Shifting perceptions of risk and reward: Dynamic selection for human development by black bears in the western United States
- Under a Creative Commons license
Open Access
Highlights
- •
- Black bear behavior was examined around three developed areas in the western U.S.
- •
- Bear selection for human development was highly dynamic across time and space.
- •
- Selection was related to natural food conditions, physiological state and age.
- •
- Black bears appear to use human development as a source of food subsidy.
Abstract
As landscapes across the globe experience increasing human development, it is critical to identify the behavioral responses of wildlife to this change given associated shifts in resource availability and risk from human activity. This is particularly important for large carnivores as their interactions with people are often a source of conflict, which can impede conservation efforts and require extensive management. To examine the adaptations of a large carnivore to benefits and risks associated with human development we investigated black bear behavior in three systems in the western United States. Our objectives were to (1) identify temporal patterns of selection for development within a year and across years based on natural food conditions, (2) compare spatial patterns of selection for development across systems, and (3) examine individual characteristics associated with increased selection for development. Using mixed effects resource selection models we found that bear selection for development was highly dynamic, varying as a function of changing environmental and physiological conditions. Bears increased their use of development in years when natural foods were scarce, throughout the summer-fall, as they aged, and as a function of gender, with males exhibiting greater use of development. While patterns were similar across systems, bears at sites with poorer quality habitat selected development more consistently than bears at sites with higher quality habitat. Black bears appear to use development largely for food subsidy, suggesting that conflicts with bears, and potentially other large carnivores, will increase when the physiological demand for resources outweighs risks associated with human activity.
Keywords
- Behavior;
- Black bear;
- Forage risk trade-off;
- Human–carnivore conflict;
- Resource selection;
- Ursus americanus
1. Introduction
As landscapes across the globe rapidly change due to increased human development (Vitousek et al., 1997 and Ellis et al., 2010), there is uncertainty about the behavioral responses of wildlife to these changes given associated shifts in resource availability and risk. Urban and exurban development (hereafter development) typically reduce native food resources for animals, but introduce novel anthropogenic foods (garbage, crops, livestock, watered landscaping, etc) and risks associated with foraging in human-dominated landscapes (mortality from vehicle collisions, lethal removal; hereafter risk). The initial response of animals to human development is typically a change in behavior, as animals have been observed to alter patterns of habitat selection (Nellemann et al., 2007), vigilance (Mccleery, 2009), daily activities (Riley et al., 2003 and Beckmann and Berger, 2003a) and foraging (Robinson et al., 2010), often in highly diverse ways (Tuomainen and Candolin, 2011). These behavioral responses reflect perceived trade-offs between the benefits of acquiring key resources and the risks associated with human activity (Frid and Dill, 2002). While these trade-offs should be dynamic in space and time as a function of habitat quality, natural food conditions and the physiological states of individuals, little is known about how animals in human-altered landscapes behaviorally adapt to such variation.
Elucidating the behavioral responses of wildlife to human development is particularly important for large carnivores as their home ranges frequently overlap with human infrastructure and activities (Mladenoff et al., 1997), and their interactions with people are often a major source of conflict (Treves and Karanth, 2003). In many cases, large carnivores avoid people (Oakleaf et al., 2006 and Nellemann et al., 2007) and exhibit antipredator behavior indicating they associate humans with risk (Ordiz et al., 2011). Some carnivores, however, have been observed to forage within human development on their natural foods (Gehrt et al., 2009 and Dellinger et al., 2013) or on anthropogenic foods (Bateman and Fleming, 2012 and Merkle et al., 2013), exploiting novel resources associated with human infrastructure. Such behavior has been linked to increased reports of human–carnivore conflicts, generating concern over human safety and property, and stymieing conservation efforts for some carnivore species (Treves and Karanth, 2003). If wildlife managers and conservation practitioners are going to be successful at reducing human–carnivore conflicts in an increasingly developed landscape, they need to understand how these animals behaviorally respond to development, and the conditions that modify their behavior.
These concerns are particularly relevant for the American black bear (Ursus americanus). Bears can readily exploit the wealth of reliable, high-calorie food resources available around human development (i.e., garbage, fruit trees, livestock), but are also susceptible to increased mortality from vehicle collisions, conflict-related euthanasia, and other human-related factors ( Beckmann and Berger, 2003a, Beckmann and Berger, 2003b and Hostetler et al., 2009 and Baruch-Mordo et al., 2014). Although studies have demonstrated that bears perceive risk associated with human activity ( Beckmann and Berger, 2003a and Nellemann et al., 2007 and Ordiz et al., 2011), human–bear conflicts have generally increased over time ( Hristienko and McDonald, 2007), displaying high temporal and spatial variation ( Baruch-Mordo et al., 2008 and Treves et al., 2010). As a long-lived species with relatively stable population dynamics ( Beston, 2011), high temporal variation in conflict activity within and across years is presumably a consequence of shifting foraging behavior, not shifting demography, as bears reassess the trade-offs of using anthropogenic foods. Factors such as natural food conditions, a bear’s physiological state (e.g., reproductive status), or degree of exposure to human activity, likely influence the benefits and risks (perceived or real) of foraging in human-dominated landscapes, driving observed variation in conflict activity. Indeed, Baruch-Mordo et al. (2014) documented a shift in bear space-use to incorporate additional human development in poor natural food years. However, no information is currently available about how bears temporally alter their use of development within a year, whether bears in different locations similarly interpret the costs and benefits of foraging in developed landscapes, or whether certain characteristics of individuals may exacerbate or inhibit their use of human development.
To understand how a large carnivore adapts to the shifting benefits and risks of foraging among human development, we examined patterns of black bear habitat selection using GPS location data from 109 bears around three developed areas in the western US (Aspen [CO], Durango [CO], and Lake Tahoe [NV]). Throughout our investigation, we interpret a bear’s selection for development as the behavioral outcome of their perceived forage/risk assessment (Frid and Dill, 2002). Specifically, our objectives were to (1) examine temporal patterns of selection for development within the active bear season and across years based on natural food conditions, (2) compare patterns of selection for development among study systems, and (3) identify individual attributes (maternal status, age, etc.) associated with selection for development. We expected that the benefits of foraging around human development would outweigh the perceived consequences when bears were physiologically stressed and needed additional food resources. Within study systems, we expected bears to increase their use of development in years when natural foods were scarce, during hyperphagia (the period of increased foraging prior to hibernation) and when females were experiencing increased energetic demands with lactation. Across study systems, we expected bears at more xeric sites with fewer natural food resources to exhibit stronger and more consistent selection for development than bears in more mesic sites with a greater abundance of natural foods.
2. Materials and methods
2.1. Study areas
We evaluated black bear habitat selection around three developed areas in the western U.S.: Aspen (CO), Durango (CO), and Lake Tahoe (NV; hereafter Tahoe). All areas have experienced high rates of human-black bear conflicts and have been the focus of long-term studies on this issue (Johnson et al., 2011 and Lackey et al., 2013 and Baruch-Mordo et al., 2014). At all three sites, grizzly bears (Ursus arctos) were absent.
Among the three study systems, Aspen is the most mesic and Tahoe is the most xeric, with associated differences in precipitation, elevation and vegetation. Over the last 10 years, annual precipitation averaged 62.8 cm in Aspen (http://ccc.atmos.colostate.edu/dataaccess.php), 48.5 cm in Durango (http://ccc.atmos.colostate.edu/dataaccess.php), and 7.5 cm in Tahoe (http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?nv4618); elevations of the towns at each site are approximately 2 405 m, 1 985 m, and 1 919 m, respectively. Vegetation around Aspen is dominated by gambel oak (Quercus gambelli), serviceberry (Amelanchier alnifolia), chokecherry (Prunus virginiana), aspen (Populus tremuloides), lodgepole (Pinus contorta), Douglas fir (Pseudotsuga menziesii), and spruce (Piceaspp.)-subalpine fir (Abies lasiocarpa) forests. Around Durango, vegetation is primarily oak, serviceberry, chokecherry, aspen, pinyon–juniper (Pinus edulis/Juniperus ssp.) woodland, and ponderosa pine (Pinus ponderosa) and spruce-fir forests. On the east side of the Lake Tahoe basin, vegetation is dominated by manzanita (Arctostaphylosmanzanita), serviceberry, aspen, Jeffrey pine (Pinus jeffreyi), ponderosa pine, lodgepole, Douglas-fir, and western white pine (Pinus monticola) forests. Key late summer and early fall mast species around Aspen and Durango include chokecherries, serviceberries and acorns (gambel oak); in Tahoe they include manzanita berries, pinyon pine nuts and other conifer seeds. Around human development, bears primarily forage on garbage and fruit trees (i.e., apple, crabapple, pear; Lewis, 2013; Nevada Department of Wildlife and Colorado Parks and Wildlife, unpublished data).
Human development at the Aspen site consisted of the towns of Aspen (6658 residents) and Snowmass (2826 residents), with additional development along the Roaring Fork valley. The Durango site consisted of the town of Durango (16,887) with additional development along adjacent valleys and mesas. In the Tahoe system, the major towns were South Lake Tahoe, CA (21,403 residents), Stateline, NV (842 residents), and Incline Village, NV (8777 residents), with additional development along the eastern shoreline of Lake Tahoe. All population sizes are reported for 2010 (United States Census Bureau, 2014). During the time period relevant to our analysis, the number of black bear harvest licenses was similar among years for the Aspen and Durango sites, and there was no hunting around Tahoe.
2.2. Black bear data
Bears were caught using culvert traps, box traps, Aldrich foot snares, and free-range techniques (Jonkel, 1993). Captured bears were immobilized, had a tooth pulled for age estimation (Willey, 1974), and were fitted with a GPS collar so that their locations could be tracked. In Aspen, collar locations were obtained from 2005 to 2010 on males and females (Lotek collars, 3300L and 4400M). In Tahoe, collar locations were collected from 2005 to 2011 on females only (Northstar, Telonics collars). In Durango, collar locations were collected in 2011 and 2012 on females only (Vectronics Globalstar collars). In Aspen and Durango locations were collected or sub-sampled on an hourly basis and in Tahoe locations were collected every 2–3 h. At all sites, trapping efforts occurred within approximately 10 km of urban development to collar a cohort of bears that experienced similar natural food conditio
