Venezuela Earthquake: Sentinel-1 Likelihood of Damaged Structures

Following the June 2026 earthquakes that struck the central coast of Venezuela, the Microsoft AI for Good Lab ran their building damage assessment AI models to map and classify the buildings affected by the earthquakes. The models were applied to satellite imagery captured on 25 and 26 June 2026, producing building footprints with predicted damage across the affected areas of Catia La Mar, La Guaira, and Caraballeda. For each assessed area, the collection provides a reference area map, a valid-area mask describing the extent of the analysis, and a GeoPackage of predicted building damage footprints. This collection mirrors the dataset published on the Humanitarian Data Exchange (HDX) by the Microsoft AI for Good Lab and made available under the Creative Commons Attribution 4.0 International (CC-BY-4.0) license. ## Methodology The Microsoft AI for Good Lab ran their damage assessment AI models on satellite images taken 25 and 26 June 2026 and have mapped out the affected buildings. ## Disclaimer This analysis represents a portion of the damage from the earthquake; other damage in the region is not included in this assessment. While these insights offer a valuable initial overview, they remain preliminary and require on-the-ground verification for a comprehensive understanding of the damage. ## Caveats Please note that the source imagery has orthorectification quality issues, which may cause building footprints to be misaligned with existing basemaps.

Summary: The product maps urban areas and building footprints in Caracas and La Guaira, Venezuela, that show substantial post-event spectral change. The damage likelihood combines increased visible brightness and increased SWIR1 reflectance, which may be associated with exposed concrete, rubble, dust, debris, roof disturbance, or collapsed structures. Approximately 46,000 buildings were identified as having a high damage likelihood based on Sentinel-2 observations and building-level screening. These estimates are intended for rapid situational awareness and should be validated using higher-resolution imagery or ground observations. For each building footprint, damage-score statistics were calculated using a 20 m scale:Mean damage score  Maximum damage score  75th percentile  90th percentile  Valid Sentinel-2 pixel count Dates of Images and Input Data:Sentinel-2 pre-event composite: 1 January 2026 – 23 June 2026 Sentinel-2 post-event composite: 24 June 2026 – 30 June 2026 Sentinel-2 collection: COPERNICUS/S2_SR_HARMONIZED ESA WorldCover: v200, built-up class 50 Building footprints: Google Open Buildings v3, confidence ≥ 0.75 Summary: This experimental product provides a rapid assessment of approximately 46,000 potentially damaged buildings following the June 2026 Venezuela earthquake using Sentinel-2 imagery and Google Open Buildings footprints. It is designed to support emergency response, situational awareness, and prioritization of areas for further investigation. Suggested Use:Rapid earthquake response and situational awareness Screening of possible building damage hotspots Prioritization of high-resolution optical or SAR image tasking Regional-scale validation against commercial imagery Emergency management planning support Product Description: Approximately 46,000 buildings were identified as having a high damage likelihood based on the experimental Sentinel-2 assessment. The product was generated from pre- and post-event Sentinel-2 median composites using cloud- and shadow-masked surface reflectance data. A damage score combining visible brightness and SWIR1 changes was computed and restricted to urban areas using ESA WorldCover. Google Open Buildings footprints were used to identify structures with elevated damage likelihood following the June 2026 Venezuela earthquake. The product is intended for rapid screening and emergency response applications.Symbology: building_damage_likelihood.tif  Damage Likelihood Color Recommended Label -0.05 Blue/white Low Probability 0.05 Yellow Moderate Probability 0.15 Red High Probability Satellite/Sensor/Resolution:• Satellite: Sentinel-2A / Sentinel-2B• Sensor: MultiSpectral Instrument (MSI)• Processing level: Level-2A Surface Reflectance• Bands used: B2, B3, B4, B11• Visible bands resolution: 10 m• SWIR1 resolution: 20 m• Building-summary scale: 20 m• Temporal Baseline: 5 days Caveats and Limitations:Sentinel-2 cannot reliably confirm individual building damage at 10–20 m resolution. Building polygons are used only to summarize damage-score signals, not to directly observe structural collapse. Spectral changes may be caused by rubble, dust, exposed soil, roof material change, shadow, moisture, cloud contamination, or other non-damage factors. ESA WorldCover urban masking reduces but does not eliminate false positives. Google Open Buildings may contain missing, outdated, or inaccurate footprints. This product should be treated as an experimental damage-likelihood layer, not a definitive building damage inventory. Validation using high-resolution commercial optical imagery, SAR coherence/backscatter products, or ground reports is strongly recommended. File names: high_damage_buildings.kml — building polygons with high possible damage  building_footprints.kml — Buildings footprints intersecting the study area  building_damage_likelihood.tif — Sentinel-2 optical damage likelihood raster S2_Pre_RGB.tif — pre-event Sentinel-2 RGB composite S2_Post_RGB.tif — post-event Sentinel-2 RGB composite Esri REST Endpoint: See Layers below. Credits Khuong Tran (Khuong.tran@nasa.gov)

Building footprints for the Caracas area of Venezuela (bounding box 67.09W-66.65W, 10.28N-10.64N), Sourced from Overture Maps (buildings theme: OpenStreetMap, Microsoft, Google, Esri). Event: M 7.5 Earthquake, west of Caracas. - Source: Overture Maps - Part of HOT disaster response exports on HDX by HOT

Location of hazardous facilities in the affected areas of the earthquake of 24 June 2026 in Venezuela.

This dataset provides a rapid experimental Sentinel-1-based damage assessment product for the broader impacted area from the 24 June 2026 Venezuelan earthquakes. It includes both the entire analyzed area of the SAR analysis and the Overture Maps building footprints identified as likely damaged. This SAR product estimates 81,437 likely damaged buildings across the analyzed impacted area. This preliminary damage proxy product is intended to support rapid screening and prioritization, targeting of follow-up assessment, triangulation with VHR optical and radar images, and other damage assessments and products.

Building footprints across the area: every feature tagged `building=*`, from residential homes to hospitals, schools, and places of worship. Useful for population estimation, damage assessment, and urban planning. Coverage reflects where volunteer mappers have been active. Urban areas are usually well represented, remote areas may be sparse. Cross-check critical decisions with local field knowledge. - Source: OpenStreetMap, a free, collaborative world map - Tags: building - Part of the country exports on HDX by HOT