NEAFWA 2018

AUTHORS: Justin R. Dion, Jacob M. Haus, Jacob L. Bowman - Department of Entomology and Wildlife Ecology, University of Delaware; Joseph E. Rogerson, Delaware Department of Natural Resources and Environmental Control

ABSTRACT. Survival and cause-specific mortality of neonate white-tailed deer has been the focus of recent research, particularly in regards to predation mortality. An understanding of the impact of predation on survival rates requires a predator-free control population. We captured 109 neonates using opportunistic capture (n = 55) and vaginal implant transmitters (VIT; n = 54) in Delaware during 2016 and 2017. Predators (i.e., black bear, bobcat, and coyotes) were functionally absent from the study area. We calculated 30-day survival using a Kaplan-Meier estimator and determined the importance of covariate on survival using Cox proportional hazard models. The overall 30-day survival estimate was 0.61 (95% CI = 0.51 – 0.72). The survival estimate for neonates captured using random searches (0.76) was greater (P < 0.01) than those for VIT neonates (0.53). Natural causes (n = 34) accounted for all of our observed mortality, including one potential predation by red fox. The top models included covariates for birth weight, doe maturity, and precipitation. Predation could be less of a limiting factor for survival than many studies have suggested. Data derived from opportunistically captured neonates may inflate estimates of survival and misrepresent cause-specific mortality. Although the influence of birth weight on survival has been reported previously, the impact of doe maturity and precipitation has not been documented. The current emphasis on predator management and the impact on deer abundance may be misplaced.

AUTHORS: Jacob M. Haus, Jacob L. Bowman - University of Delaware; Joseph E. Rogerson, Delaware Division of Fish and Wildlife

ABSTRACT. Managing male age structure in white-tailed deer populations is an important objective for state managers and private landowners seeking to improve hunter satisfaction while maintaining appropriate densities. Limiting mortality in the yearling age class is often the primary consideration, and regional differences in habitat, regulations, and hunter behavior complicate our understanding of how specific factors influence survival. We used Cox proportional hazard modeling to examine the effects of distance to road, distance to forest edge, dispersal behaviors, and landownership on the risk of mortality for yearling males (n = 61) in southern Delaware. Annual survival averaged 0.60 (95% CI = 0.49 – 0.73), with hunter harvest accounting for 79% of mortalities. The best approximating model for risk of mortality included covariates for landownership (public/private; P < 0.01) and distance to forest edge (P = 0.01), with mortality risk increasing both on public land and in closer proximity to forest edge. Increased risk of harvest due to forest fragmentation is well documented; however, the effect of landownership has not been quantified, particularly when hunter objectives and behaviors differ between landownership types. We observed annual survival rates of 0.75 (95% CI = 0.62 – 0.89) for deer exclusively on private land during the hunting season, and 0.37 (95% CI = 0.18 – 0.73) for deer that utilized public land during the hunting season. Survival rates on private lands were comparable to research from properties that actively manage male age structure, but harvest of yearlings limited male age structure on public lands within the study area.

OLD TITLE: Hunting Induced Site Level Spatial Variability in White-tailed Deer Abundance Captured Through the Lens of Wildlife Cameras

AUTHORS: Jennifer E. Kilburn, Tracy A. G. Rittenhouse - University of Connecticut; Andrew M. LaBonte, Howard J. Kilpatrick - Connecticut Department of Energy and Environmental Protection

ABSTRACT. Managed public lands balance maintaining healthy wildlife populations and human recreation opportunities. Hunting, not only as a form of recreation but as a management tool, varies across land ownership. Previous studies have shown that during the hunting season white-tailed deer alter their behavior and reduce movements to avoid hunting pressure. However, little has been done to document how these shifts affect site level local abundance, especially in areas with a diverse patchwork of land ownership and hunting pressure. We used 50 wildlife cameras evenly divided between hunted and un-hunted properties in north-eastern Connecticut and N-mixture models in a Bayesian framework to estimate local site level abundance and detection probability before, during, and after the hunting season. We found that average site level abundance increased on un-hunted properties and decreased on hunted properties after initiation of the hunting season. Averaged across the entire study period, abundance was higher on un-hunted properties (2.5 does/31 ha, (0.99,7.4)) as opposed to hunted properties (1.5 does/31 ha, (0.51,5.60)). These patterns were not statically significant, but the abundance estimates and the change over time aligns with our expectation. Detection probability increased on hunted properties during the hunting season, which conflicts with previous findings but may result from behavior during the breeding season. Understanding the relationship between local, site-level shifts in abundance in response to hunting pressure allows managers to understand deer availability during the hunting season and how that might impact attainment of management goals and expectations.

AUTHORS: Andrew M. LaBonte, Howard J. Kilpatrick - Connecticut Department of Environmental Protection, Wildlife Division

ABSTRACT. Deer (Odocoileus sp.) harvest data have been used to develop population models, set bag limits and hunting season lengths, and evaluate herd age, composition, and health. Managers use numerous methods to obtain these data, such as in-person check stations (IPC), mail questionnaires or kill cards, telephone services, and most recently, Internet harvest reporting (IHR). Agencies may struggle to determine which methods should be used and how hunters feel about using them. Our objectives were to assess current harvest reporting and data collection methods and to evaluate hunter opinions, satisfaction, and harvest reporting compliance in Connecticut. To assess harvest reporting and data collection methods, we conducted a nationwide survey of wildlife agencies and to assess hunter opinions, satisfaction, and compliance we conducted a survey of Connecticut deer hunters. All surveys were implemented through SurveyMonkey®. Many agencies are currently using telephone (41.0%), IPC (46.4), and IHR systems (64.2%). Hunters reported their harvest primarily using IHR (75%), and satisfaction with that system was high (96%). Hunters were evenly split in regards to whether reporting rates were higher with the new reporting systems than with previous systems (36% more, 33% less, 31% same). We found that most hunters (82%) reported harvesting the same number of deer on the survey as through the new system, however 14% of hunters were non-compliant. Telephone and IHR systems are becoming increasingly popular, and provide a convenient means to report harvest, have few problems, and result in high harvest reporting rates.

AUTHORS: Schuyler Pearman-Gillman, Vermont Cooperative Fish and Wildlife Research Unit, University of Vermont; Therese Donovan, U.S. Geological Survey, Vermont Cooperative Fish and Wildlife Research Unit, University of Vermont; James Murdoch, Wildlife and Fisheries Biology Program, University of Vermont; Jonathan Katz, Vermont Cooperative Fish & Wildlife Research Unit

ABSTRACT. In the Northeastern United States, population expansion, agriculture, industrial development, and urban sprawl pose serious concerns for wildlife and challenges for wildlife managers. Understanding the impacts of landscape change on species’ distributions can help inform decision-making for conservation planning. Unfortunately, empirical data on distribution is limited for many species. In this study, we used expert elicitation techniques to develop species distribution models (SDMs) for harvested species in the northeast (n = 22). This expert-based approach provided relatively inexpensive, big picture information that would have otherwise been unattainable given the spatial extent and range of species being assessed. We elicited opinions from wildlife experts on the probability of occupancy of species using a web-based survey administered in 2017. We collected ca. 6000 occupancy estimates from over 75 experts, and used mixed-model methods to develop SDMs. We present models for three species, moose (Alces alces), American black bear (Ursus americanus), and wild turkey (Meleagris gallopavo), and demonstrate their use for mapping distribution under current conditions and various scenarios of landscape change. Our results indicate the utility of expert opinion data in modeling wildlife distributions, which could be helpful for many other species, especially those that are poorly known or lack location records.