Search for a command to run...
The variability of populations over time is positively associated with their risk of local extinction. Previous work has shown that populations at the high‐latitude boundary of species’ ranges show higher inter‐annual variability, consistent with increased sensitivity and exposure to adverse climatic conditions. However, patterns of population variability at both high‐ and low‐latitude species range boundaries have not yet been concurrently examined. Here, we assess the inter‐annual population variability of 28 butterfly species between 1994 and 2009 at 351 and 18 sites in the United Kingdom and Catalonia, Spain, respectively. Local population variability is examined with respect to the position of the species’ bioclimatic envelopes (i.e. whether the population falls within areas of the ‘core’ climatic suitability or is a climatically ‘marginal’ population), and in relation to local landscape heterogeneity, which may influence these range location – population dynamic relationships. We found that butterfly species consistently show latitudinal gradients in population variability, with increased variability in the more northerly UK. This pattern is even more marked for southerly distributed species with ‘marginal’ climatic suitability in the UK but ‘core’ climatic suitability in Catalonia. In addition, local landscape heterogeneity did influence these range location – population dynamic relationships. Habitat heterogeneity was associated with dampened population dynamics, especially for populations in the UK. Our results suggest that promoting habitat heterogeneity may promote the persistence of populations at high‐latitude range boundaries, which may potentially aid northwards expansion under climate warming. We did not find evidence that population variability increases towards southern range boundaries. Sample sizes for this region were low, but there was tentative evidence, in line with previous ecological theory, that local landscape heterogeneity may promote persistence in these retracting low‐latitude range boundary populations.