Predicting the impact of global warming on the timing of spring flowering

Many plants flower in response to a change in the environment. Since one of the main goals for a plant is to complete a growth cycle in order to produce seed, flowering is a key stage in plant development. We have developed a statistical procedure for explaining the variations in flowering date, which is based on a well-accepted phenological model (growing degree-days). Our approach has several advantages over previous methods based around multiple-regression procedures, the main one being that we have a direct interpretation in terms of just two meaningful phenological parameters (thermal requirement and thermal threshold) per species. The model is used to classify 79 flowering plants. By using a statistical approach based on empirical p-values, we can decide which species can be regarded as sensitive to temperature. Our model, while a simplification of the real system, is easy to work with and enables the consequences of future temperature change to be predicted. By adopting a simple (linear), but realistic, approximation to the rise in temperature each spring, we derive a simple expression for the change in expected flowering dates under global warming. We use the expression to examine changes under three different climate change scenarios involving increasing warmth, oceanicity and continentality. Variations in flowering from species to species and year to year are explained in a straightforward manner by variations in our two parameters and the linear temperature functions, respectively. We find that the sensitivity of spring flowering dates to temperature is strongly governed by the continentality of the climate. We make predictions that will allow the assumptions used in constructing our model to be validated or repudiated. Our formulae can be used for any global warming scenario of the type we consider, whenever our basic assumptions hold. In particular, we predict the likely change in world-wide spring flowering dates under the likely climatic conditions in the 2080s as predicted under the Intergovernmental Panel on Climate Change scenario A1FI. Copyright (C) 2009 Royal Meteorological Society