, C., & Xu, X. (2012). Comment on “Spatial distribution analysis of landslides triggered by 2008.5.12 Wenchuan Earthquake, China” by Shengwen Qi, Qiang Xu, Hengxing Lan, Bing Zhang, Jianyou Liu [Engineering Geology 116 (2010) 95–108]. Engineering Geology, 133‐ 134,40–42. https://doi.org/10.1016/j.enggeo.2012.02.017 Xu, C., & Xu, X. (2014). Statistical analysis of landslides caused by the Mw 6.9 Yushu, China, earthquake of April 14, 2010. Natural Hazards, 72(2), 871–893. https://doi.org/10.1007/s11069‐014‐1038‐2 Xu, C., Xu, X., Shen, L., Dou, S., Wu, S., Tian, Y., & Li, X. (2014). Inventory of landslides triggered by the 2014 Ms6.5 Ludian earthquake and its implications on several earthquake parameters. Seismology and Geology, 36(4), 1186–1203. Xu, C., Xu, X., Shen, L., Yao, Q., Tan, X., Kang, W., et al. (2016). Optimized volume models of earthquake‐triggered landslides. Scientific Reports, 6(1). https://doi.org/10.1038/srep29797 Xu, C., Xu, X., & Shyu, J. B. H. (2015). Database and spatial distribution of landslides triggered by the Lushan, China Mw 6.6 earthquake of 20 April 2013. Geomorphology, 248,77–92. https://doi.org/10.1016/j.geomorph.2015.07.002 Xu, C., Xu, X., Shyu, J. B. H., Gao, M., Tan, X., Ran, Y., & Zheng, W. (2015). Landslides triggered by the 20 April 2013 Lushan, China, Mw 6.6 earthquake from field investigations and preliminary analyses. Landslides, 12(2), 365–385. https://doi.org/10.1007/s10346‐ 014‐0546‐1 Xu, C., Xu, X., Yao, X., & Dai, F. (2014). Three (nearly) complete inventories of landslides triggered by the May 12, 2008 Wenchuan Mw 7.9 earthquake of China and their spatial distribution statistical analysis. Landslides, 11(3), 441–461. https://doi.org/10.1007/s10346‐013‐ 0404‐6 Xu, C., Xu, X., & Yu, G. (2013). Landslides triggered by slipping‐fault‐generated earthquake on a plateau: An example of the 14 April 2010, Ms 7.1, Yushu, China earthquake. Landslides, 10(4), 421–431. https://doi.org/10.1007/s10346‐012‐0340‐x Xu, Q., Fan, X., Huang, R., & Van Westen, C. J. (2009). Landslide dams triggered by the Wenchuan Earthquake, Sichuan Province, south west China. Bulletin of Engineering Geology and the Environment, 68(3), 373–386. https://doi.org/10.1007/s10064‐ 009‐0214‐1 Xu, Z., Schwartz, S., & Lay, T. (1996). Seismic wave‐field observations at a dense, small aperture array located on a landslide in the Santa Cruz Mountains, California. Bulletin of the Seismological Society of America, 86, 655–669. Yalcin, A. (2008). GIS‐based landslide susceptibility mapping using analytical hierarchy process and bivariate statistics in Ardesen (Turkey): Comparisons of results and confirmations. Catena, 72(1), 1–12. https://doi.org/10.1016/j.catena.2007.01.003 Yamagishi, H., & Yamazaki, F. (2018). Landslides by the 2018 Hokkaido Iburi‐Tobu Earthquake on September 6. Landslides, 15(12), 2521–2524. https://doi.org/10.1007/s10346‐018‐1092‐z Yamasaki, S., Nagata, H., & Kawaguchi, T. (2014). Long‐traveling landslides in deep snow conditions induced by the 2011 Nagano Prefecture earthquake, Japan. Landslides, 11(4), 605–613. https://doi.org/10.1007/s10346‐013‐0419‐z Yan, Y., Cui, P., Chen, S., Chen, X., Chen, H., & Chien, Y. (2017). Characteristics and interpretation of the seismic signal of a field‐scale landslide dam failure experiment. Journal of Mountain Science, 14(2), 219–236. https://doi.org/10.1007/s11629‐016‐4103‐3 Yang, W., Qi, W., Wang, M., Zhang, J., & Zhang, Y. (2017). Spatial and temporal analyses of post‐seismic landslide changes near the epicentre of the Wenchuan earthquake. Geomorphology, 276,8–15. https://doi.org/10.1016/j.geomorph.2016.10.010 Yang, W., Qi, W., & Zhou, J. (2018). Decreased post‐seismic landslides linked to vegetation recovery after the 2008 Wenchuan earthquake. Ecological Indicators, 89, 438–444. https://doi.org/10.1016/j.ecolind.2017.12.006

FAN ET AL. 82 Reviews of Geophysics 10.1029/2018RG000626

Yanites, B. J., Tucker, G. E., Hsu, H.‐L., Chen, C., Chen, Y.‐G., & Mueller, K. J. (2011). The influence of sediment cover variability on long‐ term river incision rates: An example from the Peikang River, central Taiwan. Journal of Geophysical Research, 116, F03016. https://doi. org/10.1029/2010JF001933 Yanites, B. J., Tucker, G. E., Mueller, K. J., & Chen, Y.‐G. (2010). How rivers react to large earthquakes: Evidence from central Taiwan. Geology, 38(7), 639–642. https://doi.org/10.1130/G30883.1 Yi, Y., Zhang, Z., Zhang, W., Xu, Q., Deng, C., & Li, Q. (2019). GIS‐based earthquake‐triggered landslide susceptibility mapping with an integrated weighted index model in Jiuzhaigou region of Sichuan Province, China. Natural Hazards and Earth System Sciences Discussions, 2019, 1–25. https://doi.org/10.5194/nhess‐2019‐8 Yu, B., Wu, Y., & Chu, S. (2014). Preliminary study of the effect of earthquakes on the rainfall threshold of debris flows. Engineering Geology, 182, 130–135. https://doi.org/10.1016/j.enggeo.2014.04.007 Yuan, R.‐M., Tang, C.‐L., & Deng, Q. (2015). Effect of the acceleration component normal to the sliding surface on earthquake‐induced landslide triggering. Landslides, 12(2), 335–344. https://doi.org/10.1007/s10346‐014‐0486‐9 Zeng, C., Cui, P., Ge, Y., Zhang, Y., Lei, Y., & Yan, Y. (2014). Characteristics and mechanism of buildings damaged by debris flows on 11 July, 2013 in Qipangou of Wenchuan, Sichuan. Journal of Earth Sciences & Environment, 36(2), 81–91. Zeng, C., He, N., & Song, G. (2012). Analysis and assessment of methods to assess vulnerability of building in debris flow hazard. Advances in Earth Science, 27(11), 1211–1220. Zevenbergen, L. W., & Thorne, C. R. (1987). Quantitative analysis of land surface topography. Earth Surface Processes and Landforms, 12(1), 47–56. https://doi.org/10.1002/esp.3290120107 Zhang, D., & Wang, G. (2007). Study of the 1920 Haiyuan earthquake‐induced landslides in loess (China). Engineering Geology, 94(1–2), 76–88. https://doi.org/10.1016/j.enggeo.2007.07.007 Zhang, J., Wang, Y., Zhang, H., Zhu, C., Li, J., & Yan, X. (2014). Characteristics of secondary geological hazards induced by Yiliang 9.07 earthquakes in Yunnan Province. Journal of Engineering Geology, 22(2), 280–291. Zhang, L., Peng, M., Chang, D., & Xu, Y. (2016). Dam failure mechanisms and risk assessment. Singapore: John Wiley & Sons. https://doi. org/10.1002/9781118558522 Zhang, L. L., Zhang, J., Zhang, L. M., & Tang, W. H. (2011). Stability analysis of rainfall‐induced slope failure: A review. Proceedings of the Institution of Civil Engineers ‐ Geotechnical Engineering, 164(5), 299–316. https://doi.org/10.1680/geng.2011.164.5.299 Zhang, L. M., Xu, Y., Huang, R. Q., & Chang, D. S. (2011). Particle flow and segregation in a giant landslide event triggered by the 2008 Wenchuan earthquake, Sichuan, China. Natural Hazards and Earth System Science, 11(4), 1153–1162. https://doi.org/10.5194/nhess‐11‐ 1153‐2011 Zhang, L. M., Zhang, S., & Huang, R. Q. (2014). Multi‐hazard scenarios and consequences in Beichuan, China: The first five years after the 2008 Wenchuan earthquake. Engineering Geology, 180,4–20. https://doi.org/10.1016/j.enggeo.2014.03.020 Zhang, N., & Matsushima, T. (2016). Simulation of rainfall‐induced debris flow considering material entrainment. Engineering Geology, 214, 107–115. https://doi.org/10.1016/j.enggeo.2016.10.005 Zhang, S., Zhang, L., Lacasse, S., & Nadim, F. (2016). Evolution of mass movements near epicentre of Wenchuan Earthquake, the first eight years. Scientific Reports, 6(1). https://doi.org/10.1038/srep36154 Zhang, S., & Zhang, L. M. (2017). Impact of the 2008 Wenchuan earthquake in China on subsequent long‐term debris flow activities in the epicentral area. Geomorphology, 276,86–103. https://doi.org/10.1016/j.geomorph.2016.10.009 Zhang, S., Zhang, L. M., & Chen, H. X. (2014). Relationships among three repeated large‐scale debris flows at Pubugou Ravine in the Wenchuan earthquake zone. Canadian Geotechnical Journal, 51(9), 951–965. https://doi.org/10.1139/cgj‐2013‐0368 Zhang, S., Zhang, L.‐M., Chen, H.‐X., Yuan, Q., & Pan, H. (2013). Changes in runout distances of debris flows over time in the Wenchuan earthquake zone. Journal of Mountain Science, 10(2), 281–292. https://doi.org/10.1007/s11629‐012‐2506‐y Zhang, S., Zhao, L., Delgado‐Tellez, R., & Bao, H. (2018). A physics‐based probabilistic forecasting model for rainfall‐induced shallow landslides at regional scale. Natural Hazards and Earth System Sciences, 18(3), 969–982. https://doi.org/10.5194/nhess‐18‐969‐2018 Zhang, Y. (2018). Review of studies on earthquake‐induced landslides. In Yingbin Zhang, Earthquake‐induced landslides, (pp. 11–39). Singapore: Springer Singapore. https://doi.org/10.1007/978‐981‐10‐2935‐6_2 Zhang, Y., Zhang, J., Chen, G., Zheng, L., & Li, Y. (2015). Effects of vertical seismic force on initiation of the Daguangbao landslide induced by the 2008 Wenchuan earthquake. Soil Dynamics and Earthquake Engineering, 73,91–102. https://doi.org/10.1016/j. soildyn.2014.06.036 Zhang, Z., & Wang, L. (1995). Geological disasters in loess areas during the 1920 Haiyuan Earthquake, China. GeoJournal, 36(2–3), 269–274 . https://doi.org/10.1007/BF00813183 Zhang, Z., Wang, T., Wu, S., Tang, H., & Liang, C. (2017). Seismic performance of loess‐mudstone slope by centrifuge tests. Bulletin of Engineering Geology and the Environment, 76(2), 671–679. https://doi.org/10.1007/s10064‐015‐0846‐2 Zhao, T., Dai, F., & Xu, N. (2017). Coupled DEM‐CFD investigation on the formation of landslide dams in narrow rivers. Landslides, 14(1), 189–201. https://doi.org/10.1007/s10346‐015‐0675‐1 Zhou, C. H., Lee, C. F., Li, J., & Xu, Z. W. (2002). On the spatial relationship between landslides and causative factors on Lantau Island, Hong Kong. Geomorphology, 43(3–4), 197–207. https://doi.org/10.1016/S0169‐555X(01)00130‐1 Zhou, G. G. D., Cui, P., Chen, H. Y., Zhu, X. H., Tang, J. B., & Sun, Q. C. (2013). Experimental study on cascading landslide dam failures by upstream flows. Landslides, 10(5), 633–643. https://doi.org/10.1007/s10346‐012‐0352‐6 Zhou, W., & Tang, C. (2014). Rainfall thresholds for debris flow initiation in the Wenchuan earthquake‐stricken area, southwestern China. Landslides, 11(5), 877–887. https://doi.org/10.1007/s10346‐013‐0421‐5 Zhuang, J., Peng, J., Xu, C., Li, Z., Densmore, A., Milledge, D., et al. (2018). Distribution and characteristics of loess landslides triggered by the 1920 Haiyuan Earthquake, Northwest of China. Geomorphology, 314(1–12), 1–12. https://doi.org/10.1016/j.geomorph.2018.04.012 Ziotopoulou, K., & Boulanger, R. W. (2016). Plasticity modeling of liquefaction effects under sloping ground and irregular cyclic loading conditions. Soil Dynamics and Earthquake Engineering, 84, 269–283. https://doi.org/10.1016/j.soildyn.2016.02.013 Zobel, C. W., & Khansa, L. (2014). Characterizing multi‐event disaster resilience. Computers & Operations Research, 42,83–94. https://doi. org/10.1016/j.cor.2011.09.024

FAN ET AL. 83