This study aims to describe and demonstrate the applicability of a novel approach used to develop and test new methods based on species distribution models (SDMs) to establish spatial predictions of EBVs for birds based on bird diversity metrics, such as the distributions of properties of key bird habitats. A major objective of this study is to determine how to build bird SDMs that can be used to derive spatial EBVs for birds at a regional scale. We used as predictors 16 environmental variables considered ecologically meaningful for birds at 100 m spatial resolution, including two bioclimatic variables (Bio17 = precipitation of driest quarter and Bio7 = temperature annual range) for three periods: ‘current’, ‘future 2050’, and ‘future 2070’, eleven land-cover (land use) predictors (forest edge, arable land, coniferous forest, broadleaf forest, clear-cut forest, vineyard, settlement area, river, lake, meadow, and swamp forest), the normalized difference vegetation index (NDVI) and two topographic variables: slope and topography. We used multiple modelling techniques in the biomod2 package in R v3.3 to build presence-only SDMs relating bird presence to environmental features for each species. Here, we show that the suitability estimated according to the SDMs can be used as a spatial ‘species distribution’ EBV (SD EBV) and reflect the habitat quality and trends in climatic and land use impacts on populations of bird species. These developments should facilitate bird monitoring and management across space and time, ultimately helping to identify priority bird conservation areas, estimate habitat suitability and provide early warning signs regarding bird distribution trends. In general, bioclimatic variables, topography and forest structure were indicated to have an important relation to the species probability maps generated on the basis of the SDMs, signifying a dominant role of bioclimatic variable Bio17 in the development of habitat suitability patterns.