Modeling Wildlife Damage to Crops in Northern Indiana

Abstract

Comprehensive information on crop damage by wildlife species is critical if effective strategies for controlling wildlife damage are to be formulated. Discriminating how landscape composition and configuration attributes influence crop damage is important for implementing landscape management techniques to resolve human–wildlife conflicts. We analyzed crop damage data from 100 corn fields and 60 soybean fields located in the Upper Wabash River Basin of northern Indiana during 2003 and 2004. We used negative binomial regression to model the rate of damage to corn and soybean crops in response to local and landscape variables. Rate of crop damage was best predicted by a combination of local and landscape variables for both corn and soybeans. Models with landscape configuration variables were better able to explain patterns of corn damage, and models with landscape composition variables (specifically, amount of wooded areas) were better able to explain patterns of soybean damage. In general, rate of crop damage was negatively related to size of the crop field and positively related to proportion of a field’s perimeter that was adjacent to wooded areas, amount of wooded areas, amount of forest edge, and mean size of forest patches. Specific associations between local and landscape variables and rates of crop damage may serve as a guide to planting strategies and landscape management to minimize wildlife damage to crops.

Keywords

• crop damage
• human–wildlife conflicts
• indiana
• landscape composition and configuration
• marmota monax
• negative binomial regression model
• odocoileus virginianus
• procyon lotor
• raccoon
• white-tailed deer
• wildlife damage
• woodchuck