Preventing wildlife-vehicle collisions, one smartphone at a time
Whales aren’t the only animals that can suffer from colliding with vehicles. In the United States alone, an estimated one million vertebrates are killed every single day thanks to unfortunate encounters with cars, trucks, buses, and other vehicles on our streets and highways. When the animals-turned-roadkill are larger, like deer and other ungulates, those collisions also cost quite a bit of money. The 1-2 million large animals that get hit by cars each year in the US result in $8.4 billion in damages. The cost isn’t all monetary, though. Some five percent of wildlife-vehicle collisions (WVCs) result in human injuries, and around 200 people die each year either directly from collisions, or from complications arising from their injuries. For Utah State University’s Daniel D. Olson and colleagues, these statistics mean that “there is a current, critical need for accurate and standardized WVC data, because these are the foundation for mitigation projects that protect both motorists and wildlife.”
As is usually the case, there are limited resources to fund measures designed mitigate WVCs, like “exclusionary fencing” around roads, or wildlife overpasses or underpasses. The more data we have on where and when WVCs are most likely to occur, the more thoughtfully those mitigation measures can be deployed.
WVC data have been collected by ecologists and land managers for nearly 100 years, and over the course of the last century the technology for doing so has not changed one bit: pen and paper. But the last decade began to see some innovation. WVC logging software was designed for personal digital assistants (remember those?), but since they weren’t connected to the Internet, large data files still had to be manually transferred. Another entrepreneur designed a camera that could be outfitted to vehicle dashboards. While that allowed at least for photographic species identification, data still required manual transfer. In the last few years, some states have created web portals through which drivers can enter information and photos, but users still need an internet connection to do it (and broadband speeds are limited for mobile devices in rural areas), and spatial data is limited to descriptors like “Highway 345, mile marker 17” rather than precise GPS coordinates.
Olson thinks he can do better. He and his colleagues developed a smartphone app called “WVC Reporter,” and they road-tested it in Utah, where mule deer collisions cost drivers $7.5 million in repairs each year. The application is designed to be fairly simple. There’s a dropdown menu of the most commonly encountered wildlife species, and space to manually enter a species not on the list. The user also enters information about the sex (male, female, unknown) and age (adult, juvenile, unknown) of the carcass. If they have enough experience observing wildlife, they can optionally include information on fat measurement (which is a health indicator) and an ID number, if the animal has a marker or tag. In addition, the application automatically collects additional data without any effort by the user, such as the GPS coordinates for the location, which are automatically used to determine the nearest highway and mile marker. By automating this part of the process, Olson has removed any possibility for user error. In addition, the app can be used even if the device isn’t connected to the Internet; the data is simply saved and uploaded next time the phone comes online.
But that’s only half of the story. Olson and his team also created a web-based desktop application that can be used by urban planners, maintenance crews, and wildlife managers to analyze and visualize WVC data. Different viewing options allow for contextualizing the carcass data in different ways, such as based on vegetation, rivers, human developments, agricultural areas, topography, and more. The desktop version of the app also allows users to download the data in formats that can be read into GIS software, where more sophisticated spatial analyses can be performed. Managers can also manually enter reports made the traditional way – with pen and paper – from motorists who may not be quite as tech savvy or even own smartphones.
After six months of testing, the researchers determined that not only was the data more useful and reliable (average error for location data was just 4-5 meters) than if it had all been collected with pen and paper, but it was also cost effective, saving nearly $2000 that would have been spent on data entry.
But the real savings are put into context when considering both the vehicular damage that results from collisions, as well as the possibility to save more lives.
A single deer-vehicle collision costs, on average, $8388, and a moose-vehicle collision can run to more than $30,000. By comparison, the development and testing of the WVC Reporter app cost $34,000, and maintenance would run around $1500 per year. Only four deer-vehicle collisions or one moose-vehicle collision would need to be prevented for the entire system to be cost effective.
In addition, since data can be analyzed just moments after being submitted, wildlife managers and road maintenance crews should be able to save more lives. Olson explains, “if maintenance crews observe that deer carcasses are being reported in areas with exclusionary fencing over a short time period of days or weeks, they can examine the location for damaged fencing or open gates, allowing them to quickly address the problem while it is occurring to prevent further WVCs at that location. When WVC data are collected on paper forms, data can be months to years old before they are processed and examined.”
In the future, Olson hopes to include a mechanism to include photos from the cameras included on most smartphones, as well as to use the data to send push notifications to drivers’ smartphones when they find themselves in certain areas at certain times of day that are particularly prone to WVCs. – Jason G. Goldman | 06 June 2014
Source: Olson D.D., Bissonette J.A., Cramer P.C., Green A.D., Davis S.T., Jackson P.J. & Coster D.C. (2014). Monitoring Wildlife-Vehicle Collisions in the Information Age: How Smartphones Can Improve Data Collection., PloS one, PMID: 24897502
Header image: shutterstock.com