Landscape factors that contribute to animalevehicle collisions in two northern Utah canyons Ryan R. Jensen a, * , Rusty A. Gonser b,1 , Christian Joyner a a Department of Geography, Brigham Young University, Provo, UT 84602, USA b Department of Biology, Indiana State University, Terre Haute, IN 47809, USA Keywords: Animalevehicle collisions Mule deer Elk NDVI Gap data abstract Animalevehicle collisions (AVCs) are a serious problem that can result in property damage and human and animal injury and death. This study investigated the role that elevation, slope, and vegetation may have in AVC locations (mule deer Odocoileus hemionus and elk Cervus canadensis) in several canyon corridors in north central Utah. This was done by comparing these characteristics around known AVCs with those around control points where AVCs did not occur. The study found that elevation was significantly lower around AVC points, slope was significantly greater, and there was no difference in overall vegetation when measured with NDVI, but there was a difference in percentage of Sage brush (Artemisia tridentata Nutt) steppe around the points. AVCs may occur in these areas because mule deer tend to be more active in the lower elevations and use steeper slope for cover. Also, in areas with greater slope, the road will probably be curvier and provide less driver visibility than in areas with less steep slope. These results may help guide where to place deer warning signs and other AVC mitigation strategies. Ó 2014 Elsevier Ltd. All rights reserved. Introduction Animalevehicle collisions (AVCs) Humaneenvironment interactions occur when humans encroach on wild land or when humans convert wild land into usable land. Often times, roads are built through the middle of previously wildlands where they act as corridors for vehicles. These roads may intersect wildlife migration or movement patterns. Animalevehicle collisions (AVCs) represent one of the most com- mon and dramatic types of humaneenvironment interactions, and AVCs are a significant problem in many areas of the United States and throughout North America (Conover, Pitt, Kessler, DuBose, & Sanborn, 1995; McKee & Cochran, 2012). For example, in 2002 there were over 1.5 million AVCs resulting in over 1 billion dollars in damages, 150 human fatalities, and approximately 1.5 million white-tailed deer deaths (Curtis & Hedlund, 2005). In total, there are roughly 4100 AVCs per day in the United States resulting in daily damage of over 2.7 million dollars. The average insurance claim for an AVC is $3050 (Gkritza, Baird, & Hans, 2010). Unfortunately, the number of AVCs is increasing in North America (Ng, Nielsen, & St. Clair, 2008). This increase has been linked to rising human and animal populations (Found & Boyce, 2011). AVCs have been studied in a large variety of contexts and with different kinds of data. For example, Gonser, Jensen, and Wolf (2009) examined the spatial distribution of AVCs and habitat type in a western Indiana county. They found that the location of AVCs does not occur due to random chance and that habitat type prob- ably plays an important role in the AVC locations. As almost all AVCs occur at night (80e90%), Mastro, Conover, and Frey (2010) exam- ined factors that influence how motorists see deer at night using deer decoys. The authors found that deer are better detected when there are no reflective highway signs and the car’s high-beams are on. Motorists were able to detect deer equally whether they were moving or stationary. Finder, Roseberry, and Woolfa (1999) exam- ined the role of topographic features, highway construction vari- ables, and landscape metrics around areas with high numbers of deerevehicle collision in Illinois. The authors discovered that greater distances to forest cover resulted in fewer collisions. Found and Boyce (2011) discovered that deerevehicle collisions occur more in urban areas where the vegetation along the road was both dense and diverse, and that collisions are more likely to occur along roads with small-width groomed rights-of-way. Farrell and Tappe * Corresponding author. Tel.: þ1 801 422 5386. E-mail addresses: rjensen@byu.edu, ryan.jensen@byu.edu (R.R. Jensen), rusty. gonser@indstate.edu (R.A. Gonser). 1 Tel.: þ1 812 237 2395. Contents lists available at ScienceDirect Applied Geography journal homepage: www.elsevier.com/locate/apgeog http://dx.doi.org/10.1016/j.apgeog.2014.02.007 0143-6228/Ó 2014 Elsevier Ltd. All rights reserved. Applied Geography 50 (2014) 74e79