Road kill records are often a primary source of information used to identify the best places to invest in highway improvements, like crossing structures, to benefit wildlife. In Washington there are two primary sources of information; carcass removals (both WSDOT maintenance staff removals and citizen salvage of deer and elk) and crash data. On state highways, deer carcass removals are three to four times the number reported as crashes. The carcass removal dataset is therefore the more complete representation of road kill numbers. However, the geographic distributions between the two datasets are very different, with carcass removals suggesting that the vast majority of deer killed by vehicles are east of the Cascade Mountains. Crash data suggest a much more even split between western and eastern Washington. There is evidence that officers reporting crashes more commonly patrol higher volume highways, creating a bias in reporting. Additionally, higher proportions of semi-truck traffic in eastern Washington may contribute to the presence of carcasses in places where a crash report is never completed.
Even the more complete carcass removal data under-represent the true number of wildlife-vehicle collisions. Comparing carcass removals with independent observations of road kill results in the conclusion that the true number of large mammals killed in vehicle collisions is a minimum of three times what is documented by carcass removals.
Very little information is available for local roads. Eighty one percent of road centerline miles in Washington are local roads. Carcass removal data for these roads don't exist. Citizens are required to report crashes where property damage exceeds $1,000 or there is a human injury or fatality. However, citizens report less than 15% of the state's wildlife-vehicle collisions and very few are on local roads. There are, however, a reasonable number of officer-reported crashes on local roads. In total they amount to no more than 25% of the wildlife-vehicle collisions recorded during any given year. State highways, with their higher speed limits and high traffic volumes, appear to have most of the state's wildlife-vehicle collisions.
Of all of the road kill data available in Washington, carcass removal data are the most useful for understanding the numbers and locations of wildlife-vehicle collisions. Crash data are less representative of true numbers and distribution, likely because of the way officers are deployed. The broadly deployed highway maintenance personnel and the motivated citizens salvaging road-killed deer and elk produce better, though imperfect, data on road-killed wildlife in Washington State.
Terrestrial Wildlife and Ecosystem Interactions with Transportation
Carcass collision crash road kill deer wildlife