Ecosystem service mapping needs to capture more effectively the biodiversity important for service supply

•Biophysical and species-based services are ES modeled mostly with biophysical and species data, respectively.•Weighted provider richness (WPR) is a metric of the biodiversity important for 9 species-based services.•Biophysical services and WPR correlate positively at continental scale but less so at national scale.•WPRs do not show the trade-offs identified for biophysical services.•Large-scale, spatial ES assessments should include data on species-based services for effective planning. Large scale mapping of ecosystem services and functions (ES) is an important tool for researchers and policy makers to inform nature management and policies but it relies mainly on ES modelled with biophysical data such as land cover, henceforth biophysical ES. Other ES, henceforth species-based ES, are modelled at small scales based on species providers. As species-based ES are rarely included in multi-service, large-scale spatial assessments, we do not know if these assessments provide accurate information for managing the biodiversity important for species-based ES. We calculate and map weighted provider richness (WPR) for 9 species-based ES by weighting species data in Europe by their functional efficiency derived from functional trait databases. We compare WPR spatial patterns with those of 9 biophysical ES at continental and national scales in Europe. We find positive correlations at continental scale, and weaker positive correlations or neutral relationships at national scale between biophysical ES and WPR. Patterns of synergies and trade-offs for WPR are different from those of biophysical ES and change from continental to national scale. WPR for most species-based ES are synergistic with each other but WPR for existence value has the weakest synergies with other WPRs. Biodiversity data is still insufficient to truly map species-based ES at large scales but WPR can represent the next step forward for spatial ES assessments. A lack of spatial information on species-based ES in large-scale assessments leads to inaccurate information on ES distribution, and their synergies and trade-offs, which can lead to misguided management and conservation decisions.