Local and landscape effects on the butterfly community in fragmented Midwest USA prairie habitats

• Published in: Landscape ecology 2007-11, Vol.22 (9), p.1341-1354
• Main Authors: Davis, Jessica D, Debinski, Diane M, Danielson, Brent J
• Format: Article
• Language: English
• Subjects: Animal, plant and microbial ecology; Applied ecology; Biological and medical sciences; butterflies; Community composition; Environmental factors; Floral resources; Fundamental and applied biological sciences. Psychology; General aspects; habitat fragmentation; Insecta; Invertebrates; natural resource management; Ordination; Partial least squares regression; Spatial extent; Variables; Variance partitioning
• ISSN: 0921-2973; 1572-9761
• DOI: 10.1007/s10980-007-9111-9

The fragmented landscape of the Midwest USA includes prairie remnants embedded in an agricultural matrix, potentially impermeable to dispersing individuals. We examined butterfly responses to local (environmental variables measured within the prairie fragment itself such as vegetative characteristics) and landscape (environmental variables measured up to 2 km surrounding the fragment, but not the fragment itself) factors at 20 prairie remnants in Iowa. Our objectives were to: 1) document how the composition and configuration of the landscape affects butterfly community within the fragment, 2) determine whether explanatory power is gained by including both landscape and local variables rather than only local variables, and 3) analyze differences in butterfly community composition between linear and block shaped fragments. Results from partial least squares regression suggest there are significant effects of the landscape on butterfly community composition at all spatial extents investigated. The local variable that was most highly correlated with butterfly community response was percentage litter, while percentage of roads was the most important variable at all landscape spatial extents. Ordination diagrams clearly separate linear from block sites based on butterfly community composition. Variance partitioning using partial canonical correspondence analysis indicated that landscape variables at all spatial extents add additional explanatory power beyond local variables with little overlap in percentage of variation explained. Our results suggest that butterflies are making decisions based both on the local and landscape environmental factors, thus land surrounding prairie remnants should be included in management decisions.