Publications

For the study of population distribution, the ideal approach would be to have a georeferenced population file at the point-coordinate level, based on the postal addresses. This file could be aggregated at the desired resolution for a specific exercise, providing complete flexibility. In this way, we could obtain population data at the building or block level for municipal analysis—even for neighbor-hoods in large cities—, or generate high-resolution population grids, enabling extremely detailed analyses. However, this information is not publicly available.In the research carried out by Goerlich and Mollá (2025), the 2021 census population grid from the Spanish National Institute of Statistics (INE), with a resolution of 1 km x 1 km, was disaggregated into 100 m x 100 m cells using dasymetric methods based on the Spanish High Resolution Land Use Information System (SIOSEAR) for 2017.The disaggregation process by Goerlich and Mollá (2025) uses the residential buildings of SI-OSEAR2017 as the intermediate geography. It is fully consistent with the original grid—aggregating the population of the census grid cell by cell—and uses the two key types of information that the literature on dasymetric spatial disaggregation methods identifies as relevant: the typology of the buildings (residential versus non-residential) and their height (as the population lives in 3D). This study recovers information that was not previously stored and that we believe may be extraordinarily useful in practice: the population of the intermediate geography—i.e., residential buildings—prior to aggregation into the 100 m x 100 m cells carried out by Goerlich and Mollá (2025). This provides an estimate of the population per building, that may be of interest for multiple applications.