Data from: Weak evidence of provenance effects in spring phenology across Europe and North America

Forecasting the biological impacts of climate change requires understanding how species respond to warmer temperatures through inter-annual flexible variation versus through adaptation to local conditions. Yet, we often lack this information entirely or find conflicting evidence across studies. The latter is the case for shifts in spring phenology---one of the most reported and consistent impacts of anthropogenic climate change, and also one of the most critical to forecasting, given its role in carbon sequestration. Some common garden studies have found evidence of important provenance effects, which suggest there may be local adaptation in the underlying cues of spring phenology and mirrors findings for fall events, while other studies find no evidence. Here, we synthesize common garden studies across Europe and North America that reported spring event dates for a mix of angiosperm and gymnosperm tree species in the northern hemisphere, capturing data from 384 North American provenances and 101 European provenances with observations from 1962 to 2019, alongside fall event data when provided. Across continents, we find no evidence of provenance effects in spring phenology, but strong clines with latitude and mean annual temperature (MAT) for fall events. These effects, however, appear to diverge by continent and species type (gymnosperm versus angiosperm), especially for fall events where clines with latitude and MAT are much stronger in North America. Our results suggest flexible, likely plastic responses, in spring phenology with warming, and potential limits---at least in the short term---due to provenance effects for fall phenology. They also highlight that, after over 250 years of common garden studies on tree phenology, we still lack a holistic predictive model of clines across species and phenological events. This dataset included seven deciduous angiosperm and eight evergreen gymnosperm species from 17 studies and 19 gardens, encompassing 384 North American provenances and 101 European provenances, with observations from 1962 to 2019. Seven species (five in North America and two in Europe) also had fall event information available. Most species in North American gardens were gymosperms (7/11 species) while most species in European gardens were angiosperms (3/4 species). The Excel file contains one sheet of metadata and one sheet of event date data included in this study. Each row of the data sheet represents a unique provenance and its event date of a specific year. The data provided includes unique codes to identify each provenance from each common garden, the latitude, longitude, elevation, mean annual temperature (C), climate overlap percentage, standard deviation, etc. The timing of spring phenological events is recorded as day of year (DOY), and fall event timing is included when available.