Dasymetric Algorithms Using Land Cover to Estimate Human Population at Smaller Spatial Scales

Data repositories such as Eurostat and OECD provide important socioeconomic datasets useful to guide decision support towards reaching sustainable development goals. However, socioeconomic data are typically available at a limited spatiotemporal scale. In the Horizon Europe-funded AquaINFRA project, a specific scope is to make EU data more analysis ready. As part of this, transformations of data into common spatial entities are needed to facilitate cross-analysis in, for example, social-ecological modelling. This paper uses CORINE land cover as ancillary data and EUROSTAT population data to investigate binary and weighted dasymetric refinement strategies to arrive at areal interpolation algorithms to estimate population data at smaller spatial scales. Six different algorithms are presented, and their accuracies are tested with quality measures. Their limitations and further development potentials on how to make them more precise and expand their usefulness in the future to other types of socioeconomic data are discussed.