Landscape Ecology

, Volume 21, Issue 7, pp 989–1001

Occurrence pattern of Pararge aegeria (Lepidoptera: Nymphalidae) with respect to local habitat suitability, climate and landscape structure


    • Department of Community EcologyUFZ – Centre for Environmental Research Leipzig-Halle
  • Carsten F. Dormann
    • Department of Applied Landscape EcologyUFZ – Centre for Environmental Research Leipzig-Halle
  • Debra Bailey
    • FAL Swiss Federal Research Station for Agroecology and Agriculture
  • Mark Frenzel
    • Department of Community EcologyUFZ – Centre for Environmental Research Leipzig-Halle
Research article

DOI: 10.1007/s10980-005-6057-7

Cite this article as:
Schweiger, O., Dormann, C.F., Bailey, D. et al. Landscape Ecol (2006) 21: 989. doi:10.1007/s10980-005-6057-7


Distribution patterns of wild species are affected by environmental variables, such as climate, anthropogenic land use or habitat quality, which act simultaneously at different scales. To examine the relative importance of particular factors and scales on population response we investigated the speckled wood butterfly Pararge aegeria (L.) as a model organism occupying semi-natural habitats. Its distribution was recorded in 23 study sites (5×5 km) over a 2 year study period. The sites were located in agricultural landscapes within seven Temperate European countries. Environmental predictors were mapped at a local and a regional scale. Logistic regression models were then developed to represent humid (beneficial) and dry (adverse) weather conditions during larval development. The humid year model predicted that P. aegeria is equally but generally not very dependent on local and regional factors, resulting in generally high occurrence probabilities. In contrast, the dry year model predicted severe restrictions of P. aegeria to both high quality patches and landscapes with beneficial structural and climatic preconditions. As both models resulted in entirely different predictions, our study showed that the sensitivity of P. aegeria to local and landscape features might change, and that factors of less importance could easily become limiting factors. The results stress that high quality landscape is important at both the local and regional scale even for species that are considered relatively robust. They also sound a note of caution when predictions about population response for management purposes are based on just a single or a few year(s) of observation.

Key words

ButterflyDistributionFragmentationHabitat qualityLogistic regressionOccurrence probabilityPredictive habitat model
Download to read the full article text

Copyright information

© Springer 2006