Climate change affects the outcome of competitive interactions-an application of principal response curves.

It has been hypothesised that climate change may affect vegetation by changing the outcome of competitive interactions. We use a space-for-time approach to evaluate this hypothesis in the context of alpine time-of-snowmelt gradients. Principal response curves, a multivariate repeated-measurement analysis technique, are used to analyse for compositional differences in local ridge-to-snowbed gradients among 100 m altitudinal bands from 1,140 to 1,550 m a.s.l., corresponding to a temperature gradient of 2.5 degrees C (local lapse rate is 0.6 degrees C). The interaction between time-of-snowmelt and altitude is strongly significant statistically, indicating that the altitudinal gradient cannot be explained simply by the physiological responses of the species, but that there are also changes in the outcome of competitive interactions. At higher altitudes, there is a decrease in the time-of-snowmelt ranges of species which have intermediate times-of-snowmelt optima, whereas snowbed (chinophilous) species have wider time-of-snowmelt ranges. As snowbed species can survive, grow and reproduce at very early snow-free sites at high altitudes, the most likely explanation for their absence from all but the latest time-of-snowmelt habitats at lower altitudes is competitive exclusion by more vigorous lee-side species. This suggests that with future climate change snowbed species will experience, in addition to habitat fragmentation and reduced size of habitats due to increased temperature and snowmelt, an indirect effect due to competitive exclusion from late-snowmelt sites by species that have their optima outside snowbeds.