Pop-AUT: Subnational SSP Population Projections for Austria

General Information The Pop-AUT database was developed for the DISCC-AT project, which required subnational population projections for Austria consistent with the updated Shared Socio-Economic Pathways (SSPs). For this database, the most recent version of the nationwide SSP population projections (IIASA-WiC POP 2024) are spatially downscaled, offering a detailed perspective at the subnational level in Austria. Recognizing the relevance of this information for a wider audience, the data has been made publicly accessible through an interactive dashboard. There, users are invited to explore how the Austrian population is projected to evolve under different SSP scenarios until the end of this century. Methodology The downscaling of the nationwide Shared Socioeconomic Pathways (SSP) population projections to the subnational level follows a multistep procedure developed to generate spatially explicit demographic projections for Austria that remain fully consistent with the most recent national SSP projections (IIASA-WiC POP 2024). Population potential surfaces (1×1 km): Gridded population projections for Austria (1×1 km resolution, 2020-2100 in 5-year steps, for all SSPs) are obtained from Wang et al. 2023. These projections are based on population potential surfaces that represent the attractiveness of a location in terms of habitability. The population potentials are generated using a machine-learning framework (random forest models) trained on geospatial information such as land use, roads, elevation, distance to cities, and elevation (see Wang et al. 2023). Aggregation to municipality level: The 1×1 km grids are spatially aggregated to Austria's municipalities (Gemeinden). For each SSP and time step, all grid cell values within a municipality (with partial area-weighting where necessary) are summed to obtain municipality-level population counts. Deriving municipality-level growth factors: For each SSP scenario, municipality-level population change relative to the 2020 baseline is calculated. The change is expressed as a multiplicative growth factor, i.e., the ratio of the projected future population to the 2020 population. Applying growth factors to observed 2020 population counts: The municipality-specific growth factors are applied to the observed 2020 population counts from Statistik Austria. This ensures that the SSP projections are anchored to official statistics rather than to the modeled baseline values. National rescaling to ensure SSP consistency: The adjusted municipality population counts are then proportionally rescaled such that, for every time step and SSP, their sum matches the updated nationwide SSP population totals from IIASA-WiC POP (2024). This step ensures numerical consistency of the subnational projections with the latest national SSP projections. Higher-level administrative units and gridded products: District (Bezirke) and province (Bundesländer) population counts are obtained by spatial aggregation of the municipality population counts. 1×1 km gridded population counts are generated by redistributing the municipality totals back to the grid cell level, using the grid cell-specific population potentials as allocation weights. This procedure yields fully coherent population projections across all spatial scales: grid → municipality → district → province → nation. Dashboard The data can be explored interactively through a dashboard. Data Inputs Updated nationwide SSP population projections: IIASA-WiC POP (2024) (https://zenodo.org/records/14718294) Population potential surfaces: Wang, X., Meng, X., & Long, Y. (2022). Projecting 1 km-grid population distributions from 2020 to 2100 globally under shared socioeconomic pathways. Scientific Data, 9(1), 563. Shapefiles: data.gv.at Version This release corresponds to Version 2.0, built upon the updated nationwide SSP population projections (IIASA-WiC POP 2024). File Organization The SSP-consistent population projections for Austria are accessible in two formats: CSV files for administrative units: BL (Bundesländer, provinces), BEZ (Politische Bezirke, districts), GEM (Gemeinden, municipalities), NUTS3 (NUTS-3 regions) Spatial raster files (1×1 km) in GeoTIFF and NetCDF format All files contain population counts in 5-year intervals from 2020 to 2100.