• https://human-settlement.emergency.copernicus.eu/
• https://gadm.org/
• https://www.openstreetmap.org/

Libertun de Duren, Nora Ruth, et al. (2026). Human Settlements and Urban Areas in Latin America and the Caribbean by Degree of Urbanisation (DEGURBA), 1975-2020. IDB Open Data. https://doi.org/10.60966/o35b2pu5

1. English

Which countries and epochs are covered?

26 IDB borrowing member countries; epochs 1975, 1980, 1985, 1990, 1995, 2000, 2005, 2010, 2015 and 2020.

Which files are included?

For each epoch, three tables (.xlsx) and the same information as polygons in GeoPackage format (.gpkg): normal, sin_conurb and solo_conurb. The solo_conurb file contains only transboundary polygons. Normal keeps transboundary conurbations as single polygons and sin_conurb splits them by country.

What variables does the dataset contain?

polygon_id (unique identifier); GID_0 to GID_4 and NAME_0 to NAME_4 (GADM codes and names of the administrative units intersected, levels 0 to 4; multiple units are joined with “+”); name (all settlement names found in the polygon); max_name (name assigned to the polygon); pop, pop_urban_center, pop_urban_cluster, pop_rural (population, inhabitants); area_km2, area_urban_center_km2, area_urban_cluster_km2, area_rural_km2 (area, km²); density (inhabitants/km²); in_amazon (yes/no: polygon located in the Amazon region). The total population of each polygon is measured by the variable pop. It is the sum of pop_urban_center and pop_urban_cluster, except for amazonian settlements that also considers pop_rural.

How should the population figures and settlement names be read?

Population figures are modelled estimates from a population grid, not census counts. Settlement categories in the Amazon region rely on calibrated thresholds and are not directly comparable with those outside it. Settlement names come from OpenStreetMap and Azure Maps and may be incomplete; names preceded by “Cerca de” identify unnamed settlements labelled after the nearest named locality.

What attribution applies?

Contains information from OpenStreetMap and its contributors, available under the Open Database License (ODbL); administrative boundaries from GADM.

Which step of the Degree of Urbanisation does it apply?

The dataset applies Step 1 (grid-cell classification) of the Degree of Urbanisation. Step 2 (classification of local administrative units) was not applied: administrative units in the region are large and heterogeneous, and classifying them would under-represent high- and moderate-density settlements located inside units classified as rural. Because the classification is based on a population grid rather than on administrative units, whose size and shape vary widely across the region, it captures settlements regardless of the administrative boundaries they cross, including transboundary conurbations.

How were the settlements identified and named?

(1) Grid cells are classified by population density and by the total population of contiguous cell clusters. Outside the Amazon region the original DEGURBA thresholds are kept (urban centre: at least 1,500 inhabitants/km² and 50,000 inhabitants; urban cluster: at least 300 inhabitants/km² and 5,000 inhabitants), and settlements are retained from 300 inhabitants with at least one cell of 200 inhabitants/km². In the Amazon region, the rural cluster threshold was lowered to 150 inhabitants/km² and 200–5,000 inhabitants. (2) Cells classified as low- or very-low-density rural are removed and the remaining adjacent cells (eight-neighbour connectivity) are merged into polygons, whose population is summed. (3) Polygons are named with OpenStreetMap place points following the hierarchy city > town > village > hamlet; unmatched polygons are named through Azure Maps reverse geocoding and, failing that, after the nearest named settlement preceded by “Cerca de”. Transboundary polygons take the highest-ranking settlement of each country. (4) Urban areas (at least 5,000 inhabitants) and cities (at least 50,000 inhabitants) are identified for each epoch.

What data sources were used?

Global Human Settlement Layer (GHSL), European Commission Joint Research Centre — GHS-POP, GHS-BUILT and GHS-SMOD, R2023A; GADM, Database of Global Administrative Areas, version 4.1; OpenStreetMap (© OpenStreetMap contributors); Azure Maps (Microsoft).

What is the methodological reference?

European Union/FAO/UN-Habitat/OECD/World Bank (2021). Applying the Degree of Urbanisation — A methodological manual to define cities, towns and rural areas for international comparisons, 2021 edition. doi:10.2785/706535.

Under what licence is it published?

Creative Commons Attribution 4.0 International (CC BY 4.0).

How do I cite this dataset?

Use the citation shown on this page, which carries the dataset's DOI.