Coalition for Sonoran Desert Protection

By Jessica Moreno, Conservation Science Director

One of the best things about doing roadkill surveys is watching the sunrise. During the monsoons, the sun breaks across the eastern horizon and lights up the moisture laden air and morning cloud cover with light and throws the foothills into brilliant color. The sun feels hopeful and the day new. The second-best thing is knowing that every datapoint is helping us make things better.

The 20-mile stretch of Interstate-10 (I-10) between Vail and Benson, east of Tucson, Arizona, divides the regionally important Rincon-Santa Rita-Whetstone Mountains Wildlife Linkage. This wildlife pathway is one of only a few – somewhat safe – crossing points across I-10 found between Tucson and New Mexico, and it encompasses several protected areas and important waters, including Davidson Canyon and Cienega Creek, making it critically important for desert wildlife in the face of climate change.

Thanks to aerial support provided by LightHawk, in February 2020 we were able to take to the air and photograph the project study area. The above photo is a view of the Rincon Mountains to the north with Cienega Creek and I-10 in the foreground. The Cienega Creek bridge, on the far center right, is one of the project’s wildlife camera monitoring sites, and cameras have been placed at each of the tributary drainages where they cross I-10. Roadkill surveys stretched from SR 83 to SR 90. Photo by Jessica Moreno, CSDP.

This wildlife pathway is protected partially by federally protected lands including Saguaro National Park, Rincon Wilderness, and Las Cienegas National Conservation Area. Pima County has invested in securing additional conservation lands in this linkage as mitigation under the Multi-Species Conservation Plan (MSCP), including Bar V Ranch, Cienega Creek Natural Preserve, Rancho Agua Verde, and Colossal Cave Mountain Park, which are managed under the Cienega Creek Management Plan.

However, I-10 cuts through this area of wildlife movement, resulting in more wildlife-vehicle collisions for large animals such as black bear, mule deer, whitetail deer, javelina, coyote, and mountain lion, as well as smaller species like box turtles, opossum, Antelope jackrabbit, white-nose coati, raccoon, badger, and bobcat. It also serves as a potential movement area for jaguars and ocelots. Existing culverts and right of way fencing are not currently adequate at reducing wildlife-vehicle collisions or ensuring safe wildlife passage in this area. Future growth and increases in traffic volumes will only worsen the frequency of vehicle-wildlife collisions.

Since 2006, several qualitative assessments and reports have been produced that highlight the area’s importance for wildlife movement, but there has not been a study that has collected data to build a high accuracy model of wildlife movements and roadkill locations along this corridor. With the lack of such a product, nothing has been done on the ground to reduce wildlife-vehicle collisions and to improve safe wildlife passage by employing effective mitigation structures like winged highway fencing or wildlife overpasses. The Coalition for Sonoran Desert Protection, in conjunction with partners, is spearheading the effort to collect data to provide the information necessary to make this stretch of I-10 a safer passage for both wildlife and people in this important corridor.

Raynor VanDeven captured this photo of a mule deer near one of our culvert study sites under I-10, just east of Davidson Canyon, with bright tail lights on the highway above appearing to streak across the night sky. He uses a custom built wildlife camera trap to gather these professional photos. Raynor’s work has been a tremendous contribution to this project!

In the spring of 2017, CSDP conducted comprehensive assessments and wildlife surveys of the nearly 80 existing concrete box culverts and metal plate pipe structures between milepost 277-302 (Houghton Road to SR90), with participants from several government, nonprofit, and community partners, including the Arizona Game and Fish Department, Cochise County, Pima County, Tucson Audubon Society, Sky Island Alliance, Defenders of Wildlife, and the Wildlands Network. Our results indicated that this wildlife linkage could be made safer for wildlife and motorists by 1) installing wildlife funnel-fencing to keep animals off the highway and to direct wildlife toward existing crossing points; 2) retrofitting and widening existing drainage culverts located in high volume areas; and 3) construction of an additional wildlife crossing between Cienega Creek and the railroad underpass near Empirita Rd Exit 292.

Roadkill surveys were part of the next phase of this effort: to gather data on roadkill hotspots, together with wildlife passage rates using wildlife cameras in the best existing culverts. We hope this data will inform State and County highway and wildlife officials on where to focus mitigation efforts to improve highway safety and minimize wildlife-vehicle collisions with site-specific wildlife funnel-fencing installation, existing culvert retrofits, and new wildlife crossing structures. Our results illustrate the need to implement changes and provides baseline information to evaluate the success of future mitigation measures.

Our Desert Roadies project began with a team of volunteers to help collect wildlife-vehicle mortality data on Interstate-10 between SR83 and SR90, east of Tucson. Desert Roadies volunteers worked in teams of 2-4 people, including myself, to conduct driving roadkill surveys. Observations were recorded by our notetaker with the GPS coordinates. Surveys were conducted in the morning, starting up to 30 minutes before sunrise, every week for 6 weeks during the monsoon season in 2019 and 2020, beginning in July, except when conditions were bad for visibility or driving such as rain, dust storms, or other low visibility weather. In addition to these formal roadkill surveys, we collected other data on iNaturalist.org from community members and personally investigated reports about black bear mortalities. We also requested ADOT records on reported animal-vehicle crashes in the study area. This data was analyzed separately.  

Due to the onset of the Covid-19 pandemic in March 2020, carpooling with volunteers was no longer a safe option. So, we adapted as best we could by canceling the planned spring 2020 survey and having another monsoon season survey instead, which was conducted by myself and my family, Eddie Moreno (who is also a biologist experienced in roadkill surveys) and my two toddlers munching on donuts and “helping” look for animals from the backseat. This survey was done when the state was under curfew and lockdown, resulting in reduced traffic volumes. This likely was one reason for fewer observations of roadkill during this survey window.

Walking surveys are the ideal method for complete and accurate data. But safety concerns eliminated that option because we are using volunteer participation and this is a busy interstate highway with narrow right of ways. Therefore, the roadkill survey portion of this project was designed to be a broad-brush stroke only and we anticipated a small sample size. Recording categories of small, medium, large animals and/or by clade (reptile, bird, small/med/large mammal) are adequate for our purpose. Because of the project design, we have near zero detectability of small animals, and instead our target species are deer, bear, coyote, bobcat, gray fox, javelina, and those larger animals that the Arizona Department of Transportation consider most hazardous to drivers. The data will hopefully serve to provide a preliminary look at potential roadkill hotspot areas. For best accuracy 1) our GPS units had a one-button click to mark points quickly, 2) volunteers could also use maps and mileposts to confirm locations, 3) we analyzed the data in 1-mile segments, which better informs the fencing solution strategy and somewhat addresses imprecise location information, and 4) a staff biologist was always present to provide corroborating identification.

We completed two monsoon survey seasons for roadkill observations in 2019 and 2020, resulting in 78 data points of at least 14 different species. We were able to detect animals as small as a rock squirrel or cottontail rabbit. Skunks, raptors, and deer were not identified to specific species. All of our deer mortality observations (n=2) occurred between mileposts 296-297, near the Pima-Cochise County line. A hotspot of roadkill observations occurred between mileposts 291-295 (between Cienega Creek and Empirita Rd), in an area that coincides with a gap in available culverts and crossing points, and near several culvert locations between mileposts 283-287.

Coyotes, cottontail rabbits, and javelina were the most abundant roadkill species observed during our surveys. We also observed skunks, badger, opossum, raccoons, mule deer, jackrabbits, gray fox, domestic dogs, and raptors. Our 2020 monsoon survey occurred during a state lockdown at the beginning of the Covid-19 pandemic, and less traffic may be one reason for the reduction of roadkill during this survey period. Bruce Jacobsen created this “heat” map showing hotspots of our observations.

Our roadkill survey results for 2019 and 2020 monsoon survey seasons show hotspots around culverts east of Davidson Canyon, including an eastbound lane culvert that opens up to the median without a way across the westbound lane at milepost 285. Another larger hotspot appears between Cienega Creek and Empirita Road, which coincides with an area absent of available culverts for wildlife to use as crossing points.

The Arizona Department of Transportation (ADOT) collects crash records from police reports of vehicle crashes. Vehicle crashes caused by “animals” in recent years align with where we are seeing larger animals (deer, black bear) being hit most frequently on the interstate. Crash records can help make the case for increasing human safety by improving safe passage for wildlife – in addition to the goal of protecting healthy wildlife populations.

I-10 has been identified as a significant barrier to black bears in southern Arizona. In 2011, Todd Atwood et al published a study describing I-10 as a more significant barrier to functional gene flow for black bears than the U.S.-Mexico border, identifying I-10 as the dividing line between the Border subpopulation of black bears to the south, and the White Mountain subpopulation to the north. In our study area, these subpopulations meet and Romeo and Juliet romances unfold, ensuring healthy black bear populations in Arizona and in Mexico. For black bears, I-10 may be a barrier to gene flow as well as a “population sink” due to wildlife-vehicle deaths. This could pose a significant issue for the Border subpopulation, which additionally faces barriers to movement at the U.S.-Mexico border due to the construction of the border wall and other border-related infrastructure such as roads and lighting. Without safe passage both north of I-10 or south of the border, our southern Arizona black bears are at risk of genetic isolation and disappearing all together (what we call extirpation).

We have gathered four records of black bear mortalities in recent years, and three of these are verified with photos. These records indicate that the bears attempted to cross I-10 by following higher elevation ridgelines closely associated with the two largest underpass structures at Davidson Canyon and Cienega Creek. We also have animal track and camera trap data showing successful passage under I-10 at a few specific culverts. Most of our observations show bears moving south. April-May and August-September appear to be peak periods of activity for black bears in this area according to our data.

Black bears are just one example of the impact I-10 is having on native wildlife. In addition to large animals like bear, deer, and jaguar, smaller animals are at risk as well. Saguaro National Park has been monitoring a disturbing decline in “lost carnivores” over the last decade, including skunks, foxes, and badgers. Biologists are also concerned about the local disappearance of smaller range habitat specialist species affected by habitat fragmentation, including box turtles, Antelope jackrabbits, skunks, kit foxes, badgers, and white-nose coati. Increasing wildlife connectivity conditions for black bears and deer should allow struggling populations to repopulate from connected habitat areas and also benefit a wide array of other species.

In addition to our roadkill survey data, we were also able to collect four records of black bears killed by vehicle collisions in our study area. Bear roadkill occurred near mileposts 285 and 289-290, associated with the higher ridge lines close to Davidson Canyon and Cienega Creek. Deer collisions, on the other hand, were largely occurring near mileposts 296-297, where the landscape is naturally flatter.

With the data we’ve collected, we have solid evidence to back up the need for funding wildlife funnel fencing where it is most needed and make the case for new or improved wildlife crossings.

Read CSDP’s final project report to Arizona Game and Fish Department here.
(Please note: Photos of roadkill animals are included in the report)

Want to help?

Although collecting information on roadkill is not for the faint of heart, the data it provides is valuable. In the future, CSDP will be looking to continue our Desert Roadies program in more areas. In the meantime, folks can also document sightings of wildlife on roads in Pima County at https://www.inaturalist.org/projects/csdp-safe-passages.

Interstate 10 is a busy roadway, with 4 divided lanes and frequent commercial truck traffic. Adding wildlife funnel fencing to existing crossing structures is one simple solution to improve safe wildlife passages in this linkage. Photo by Matt Clark.

Acknowledgements

This work was made possible thanks to the time, effort, and skill of the following volunteers: Matt Clark, Sami Hammer, Bruce Jacobsen, Ken Lamberton, Eduardo Moreno, D’angelo Padilla, Raynor VanDeven, Althea Weeks, and Daisy Weeks.

We also want to thank Scott Sprague, Jeff Gagnon and Brit Oleson of the Arizona Game and Fish Department, and give special thanks to our collaborating partners: Myles Traphagen with The Wildlands Network and Brian Powell with Pima County Natural Resources, Parks and Recreation.

This project was funded in part by the Arizona Game and Fish Department Heritage Fund.