Wildlife Damage Management II

Contributed Paper
ROOM: HCCC, Room 20

12:50PM Perceptions of Wildlife Damage on Agriculture; Management Implications
Michael R. Conover; Erin Butikofer; Daniel J. Decker
A national survey of wildlife agencies was conducted during 1957 to determine perceptions of wildlife damage to agriculture. This study was repeated in 1987, expanding the sample to include state Farm Bureaus and state Wildlife Extension Specialists. During 2017, we replicated these studies to determine how perceptions had changed over the last 30 and 60 years. During 2017, 22 different wildlife species were listed by >1 respondent as causing the most damage in their state. Several of these species, including moose (Alces alces), bison (Bison bison), feral hogs (Sus scrofa), bears (Ursus spp.), wild turkeys (Meleagris gallopavo), and ravens (Corvus spp.), were not listed by any respondents in either of the prior surveys, suggesting that problems caused by these species have increased in severity over the last 30 years. Despite this growing list of species causing damage, deer were listed by most respondents as producing the most agricultural damage in their state. State wildlife agencies, state Farm Bureaus, and state Wildlife Extension Specialists largely were in agreement about the level of damage caused by wildlife, but state wildlife agencies tended to rank damage by big game species higher than the other 2 respondent groups. Wildlife Extension Specialists considered damage by muskrats (Ondatra zibethica), woodchucks (Marmota monax), cottontails (Sylvilagus spp.), tree squirrels (Sciuridae), raccoons (Procyon lotor), and songbirds, to be more severe than Farm Bureau and wildlife agency respondents. Most respondents believed that wildlife damage to agriculture had increased during the last 30 years, but 48% of respondents from Farm Bureaus felt it had increased greatly versus only 14% of respondents from state wildlife agencies. State Farm Bureaus and Wildlife Extension Specialists had an excellent understanding of the services and products available from governmental sources to help mitigate wildlife damage on farms and ranches.
1:10PM Non-Linear Relationships of Anthropogenic Features Influence Wolf-Livestock Depredation in Michigan’s Upper Peninsula
Nicholas L. Fowler; Jerrold L. Belant; Dean E. Beyer, Jr.
Large carnivore populations are threatened by habitat loss and human retaliation in the wake of real and perceived threats to safety, property, and livelihood. Through legal protection many large carnivore populations in North America are increasing, but control of livestock depredation remains divisive among stakeholders, undermining conservation and management strategies. Evaluating interactions based on anthropogenic features has provided insights into behavioral ecology and the development of mitigation strategies facilitating coexistence as opposed to further loss of carnivore habitat. Pursuant to this, we evaluated anthropogenic characteristics influencing wolf (Canis lupus)-livestock depredations (hereafter depredations) in Michigan’s Upper Peninsula (the UP). Using case-control generalized linear models in a multi-scale approach we tested our hypothesis that humans and anthropogenic features may facilitate and inhibit depredations across their range of magnitude. Across scales we found support of non-linear, concave quadratic relationships between cattle density, human density, and agricultural lands to occurrence of wolf-livestock depredations (n = 260). We demonstrated that at the wolf territory scale road density reduces the probability of depredations, likely a mechanism of decreased wolf survival and density. Accuracy of models predicted 90 and 84% (n = 28 and 26) of depredations not involved in model construction at the land section and core wolf territory scales. In comparison to other works, we provide commentary on how our hypotheses and results are more in line with large carnivore ecology. We provide improved, updated models and risk maps of depredations in the UP robust to wolf distribution and previous depredations. This makes our approach applicable to areas where wolves are not present but may be in the future. Following recommendations of other works we incorporate our results and depredation risk maps into easily accessible, searchable, and open access formats for wildlife managers and the public.
1:30PM Evaluating Blackbird Behavioral Response Toward Unmanned Aircraft Systems (Uas)
Conor C. Egan; Bradley F. Blackwell; Esteban Fernández-Juricic; Page E. Klug
Animals respond to nonlethal forms of human disturbance using behaviors adapted to avoid predators. Wildlife managers can potentially exploit these behaviors when using disturbance stimuli to increase perceived predation risk and possibly encourage animals to abandon a resource patch. A promising tool in the field of wildlife damage management is the unmanned aircraft system (UAS), able to overcome the mobility limitations of other hazing strategies. In particular, multirotor UAS are ideal in that they are a multi-functional tool in precision agriculture and relatively easy to fly. Thus, multirotor UAS have been proposed as a hazing tool to minimize blackbird damage to agriculture crops and bird strikes in the aviation industry. Our specific objectives were to 1) evaluate the antipredator response of captive red-winged blackbirds (Agelaius phoeniceus) to three UAS platforms (i.e., multirotor, fixed-wing, and predator model) approaching at direct and overhead trajectories; and 2) conduct a field evaluation of UAS effectiveness as a hazing tool. We did not observe an effect of trajectory on alert response, however, blackbirds alerted to the predator model (mean ± SD seconds before the UAS reached the bird; 15.50 ± 7.98) approximately 8 seconds earlier than the fixed-wing (7.51 ± 6.97), and approximately 13 seconds earlier than the multirotor (2.49 ± 2.11). Additionally, blackbirds returned to foraging earlier and alarm-called and took flight less frequently in response to multirotor approaches compared to the predator model. Overhead approaches failed to elicit flight, suggesting UAS hazing may be most effective at low altitude, direct approaches. In direct approaches, only the multirotor failed to elicit an escape response. In the field blackbird flocks responded to all three platforms by taking flight. Compared to a simulated predator approach, a multirotor may be a suboptimal hazing tool as evaluated through the antipredator responses of individual blackbirds in semi-natural settings.
1:50PM Will Zinc Phosphide-Coated Cabbage Allow for Effective Management of Belding’s Ground Squirrels?
Roger A. Baldwin; Heather Halbritter; Ryan Meinerz; Laura K. Snell
Belding’s ground squirrel (Urocitellus beldingi) is a significant pest of alfalfa in northeastern portions of California and southeastern Oregon. In 2014, a zinc phosphide-coated cabbage bait was approved for use against Belding’s ground squirrels in Oregon, with subsequent approval obtained in California in 2015. However, no data existed as to the efficacy or safety of this approach. We established a study in alfalfa fields in northeastern CA during winter/spring 2016 and 2017 to address these questions. Our specific objectives included determining: 1) efficacy of zinc phosphide-coated cabbage bait, 2) optimal time for bait application, 3) impact of prebaiting on efficacy, and 4) potential non-target risks from bait application. To determine efficacy, we used visual counts to monitor ground squirrel activity at sites before and after treatment. Half of the sites were prebaited with nontoxic cabbage before toxic-bait application, while the other half were not prebaited. We used remote-triggered cameras to monitor timing associated with ground squirrel consumption of untreated cabbage, as well as potential nontarget consumption of cabbage. Efficacy of the cabbage bait exhibited substantial variability (25–91%) depending on where bait was applied (western portions exhibited a 23% increase in efficacy compared to eastern study sites) and the density of ground squirrels present at the application site (6.5% increase in efficacy per 10 ground squirrels). Prebaiting also increased efficacy by 18%. Peak consumption of cabbage occurred at mid-morning and early afternoon. The only observation of nontarget consumption was by California kangaroo rats (Dipodomys californicus) and ravens (Corvus corax). Raven consumption can be eliminated through hazing activities, while kangaroo rat risk can be substantially reduced by applying bait early in the morning. The use of zinc phosphide-coated cabbage currently appears to be a viable component of an integrated pest management program for managing Belding’s ground squirrels in forage crops.
2:10PM Chemical Repellents for Reducing Blackbird Damage: Further Evaluation of an Anthraquinone-Based Repellent on Mature Sunflowers
Brandon Kaiser; Burton Johnson; Page Klug; Page Klug
Blackbirds (Icteridae) cause significant damage to sunflower (Helianthus annuus) with damage estimates of $US3.5 million annually in the Prairie Pothole Region of North Dakota, the largest sunflower producing state. Chemical repellents may be a tool for reducing bird damage if application strategies can be optimized for sunflowers. Anthraquinone-based repellents have been shown to reduce feeding on loose sunflower achenes by more than 80% in lab studies, but field results are inconclusive due to application issues where floral components of sunflower result in low repellent contact with achenes. We evaluated AV-5055 (a.i. 13% 9,10 anthraquinone) as a red-winged blackbird (Agelaius phoeniceus) repellent applied directly to mature sunflowers in both a concentration-response (no choice) and preference (two choice) feeding experiment. We tested four tank mixtures (2.5%, 5%, 10%, and 20% repellent) applied at 126.3 L/ha using a spraying system for uniform distribution of repellent across the sunflower face. We analysed residues on florets (40-294 ppm) and achenes (0.4-2.8 ppm) and identified the role of florets as a barrier to repellent reaching the achenes. In both feeding experiments, we found no significant difference in consumption between repellent concentrations. Results from our concentration response showed no significant differences in consumption after one day of repellent treatment. We found no significant difference between treatments in the preference experiment (F3,34 = 0.37, p < 0.77, ηG2 = 0.02), but did observe a significant influence of day (F1.9,65.7 = 28.76, p < 0.0001, ηG2 = 0.21. Although total consumption decreased (31-46%) after the second test day for all treatments, we found no significant difference in consumption of treated (= 17.41, SD = 10.84) versus untreated (= 13.28, SD = 11.71) sunflowers; t 37= -1.78, p = 0.08. Thus, birds interact with repellent applied directly to sunflowers differently than loose achenes for which the repellent was formulated.
2:30PM Refreshment Break
3:20PM Porcine Zona Pellucida Contraception to Achieve Fertility Control in Free-Ranging African Elephants
Henk J. Bertschinger; Audrey Delsink
Elephant populations on smaller fenced game reserves in South Africa grow at 8-10% per annum. Uncontrolled population growth rapidly leads to habitat destruction, break outs and conflict situations. Because very little new habitat is available for elephants, the only acceptable solution is to control the rate of reproduction. The objective of this ongoing study was to test the immunocontraceptive efficacy of the native porcine zona pellucida (pZP) vaccine as a means of population control in free-ranging elephants on 7 game private reserves in South Africa. From 2000 to 2005 (year of commencement) total of 108 individually identified cows were immunised either 3, or later 2 times, with pZP formulated with Freund’s adjuvants in Year 1. This followed by annual boosters. Initially 600 and 400 µg pZP was used for primary and booster vaccinations, respectively. After the first 3 years, this was reduced to 400 and 200 µg. Initially the vaccine was delivered remotely with drop-out darts from the ground but from 2003 and onwards helicopter delivery was used. In Year 3 and onwards in each reserve, no calves were born to treated cows. Retrospective analyses showed that, on the day of primary vaccination, 62 cows were pregnant. They gave birth to apparently normal healthy calves. Furthermore, contraceptive efficacy was attained soon after the first booster in Year 1. The minimal side effects observed were injection site reactions which resolved spontaneously without treatment. The reduction in dose and number of boosters in Year 1 did not affect efficacy. The study demonstrated the contraceptive efficacy of pZP vaccine and safety in pregnant and non-pregnant cows. Since 2005 another 18 game reserves (>750 cows) have joined the program. In 7 of these reserves, with cow populations ranging from 40 to 157, individual identification is not possible and so, blanket treatment is being used.
3:40PM Foraging Habitat Use of Double-Crested Cormorants and Great Egrets Wintering in the Mississippi Delta: Aquaculture Versus Natural Habitat
Paul C. Burr; Jimmy L. Avery; Garrett M. Street; Bronson K. Strickland; Brian S. Dorr
Double-crested cormorants (Phalacrocorax auritus) and great egrets (Ardea alba) are two piscivorous avian species that have an extensive history of human-wildlife conflict with the aquaculture industry of Mississippi due to their depredation of cultured catfish. Although dominated by aquaculture, Mississippi also contains a significant amount of natural water bodies that offer alternative foraging resources to these species. How either cormorants or egrets distribute themselves among these two foraging habitats is unknown, but it has generally been assumed each species uses aquaculture disproportionately more due to the higher density of available prey. To test this assumption, we surveyed these species on aquaculture and natural habitat using aerial surveys during the winter months of 2015, 2016, and 2017. We used logistic regression to model proportions of each species on aquaculture against the variables of month, year, and average total rainfall and maximum temperature. Estimated proportional use was tested against the proportional availability of each habitat to examine possible preferential use. Increasing precipitation resulted in higher proportional use of aquaculture for both species, and increasing temperature showed higher proportional use of aquaculture for cormorants. Model averaged estimates between years show cormorants to use aquaculture less than what was proportionally available during the months of October – February, and more during March and April. Egret’s use of aquaculture was more than what was proportionally available year round, but the highest estimates occurred in December and January. For both species, models show a distribution shift towards aquaculture in the months immediately prior to their migration. We also found higher proportional use of aquaculture by cormorants in 2016, a year when producers were not allowed to use lethal control measures against cormorants. Although only correlative, this relationship indicates a possible change in cormorant foraging distribution due to the recent reversal of the double-crested cormorant depredation order.
4:00PM Evaluating Surgical Sterilization as a Management Technique for Overabundant Suburban Deer Populations
Anthony DeNicola; Vickie DeNicola
Overabundant suburban deer (Odocoileus spp.) are a source of human-wildlife conflict in many communities throughout the United States. Deer-vehicle collisions, tick-borne pathogens, impacts on local vegetation, and other negative interactions are the typical reasons cited for initiating a deer management program. Social attitudes, legal constraints and perceived safety concerns lead many communities to examine nonlethal management options. Surgical sterilization is currently the only nonlethal method available to permanently sterilize females with a single treatment. There are limited data demonstrating methods and outcomes in high percentage (>90%) surgical sterilization programs; particularly impacts of immigration on non-isolated populations. We present data from five surgical sterilization sites with open populations (not fenced or island environments) in California, Maryland, Ohio, Virginia, and New York, USA. From 2012 to 2017, we sterilized 403 deer via tubal ligation and ovariectomy. Annual or periodic population estimates were conducted using camera surveys, road-based distance sampling, and intensive field observations to assess population trends. We noted an average reduction in deer abundance of approximately 27% from Year 1 to Year 2. Initial populations ranged from approximately 38-404 deer/km2 (15-159 deer/mi2). On sites with 3-4 years of sterilization treatment, at sites that were active that long, we noted an average total reduction of ~49%, resulting in an estimated 20-221 deer/km2 (8-87 deer/mi2) at each location. These projects clearly demonstrate that significant reductions in local deer densities using high percentage surgical sterilization programs can be achieved. Finally, we provide an overview of these and other programs, site considerations, costs, immigration rates, and program maintenance strategies.
4:20PM Predation Risk of Double-Crested Cormorants on Commercial Catfish Production in the Mississippi Delta
Terrel W. Christie; J. Brian Davis; Brian S. Dorr; Katie C. Hanson-Dorr; Luke A. Roy; Anita Kelly; Carole Engle
Double-crested Cormorants (Phalacrocorax auritus) impact United States commercial aquaculture and are considered the greatest avian predators at catfish (Ictalurus spp.) aquaculture facilities. Cormorants are especially problematic in western Mississippi, the Delta, where catfish production is concentrated providing ideal wintering and foraging areas. Although cormorant/aquaculture dynamics have been studied in the past, recent changes (e.g., decreased overall hectares in production) in aquaculture practices and regulatory policies merit contemporary research. Therefore, we estimated abundance and distribution of cormorants at their night roosts and assessed diet related to catfish consumption. We used aerial point count surveys flown over night roosts from October through April during two winters, 2016-2018. Following each survey three active night roosts were randomly selected for harvesting cormorants for later necropsy and stomach contents assessment. We completed 25 total surveys and counted 242,923 cormorants. Mean number of cormorants detected per survey, pooled over years, was 9,717 (range 2,044 to 17,672). We collected 730 cormorants from 27 different night roosts across years. In winter 2016-2017, catfish comprised 55% of the prey biomass detected with shad (Dorosoma spp.) being the other predominate prey species. Evidence suggests that the distance between a night roost and the nearest catfish aquaculture facility is an important predictor for a bird’s relative amount of catfish consumption. These results will inform wildlife management about relationships between cormorant night roost locations in the Delta and disproportionate consumption of catfish, aiding techniques to help ameliorate fish losses on aquaculture facilities.
4:40PM Non-Lethal Management of Feral Livestock: Fertility Control for Free-Living Cattle in Hong Kong
Giovanna Massei; Koon K. Kathy; Law Lucia; Gomm Matthew; Mora S.O. Darcy; Callaby Rebecca; Eckery C. Douglas
Free-living cattle (Bos taurus/Bos indicus) in Hong Kong are valued as local heritage but are also associated increasingly with traffic disturbance, accidents and crop damage. As lethal control is not publicly supported, contraception may offer an alternative for reducing cattle numbers. We evaluated the effect of the immunocontraceptive GonaCon on health and reproduction of free-living cattle. In summer 2015 we assigned female cattle to treated (T, n = 42) or control (C, n=18) groups and injected T with GonaCon and C with a saline solution. We administered a booster dose (GonaCon or saline) 3-6 months later. We assessed the effectiveness of GonaCon by a pregnancy test, by recording the calves produced, and by quantifying anti-Gonadotropin Releasing Hormone (GnRH) antibody titres in serum samples. No site-injection reaction, limping or any other abnormal behaviour were observed in T animals. GonaCon did not affect ongoing pregnancies: T and C cattle that were pregnant when vaccinated delivered calves in similar proportions. GonaCon affected reproduction. At vaccination, 76% of the cattle in the T group versus 71% in the C group were pregnant; one year later, only 6% of the cattle in the T group versus 62% in the C group were pregnant. Cattle with the highest levels of anti-GnRH antibody titres were less likely to be pregnant than those with lower levels. Among the latter, some animals with relatively low antibody titres were not pregnant one year after vaccination. This study confirmed that GonaCon is a safe, effective option to induce infertility in cattle, thus warranting a long-term monitoring to assess the longevity of the contraceptive effect. Future studies should evaluate the effectiveness, feasibility and sustainability of immunocontraception to manage other populations of feral livestock, such as buffaloes, goats and horses, for contexts where culling is unfeasible or publicly unacceptable.


Contributed Paper
Location: Huntington Convention Center of Cleveland Date: October 11, 2018 Time: 12:50 pm - 5:00 pm