Species’ functional traits play an important role in ecosystem services and functioning. Functional diversity in bee communities enhances pollination services and crop yield, as interspecific morphological variation is related to resource partitioning such as differential flower use. The strong relationships between ecosystem services and biodiversity mean that we should be able to map ecosystem service potential using functional diversity estimates, but few studies have attempted this. Here we try this strategy by using trait hypervolume analyses and species distribution models to map functional diversity patterns for bee communities in Japan and determine their environmental drivers. Specifically, our workflow is to estimate bee community composition per grid cell by stacking species distribution models, then build hypervolume models using functional traits for bees to estimate functional diversity metrics for each community type. Distribution models for 17 Japanese bumblebee species show that current ranges are mainly driven by mean temperature of coldest quarter, annual precipitation, precipitation seasonality, precipitation of driest quarter, land cover, and elevation. Further, results for 12 measured traits show that functional diversity in Japanese bumblebees is mainly driven by traits related to proboscis length, pollen storage, body size, and nest size. This approach opens up new possibilities for quantifying and mapping ecosystem service provisions at regional scales.