Perennial bracket fungi show that recorder timing causes the apparent widening of the autumn mushroom season in GBIF records of 11 European countries, 1990–2025; corrected dates moved later

Earlier studies of opportunistic records reported that the European fungal fruiting season became wider and later (Kauserud et al. 2012, 1970–2007). Record dates also change when recorders change when they go out. We used 11.8 million day-dated Agaricomycetes records from the GBIF snapshot of 1 October 2026 for 11 countries, 1970–2025. Species with perennial fruitbodies (for example Fomes fomentarius) are present all year, so their record dates measure recorder timing. For each fleshy autumn species we calculated 10th, 50th and 90th percentile record dates from equal-size subsamples, raw and standardized to reference-period recorder timing. From 1990–2007 to 2017–2025 (4,719 species-country series; means of 11 country means, 95% intervals across countries), the perennial record window became 43 days wider. In raw records of fleshy fungi, the end of season (90th percentile) moved 6.1 days later (3.8 to 8.4) and the width increased by 3.3 days (−1.4 to 8.0; wider in 10 of 11 countries). After standardization, the end moved 2.8 days later (0.5 to 5.1), the start 7.2 days later (1.9 to 12.5), the median 3.3 days later, and the width changed by −4.4 days (−9.1 to 0.2; wider in 3 of 11). A split-half test with perennial species showed that the standardization removes about 85% of a shared recorder-timing shift. A model fitted to 1971–2007 year-to-year changes predicts start +7.3, median +3.8, end +4.2 and width −3.0 days, in agreement with the standardized values. Year to year, the end of season moved about 3 days later per °C of September–November warming. Standardization with the same year's recorder timing did not remove this response (2.9 versus 3.1 days/°C). Thus the season moved later, but it did not become wider after correction for recorder timing. Limitation: recorders also follow the fungi and warm autumns, so the standardized changes are probably conservative. Our design did not reproduce the 1970–2007 widening in the four original countries.

fungi · fruiting phenology · mushrooms · GBIF · citizen science · sampling effort · recorder bias · climate change · Europe · negative control · perennial polypores · out-of-sample test