Conservation of butterfly populations in dynamic landscapes: The role of farming practices and landscape mosaic

In a dynamic landscape the rate of change in landscape structure can be even more important than the degree of patch isolation in determining population survival and abundance. If the changes in landscape structure occur at an "extremely" high rate (as in some anthropogenic changes), dispersal may not be able to keep up with the high rates of local extinction. Understanding impacts of such changes is thus crucial for determination of the driving factors for species survival in agricultural landscapes and for elaboration of conservation plans. Here we studied the effects of landscape dynamics under local farming practices on movements and population dynamics of a diurnal butterfly Maniola jurtina L. (Satyridae), specifically the impacts of regular yearly mowing on butterfly movements, distribution and abundance over many years. We used an existing simulation model, extended it by the effect of the intensity of disturbance (amount of mown habitat) and timing of disturbance within the reproductive season on the butterfly population dynamics, and validated on our data from a field experiment using a mark-release method. Increase in the amount of disturbed habitats in the landscape led to an increasing isolation of the remnant habitat patches for butterflies. This resulted in decreasing movements between habitat patches and ultimately to population decline, especially in less accessible patches. In the past, influence of landscape dynamics on species survival was usually considered at the long-term scale. our results demonstrate that the short time scale landscape dynamics induced by farming practices should not be neglected. The novelty of this paper stems in the combination of inclusion of landscape dynamics, of realistic dispersal strategies of individuals, and of considering real landscapes. The effect of man-induced landscape changes on population persistence of a real species in a real landscape has not been possible to be studied by any of the previously developed models. (c) 2007 Elsevier B.V. All rights reserved.