What Beekeepers See and Weather Stations Miss: Climate Perception in Two Italian Land Systems

Whether local observers and instrumental records describe environmental change alike is a recurrent question in climate-perception research. We address it for beekeeping, whose output depends on a variable that conventional monitoring does not record: whether plants secrete nectar during their short flowering windows. Ten recurrent observations reported by 45 beekeepers in two contrasting Northwestern Italian land systems, an Alpine province (Trentino, 15 respondents, 2025) and a plain–hill–coastal system comprising Piedmont and Liguria (30 respondents, 2022), were compared with published regional climate analyses for 1985–2024 and with annual series computed from ERA5-Land (the land component of the fifth-generation Re-Analysis of the European Centre for Medium-Range Weather Forecasts, gridded at about 9 km resolution). The design is comparative and exploratory. Three items were confirmed by the record, one was confirmed but under-estimated (about 1.5 °C of perceived warming against a fitted 2.1 °C for Trentino), two were partially confirmed, two were not detectable at the resolution of the available record, and two were contradicted, one on the attribution of parasite pressure to climate and one at regional scale. The most apparent disagreements reflect a mismatch between a regional, annual record and observations made over ten-day flowering windows within a few kilometres of the hive. The three sub-populations reported different classes of change: bloom timing in Trentino, spatial displacement of forage and non-climatic drivers in Piedmont, and slow floristic change in Liguria. We propose, as hypotheses, that these differences follow from the space–time sampling geometry of each production system, that landscape structure has epistemic as well as ecological affordances, and that supplemental feeding may reduce the visibility of landscape change (“adaptive blinding”). Local ecological knowledge and instrumental monitoring emerge as complementary observation systems; we specify phenologically indexed indicators for comparing them at the scale where losses occur.