The August 14, 2021, magnitude 7.2 Nippes, Haiti, earthquake triggered thousands of landslides on the Tiburon Peninsula. The landslides directly caused fatalities and damage and impeded response efforts by blocking roads and causing other infrastructure damage. Adverse effects of the landslides likely will continue for months to years. This report presents an assessment of potential postearthquake landslide-related geologic hazards for the Tiburon Peninsula and a preliminary map of the landslides triggered by the earthquake. This hazard assessment is based on an emergency analysis of the currently available, postearthquake satellite imagery. In this report, we highlight specific areas of concern that may benefit from more detailed assessment and longer-term monitoring. Our mapping efforts revealed that at least 4,893 landslides were triggered across the Tiburon Peninsula by the earthquake and subsequent rainfall from Tropical Cyclone Grace. We also observed hundreds of landslide deposits potentially restricting flow in rivers and streams. In addition, we observed landslides that likely affected roads by rendering them impassable or susceptible to subsequent damage from existing landslides. Because of the preliminary nature of this report and the limits of remote analyses, additional investigation and monitoring would be beneficial to accurately determine the threat posed by these hazards to people and infrastructure.
For more information on the USGS—the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment—visit
For an overview of USGS information products, including maps, imagery, and publications, visit
Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Although this information product, for the most part, is in the public domain, it also may contain copyrighted materials as noted in the text. Permission to reproduce copyrighted items must be secured from the copyright owner.
Funding for the preparation of this report was provided by the U.S. Agency for International Development’s Bureau for Humanitarian Assistance and the U.S. Geological Survey. We thank Robert Mason, Karl Winters, and Allen Gellis for their insights on channel sedimentation and flooding and Jeff Coe, Randall Jibson, David Wald, and Ryan Gold for reviewing the report (U.S. Geological Survey).
Multiply | By | To obtain |
meter (m) | 3.281 | foot (ft) |
kilometer (km) | 0.6214 | mile (mi) |
kilometer (km) | 0.5400 | mile, nautical (nmi) |
meter (m) | 1.094 | yard (yd) |
square kilometer (km2) | 247.1 | acre |
square kilometer (km2) | 0.3861 | square mile (mi2) |
Coordinate information is referenced to the World Geodetic System 1984 (WGS 84) Universal Transverse Mercator (UTM) Zone 18N projection.
disaster assistance response team
digital elevation model
acceleration of gravity
magnitude
response management team
United Nations Children's Fund
U.S. Agency for International Development
U.S. Geological Survey
coordinated universal time
Universal Transverse Mercator
World Geodetic System
The magnitude (
Overview of the Tiburon Peninsula showing contours of estimated peak ground acceleration (PGA) for the 2021 Nippes, Haiti, earthquake from the U.S. Geological Survey (USGS) ShakeMap (ver. 14) overlain on the USGS Ground Failure (GF) product landslide probability model (ver. 13). The landslide probability model provides an estimate of areas where earthquake-triggered landslides were most likely to have occurred, based on the ShakeMap shaking estimates. The model is intended primarily for near-real-time use and is available at
Figure 1. Overview of the Tiburon Peninsula showing contours of estimated peak ground acceleration for the 2021 Nippes, Haiti, earthquake.
During the week following the earthquake, a team of USGS scientists rapidly mapped landslides using remote methods and created a landslide inventory. By the end of this effort, 3,625 landslides had been mapped. This inventory, with critical daily observations, was shared with the U.S. Agency for International Development’s (USAID’s) Disaster Assistance Response Team (DART) and Response Management Team (RMT) for situational awareness and emergency response. The inventory also was made publicly available as an ArcGIS online web map for use by humanitarian aid organizations for situational awareness and emergency response. The original rapid inventory that was available immediately after the earthquake for situational awareness only is now superseded by the inventory associated with this report (
For this report, we mapped additional landslides that were missed in the initial inventory. We carefully reviewed the inventory in its entirety to identify landslides that may pose ongoing threats to people and property (
Overview map of the 2021 Nippes, Haiti, earthquake landslide inventory showing landslide dams, road hazards, and flagged landslides detailed in
Figure 2. Overview map of the Haiti 2021 earthquake landslide inventory showing landslide dams, road hazards, and flagged landslides.
In addition to landslide headscarp locations, we present an assessment of potential postearthquake landslide-related geologic hazards for the Tiburon Peninsula. We highlight specific areas of concern that may benefit from more detailed assessment and longer-term monitoring because of the potential effects on people and infrastructure. Potential postearthquake landslide hazards include (1) sudden flooding from the failure of landslide dams; (2) continued landsliding of damaged slopes, especially along roads and river channels; (3) an overall increase in landslide activity for multiple years; and (4) changes in river behavior from an influx of landslide sediment that can lead to long-term changes in flooding risks.
As this event shows, earthquakes can set in motion a cascade of geologic events that can affect the landscape for months (or years) following an event (
To map the landslides, we used mid- to high-resolution satellite imagery including Sentinel-2 (10-meter [m] resolution), WorldView (0.3–0.5-m resolution), Planet (2.7–4.0-m resolution), and a high-resolution (1.5-m) digital elevation model (DEM) that was derived from light detection and ranging (lidar) collected from 2014 to 2016 (HaitiData and The World Bank, 2021). Postearthquake images were compared with pre-earthquake images to ensure the landslide feature was caused by the earthquake or the cyclone. The positional accuracy of satellite imagery can vary, especially in hilly areas or when using low-resolution imagery. Because of the varying quality of imagery used and our rapid mapping for the response, we estimate our accuracy of landslide headscarp points to be within tens of meters of each landslide. In many cases, when using poorly orthorectified images, we adjusted the location using the high-resolution DEM. For one of our more poorly orthorectified images, the root mean square error was calculated to be 45 m. This error is not representative of all images used, but it provides an upper limit on the positional accuracy of our landslide mapping. As there was a large quantity of images utilized in our rapid mapping effort, a formal and systematic assessment on the positional accuracy of the data has yet to be completed.
To determine the potential for human and infrastructure effects, we referenced a grid of population data from the
Table 1. Notable landslide dams and road obstructions after the 2021 Nippes, Haiti, earthquake.
[Notable landslide dams and road obstructions are shown by identification numbers that correspond to point features within the associated inventory available (
ID no. | Coordinates | Commune | Description | Threat | Figure no. |
D2 |
73.8475146°W 18.3767446°N | Camp-Perrin | Landslide dams blocking river; water is pooling. | These landslide dams are upstream (0.2–0.4 km) from a populated area with several structures adjacent to the river. | 3 |
D4 |
73.8476736°W 18.3799054°N | Camp-Perrin | Landslide dams with water pooling present. | These landslide dams are upstream (0.8 km) from a populated area with several structures adjacent to the river. | 3 |
R1 | 73.8788310°W 18.3703507°N | Camp-Perrin | A series of landslides along main road, RN-7, that initially blocked this key route after the earthquake. It has since been cleared (as observed in imagery), but material will likely continue to move downslope at this location. | This is a main road that connects Les Cayes and Jérémie and provides access to interior communities. | 3 |
R7 | 73.8387131°W 18.3719796°N | Camp-Perrin | Potentially deep landslide caused severe damage to a road in a settled area. | This poses an ongoing hazard to the road and access may be limited in this area. | 3 |
D10 |
73.8309072°W 18.3859385°N | Camp-Perrin | Two landslide dams blocked a drainage that appeared dry at the time of the image, but a small amount of pooled water is visible northeast of the easternmost dam. | These landslide dams are 3.5-km upstream from a populated area and structures that are adjacent to river floodplain. | 3 |
D27 | 73.8495048°W 18.3369706°N | Camp-Perrin | Landslide dam with visible pooling. | Immediately upstream (0.05 km) from a populated area and structures. This dam is of high concern because of its proximity to structures. | 3 |
D28 |
73.8227571°W 18.3359879°N | Camp-Perrin | Landside dam with visible pooling. | This landslide dam is directly adjacent to a populated area and several structures. | 3 |
D24 | 73.8996369°W 18.4269625°N | Pestel | Landslide dam blocking 450 m of the river channel. Pool is about 300 m long and appears to be growing in recent imagery. | Scattered settlements downstream and along the river. River drains into a closed basin or maybe a karst cavern. Adjacent to RN-7, so there is likely awareness of the growing pool. | 4 |
D39 |
73.9637699°W 18.3832639°N | Beaumont | Landside dams with visible pooling. | No densely populated areas or structures are downstream from these dams. | 4 |
D22 | 73.9681267°W 18.3697023°N | Chantal | Landslide dams; unknown if they are completely or partially blocking river. | No populated area or structures are downstream from these dams. | 4 |
D18 |
74.0443117°W 18.4109764°N | Roseaux | Landslide dams with pooling visible. | No populated area or structures are downstream from these dams. | 4 |
D8 | 73.7247343°W 18.3867478°N | Maniche | Landslide dam with no pooling currently visible; water is not clearly visible in the channel. | This landslide dam is 0.7-km upstream from a populated area and structures that are adjacent to river floodplain. A road parallels the river downstream. | 5 |
D9 | 73.7662063°W 18.3925709°N | Maniche | Landslide dam with no pooling currently visible, but no water is visible in the channel at the time of the image. Landslide is large but is close to the top of its tributary, so less area can contribute water than for some other dams. | This landslide dam is 2.6-km upstream from a populated area and structures that are adjacent to river floodplain. | 5 |
D14 | 73.7512845°W 18.4118467°N | Maniche | Landslide dam with no pooling currently visible; channel dry at time of the image. | This landside dam is 1.6-km upstream from a populated area and structures that are adjacent to river floodplain. | 5 |
D16 |
73.8011201°W 18.3864305°N | Maniche | Landslide dams with pooling visible behind the larger dam. | This landslide dam is directly adjacent to a populated area and several structures. More populated regions are downstream. Roads are adjacent to and intersect the river. | 5 |
R23 | 73.7172863°W 18.3685561°N | Maniche | Partial landslide dam adjacent to road and river. | Adjacent to road, which presents an ongoing road hazard. Upstream (1.6 km) from a populated area adjacent to river. | 5 |
D21 | 73.5362199°W 18.3865591°N | L’Asile | Possible deep landside damming a stream with visible pooling. | This landside dam is 3.5-km upstream from populated areas and structures that are adjacent to Grand Rivière de Nippes floodplain. | 6 |
D25 | 73.5479989°W 18.3475403°N | L’Asile | Landslide dam; unknown if the dam is completely or partially blocking river channel. | Upstream (1.6–2 km) from clusters of structures. | 6 |
D37 | 73.4468978°W 18.3712835°N | L’Asile | Landslide dam; unknown if the dam is completely or partially blocking river, but some pooling is visible. | Directly upstream from roads (0.4–0.7 km) that intersect river and upstream (1.2 km) from populated areas with structures adjacent to river. | 6 |
D41 | 73.5263153°W 18.3857851°N | L’Asile | Landside dam with visible pooling but not complete restriction of flow. | Upstream (3 km) from clusters of structures. | 6 |
R42 | 73.4908007°W 18.3811946°N | L’Asile | Potential road damage from landslide; would need higher resolution imagery to confirm. | Potential ongoing road hazard. | 6 |
D26 | 73.6485411°W 18.4249629°N | Cavaillon | Landslide dam; unknown if the dam is completely or partially blocking river channel. | Upstream (0.6 km) from clusters of structures and a road. | 7 |
D30 |
73.5438633°W 18.3318873°N | Saint Louis du Sud | Landslide dams; unknown if dams are completely or partially blocking river channel. | Upstream (0.7 km) from a cluster of structures. | 7 |
D33 |
73.6196799° W 18.3886041° N | Cavaillon | Landside dams with visible pooling. | Upstream (3.5 km) from clusters of structures and a road. | 7 |
D36 | 73.6237611° W 18.3887233° N | Cavaillon | Landside dam with visible pooling. | Upstream (2 km) from clusters of structures and a road. | 7 |
D38 | 73.6848976° W 18.4241081° N | Baradères | Landslide dam; unknown if the dam is completely or partially blocking river. | Upstream (1.4 km) from densely populated areas and structures. Also upstream from roads that are adjacent to and intersect the river channel farther downstream. | 7 |
Flagged landslide features in Camp-Perrin commune, identified from satellite imagery after the 2021 Nippes, Haiti, earthquake. Refer to
Figure 3. Flagged landslide features in Camp-Perrin commune, identified from satellite imagery after the 2021 Haiti earthquake.
Flagged landslide features in Pestel, Beaumont, Chantal, and Roseaux communes, identified from satellite imagery after the 2021 Nippes, Haiti, earthquake. Refer to
Figure 4. Flagged landslide features in Pestel, Beaumont, Chantal, and Roseaux communes, identified from satellite imagery after the 2021 Haiti earthquake.
Flagged landslide features in Maniche commune, identified from satellite imagery after the 2021 Nippes, Haiti, earthquake. Refer to
Figure 5. Flagged landslide features in Maniche commune, identified from satellite imagery after the 2021 Haiti earthquake.
Flagged landslide features in L’Asile commune, identified from satellite imagery after the 2021 Nippes, Haiti, earthquake. Refer to
Figure 6. Flagged landslide features in L’Asile commune, identified from satellite imagery after the 2021 Haiti earthquake.
Flagged landslides in Cavaillon, Saint Louis du Sud, and Baradères communes, identified from satellite imagery after the 2021 Nippes, Haiti, earthquake. Refer to
Figure 7. Flagged landslides in Cavaillon, Saint Louis du Sud, and Baradères communes, identified from satellite imagery after the 2021 Haiti earthquake.
The most intense landsliding occurred at the western end of the fault, where the terrain is most rugged, especially in the Pic Macaya National Park (
Most landslides were in remote areas, although many were close to small human settlements. The green dots in
Many roads were blocked by landslides, which hindered travel and the movement of humanitarian aid.
Simple models based on global data indicate that an
Although it affected a different part of Haiti, the August 2021 earthquake was in terrain similar to the
Hazards associated with the landslides triggered by this earthquake may pose a risk to people on the Tiburon Peninsula for several years. The following hazard assessment is based on our preliminary and rapid analysis of satellite imagery coupled with lessons learned from scientific studies of postearthquake hazards in other locations. Remotely sensed data alone cannot be used to accurately determine hazard; ideally, a remote investigation would be accompanied by field reconnaissance. This preliminary report provides information that can be used to guide future field investigations.
Thousands of landslides deposited debris in channels, some of which created landslide dams that impounded water (
Most landslide dams fail within 10 days of formation, usually by overtopping (
Satellite imagery from Haiti after the 2021 earthquake and examples of two landslide dams that were flagged as warranting continued monitoring.
Figure 8. Satellite imagery from Haiti after the 2021 earthquake and examples of two landslide dams that were flagged as warranting continued monitoring.
Continued landslide activity adjacent to roads is likely to occur where destabilized debris or areas of cracking are present above or below the roadway. Continued landslides are likely during or after periods of rainfall or earthquake aftershocks. Numerous areas of minor rockfall along roads throughout the affected region may continue to affect roads (
Based on global studies of past earthquakes summarized by
Debris flows, not commonly triggered directly by earthquakes, are likely to pose a greater hazard in coming years, as they can mobilize the landslide debris that was deposited in channels. Debris flows are destructive flows of rock, soil, and water that can travel down steep channels for long distances at high speed and generally are triggered by intense rainfall. Debris flows can pose risks to low-lying settlements and infrastructure along channels and near rivers. After an earthquake, debris flows tend to be triggered by lower-intensity rainfall thresholds than before the earthquake, and this increased sensitivity can last from a few years to a decade or longer (
Landslides can occur without a specific obvious trigger at any time but are more likely to be concentrated during periods of intense or prolonged precipitation or aftershock activity. Debris flows also can be triggered by catastrophic landslide dam failures. The Atlantic tropical storm season (June–November), which peaks around mid-September through mid-October, is a period of significant concern for both landslide and debris flow hazards because of intense rain associated with tropical storms.
The likely footprint for increased landslide hazards is the area where earthquake-triggered landslides most occurred, as defined by (1) the USGS Ground Failure product landslide model and (2) the mapped landslide locations (
Debris flows are most likely to occur in steep drainages where the landslides triggered during the earthquake deposited sediment in and along channels (
The large amount of landslide debris in and above rivers and streams is a potential long-term hazard of riverbed sedimentation, aggradation, and consequent increased downstream flooding. Many landslides triggered by the earthquake deposited sediment in stream and river channels, and this debris may migrate down the river networks for several years or longer (
Map showing the relative density of landslides along 1,000-meter (m) river segments on the Tiburon Peninsula, Haiti, after the 2021 Nippes earthquake. For each segment, we counted all landslides within 500 m on both sides of the channel and assigned a relative landslide density for that segment. The analysis may double-count landslide points in adjacent 1,000-m segments and does not take into consideration the size of the landslide or whether it reached the channel. Low density is defined as 0–5 landslides per segment, moderate is 6–10, moderate–high is 11–25, high is 26–50, and very high is greater than 50. This map does not show the hazard or risk associated with large, long-traveled debris flows. The relative density of landslides that may have directly contributed sediment to channels illustrates the drainages most likely to experience changes to flooding behavior. This map does not depict potential hazards to downstream segments. N, north; W, west; km, kilometer; >, greater than.
Figure 9. Map showing the relative density of landslides along 1,000-meter river segments on the Tiburon Peninsula, Haiti, after the 2021 earthquake.
Satellite imagery was used to map landslides associated with the 2021 magnitude 7.2 Nippes, Haiti, earthquake. We performed a preliminary hazard analysis by identifying landslides that may pose a threat to people and property. There are limitations to performing a hazard analysis remotely, and thus, further investigation and monitoring would be beneficial to accurately determine the threats posed by these hazards. Our findings are summarized as follows:
The Tiburon Peninsula will continue to experience ongoing hazards related to the landslides triggered by the earthquake.
A minimum of 4,893 landslides were triggered by the earthquake and possibly Tropical Cyclone Grace.
We observed 292 landslide dams, 35 of which may warrant additional investigation and continued monitoring (see
We identified 116 landslides, 4 of particular concern, that likely affected roads by rendering them impassable or susceptible to subsequent damage from existing landslides (see
There will be an increased likelihood of flooding, including water and debris surges, in watersheds that experienced numerous landslides.
There will be an increased likelihood of landslides triggered by rainfall and aftershocks for several years.
There will be an increased likelihood of flood surges from landslide dam failures, some of which may not have been identified in this iteration of mapping.