Published Fall 2022
Nick Andrews, Dana Sanchez, Jenifer Cruickshank, Vee Blackstone, and Nik Wiman
In 2018 and 2019, vole populations irrupted in the Willamette Valley and remained very high through 2021. There are many different species of vole in the Pacific Northwest, but in the Willamette Valley the grey-tailed vole (Microtus canicaudus) stands out as especially important. They feed on roots, shoots,and bark, sometimes causing severe damage in vegetables, pastures, young orchards, vineyards, grass seed, and other crops. Conventional farmers often use rodenticides to manage voles, but those materials can have unintended impacts on non-target species and don’t provide complete control. Synthetic rodenticides are not allowed on organic farms, so organic farmers mainly rely on snap traps or other management techniques like crop rotation and tillage. The photos below show some of the extreme damage seen in the Willamette Valley during the recent vole irruption. Conner Voss and Sarah Brown’s pasture was decimated by November 2020 (Figure 1), and their adjacent vegetable crops were also badly damaged. They reported near total loss of root crops and even saw rodents eating peppers off the plant. In 2021, their local vole population declined quickly. Their root crops are now free of damage and their unirrigated pasture is beginning to recover. Voles are still present on their farm but seem to be in far lower numbers.
They are easy to distinguish from gopher and mole holes because they don’t create mounds. In 2021, we received a Western Sustainable Agriculture Research and Education grant (WSARE Project #OW21-364) to explore canine detection as a potential new tool for vole management on Willamette Valley farms. Vole populations rise and fall over time spans of several years and the causes of these multi-year cycles aren’t well understood or very predictable. Vole numbers more consistently fluctuate annually: increasing in the spring and summer, then decreasing over the fall and winter. Over the winter, voles significantly slow their reproduction. Based on this cycle, it makes sense that further reducing volenumbers in the winter before they breed should reduce the ability of a local population to increase enough to damage crops in the summer. Vole holes lead to networks of interconnected tunnels. In our preliminary investigations, we used an
insecticide fogger to blow vegetable oil smoke down holes (Figure 2a) to get a sense of how extensive and interconnected the tunnels are. At one location we found a network that spanned 11 feet with at least 40 entrance holes (Figure 2b) at the edge of a pasture. These voles were inflicting severe damage in adjacent vegetable crops.
Scouting fields for voles can be time-consuming. Burrow entrances persist over months and even years, so there can be thousands of vole holes per acre, with large tunnel networks, but the holes and runs themselves cannot really indicate number of animals alive at a given time. However, by looking closely for signs of current activity, such as fresh scat at burrow entrances and in runways, and recent clipping from feeding, you can assess how extensive the colony is, and whether the population is very low or high. Looking for these fresh signs is vital to maximize effectiveness when setting traps. Because setting and clearing traps takes even more time than the scouting, it would be helpful to have more confidence
that there is a vole in a particular tunnel before investing time in the effort. Dogs inherited from their wild for bearers tremendous abilities to scent, process, and track odors. These abilities, typically called “canine detection” when put to work by humans, are being put to use for many security, medical, social, natural resource, and sporting purposes. Agriculturists are beginning to explore this practice for different applications, but it is not yet utilized to full advantage. Our central hypothesis is that canine detection of live voles can make snap-trapping a feasible, non-toxic option to effectively manage vole populations on farms. To test this we are comparing canine-assisted trappers to unassisted human trappers. Our initial approach was to cross-train detection teams that have been successful in National Association of Canine Scent Work (NACSW) competitions, because those dogs have a proven ability to acquire a target odor and definitively “alert” or signal the handler at the location of that scent. Likewise, the dog handler has learned to recognize their dogs’ individual alert signs and identify the location of the scent.
David Leer and Daphne (CreekWalker Photography, Philomath, OR) in the foreground, and Donna Yanik with Sierra (QuarterWay Ranch, Lebanon, OR) in the background are Elite competitors in NACSW trials, and are collaborating in our research. Here they are training with their dogs to alert on active vole holes at the OSU Dairy Farm in February, 2022. Photo by Vanessa Blackstone © Oregon State University.
Our first year taught us a lot! To allow the dogs to acquire scent-signature for live voles, we placed field-captured voles in perforated PVC tubes (design adapted from the Barn Hunt Association rulebook) and placed them in trenches and holes we prepared in a mowed grass field. Sharon Gakstatter (For the Love of Dog) coached our first-year cohort as we trained our dogs to alert on holes and tunnels containing the live voles. We used Sherman traps to capture the live voles during November 2021 for this training. Although six animals were securely and comfortably housed in an OSU lab animal facility (OSU ACUP # 2021-0187), two animals died in the traps. Upon necropsy, our Attending Veterinarian (Dr. Jennifer Sargent) discovered those animals were infected with Leptospirosis. A pooled urine sample from the six - surviving voles also tested positive, therefore we halted work and developed bio-hazard handling protocols for the voles and biohazard management procedures for each stage in a day’s work, from transport to setting animal tubes in the training area. Leptospirosis can also infect any mammalian species, including dogs and humans. After consulting with Dr. Sargent and a Biosafety Officer (Dr. Matthew Philpott) we concluded that infection risk was too high to continue work with live animals. Therefore, we euthanized the captive voles and collected vole scent for training. We suspended cotton swabs in the vole cages for one week, and rubbed cotton swabs directly on the animals immediately after death. We also went to great lengths to develop a biohazard protocol and safely freeze dry the voles as a training tool. These alternatives (air swab, rubbed swab, freeze-dried vole) were placed in glass jelly jars with pierced metal lids, but subsequent dog-training work was unsatisfactory in a field setting. Our next approach was to identify OSU and privately-owned pastures that still had current vole activity. We moved our training and accuracy testing to the pasture where we had trapped lives voles a few months earlier. However, a further complication in our first year of work was that after 2-3 years of extremely high population densities, the regional populations of grey-tailed voles had “crashed” – as is typical of irruptive, or “boom-and-bust” species. The vole population in that pasture had fallen dramatically so we moved to another pasture where we found a large enough vole population for training. We tested dog/handler accuracy (number of holes identified compared to number of individuals and species trapped) at a third pasture. Based on those results, we narrowed our teams to David/Daphne, Donna/Sierra, and Nick/Bijou for the on-farm trials.
These preliminary results are promising. Winter snap-trapping assisted by canine detection teams may yet prove to be a viable, non-toxic tool in managing populations of voles on Willamette Valley farms. This year we are refining our canine detection training approach by collaborating with a professional conservation detection trainer/handler in order to benefit from subtleties and practices used in that
sector of canine scent work. We are also narrowing our research to focus on perennial crops (e.g., pastures, hay, and hazelnuts) because in the first year we realized that tillage and complex vegetable crop rotations precluded a fair comparison of canine and human vole detection on vegetable farms.However, we fully expect that our final results will be helpful to vegetable farmers and all others whose crops are perennially challenged by vole damage.
This article is based upon work that is supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, under award number 2020-38640-31523 through the Western Sustainable Agriculture Research and Education program under project number OW21-364. USDA is an equal opportunity employer and service provider. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture.