Data from: Flexible response to short-term weather in a cold-adapted songbird

To improve survival during winter, temperate species use a variety of behavioural and physiological adaptations. Among songbirds, the maintenance of lipid reserves is a widely-used strategy to cope with the severity of winter; however, little is known regarding how multiple synchronously acting environmental mechanisms work together to drive these effects. In a context where temperate winter conditions are becoming more variable, it is important to widen our understanding regarding the flexible adaptations that may allow wintering species to adjust to projected climate change. Using a long-term dataset collected across multiple wintering populations (7 years; 8 locations), we analyzed the effects of daily variation in weather (e.g., temperature, snowfall) on the variation in energy reserves (i.e., fat stores) of wintering snow buntings (Plectrophenax nivalis). Our results support the prediction that birds carry more reserves to increase the safety margin against starvation when conditions are energy-demanding and access to food is unpredictable (i.e., colder, snowier conditions). Birds responded to daily changes in weather by increasing their reserves as conditions deteriorated, with maximal temperatures and snow depth being the most important predictors of fattening decisions. We also found that females consistently exhibited higher fat reserves than males relative to their body size, suggesting that differential physiological adaptations among sexes or social dominance may play an additional role in explaining variation in energy reserves across individuals in this species. Overall, our findings increase knowledge on phenotypic adjustments used by species wintering in temperate zones to match variation in their environment.Snow bunting winter banding data and associated daily weather variables for each individual captureSnow bunting winter (Nov 1 to Mar 20) banding data (including age, sex, banding location, body mass, wing chord, fat score and time of capture) merged to associated daily weather data (including mean temperature, minimal temperature, maximal temperature, snow depth, total snowfall, absolute humidity, maximal wing gust and cloud cover (for the period 2009-2015 for 8 locations in eastern Canada. Banding data were obtained from both the Canadian bird banding office and the citizen science project Canadian Snow Bunting Network. Daily weather variables were merged to banding data and were extracted from the following three sources : 1) environment and climate change weather office online, 2) ministère du développement durable de l'environnement et de la lutte contre les changements climatiques et 3) National snow and ice data center. Only individual record for which a complete set of information was available for every individual (i.e. sex, age, wing chord, fat score, body mass, time of capture) and only banding entries with an associated complete set of weather variables were kept in the dataset. Excel was used to calculate weather averaged over the three days preceding capture. R Software (3.2.1) was used to create the datafile and do all other manipulations.snbudata.cs