The cues, responses to temperature and potential for mismatch in UK plant phenology
Changes in phenology are often cited as a key biotic impact of climate change. Consequently, understanding the major environmental cues and responses to those cues in different species is important for making predictions about the future impacts and ecological implications of changing phenology. In this thesis, I set out to explore the phenological cues, mechanisms of response to temperature and the potential for interacting species to experience phenological mismatch in a range of UK plants. To do this, I utilised phenological records from two citizen science schemes; the well-established Nature’s Calendar, which collects observations for the UK Phenology Network (UKPN), and Track a Tree, a novel project I set up specifically to examine the phenology of interacting plant species in UK woodlands. I first assessed the ability of plasticity to track shifts in the optimum phenology for 22 plant species. I employed a statistical approach to estimate the plasticity and temperature sensitivity of the phenological optimum for leafing and flowering dates obtained from the UKPN. In identifying the most important cues I found that all species are sensitive to spring forcing temperatures, with plastic responses ranging from -3 to -8 days °C-1. Chilling temperatures in autumn/winter and photoperiod were important in species with early and late phenology, respectively. In seven species, plasticity was sufficient to track geographic variation in the optimum phenology. In four species, plasticity did not track the optimum, which is consistent with clinal local adaptation to temperature, and which could place phenology under directional selection in a changing climate. I then performed a phylogenetic comparative analysis on the median phenology and estimates of plasticity and local adaptation for the 22 species analysed previously. I found that phenological event (leafing or flowering) and growth form (woody or herbaceous perennial) predicted plasticity in phenological response. These traits may help inform future predictions of phenological responses to temperature. In contrast, the median date of phenology and clinal local adaptation over latitude were not predicted by any of the ecological traits considered. I next used records from the Track a Tree project to examine the relative phenology of canopy tree and understorey flowering species across UK woodlands. I found that first leafing and peak flowering of focal species pairs were correlated over space, and that the time between canopy leafing and the ground flora flowering (relative phenology) was spatially consistent. Relative phenology of two canopy tree species pairs was spatially consistent, but for a native versus non-native tree species pair the relationship varied over space (with a slope close to 0). If temperature-mediated plasticity determines these species’ phenology, my results suggest understorey flowering may be able to track canopy leafing in future, maintaining shading interactions. Finally, I used the Track a Tree data to partition the variance in phenology for seven tree species, and test what predicts variation in oak and birch. I found that the contributors to variance differ among tree species, with spatial variables important, and within site variance low, for all species except sycamore. The low intraspecific within-site variance suggests that some species may have a limited capacity for phenological buffering. These findings contribute to understanding what impacts on the phenological distribution of different species, an important requirement for assessing the phenological buffering of mismatch. In this thesis, I broadened the range of approaches that can be used to understand plant phenology in a changing climate. I demonstrated the value of employing novel statistical methods to analyse existing phenology data and the utility of hypothesis driven citizen science for predicting phenological shifts and the subsequent ecological implications for interacting species.