New Open Access Geospatial Datasets to Enhance Global Population Mapping

A groundbreaking collection of 73 global, high-resolution, multi-temporal geospatial datasets has been released, poised to significantly enhance the accuracy and detail of human population distribution modelling worldwide. Produced by WorldPop at the University of Southampton, in collaboration with the Università degli Studi di Milano Scienze della Terra and funded by the Bill and Melinda Gates Foundation, these datasets are now openly accessible to researchers, policymakers, and humanitarian organisations.

The extensive collection offers annual time series data from 2015 up to at least 2020, and as recent as 2023 where source data permit, covering the globe at a spatial resolution of approximately 100 meters (3 arc-seconds). This level of detail is crucial for understanding the complex spatial and temporal variability of population distributions, which is influenced by a range of dynamic factors.

These new datasets represent significant advancements over previous repositories, incorporating new and higher spatial resolution source data, such as an expansion to seven source datasets for the built-up environment, and improved processing algorithms. For instance, a new outlier removal algorithm was developed for nighttime lights, and a series of seven inland water masks based on different water coverage thresholds were created.

While primarily motivated by the need for a new global time series of high-resolution age- and sex-structured population estimates by WorldPop, the applications of these covariates extend broadly across environmental, economic, and health sectors, supporting informed policymaking and resource allocation for sustainable development.