Human Dimensions of Wildlife, 10:201–212 Copyright © 2005 Taylor & Francis Inc. ISSN: 1087-1209 print / 1533-158X online DOI: 10.1080/10871200591003445
Bear and Deer Hunter Density and Distribution on Public Land in Pennsylvania
1533-158X 1087-1209 UHDW Human Dimensions of Wildlife Wildlife, Vol. 10, No. 03, May 2005: pp. 0–0
Hunter D. R. Diefenbach Density and Distribution
DUANE R. DIEFENBACH1, JAMES C. FINLEY2, A. E. LULOFF3, RICHARD STEDMAN4, CRAIG B. SWOPE5, HARRY C. ZINN6 AND GARY J. SAN JULIAN7 1
USGS Pennsylvania Cooperative Fish and Wildlife Research Unit, The Pennsylvania State University, University Park, Pennsylvania, USA 2 School of Forest Resources, The Pennsylvania State University, University Park, Pennsylvania, USA 3 Department of Agricultural Economics and Rural Sociology, The Pennsylvania State University, University Park, Pennsylvania, USA 4 Department of Agricultural Economics and Rural Sociology, The Pennsylvania State University, University Park, Pennsylvania, USA 5 School of Forest Resources, The Pennsylvania State University, University Park, Pennsylvania, USA 6 Department of Recreation, Park & Tourism Management, The Pennsylvania State University, University Park, Pennsylvania, USA 7 School of Forest Resources, The Pennsylvania State University, University Park, Pennsylvania, USA Little research has focused on density, distribution, and habitat use of hunters. Here, field methods and statistical techniques developed for estimating density and habitat use of wildlife species are applied to big game hunters in Pennsylvania. Using aerial surveys, a large public land area (45,907 ha) in north-central Pennsylvania was studied to estimate hunter density and distribution during black bear (Ursus americanus) and white-tailed deer (Odocoileus virginianus) rifle seasons. This area was dominated by deciduous forest, had slopes that ranged from 0 to 61 degrees, and a road network that provided access to within 2.5 km of any location in the study area. The authors found a hunter density of 4.0 hunters/1,000 ha during the second day of bear season. During the second and third day, and first Saturday of deer season, the hunter density was 0.4–1.9 hunters/1,000 ha. Relative deer hunter density was predicted by distance from nearest road and slope but these were not strong predictive variables for the
The authors thank Audubon Pennsylvania, Pennsylvania Habitat Alliance, DCNR Bureau of Forestry, and the PGC for support for this project. In addition, the authors thank J. Sinclair, S. Murphy, T. Hillegas, and S. Repasky who conducted aerial surveys, J.T. McQuaide who developed the software for the tablet PCs, and J.A. Mattice who extracted data from the GIS. Finally, the authors thank the members of the PGC Deer Management Team for their ideas that led to the development of this project: G. L. Alt, M. Grund, G. Kelly, C. S. Rosenberry, B. P. Shissler, and B. D. Wallingford. Address correspondence to Duane R. Diefenbach, USGS Pennsylvania Cooperative Fish and Wildlife Research Unit, Merkle Laboratory, The Pennsylvania State University, University Park, PA, 16802. E-mail:
[email protected]
201
202
D. R. Diefenbach distribution of bear hunters. Deer hunters were nearly three times less likely to hunt in an area for every increase of 500 m from a forest road, and for every 5 degree increase in slope, an area was 1.5 times less likely to be hunted. It was estimated that 87% of deer hunters hunted within 0.5 km of a road. Even if hunter densities during deer season were equal to those in bear season, hunters likely would have harvested an insignificant proportion of the deer population. Under current regulations and low hunter densities, recreational hunting may not be an effective tool for controlling deer populations on large tracts of public land in Pennsylvania with similar topography and road networks. Keywords black bear, hunter density, hunter distribution, Odocoileus virginianus, Pennsylvania, Ursus americanus, white-tailed deer
Introduction Despite the importance of humans in managing game populations, far less is known about their behavior as predators (e.g., their distribution and habitat use) than other predators or the game species they hunt. Game species have been studied extensively in North America by wildlife biologists; for example, contemporary applied research topics on black bear and white-tailed deer include population dynamics, population estimation, habitat use and modeling, migratory movements, dispersal, and seasonal movements. As deer managers have begun to address the control of deer populations in urban and/or protected areas (Doerr, McAninch, & Wiggers, 2001; Gogan, Barrett, Shook, & Kucera, 2001; Hansen & Beringer, 1997), the future of recreational hunting as an effective tool for controlling deer populations has been questioned (Brown et al., 2000). The long-term decline in number of hunters is well documented (U.S. Dept. of the Interior and U.S. Dept. of Commerce, 2002) and has motivated much human dimensions research on issues such as hunting initiation (Decker, Purdy, & Brown, 1986), motivations (Hammit, McDonald, & Patterson, 1990), factors influencing participation (Zinn, 2003), and desertion (Applegate, 1989). However, relatively little is known about hunter field behavior, especially with respect to abundance, distribution, and habitat use. Information about hunter density and distribution could be used by wildlife managers to develop regulations to increase the effectiveness of recreational hunting to manage wildlife populations. Most published studies of hunting effort are limited to hunter density (e.g., Kubisiak et al., 2001; Sparrowe & Springer, 1970) with little or no information about distribution, and the few published studies on how and where hunters hunt have relied on hunter ability to accurately mark their locations on maps (e.g., Thomas, Gill, Pack, Healy, & Sanderson, 1976). Radiotelemetry and global positioning system (GPS) technology has been recognized as a means to monitor hunter movements (Lyon & Burcham, 1998) but is only beginning to be applied. For example, Broseth and Pedersen (2000) monitored the behavior of nine willow grouse hunters on a relatively small parcel of private land in Norway using GPS units. Large tracts of public land in Pennsylvania (>1 million ha) contain important natural resources. The effective management of wildlife, particularly bear and deer, is important for the sustained use of these resources but no information is available on abundance and distribution of hunters. The present authors used aerial surveys and distance sampling methods to estimate hunter density (Buckland et al., 2001) and resource selection methods to model distribution on the landscape (Manly, McDonald, Thomas, McDonald, & Erickson, 2002).
Hunter Density and Distribution
203
Study Area The study area is located in north-central Pennsylvania (Clinton and Centre counties) and part of the Allegheny Plateau. North-central Pennsylvania is known as the “Big Woods” habitat because of the large tracts of largely forested public land (>800,000 ha). The 45,907-ha study area consists of the southern portion of the Sproul State Forest, which is managed by the Department of Conservation and Natural Resources (DCNR) and State Game Lands 100 managed by the Pennsylvania Game Commission (PGC). A north-south paved, two-lane state highway bisects the study area; numerous hiking and snowmobile trails, gas wells, and natural gas and electric utility corridors are located throughout the area. Despite these access points, 16% of the study area was >1 km from the nearest road open to the general public. Forest cover primarily is second- and third-growth mature hardwood forest and regenerating stands containing red maple and red and white oak species with lesser amounts of black cherry, American beech, and hickory. Evergreens are scarce (1%), and generally consist of hemlocks in drainages and wetlands. Regeneration is sparse and a distinct deer browse line marked the area. This forest has been affected by large forest fires and tornadoes in the late 1980s and early 1990s as well as forest insect outbreaks. The Big Woods has a long history of deer hunting. By 1900, this area of the state contained the only huntable populations of black bear and white-tailed deer (Kosack, 1995). In the early 1900s, the state Bureau of Forests and Waters leased hundreds of small plots of land on state forestlands to Pennsylvania residents to build camps and others were developed on private in-holdings. Consequently, generations of hunters have evolved a tradition of hunting bear and deer, primarily antlered bucks, during the regular rifle seasons in the Big Woods of Pennsylvania. Deer densities in this region of Pennsylvania in 2002 were estimated to be 81 deer/1,000 ha prior to the hunting season (PGC, unpublished data) and bear densities were approximately 2.7 bear/1,000 ha (D. R. Diefenbach, unpublished data).
Methods Data Collection The authors randomly selected a location within the study area and from this point systematically placed east-west transect lines approximately 1.8 km apart across the study area in a Geographic Information System (GIS). This provided 13 transects, 5.6–19.4 km long, across the study area. The aerial surveys were conducted from a fixed-wing aircraft that flew at approximately 90 knots airspeed and 1,100 m elevation (∼450 m above ground level). To navigate transects, the authors used a handheld PC running ArcPad 6.0 and linked to a Garmin GPS III+ global positioning system (GPS) unit (Diefenbach, McQuaide, & Mattice, 2002). A geo-referenced image of the study area was overlaid with a map containing the transect lines and boundary of the study area as a visual aid in navigating the predetermined transects.1 Locations of hunters2 were recorded on a tablet PC equipped with a digitizing pen and integrated GPS. A software program used input from the GPS to display either a geo-referenced topographic 1:24000 USGS quadrangle or a digital orthophoto quadrangle (DOQ) image referenced to the observer’s view out the aircraft window.3 Observers were trained prior to the hunting seasons so they were familiar with use of equipment and software as well as navigating transects.
204
D. R. Diefenbach
Weather conditions only permitted conducting of aerial surveys during the afternoon of the second day of bear season (26 November 2002), morning and afternoon of the second day of deer season (3 December 2002), the afternoon of the third day of deer season (4 December 2002), and the morning and afternoon of the first Saturday of deer season (7 December 2002). Legal hunting hours at this time of year began approximately 0645 and concluded at 1645 hours. Data Analysis Distance sampling methods (Buckland et al., 2001) were used to estimate the density and abundance of hunters on the study area. The authors used Distance software to analyze the data (Thomas et al., 2003). The actual recorded flight path was used for each transect, rather than the predetermined transect lines, when calculating the distance of hunters from the transect line because it was impossible to navigate the transect lines exactly. Because the viewing angle from the aircraft window prevented observers from detecting hunters within approximately 122 m, 122 m were subtracted from each distance measured between the transect line and the observed hunter. Procedures outlined in Buckland et al. (2001) were followed for truncating observations at the greatest distances from observers and modeling the detection function. Truncating observations at the greatest distance has little effect on density estimates, but can improve precision because truncation usually results in a simpler model of the detection function (Buckland et al. 2001). Detection functions for each observer and a single detection function for both observers were developed and the most parsimonious model of detection probability using Akaike’s Information Criterion (AIC) was selected. Observations were used from all surveys combined to estimate a global detection function (for each observer), but calculated a density and abundance estimate of hunters for each survey (defined as a stratum in Distance). Bootstrap methods were used in Distance to estimate the precision (variances and 95% confidence intervals) of density and abundance estimates. The authors used all observations of hunters to develop a model of hunter distribution based on distance from road (km) and slope (degrees). Using a GIS, a 100 m × 100 m grid was placed over the study area and a 30 m digital elevation model of the study area was used to assign each grid cell a slope value. In addition, a shapefile of all roads open to public travel was added so that the distance to the nearest road for each grid cell could be calculated. Available slope and distance from road conditions were estimated from a systematic random sample of every ninth grid cell in the study area (Erickson, McDonald, & Skinner, 1997; Manly et al., 2002). The sample of grid cells associated with locations of hunters observed during aerial surveys provided SLOPE and ROAD characteristics of sites used by hunters. Hunters may have used habitat differently for these two hunting seasons, although bear and deer occurred throughout the study area and hunting regulations were the same with respect to pursuit of these two species.4 To assess whether the distribution of bear and deer hunters differed, bear and deer hunter locations were categorized by distance from road into 200 m intervals (
