Predictive models of herpetofauna road mortality hotspots in extensive road networks: Two approaches and a general procedure for creating hotspot models that are useful for environmental managers

Background/Question/Methods
Amphibian and reptile road-kill can result in serious population declines. Spatially-explicit models of road-kill risk to herpetofauna indicate that effective planning and mitigation of road mortality necessitates attention to the entire regional road network. The time and expense to adequately survey an extensive road network may be prohibitive to agencies, however, so there is a need for accurate and efficient models to prospectively identify the most promising sites for monitoring and mitigation. I review two studies, one focused on all herpetofauna and a second focused on turtles, which were designed to identify valid predictors of hotspots of road mortality in northeastern New York State. Road survey methods for amphibians and reptiles were developed and validated. I then surveyed a representative sample of rural highways using these methodologies to measure spatial patterns of road mortality. Using GIS, I created predictive models of road mortality hotspots by comparing the spatial patterns of road-kill to road traffic volumes, wetland distribution and configuration, and surrounding land-use and land cover at different spatial scales. I validated these models by locating segments of highway that were predicted by my models to be road-kill hotspots, and matched points that were predicted to be cold-spots.

Results/Conclusions
Road-kill was highly clustered, and localized to short (under 300m) segments of road. Road-kill sites were most strongly associated with nearby (within 100 m) large wetlands embedded within a forest matrix, along highways with relatively high traffic volumes. Wetland configuration was important: causeways (nearby wetlands on both roadsides) had higher densities of road mortality than similar sites where surface water was limited to one roadside, and causeways with certain landscape and road attributes had the highest road-kill toll overall. Validation surveys using the predictions of the models indicated that these models provide an accurate tool for predicting the most severe hotspots on a road network. I will discuss some of the informational and logistical challenges to developing hotspot models that are useful for management agencies. I argue that predictive hotspot models are tools that are essential for effective and economical whole road-network survey and mitigation, and for planning new road routes that avoid areas of high road-kill risk. Such models may prove to be especially effective for reptiles and amphibians, which have hotspots that are short in length but severe in road-kill toll.