The Global Landslide Catalog (GLC) was developed with the goal of identifying rainfall-triggered landslide events around the world, regardless of size, impacts or location. The GLC considers all types of mass movements triggered by rainfall, which have been reported in the media, disaster databases, scientific reports, or other sources. The GLC has been compiled since 2007 at NASA Goddard Space Flight Center. This is a unique data set with the ID tag “GLC” in the landslide editor. This dataset on data.nasa.gov was a one-time export from the Global Landslide Catalog maintained separately. It is current as of March 7, 2016. The original catalog is available here: http://www.arcgis.com/home/webmap/viewer.html?url=https%3A%2F%2Fmaps.nccs.nasa.gov%2Fserver%2Frest%2Fservices%2Fglobal_landslide_catalog%2Fglc_viewer_service%2FFeatureServer&source=sd To export GLC data, you must agree to the “Terms and Conditions”. We request that anyone using the GLC cite the two sources of this database: Kirschbaum, D. B., Adler, R., Hong, Y., Hill, S., & Lerner-Lam, A. (2010). A global landslide catalog for hazard applications: method, results, and limitations. Natural Hazards, 52(3), 561–575. doi:10.1007/s11069-009-9401-4. [1] Kirschbaum, D.B., T. Stanley, Y. Zhou (In press, 2015). Spatial and Temporal Analysis of a Global Landslide Catalog. Geomorphology. doi:10.1016/j.geomorph.2015.03.016. [2]

The Global Landslide Catalog (GLC) was developed with the goal of identifying rainfall-triggered landslide events around the world, regardless of size, impacts or location. The GLC considers all types of mass movements triggered by rainfall, which have been reported in the media, disaster databases, scientific reports, or other sources. The GLC has been compiled since 2007 at NASA Goddard Space Flight Center. This is a unique data set with the ID tag “GLC” in the landslide editor. This dataset on data.nasa.gov was a one-time export from the Global Landslide Catalog maintained separately. It is current as of March 7, 2016. The original catalog is available here: http://www.arcgis.com/home/webmap/viewer.html?url=https%3A%2F%2Fmaps.nccs.nasa.gov%2Fserver%2Frest%2Fservices%2Fglobal_landslide_catalog%2Fglc_viewer_service%2FFeatureServer&source=sd To export GLC data, you must agree to the “Terms and Conditions”. We request that anyone using the GLC cite the two sources of this database: Kirschbaum, D. B., Adler, R., Hong, Y., Hill, S., & Lerner-Lam, A. (2010). A global landslide catalog for hazard applications: method, results, and limitations. Natural Hazards, 52(3), 561–575. doi:10.1007/s11069-009-9401-4. [1] Kirschbaum, D.B., T. Stanley, Y. Zhou (In press, 2015). Spatial and Temporal Analysis of a Global Landslide Catalog. Geomorphology. doi:10.1016/j.geomorph.2015.03.016. [2]

This data set is an inventory of some 2800 landslides that occurred in the High Mountain Asia (HMA) study area between 5 January 2007 and 31 December 2018 (plus one event from 28 January 1990). The catalog includes dates and locations of landslides, plus additional characteristics such as event triggers, country, length and area of the slide, and the number of injuries and fatalities.

The events in this catalog represent an HMA-specific subset of the Cooperative Open Online Landslide Repository (COOLR), a project that was created to build a more robust, publicly available inventory of landslides by supplementing data in the NASA Global Landslide Catalog with citizen science reports.

The Global Landslide Hazard Distribution is a 2.5 minute grid of global landslide and snow avalanche hazards based upon work of the Norwegian Geotechnical Institute (NGI). The hazards mapping of NGI incorporates a range of data including slope, soil, soil moisture conditions, precipitation, seismicity, and temperature. Shuttle Radar Topography Mission (SRTM) elevation data at 30 seconds resolution are also incorporated. Hazards values less than or equal to 4 are considered negligible and only values 5 through 9 are utilized in further analyses. To ensure compatibility with other data sets, value 1 is added to each of the values to provide a hazard ranking ranging 6 through 10 in increasing hazard. This data set is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), Norwegian Geotechnical Institute (NGI), and Columbia University Center for International Earth Science and Information Network (CIESIN).

The Global Landslide Hazard Distribution is a 2.5 minute grid of global landslide and snow avalanche hazards based upon work of the Norwegian Geotechnical Institute (NGI). The hazards mapping of NGI incorporates a range of data including slope, soil, soil moisture conditions, precipitation, seismicity, and temperature. Shuttle Radar Topography Mission (SRTM) elevation data at 30 seconds resolution are also incorporated. Hazards values less than or equal to 4 are considered negligible and only values 5 through 9 are utilized in further analyses. To ensure compatibility with other data sets, value 1 is added to each of the values to provide a hazard ranking ranging 6 through 10 in increasing hazard. This data set is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), Norwegian Geotechnical Institute (NGI), and Columbia University Center for International Earth Science and Information Network (CIESIN).

Summary of fields included in the Cooperative Open Online Landslide Repository (COOLR).

From NASA's Cooperative Open Online Landslide Repository (COOLR). The Cooperative Open Online Landslide Repository (COOLR) is a worldwide database of landslide events. It currently includes NASA’s Global Landslide Catalog (GLC) and landslide events contributed by citizen scientists. In a future release of COOLR, collated landslide inventories will be added by REST API or manually.

List of all knowna landslide citizen science projects.

The Global Landslide hazard map is a gridded dataset of landslide hazard produced at the global scale. Landslides happen around the world and have devastating impacts on people and the built environment. To better understand the spatial and temporal distribution of landslide hazard worldwide, the World Bank and the Global Facility for Disaster Reduction and Recovery (GFDRR) commissioned Arup to undertake a landslide hazard assessment at a global scale. Using a global landslide inventory, landslide susceptibility information provided by NASA, and an innovative machine learning model, our geohazard and risk management experts produced a state-of-the-art quantitative landslide hazard map for the whole world.

This dataset constitutes one of the outcomes of the Master's Thesis titled 'Landslide identification using deep learning-based change detection and the DeepESDL collaborative cloud platform,' authored by Julia Anna Leonardi, conducted under the supervision of Prof. Maria Antonia Brovelli and Dr. Vasil Yordanov at Politecnico di Milano. The authors extracted the image patches through the DeepESDL platform, which the ESA NoR sponsorship provided access to.

The resource contains the .csv database containing landslide information of 174 events, such as the date, location, source, and the dates of the Sentinel-2 images from before and after the event. The sources of the data included in this inventory are:

• Inventario dei Fenomeni Franosi in Italia (IFFI) compiled by ISPRA
• NASA Global Landslide Catalog (GLC) [1],[2].
• Copernicus Emergency Management Service (© European Union, 2012-2024) [3]
• Geological Survey Ireland | Geohazards | National Landslide Susceptibility Map
• Monsoon triggered landslides in Nepal timed with Sentinel-1 for 2015, 2017, 2018, and 2019, courtesy of Katy Burrows [4]
• Landslides after 2023 earthquakes in Turkey compiled by the NASA GSFC team
• University of Twente online landslide course
• Manually extracted sources from [5], [6], [7], [8].

The .zip file contains the TRAIN dataset with the pre-event Sentinel-2 image patches in folder PRE and the post-event Sentinel-2 image patches in folder POST, and the TEST set with 17 bi-temporal pairs in the PRE and POST folders following the above convention and the labels in the form of change maps in the CM folder. In this updated version the full 13-band Sentinel-2 images are published. (In the previous version only 5 bands (B02, B03, B04, B08, and CLM) were available). All the image patches and the ground truth annotations are in the GeoTIFF format.

The authors developed the dataset for change detection workflows.

Project supported by the ESA Network of Resources Initiative.

This work is funded by the Italian Ministry of Foreign Affairs and International Cooperation within the project “Geoinformatics and Earth Observation for Landslide Monitoring” CUP D19C21000480001

[1] Kirschbaum, D.B., Stanley, T., & Zhou, Y. (2015). Spatial and temporal analysis of a global landslide catalog. Geomorphology, 249, 4-15. doi:10.1016/j.geomorph.2015.03.016

[2] Kirschbaum, D.B., Adler, R., Hong, Y., Hill, S., & Lerner-Lam, A. (2010). A global landslide catalog for hazard applications: method, results, and limitations. Natural Hazards, 52, 561-575. doi:10.1007/s11069-009-9401-4

[3] Copernicus Emergency Management Service. (n.d.). Retrieved March 12, 2024 from https://emergency.copernicus.eu/

[4] K. Burrows, O. Marcand C. Andermann, “Monsoon triggered landslides in Nepal timed with Sentinel-1 for 2015, 2017, 2018 and 2019”. Zenodo, May 25, 2023. doi: 10.5281/zenodo.7970874.

[5] F. Wang et al., ‘Coseismic landslides triggered by the 2018 Hokkaido, Japan (Mw 6.6), earthquake: spatial distribution, controlling factors, and possible failure mechanism’, Landslides, vol. 16, no. 8, pp. 1551–1566, Aug. 2019, doi: 10.1007/s10346-019-01187-7.

[6] P. Amatya, D. Kirschbaum, and T. Stanley, ‘Rainfall-induced landslide inventories for Lower Mekong based on Planet imagery and a semi-automatic mapping method’, Geoscience Data Journal, vol. 9, no. 2, pp. 315–327, 2022, doi: 10.1002/gdj3.145.

[7] ]‘Vietnam – At Least 12 Killed in Flash Floods and Landslides in North – FloodList’. Accessed: Feb. 18, 2024. [Online]. Available: https://floodlist.com/asia/vietnam-floods-landslides-yen-baison-la-august-2017

[8] L. Colombo, ‘Crollata, causa frana, la volta di una galleria lungo la SP72 a Fiumelatte a Varenna - GLI AGGIORNAMENTI’, Lecco Notizie. Accessed: Feb. 18, 2024. [Online]. Available: https://lecconotizie.com/cronaca/crollata-la-volta-di-una-galleria-lungo-la-sp72-a-fiumelatte-avarenna/

The Global Landslide Mortality Risks and Distribution is a 2.5 minute grid of global landslide mortality risks. Gridded Population of the World, Version 3 (GPWv3) data provide a baseline estimation of population per grid cell from which to estimate potential mortality risks due to landslide hazard. Mortality loss estimates per hazard event are caculated using regional, hazard-specific mortality records of the Emergency Events Database (EM-DAT) that span the 20 years between 1981 and 2000. Data regarding the frequency and distribution of landslide hazard are obtained from the Global Landslide Hazard Distribution data set. In order to more accurately reflect the confidence associated with the data and procedures, the potential mortality estimate range is classified into deciles, 10 classes of increasing risk with an approximately equal number of grid cells per class, producing a relative estimate of landslide-based mortality risks. This data set is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), International Bank for Reconstruction and Development/The World Bank, and Columbia University Center for International Earth Science Information Network (CIESIN).

The Global Landslide Proportional Economic Loss Risk Deciles is a 2.5 minute grid of landslide hazard economic loss as proportions of Gross Domestic Product (GDP) per analytical Unit. Estimates of GDP at risk are based on regional economic loss rates derived from historical records of the Emergency Events Database (EM-DAT). Loss rates are weighted by the hazard's frequency and distribution. The methodology of Sachs et al. (2003) is followed to determine baseline estimates of GDP per grid cell. To better reflect the confidence surrounding the data and procedures, the range of proportionalities is classified into deciles, 10 class of an approximately equal number of grid cells of increasing risk. This dataset is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), International Bank for Reconstruction and Development/The World Bank, and Columbia University Center for International Earth Science Information Network (CIESIN).

From NASA's Cooperative Open Online Landslide Repository (COOLR).

The Cooperative Open Online Landslide Repository (COOLR) is a worldwide database of landslide events. It currently includes NASA’s Global Landslide Catalog (GLC) and landslide events contributed by citizen scientists. In a future release of COOLR, collated landslide inventories will be added by REST API or manually.

Background: The global Landslide Hazard Assessment for Situational Awareness (LHASA) model is developed to provide situational awareness of landslide hazards for a wide range of users. Precipitation is a common trigger of landslides. The Integrated Multi-satellitE Retrievals for GPM (IMERG) data shows recent precipitation, updated every thirty minutes. A LHASA landslide “nowcast” is created by comparing GPM data from the last seven days to the long-term precipitation record provided by the Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA). Because IMERG data is only available starting in 2014, the record of historical rainfall was established by TMPA, comparing 2001-present. The TMPA rainfall probability distributions were then compared to that of IMERG and the rainfall thresholds were adjusted so that the IMERG data more closely mapped to those of the TMPA archive. The past 7 days of rainfall are considered, with each day is weighted according to their date before present, with the last twenty-four hours having the most impact.Data Visualization: Landslide Nowcast is shown using a Two Color Ramp (yellow and red). Values represent Moderate (yellow) and High (red) risk of potential landslide.Update Frequency: 3 Hour latency Suggested Usage: In places where precipitation is unusually high, the susceptibility of the terrain is evaluated, which includes quantitative information on if: roads have been built; trees have been cut down or burned; a major tectonic fault is nearby; the local bedrock is weak; the hillsides are steep.Further Reference: Learn more about Global Landslide Model Learn more about GPM's science objectives. Learn more about GPM's applications.

The Global Landslide Total Economic Loss Risk Deciles is a 2.5 minute grid of global landslide total economic loss risks. A process of spatially allocating Gross Domestic Product (GDP) based upon the Sachs et al. (2003) methodology is utilized. First the proportional contributions of subnational Units to their respective national GDP are determined using sources of various origins. The contribution rates are then applied to published World Bank Development Indicators to determine a GDP value for the subnational Unit. Once the national GDP has been spatially stratified into the smallest administrative Units available, GDP values for grid cells are derived using Gridded Population of the World, Version 3 (GPWv3) data of population distributions. A per capita contribution value is determined within each subnational Unit, and this value is multiplied by the population per grid cell. Once a GDP value has been determined on a per grid cell basis, then the regionally variable loss rate as derived from the historical records of EM-DAT is used to determine the total economic loss risks posed to a grid cell by landslide hazards. The final surface does not present absolute values of total economic loss, but rather a relative decile (1-10 with increasing risk) ranking of grid cells based upon the calculated economic loss risks. This data set is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), International Bank for Reconstruction and Development/The World Bank, and Columbia University Center for International Earth Science Information Network (CIESIN).

The Global Landslide Nowcast addresses the need for real-time situational awareness of landslide hazard. The Landslide Hazard Assessment for Situational Awareness model (LHASA) combines satellite rainfall estimates from the Global Precipitation Measurement mission (GPM) with soil moisture estimates from the Soil Moisture Active Passive (SMAP) satellite and other factors to produce a map of locations where rainfall-triggered landslide activity is probable. Due to the latency of the rainfall data, the nowcast is a near-real time product with a minimum latency of 5 hours. Although the model could be run every half hour, this archive contains a daily record derived from a retrospective model run.The Global Landslide Nowcast version 2.0.0 retains replaces the heuristic decision tree from version 1.0 with a machine learning model. Instead of merging all factors other than precipitation into a susceptibility map, LHASA 2.0 takes in each variable as a separate input layer. The most important change is the replacement of the categorical nowcast with a probabilistic output. This will enable users to adjust the threshold to suit their specific application and geographic location.

The Landslide Hazard Assessment for Situational Awareness (LHASA) model identifies locations with high potential for landslide occurrence at a daily temporal resolution. LHASA combines satellite‐based precipitation estimates with a landslide susceptibility map derived from information on slope, geology, road networks, fault zones, and forest loss. When rainfall is considered to be extreme and susceptibility values are moderate to very high, a “nowcast” is issued to indicate the times and places where landslides are more probable. Although the model could be run every half hour, this archive contains a daily record derived from a retrospective model run and spatial coverage is from 60°N to 60°S .

The Landslide Hazard Assessment for Situational Awareness (LHASA) model identifies locations with high potential for landslide occurrence at a daily temporal resolution. LHASA combines satellite‐based precipitation estimates with a landslide susceptibility map derived from information on slope, geology, road networks, fault zones, and forest loss. When rainfall is considered to be extreme and susceptibility values are moderate to very high, a “nowcast” is issued to indicate the times and places where landslides are more probable.This archive contains GeoTIFF Rasters that are a 16-year average (beginning of 2001 - end of 2016). The spatial coverage is from 72°N to 60°S latitude, and 180°W to 180°E longitude, based on IMERG Ver06B from the aforementioned time interval. The provided global maps of exposure to landslide hazards, are at a 30x30 arc-second resolution. These maps show the estimated exposure of population, roads, and critical infrastructure (hospitals/clinics, schools, fuel stations, power stations & distribution facilities) to landslide hazard, as modeled by the NASA LHASA model.The data collection consists of eight files, covering the aforementioned spatial and temporal ranges, totaling approximately 20.3 GB (~2.5 GB each): (1): Landslide hazard (annual average; Units: Nowcasts.yr-1) (2): Landslide hazard (annual standard deviation; Units: Nowcasts.yr-1) (3): Population exposure (annual average; Units: Person-Nowcasts. yr-1. km-2) (4): Population exposure (annual standard deviation; Units: Person-Nowcasts. yr-1. km-2) (5): Road exposure (annual average; Units: Nowcasts.km.yr-1.km-2) (6): Road exposure (annual standard deviation; Units: Nowcasts.km.yr-1.km-2) (7): Critical infrastructure exposure (annual average; Units: Nowcasts.element.yr-1.km-2) (8): Critical infrastructure exposure (annual standard deviation; Units: Nowcasts.element.yr-1.km-2)

The Global Landslide Mortality Risks and Distribution is a 2.5 minute grid of global landslide mortality risks. Gridded Population of the World, Version 3 (GPWv3) data provide a baseline estimation of population per grid cell from which to estimate potential mortality risks due to landslide hazard. Mortality loss estimates per hazard event are caculated using regional, hazard-specific mortality records of the Emergency Events Database (EM-DAT) that span the 20 years between 1981 and 2000. Data regarding the frequency and distribution of landslide hazard are obtained from the Global Landslide Hazard Distribution data set. In order to more accurately reflect the confidence associated with the data and procedures, the potential mortality estimate range is classified into deciles, 10 classes of increasing risk with an approximately equal number of grid cells per class, producing a relative estimate of landslide-based mortality risks. This data set is the result of collaboration among the Columbia University Center for Hazards and Risk Research (CHRR), International Bank for Reconstruction and Development/The World Bank, and Columbia University Center for International Earth Science Information Network (CIESIN).

The Cooperative Open Online Landslide Repository, or COOLR, is an open platform where scientists and citizen scientists can share landslide reports. See all landslide data from COOLR with other scientific data using the Landslide Viewer application. For more information: https://gpm.nasa.gov/landslides/index.html Landslides @ NASA is a project of NASA's Precipitation Measurement Missions (PMM) which use satellite data and citizen science data to model and inventory landslides around the world. We are based at NASA Goddard Space Flight Center (GSFC). Learn more about our publications and projects on the Resources page.