Researchers combined 15 years of fungal sampling from 103 plots in Catalonia with machine-learning models and detailed climate data. The plots covered pine forests from about 300 to 2,000 metres in elevation and several climate regions.

The projections differed by elevation and by fungal lifestyle. Species richness and abundance-weighted diversity were expected to decline most at higher elevations, while the largest changes in community composition were projected at lower elevations. Effects were stronger under the high-emission RCP 8.5 scenario than under the moderate-emission RCP 4.5 scenario.