Temperature and photoperiod drive spring phenology across all species in a temperate forest community

Accurate predictions of spring plant phenology with climate change are critical for projections of growing seasons, plant communities and a number of ecosystem services, including carbon storage. Progress towards prediction, however, has been slow because the major cues known to drive phenology - temperature (including winter chilling and spring forcing) and photoperiod - generally covary in nature and may interact, making accurate predictions of plant responses to climate change complex and nonlinear. Alternatively, recent work suggests many species may be dominated by one cue, which would make predictions much simpler. Here, we manipulated all three cues across 28 woody species from two North American forests. All species responded to all cues examined. Chilling exerted a strong effect, especially on budburst (-15.8 d), with responses to forcing and photoperiod greatest for leafout (-19.1 and -11.2 d, respectively). Interactions between chilling and forcing suggest that each cue may compensate somewhat for the other. Cues varied across species, leading to staggered leafout within each community and supporting the idea that phenology is a critical aspect of species' temporal niches. Our results suggest that predicting the spring phenology of communities will be difficult, as all species we studied could have complex, nonlinear responses to future warming.