Investigating social inequality of urban green spacedistribution using Sentinel-2: the case of Vienna

Urban green space (UGS) is known to provide several benefits for the local population, including regulating local climate and improving human health. The inequality hypothesis claims that these environmental amenities are unequally distributed across space and among different social groups. We propose using a continuous vegetation index derived from satellite imagery to investigate environmental inequality (EI) in UGS distribution. We used spatial autoregressive models to describe the relationship between the normalized difference vegetation index (NDVI) and socioeconomic variables in a case study on the city of Vienna at an unprecedented level of detail (250 m resolution). We show statistically significant evidence for the existence of EI in Vienna. Neighborhoods with a higher share of foreigners have significantly less UGS. Results are robust across spatial aggregation levels and alternative spatial and non-spatial model specifications. We find that our model outperforms alternative ground measure for UGS, as NDVI does not cluster around extreme values. We demonstrate the potential of satellite imagery to investigate complex social problems related to EI in urban areas.