AbstractGlobal warming is pushing populations outside their range of physiological tolerance. According to the environmental envelope framework, the most vulnerable populations occur near the climatic edge of their species' distributions. In contrast, populations from the climatic center of the species range should be relatively buffered against climate warming. We tested this latter prediction using a combination of linear mixed effects and machine learning algorithms on an extensive, citizen‐scientist generated dataset on the fruitbody productivity of the Burgundy (aka summer) truffle (Tuber aestivum Vittad.), a keystone, ectomycorrhizal tree‐symbiont occurring on a wide range of temperate climates. T. aestivum's fruitbody productivity was monitored at 3‐week resolution over up to 8 continuous years at 20 sites distributed in the climatic center of its European distribution in southwest Germany and Switzerland. We found that T. aestivum fruitbody production is more sensitive to summer drought than would be expected from the breadth of its species' climatic niche. The monitored populations occurring nearly 5°C colder than the edge of their species' climatic distribution. However, interannual fruitbody productivity (truffle mass year−1) fell by a median loss of 22% for every 1°C increase in summer temperature over a site's 30‐year mean. Among the most productive monitored populations, the temperature sensitivity was even higher, with single summer temperature anomalies of 3°C sufficient to stop fruitbody production altogether. Interannual truffle productivity was also related to the phenology of host trees, with ~22 g less truffle mass for each 1‐day reduction in the length of the tree growing season. Increasing summer drought extremes are therefore likely to reduce fruiting among summer truffle populations throughout Central Europe. Our results suggest that European T. aestivum may be a mosaic of vulnerable populations, sensitive to climate‐driven declines at lower thresholds than implied by its species distribution model.