Annual Report: wildlifecrossing.net/california

Transcription

1 2011 Annual Report: wildlifecrossing.net/california Photo: Steve Bobzein & Felidae Conservation Fund Road Ecology Center & Information Center for the Environment September, 2011

2 California Roadkill Observation System Program Update (September, 2011) As some of you know, the third week in September was national Watch Out for Wildlife week. We hope you joined us in both watching out for wildlife so you didn t hit them and for those that do get hit, reported them to the CROS system online ( This will help us with our goal of building the case for solutions to wildlife-vehicle collision to benefit both animals and people. We are starting to write up the first phase of this work to submit to a scientific journal. We welcome collaboration in this endeavor from frequent and otherwise enthusiastic roadkill observers. Summary Roadkill occurs constantly on California roads and highways, affecting individual animals, populations, and species, as well as the people involved in the wildlife-vehicle collision (WVC). The California Roadkill Observation System (CROS, is the first statewide roadkill reporting web site and is a way for people throughout the state to record their observations of the dead animals and of their environmental context. By recording these roadkills, we collectively increase our knowledge of the potential and actual impacts of WVC and how we might reduce these impacts. Over the last 2 years, more than 650 observers recorded over 14,000 observations of roadkills from one end of the state to the other. Roadkills from 316 species were reported on all types of roads and highway, representing about half of the part and full-time vertebrate wildlife species in California. Some observers only recorded a single observation, others recorded hundreds. Based on photographs observers upload, species identification is >95%. Intensive studies of individual roads are providing estimates of the number of animals killed per mile and per year. Last year, we also developed a similar web site for the state of Maine, in partnership with Maine Audubon and their state agency collaborators ( One goal of the Maine program is to report both live and dead animals along roads and highways and to compare these occurrences to a statewide connectivity model that was recently constructed. We have also built a site, in collaboration with partners, in Colorado and we are in discussions with organizations in New York and New England about setting up wildlife observation websites for them too. As we collect more data in California and other states, we improve our ability to understand causes of roadkill, changes in wildlife distribution, and potential solutions to conflicts with traffic. "This [roadkill] is a relatively new source of fatality; and if one were to estimate the entire mileage of such roads in the state, the mortality must mount into the hundreds and perhaps thousands every 24 hours." (Grinnell, 1920)

3 How the System Works If you have registered on the system before, but haven t yet reported a roadkill, please consider registering again and join hundreds of other people in California who are watching out for wildlife. To use the system, observers register on the website ( and start entering their observations. Entries are made using an online form and include information about the type of animal, location, time of observation, approximate age of the carcass, and the environmental context of the carcass (e.g., type of road, nearby landscape). Registered users have the ability to store their own observation data online--as part of the greater study--and download it anytime as a comma-delimited text file (.csv). Users have the ability to generate a map of their own observations, to see the spatial distribution across the state, or map all roadkill occurrences in the system. We can also accept lists of past observations in table form. Just ask for a template table if you want to do your reporting this way. A new development is the creation of iphone and Droid applications that should be available in late As is the case for the rest of CROS, both applications are the result of volunteer effort on the part of programmers in different parts of the country. Transect Roads If you regularly monitor roadkill on a particular road or highway, consider joining our transect crew. Even after two years and thousands of observations, only 10% (>12,000 km) of the state s paved roads and highways have at least one reported roadkill, meaning that we are far from a true census of roadkill and estimation of impacts to individual species and biodiversity in general. To measure the ecological impacts of wildlife-vehicle collisions, we have developed index roads in several ecosystems where reporters regularly do sampling runs to intensively monitor for a roadkill on a monthly or weekly basis. These transects represent a sample of the population of road and ecosystem types. Results to Date When all observations to date are mapped, it is obvious that there areas where there are many reported roadkills and others where there are few reports (see map below). This represents a combination of uneven distribution of observers and of roadkills. We invite people in the areas with few reports to participate. So far, the largest identifiable group of reporters using the CROS are professional scientists (35%); next in line are natural history enthusiasts from other professions (20%). Finally, we have a large selection of individuals from all walks of life. We would like to improve our connections to all of these groups, so if you know people who can identify animals, or at least upload good pictures of roadkill for us to identify, please let them know about the website. We are also finding that there are few places where one could confidently state: Here is a hotspot. What this means is that reducing the number of roadkilled animals will take more than building wildlife passages under roads (good for urban areas), it will probably take large-scale public education. In the maps and tables below, we summarize findings from all records to date, as well as for specific wildlife groups.

4

5 Top 10 Species Observed Not surprisingly, the animals killed most-often on California roads and highways are also some of the most common and easy-to-spot species in California, at least among those tolerant of road-side environments. Scientific Name Common Name Observations Procyon lotor Raccoon 1693 Mephitis mephitis Striped Skunk 1372 Spermophilus beecheyi California Ground Squirrel 845 Didelphis virginiana Virginia Opossum 763 Odocoileus hemionus Mule (or Black tailed) Deer 761 Sciurus griseus Western Gray Squirrel 715 Sylvilagus audubonii Desert Cottontail 629 Tyto alba Barn Owl 486 Lepus californicus Black-Tailed Jackrabbit 466 Pituophis catenifer Gopher Snake 363 Single/Few-Occurrences Species Some of the least-commonly observed roadkill animals are a mix of rare and not so rare California native fauna. For certain species, observers unfamiliarity with the species could impede identification at other times the species are observed. Scientific Name Common Name Observations Aechmophorus clarkii Clark s Grebe 1 Aquila chrysaetos Golden Eagle 1 Buteo swainsoni Swainson s Hawk 7 Calidris alba Sanderling 1 Carduelis lawrencei Lawrence s Goldfinch 1 Cervus elaphus Tule elk 1 Chamaea fasciata Wrentit 1 Coleonyx variegatus Desert Banded Gecko 3 Crotalus cerastes Pitviper 2 Dicamptodon tenebrosus/ensatus Coastal/Pacific giant salamanders 8 Falco peregrinus Peregrine falcon 1 Gambelia wislizenii Long-nose leopard lizard 3 Glaucidium gnoma Mountain pygmy owl 1 Laterallus jamaicensis Black rail 1 Marmota flaviventris Yellow-bellied marmot 2 Martes pennanti Fisher 1 Mirounga angustirostris Northern elephant seal 1 Ovis canadensis Bighorn sheep 1 Picoides pubescens Downy woodpecker 1 Plegadis chihi White-faced ibis 1 Quiscalus mexicanus Great-tailed grackle 2 Scapanus orarius Coast mole 2 Uria aalge Common murre 1 Vermivora celata Orange-crowned warbler 1

6 Observer Participation A critical aspect of citizen-science projects like CROS is the recruitment and retention of observers. We have found two main kinds of events that encourage new and renewed participation in CROS messages to our membership and news stories about CROS. The graph below shows the rate of new observers/month and the cumulative total number of observers, as well as the correlation between participation spikes and and press exposure. Observer retention is harder to measure, but is an important component of a long-term successful program. One way to measure retention is to find out how long people who made recent observations have been members of CROS. For the most recent 200 observations made (while writing this report), the 29 observers who made those observation had been members for between 2 weeks and 2 years, with an average of 12.2 months (+ 6.4 standard deviation). This means that people reporting roadkill currently are a mix of long-term to recent observers, which is what you would expect with a high retention-rate system. Observer Accuracy One common question that CROS scientists receive is: Can we trust volunteer-observers and their data? This is an important question for these data to be used in ecological studies. People seemed to be very confident about most of their observations: 78% reported 100% certain and only 5% reported Best Guess. It looks like they were right. In our assessment of accuracy of species identification (based on >2,500 uploaded photographs of roadkills), >95% of identifications were correct. This is on par with accuracy rates for any roadkill study, where some carcasses are just hard to identify. Similarly, when we looked at an indicator for accuracy of location how close the points observers placed on the map were to the road, 88% of observations were placed on the road, and the average distance from the road for a sample of 100 points was 2.5 m.

7 Roadkill as Data about Wildlife Distribution Besides being an avoidable impact on wildlife, every time an animal is killed in a collision with a vehicle, it becomes a sample of that species. As roadkill are reported, they tell us about the distribution of that animal on the landscape. They are also indicative of potential places that animals are moving on the landscape. This relates to the property of landscape connectivity. Although roadkill occurrences don t indicate where animals successfully crossed roads, they can still tell part of the story about roadkill. Because connectivity modeling in California at the regional and state levels have not included actual animal occurrences, we are using roadkill occurrences to test the various connectivity maps that have been developed in parts of California. We have begun to analyze the roadkill data you have all provided from these perspectives and welcome your ideas and collaboration on these endeavors. Alien Invasion Over the last few decades, Eastern Fox Squirrel have been migrating from their original sites of invasion in California (Bay Area and Los Angeles) toward other areas, where they tend to displace the native Western Grey Squirrel. Because these two species are commonly-reported as roadkill, we can study their codistribution and where Eastern Fox Squirrel have successfully invaded or are invading the habitat of Western Grey Squirrel. As you can see from the map below, there are several areas where only Eastern Fox Squirrel or Western Grey Squirrel occur and other areas, such as Sacramento, Santa Cruz, and Monterey Bay where both species occur. These mixing zones may indicate active invasion by the non-native species.

8

9 Roadkill All-Stars There is a cadre of Caltrans biologists and maintenance staff who regularly report roadkill to the Caltrans database. We welcome all Caltrans staff to join the CROS community and report their roadkill here too. Although we have experienced a lot of media exposure for certain roadkill reporters, anyone who enters a record into CROS is an all-star for wildlife and for helping to reduce wildlife-vehicle collisions. We welcome anyone to join in reporting roadkill who has an interest in developing ways to reduce impacts from roadkill. What the Future Holds We are collaborating with wildlife agencies and advocates in various states (e.g., Maine, Vermont, Wisconsin) in developing state of the art methods for recording roadkill and live wildlife observations and using these data to evaluate proposed connectivity maps. This effort involves trained citizen scientists, expanding both the number of sites that can be surveyed and the size of constituency interested in solving the problem. We have similar partnerships under development in our home state too, with individuals and non-governmental organizations that recognize the severity of impacts to wildlife from roads and traffic and want to understand how we can reduce these impacts. We have provided data to a dozen people who have requested roadkill data for specific roadways or regions of California. To accomplish some of our goals, we are expanding our wildlife movement studies to include more highways and roads. We have developed roadkill transects where people can periodically census all roadkills occurring along specific roadways. Finally and most importantly, we are training observers and reporters to record live and dead wildlife occurrences so that private and public entities have the information needed to plan for reduced wildlife impacts, instead of increased impacts from development. Our staff, students, and collaborators use computer-based modeling tools to understand and predict wildlife use of habitats and conflicts that can arise with transportation and land-use. We are beginning to combine these modeling efforts with our improving understanding of wildlife occurrence and movement in California to inform smart and environmentally-sustainable land-use and transportation planning. Joins us in our collaborative efforts to measure and reduce the impacts to wildlife from human activity. Contacts Fraser Shilling, Lead Scientist, fmshilling@ucdavis.edu Dave Waetjen, Project Programmer, dwaetjen@ucdavis.edu Jim Quinn, Advising Scientist, jfquinn@ucdavis.edu Department of Environmental Science and Policy University of California, Davis This system is a collaborative project of the Road Ecology Center & Information Center for the Environment at UC Davis