Colombia buildings and infrastructure data

Creamos evaluaciones de daños a nivel de edificio tras el terremoto en Colombia mediante el entrenamiento y la posterior ejecución de un modelo de inteligencia artificial sobre imágenes satelitales adquiridas después del desastre. El modelo de IA clasifica cada edificio identificado en las imágenes como "sin daños", "afectado" o "desconocido". La categoría "desconocido" se utiliza cuando el edificio no puede ser evaluado adecuadamente, por ejemplo, debido a la presencia de nubes. Utilizamos los polígonos de edificios de Overture Maps, que representan el estado de las edificaciones sobre el terreno antes del evento. Los resultados se distribuyen como un archivo vectorial en formato GeoPackage, con los siguientes atributos para cada edificio: - `id` – identificador único de Overture Maps y Google para cada edificio. - `damaged` – valor 1 si el edificio está dañado; de lo contrario, 0. - `unknown` – valor 1 si el edificio está cubierto por nubes, neblina, humo o si, por alguna otra razón, no fue posible clasificarlo; de lo contrario, 0. - `area` – área del edificio en metros cuadrados. We create building level damage assessments by training and then running an AI model on the post-disaster imagery. The AI model predicts whether each footprint in the imagery is "building", "damaged", or "unknown" (i.e. cloudy). We use Overture Maps and Google building footprints which represent the state on the ground pre-event and distribute the resulting data as a vector file GeoPackage with the following per-footprint attributes: - `id` – the Overture Maps unique ID for each footprint. - `damaged` – 1 if the building is damaged, else 0. - `unknown` – 1 if the building was covered by clouds/haze/smoke or otherwise unable to be classified, else 0. - `area` – area of the building in sq meters.

This dataset contains building footprints mapped in OpenStreetMap as part of the LATAM Regional Mapatón 2026 in Colombia. The data were generated by volunteers participating in the regional mapping campaign conducted from 13 June to 31 July 2026. The mapping focused on priority areas identified according to disaster risk exposure and vulnerability. The resulting open geospatial data can support disaster preparedness, risk analysis, local planning and climate change adaptation

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

Education facilities: schools, kindergartens, colleges, and universities tagged via `amenity` or `building`. Useful for school-access mapping, education planning, and resource allocation. 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: amenity, building - Part of the country exports on HDX by HOT

Microsoft AI for Good Lab ran their damage assessment AI models over an extended area of Pereira. Google summary: Out of a total of 152,128 building footprints with predictions in the study area, 8,656 were fully obscured by clouds. Of the remaining 143,472 non-cloudy footprints, the model predicted that 752 (0.5%) were damaged to some extent. Overture summary: Out of a total of 77,589 building footprints with predictions in the study area, 6,582 were fully obscured by clouds. Of the remaining 71,007 non-cloudy footprints, the model predicted that 441 (0.6%) were damaged to some extent. Microsoft summary: Out of a total of 123,924 building footprints with predictions in the study area, 7,731 were fully obscured by clouds. Of the remaining 116,193 non-cloudy footprints, the model predicted that 650 (0.6%) were damaged to some extent.

Microsoft AI for Good Lab ran their damage assessment AI models in Cali. The Lab utilized two building footprint sources, Overture and Google (posted in HDX). Results varied: Google - 320,178 buildings identified - 621 identified as damaged Overture: - 97,085 building identified - 266 identified as damaged Both geopackages are available.

Muestra las calles existentes en la Comunidad "Isla de León" en Cartagena. Estas calles no están pavimentadas y la superficie es de escombros. Capas vectoriales de lineas.

Muestra los cuerpos de agua dentro de la Comunidad "Isla de León" en Cartagena. Estos son naturales y artificiales (creados por el hombre). El agua es no potable. Capas vectoriales de poligonos.

Muestra las cercas que delimitan los terrenos entre vecinos en la Comunidad "Isla de León" en Cartagena. Capas vectoriales de lineas.

Muestra los palos de luz existentes en la Comunidad "Isla de León" en Cartagena. Este servicio de energía es social pero muchas de las viviendas tienen conexión irregular. Capas vectoriales de puntos.

Muestra los limites geográficos de la Comunidad "Isla de León" en Cartagena. Capas vectoriales de polígonos.

Muestra los puntos de agua colectivos en la Comunidad "Isla de León" en Cartagena. Este se maneja a través de mangueras comunitarias para quienes no tienen acceso de agua dentro de la vivienda. Capas vectoriales de puntos.

Muestra los caminos creados por las personas en la Comunidad "Isla de León" en Cartagena. No tienen el ancho suficiente para el paso de vehículos. La superficie es tierra y vegetación. Capas vectoriales de lineas.

Muestra las viviendas, iglesias y tiendas presentes en la Comunidad "Isla de León" en Cartagena. Resalta las viviendas de emergencia de TECHO COL en pilotis (elevadas a 1 metro y 50 centímetros) ya que se presentan inundaciones frecuentes. Capas vectoriales de poligonos.

Following a large mudflow that hit the city of Mocoa (Department of Putumayo, Colombia) on 31 March 2017 as a consequence of the heavy rains, the International Charter Space and Major Disasters has been activated upon the request of the National Unit for Disaster Risk Management (UNGRD) / Unidad Nacional para la Gestion del Riesgo de Desastres, IDEAM. In response to this emergency, UNITAR-UNOSAT has carried out satellite based damage analysis to assess (visible) building structural damage caused by mudflows. This map illustrates satellite-detected damaged structures including building damage density in Mocoa city and its surroundings. Analysis has been undertaken by using post satellite imagery acquired by Pleiades and Geoeye-1 satellites on the 7 and 10 April 2017 and pre satellite images acquired by Worldview-2 & 1 satellites on the 21 December 2016 and 26 December 2013. As result of UNOSAT satellite analysis, a total of 1,082 buildings were detected as damaged of which 736 observed as destroyed or washed out by mudflow and 346 observed as severely damaged. Due to the pronounced incident angle and cloud cover of the satellite images, UNITAR-UNOSAT has also used orthophotos collected by Corpoamazonia in order to validate satellite based analysis. Orthophotos and additional baseline data were provided by OSM Humanitarian Mapping Unit and UMAIC (Unidad de Manejo y Analisis de Informacion de Colombia) who is also providing information management coordination support for this event. Kindly note that the number of damaged structures could have been underestimated due to the parameters of the images. This analysis has not yet been validated in the field. Please send ground feedback to UNITAR-UNOSAT.

Archivo contiene las construcciones mapeadas durante el mapatón nacional de Colombia 2025

**UNOSAT code: FL20241112COL, GDACS ID: 1102988** This map illustrates potentially affected structures in Riosucio Municipality, Chocó Department, Colombia as observed from a Pleiades satellite image acquired on 20 November 2024 at 11:07 Local Time. Within the analysis extent, UNOSAT identified 60 potentially affected structures amongst a total of 2,085 structures observed. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to the United Nations Satellite Centre (UNOSAT).

**UNOSAT code: FL20241112COL, GDACS ID: 1102988** This map illustrates potentially affected structures in Bojayá Municipality, Chocó Department, Colombia as observed from a Pleiades satellite image acquired on 17 November 2024 at 10:40 Local Time. Within the analysis extent, UNOSAT identified 13 affected structures and 3 potentially affected structures amongst a total of 2,165 structures observed. This is a preliminary analysis and has not yet been validated in the field. Please send ground feedback to the United Nations Satellite Centre (UNOSAT).

**UNOSAT code: TC20201123COL** This map illustrates potentially damaged structures and buildings in Providencia Island, Providencia and Santa Catalina Department of Colombia as detected by satellite image acquired after landfall of the Tropical Cyclone IOTA on 17 November 2020. UNITAR-UNOSAT analysis used a WorldView-2 image acquired on 21 November 2020 as a post-event image. Within the Island boundary, UNITAR-UNOSAT identified in the cloud-free zones 1,110 damaged structures. Taking into account the pre-building footprints provided by Humanitarian OpenStreetMap, this represents about 90% of the total number of structures within the analyzed cloud-free areas.

**UNOSAT code: TC20201123COL** This map illustrates potentially damaged structures and buildings in San Andres Island, Providencia and Santa Catalina department of Colombia as detected by satellite image acquired after landfall of the Tropical Cyclone IOTA on 17 November 2020. UNITAR-UNOSAT analysis used a WorldView-2 image acquired on 18 November 2020 as a post-event image. Within the Island boundary, UNITAR-UNOSAT identified in the cloud-free zones 220 Potentially damaged structures. Taking into account the pre-building footprints provided by Humanitarian OpenStreetMap, this represents about 6% of the total number of structures within the analyzed cloud-free areas.

**UNOSAT code: EQ20260810COL, GDACS ID: 1557236** This map illustrates structures and buildings affected by the magnitude 7.4 earthquake with an epicentre south of San José del Palmar town in Colombia, with analysis focused on Viterbo in Caldas Department, approximately 50 km from the mainshock epicentre. Damage and destruction were identified primarily using Pleiades very-high-resolution satellite imagery acquired on 12 August 2026. In this area 55 buildings were identified as damaged and 99 were identified potentially damaged buildings. This is a preliminary analysis and has not yet been validated in the field. Ground verification is encouraged, and feedback can be sent to the United Nations Satellite Centre.

This dataset contains OpenStreetMap data exported from the HOTOSM disaster mapping response to the earthquake in Colombia in August 2026 (Glide: EQ-2026-000146-COL). On Monday, August 10, 2026, a 7.4-magnitude earthquake struck western Colombia, the strongest earthquake recorded in the country in over a century. The quake was centered in the Chocó department and was felt across much of the country, as well as in parts of neighbouring Ecuador and Panama, causing widespread destruction in Cali, Pereira, Manizales, Quibdó, and other cities. HOT worked with OSM Colombia, UNGRD (Colombian National Disaster Risk Management Unit), UN Maps, UNDP and IGAC to define priority areas and features. The data has been contributed by volunteers through the Humanitarian OpenStreetMap Team's Tasking Manager and represents buildings and roads. See the list of projects in the Tasking Manager here. For more information see the HOTOSM activation page. As of 8 September 2026: over 450 contributors.