ARTICLE

Black bear (Ursus americanus) functional resource selectionrelative to intraspecific competition and human riskJared F. Duquette, Jerrold L. Belant, Clay M. Wilton, Nicholas Fowler, Brittany W. Waller,Dean E. Beyer, Jr., Nathan J. Svoboda, Stephanie L. Simek, and Jeff Beringer

Abstract: The spatial scales at which animals make behavioral trade-offs is assumed to relate to the scales at which factors mostlimiting resources and increasing mortality risk occur. We used global positioning system collar locations of 29 reproductive-agefemale black bears (Ursus americanus Pallas, 1780) in three states to assess resource selection relative to bear population-specificdensity, an index of vegetation productivity, riparian corridors, or two road classes of and within home ranges during spring–summer of 2009–2013. Female resource selection was best explained by functional responses to vegetation productivity acrossnearly all populations and spatial scales, which appeared to be influenced by variation in bear density (i.e., intraspecificcompetition). Behavioral trade-offs were greatest at the landscape scale, but except for vegetation productivity, were consistentfor populations across spatial scales. Females across populations selected locations nearer to tertiary roads, but females inMichigan and Mississippi selected main roads and avoided riparian corridors, whereas females in Missouri did the opposite,suggesting population-level trade-offs between resource (e.g., food) acquisition and mortality risks (e.g., vehicle collisions). Ourstudy emphasizes that female bear population-level resource selection can be influenced by multiple spatially dependent factors,and that scale-dependent functional behavior should be identified for management of bears across their range.

Key words: black bear, food, hunting, mixed models, riparian, roads, Ursus americanus.

Résumé : Il est présumé que les échelles spatiales auxquelles les animaux font des compromis comportementaux sont reliéesaux échelles auxquelles se manifestent les facteurs qui limitent le plus les ressources et accroissent le plus le risque de mortalité.Nous utilisons des emplacements obtenus par colliers GPS de 29 ours noirs (Ursus americanus Pallas, 1780) en âge de procréer danstrois États pour évaluer la sélection des ressources par rapport a la densité de la population d’ours, a un indice de la productivitéde la végétation, aux corridors rivulaires ou a deux classes de routes associés dans les domaines vitaux durant les printemps etétés de 2009 a 2013. La sélection des ressources par les femelles s’expliquait le mieux par des réponses fonctionnelles a laproductivité de la végétation pour presque toutes les populations et échelles spatiales, qui semblaient être influencées par desvariations de la densité d’ours (c.-a-d., concurrence intraspécifique). Les compromis comportementaux étaient les plus grands al’échelle du paysage, mais, a l’exception de la productivité de la végétation, étaient cohérents pour les populations a toutesles échelles. Les femelles de différentes populations choisissaient des emplacements plus proches de routes tertiaires, maisles femelles au Michigan et au Mississippi choisissaient des routes principales et évitaient les corridors rivulaires, alors que lesfemelles aux Missouri faisaient le contraire, indiquant des compromis au niveau de la population entre l’acquisition desressources (p. ex. de la nourriture) et les risques de mortalité (p. ex. collisions avec des véhicules). L’étude souligne le fait que lasélection des ressources par les ourses au niveau de la population peut être influencée par différents facteurs dépendants del’emplacement et que la gestion des ours dans l’ensemble de leur aire de répartition requiert une connaissance des comporte-ments fonctionnels dépendants de l’échelle. [Traduit par la Rédaction]

Mots-clés : ours noir, nourriture, chasse, modèles mixtes, rivulaire, routes, Ursus americanus.

IntroductionAnimals improve their reproductive success by selecting the

most beneficial resources available across a landscape (Rettie andMessier 2000; Manly et al. 2002). But resources may be limited byfactors including habitat fragmentation (e.g., roads: Shepherd andWhittington 2006; land-use change: Hiller and Belant 2015; Hilleret al. 2015), intra- and inter-specific competition (Ciarniello et al.2007; Watts and Holekamp 2009), or direct and indirect mortalityrisks (Creel et al. 2008; Steyaert et al. 2013; Kiffner et al. 2014) that

can vary across landscapes (Johnson and Seip 2008). Imperfectknowledge or cues to resource availability or quality can also limitresource accessibility (Basille et al. 2013). Therefore, animals im-prove their reproductive success by using behavioral trade-offsbetween resource selection and mortality risk avoidance (Martinet al. 2010; Latif et al. 2011).

The spatial scales at which animals adapt their behavior aregenerally assumed to relate to the scales where access to resourcesis maximized and mortality risk is minimized (Rettie and Messier2000). A factor (e.g., predation risk) that limits resources and in-

Received 9 February 2016. Accepted 9 December 2016.

J.F. Duquette,* J.L. Belant, C.M. Wilton, N. Fowler, B.W. Waller, N.J. Svoboda, and S.L. Simek. Carnivore Ecology Laboratory, Forest and WildlifeResearch Center, Mississippi State University, Mississippi State, MS 39762, USA.D.E. Beyer, Jr. Michigan Department of Natural Resources, Wildlife Division, Marquette, MI 49855, USA.J. Beringer. Missouri Department of Conservation, Columbia, MO 65201, USA.Corresponding author: Jared F. Duquette (email: jfd2017@illinois.edu).*Present address: Illinois Natural History Survey, 1816 South Oak Street, Champaign, IL 61820, USA.Copyright remains with the author(s) or their institution(s). Permission for reuse (free in most cases) can be obtained from RightsLink.

203

Can. J. Zool. 95: 203–212 (2017) dx.doi.org/10.1139/cjz-2016-0031 Published at www.nrcresearchpress.com/cjz on 1 February 2017.

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

creases mortality risk modifies animal behavior at successivelyfiner scales until another factor (e.g., food) becomes more limitingand further modifies behavior. Intraspecific population densitycan be a major factor influencing the behavioral trade-offs ofanimals because intraspecific competition increases with popula-tion density (e.g., Ciarniello et al. 2007). Greater intraspecific com-petition can decrease the availability of ideal resources and limitanimal fitness (Zedrosser et al. 2006; van Beest et al. 2014). Al-though identifying factors that influence the behavioral trade-offsof animals across multiple spatial scales is common (e.g., Basilleet al. 2013), generalizing those conclusions across the geographicrange of a species can introduce bias from variability in biologicaland environmental factors found within each population. Assess-ing behavioral trade-offs among population density, resource se-lection, and mortality risks within and across populations couldprovide insights into factors affecting the range-wide reproduc-tive success of a species (McLoughlin et al. 2000; Bojarska andSelva 2012).

Animals maximize their fitness by adapting to environmentalvariability in a functional nonlinear manner (Matthiopoulos et al.2011), often at multiple spatial scales (Moreau et al. 2012). Forexample, animals living in landscapes with patchy and fluctuat-ing food resources and mortality risks may functionally selecthabitat patches based on a trade-off between perceived resourcesand mortality risks (Godvik et al. 2009; Moreau et al. 2012). Thisconcept can be extended to variability in other factors, such asresource competition and anthropogenic disturbance, which di-rectly or indirectly influence animal fitness. Understanding thefunctional behaviors of animals relative to their population den-sity (i.e., competition) can indicate the habitat quality of the land-scape (McLoughlin et al. 2010) and assist in guiding populationmanagement.

Behavioral trade-offs are important for many large carnivoresduring spring, when young are dependent on maternal nutrition(Elowe and Dodge 1989) and vulnerable to intra- or inter-specificcompetition and predation (Palomares and Caro 1999; Garrisonet al. 2007). Large carnivores in human-dominated landscapes of-ten adjust behaviors to reduce mortality risks or disturbance as-sociated with roads, including vehicles (Brody and Pelton 1989;Dixon et al. 2006) and hunters (Milner et al. 2007; Stillfried et al.2015). Movement corridors are important to large carnivores be-cause they facilitate navigation through fragmented landscapeswhile minimizing mortality risks (Shepherd and Whittington2006). Although large carnivores typically avoid roads (e.g., Simeket al. 2015), roads with low traffic volume (Dixon et al. 2006) mayserve as movement corridors or areas with plentiful food (Maceet al. 1996; Roever et al. 2008; Hiller and Belant 2015, Hiller et al.2015). Additionally, riparian areas can allow large carnivores tomove among patches of food and cover (Naiman and Rogers 1997)while often providing food within the corridor itself (Lyons et al.2003). Investigating frequency of large carnivore use of differentmovement corridors help reveal the relationship between spatialvariation in resources and mortality risks.

Black bears (Ursus americanus Pallas, 1780) are omnivorous carni-vores found throughout most forested regions of North America(Scheick and McCown 2014) and are thought to use the smallestamount of space necessary to acquire resources for reproduction(Mitchell and Powell 2007). To access resources, bears must con-tend with variation in intraspecific competition related to beardensity, with individuals in denser populations typically havinggreater despotic behavior (Beckmann and Berger 2003). Bears mayreduce intraspecific competition or human-related risks (Ordizet al. 2011) by using habitat corridors (e.g., riparian areas; Lyonset al. 2003) to move among patches of resources (Dixon et al. 2006)and adjusting their fine-scale resource selection behaviors to thoserisks. Bears may also occasionally use roads with low anthropogenicrisk (Brody and Pelton 1989; Mace et al. 1996; Reynolds-Hogland andMitchell 2007) as movement corridors, but roads typically hinder

bears by introducing mortality risks (e.g., vehicle collisions) andfragmenting resources across the landscape (Reynolds-Hoglandand Mitchell 2007; Waller et al. 2014). Anthropogenic mortalityrisk may especially influence resource selection of adult femaleswith dependent young while meeting increased nutritional de-mands and avoiding vehicles (Wilton et al. 2014a), hunters(Reynolds-Hogland and Mitchell 2007), or intra-specific predation(LeCount 1987; Garrison et al. 2007). An assessment of resourceselection and mortality risk trade-offs of reproductive-aged femaleblack bears may therefore reveal factors that limit population-levelfitness (Mitchell and Powell 2007). Although numerous aspects ofblack bear resource selection relative to mortality risks have beenreported for individual populations (e.g., Belant et al. 2010; Walleret al. 2014), little is known about bear trade-offs between re-sources and mortality risks across populations with varying pop-ulation density and environmental characteristics.

Our objective was to assess reproductive-aged female black bearresource selection relative to variation in population density, veg-etation productivity, riparian corridors, and two classes of roadsof and within home ranges during spring–summer for popula-tions along a latitudinal gradient of the United States. We focusedon reproductive-aged females because this demographic is likelysensitive to behavioral trade-offs between resources and mortalityrisks (Martin et al. 2010) while maintaining nutritional condi-tion for reproductive success (Elowe and Dodge 1989; Dahle andSwenson 2003; Ayers et al. 2013). We established four hypothesesrelated to differences in population density, latitude, and spatialscale. First, we hypothesized that vegetation productivity wouldbest explain female resource selection across populations andspatial scales because reproductive-age females require increasedfood intake to support their nutritional condition for breeding,gestation, and care of dependent young (Dahle and Swenson 2003;Ayers et al. 2013). Second, we hypothesized that females wouldreduce their selection for profitable resources as populationdensity increased because females would avoid intraspecific com-petition (van Beest et al. 2014). Third, we hypothesized femaleresource selection behaviors to be more influential to the estab-lishment of home ranges across the landscape than selectionwithin home ranges because variation in resources and mortalityrisks would be more predictable at this scale (Rettie and Messier2000; Ordiz et al. 2011). Fourth, we hypothesized females to ex-hibit greater resource–risk trade-offs from Mississippi to Michi-gan because of increasing intraspecific competition (Dixon et al.2006), shorter vegetation growing season (Ferguson and McLoughlin2000; Bojarska and Selva 2012), and greater mortality risks fromvehicles and hunters (Reynolds-Hogland and Mitchell 2007; Stillfriedet al. 2015).

Materials and methods

Study areasWe conducted the study in Michigan, Missouri, and Mississippi,

USA. The study area in the south-central Upper Peninsula of Mich-igan (528.9 km2; 45°34=N, 87°20=W; Fig. 1) has a mean elevation of185 m above sea level and topography is flat. Predominant vege-tation included upland and lowland hardwoods, lowland coniferswamps, upland conifers, aspen (species of the genus Populus L.)stands, row-crop and livestock agriculture, herbaceous wetlands,and intermittent patches of berry-producing shrubs (e.g., raspber-ries (species of the genus Rubus L.) and blueberries (species ofthe genus Vaccinium L.)). Human development was low density(0.09 km/km2) residential and recreational camps and main andtertiary road densities were 0.03 and 1.69 km/km2, respectively(United States Bureau of the Census 2011). Riparian corridor (i.e.,rivers and streams) density was 1.17 km/km2 (United StatesGeological Survey 2014). During May–August, daily mean temper-ature ranged from 3.0 to 32.2 °C and rainfall during the studyperiod was 12.2–14.5 cm based on a weather station that we de-

204 Can. J. Zool. Vol. 95, 2017

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

ployed in the middle of the study area. Black bear density was14–19 bears/100 km2 (J.L. Belant, unpublished data). Black bears inMichigan were harvested annually during September and October,but only females without dependent young were legal for harvest(Belant et al. 2011).

The study area in the Ozark Highlands ecological region ofsouth-central Missouri (8033.8 km2; 35°57=N–40°35=N, 89°08=W–95°46=W; Fig. 1) consisted of mostly hilly and steep topographywith elevation from 70 to 540 m and the highest elevations in theOzark Highlands (Wilton et al. 2014b). Dominant landcover typesin the Ozark Highlands included forest, crop and pasture, grass-land, and human-developed areas (Fry et al. 2011). Forests wereprimarily upland oak (species of the genus Quercus L.) – hickory(species of the genus Carya Nutt.) and oak – pine (species of thegenus Pinus L.) (Raeker et al. 2010). Landownership included pri-vate homesteads and farms and public lands. Main and tertiaryroad densities were 0.18 and 1.28 km/km2, respectively (UnitedStates Bureau of the Census 2011). Riparian corridor density was0.08 km/km2 (United States Geological Survey 2014). Mean dailytemperature ranged from 7.9 to 33.3 °C and mean rainfall was12.7 cm during the study period (NOAA 2014). Black bear densitywas 1.7 bears/100 km2 (Wilton et al. 2014b) and are recolonizing the

southern and central regions of the state (Puckett et al. 2014;Wilton et al. 2014a).

The study area in Mississippi (7004.8 km2; 33°20=N, 90°93=W;Fig. 1) has elevations from 0 to 114 m above sea level and topogra-phy is generally flat. Land cover included agricultural (39%), waterand wetland (35%), forested (24%), and urban areas (2%) (Simeket al. 2015). Main and tertiary road densities were 0.14 and0.31 km/km2, respectively (United States Bureau of the Census2011). Riparian corridor density was 0.33 km/km2 (United StatesGeological Survey 2014). Climate is subtropical with mean monthlytemperature ranging from 7.5 to 27.0 °C (Simek et al. 2015). Beardensity was likely <1 bears/100 km2 (R. Rummell, Mississippi Depart-ment of Wildlife, Fisheries, and Parks, personal communication), buthas been increasing due to recolonization from expanding popu-lations in adjacent states, increased restoration of native habitat,and no legal harvest of bears (Simek et al. 2012).

Capture, handling, and monitoringWe live-captured female bears in Michigan from late May to

early July of 2009–2011, in Missouri from late May to early Octoberof 2010–2013, and in Mississippi year-round in 2009–2011 usingAldrich foot snares and culvert traps (Johnson and Pelton 1980;

Fig. 1. Reproductive-age female black bear (Ursus americanus) study areas (black polygons) and locations within study areas in Michigan,Missouri, and Mississippi, USA, during 2009–2013.

Duquette et al. 205

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

Reagan et al. 2002) baited with commercial lures and attractants.We checked traps at least daily. We immobilized females in thethree study areas intramuscularly with Telazol® (1:1 mixture oftiletamine hydrochloride (HCL) and zolazepam; Fort Dodge Ani-mal Health, Fort Dodge, Iowa, USA; Kreeger and Arnemo 2007)using a syringe pole or CO2-powered dart pistol or rifle (Pneu-Dart,Inc., Williamsport, Pennsylvania. USA). After induction, we re-corded morphometrics, determined sex, ear-tagged individuals,and collected blood, hair, and tissue samples. We fit femaleswith various Global Positioning System (GPS) telemetry collars inMichigan (Lotek 7000MU; Lotek Wireless, Newmarket, Ontario,Canada), Missouri (Northstar NSG-LD2: RASSL Globalstar, KingGeorge, Virginia, USA; ATS M2610B: Advance Telemetry SystemsInc., Isanti, Minnesota, USA; Lotek Wireless 7000MU: Lotek Wire-less, Newmarket, Ontario, Canada), and Mississippi (Telonics, Inc.,Mesa, Arizona, USA; Advanced Telemetry Systems Inc., Isanti,Minnesota, USA; Northstar, King George, Virginia, USA). We re-leased all females at their sites of capture. All capture and han-dling complied with the American Society of Mammalogistsguidelines (Sikes et al. 2011) and the Institutional Animal Care andUse Committee at Mississippi State University (protocol Nos. 08-052, 09-004, and 10-037). We programmed GPS radio collars inMichigan, Missouri, and Mississippi to record a location every 15,10, or 60 min from the time of deployment, respectively.

Spatial and habitat selectionWe estimated the likelihood of female resource selection of and

within home ranges using second- and third-order selection anal-yses (Manly et al. 2002) with designs 2 and 4 (Thomas and Taylor2006), respectively. We first selected female GPS locations thatwere separated by ≥60 min to reduce spatial autocorrelation,which was the minimum relocation time across populations. Westandardized step length estimates from locations >60 min apartto a 60 min interval to compare parameter estimates across pop-ulations and selection of home ranges across the landscape. Wethen defined resource use as female locations from 1 day after denexit or capture to cessation of monitoring (e.g., dropped collar), or31 August, which we imported into ArcGIS version 10.1 (ESRI, Inc.2011) as a shape file. We used 31 August as our cut-off date becauseof senescence of primary vegetation productivity and limited GPSlocation data after this time. To estimate female selection homeranges across the landscape, we first created a 100% minimumconvex polygon around all locations in each population usingArcGIS. We then defined study-area availability by using the Gen-erate Conditional Random Points tool in the Geospatial ModelingEnvironment (version 0.7.1.0; Beyer 2012) to plot the same numberof random points as GPS locations for each population withineach of the polygons of used locations.

To estimate female selection of resources within home ranges,we estimated the mean step length of successive locations foreach female in each population using Calculate Movement PathMetrics tool in the Geospatial Modeling Environment and thencalculated mean step length across populations. We then definedresource availability using the Generate Conditional RandomPoints tool in the Geospatial Modeling Environment to establish apaired random point within the determined standardized stepdistance (1054 m) of each female location across populations. Wethen copied the individual female and year information fromused locations to respective random locations for each populationfor model development.

We used the 16-day composite normalized difference vegetationindex (NDVI) data from the MODIS server (250 m resolution;United States Geological Survey 2013) as a metric of vegetationproductivity and quality (i.e., food) for all populations, which isassumed to be of great nutritional value during spring and sum-mer (Milakovic et al. 2012; Fowler 2014). We obtained NDVI rasterlayers with the closest date to the beginning of each monthfor May–August of 2009–2013 and used the Raster Calculator in

ArcGIS version 10.1 to estimate a mean NDVI score for each year.We imported the composite NDVI layers into ArcGIS, extractedthe NDVI value of each female location corresponding to the yearwhen the location was recorded, and exported this data to aspreadsheet.

We assessed rivers and nonintermittent streams as a metric ofriparian movement corridors for all populations using the UnitedStates Geological Survey National Hydrography Data (UnitedStates Geological Survey 2014). We imported river and stream datafor each study area into ArcGIS and then estimated the distanceof each female or random location to the nearest river or nonin-termittent stream. Similarly, we used distance to primary andsecondary (main roads) and tertiary roads as two metrics of an-thropogenic mortality risk (Reynolds-Hogland and Mitchell 2007;Martin et al. 2010; Waller et al. 2014), including hunting (Ciarnielloet al. 2007; Czetwertynski et al. 2007; Ordiz et al. 2011). We obtainedspatial road data in Michigan from the United States Bureau of theCensus (2011), in Missouri from the Missouri Spatial Data InformationService (Curators of the University of Missouri–Columbia 2011), and inMississippi from the Mississippi Automated Resource InformationSystem (2014) and estimated distance of each female location or ran-dom point to the nearest main and tertiary road in each study area byconducting a spatial join in ArcGIS.

We standardized all resource variables to z scores and centeredscores to provide equal weight in multiple regression analyses(Zar 1999). We used variance inflation factor (VIF) analysis to assessmulticollinearity (collinearity considered ≥7) among resourcevariables (Quinn and Keough 2002); no metrics were correlated(VIF = 1.05–1.13). We used package lme4 (Bates et al. 2011) in pro-gram R (R Core Team 2013) to develop six generalized linearmixed-effects models for each spatial scale in each state using amaximum likelihood estimator and binomial distribution usingfemale radiolocations (1) and random points (0) as the responsevariable. We assessed a null model incorporating only an inter-cept and a naïve model incorporating main roads, tertiary roads,riparian corridors, and NDVI as fixed effects. We structured theremaining four models to assess a potential functional responseto each of the variables using the following notation:

g(x) � �0 � �1xij � … �nxnij � �njxnj � �0j

where �0 is the mean intercept, �n are the fixed regression coeffi-cients with covariates xn, ij are individual females clustered withineach year, �nj is the random coefficient of covariate xn for eachfemale (j), and �0j is the random intercept (Gillies et al. 2006). Eachof the four functional models incorporated the fixed parametersused in the naïve model, but incorporated either main roads,tertiary roads, riparian corridors, or NDVI as a random covariate.Individual females were clustered within year as a random effectto account for variation in female behavior among years (Gillieset al. 2006). We considered parameter significance as � = 0.05. Weverified model fit by examining standardized versus fitted resid-ual plots and ranked models using Akaike’s information criterioncorrected for small sample size (AICc) (Burnham and Anderson2002). We considered models competing if the difference in their�AICc were ≤2 and presented models best explaining bear re-source selection.

ResultsWe obtained a median of 1 677 (range = 658 – 5 051) locations

from 7 females in Michigan, 3 153 (range = 68 – 6 127) locationsfrom 11 females in Missouri, and 3 377 (range = 103 – 10 363) loca-tions from 11 females in Mississippi. Females in Mississippi usedlocations farther from tertiary roads than did females in Michiganor Missouri (Table 1). Females in Missouri used locations fartherfrom main roads, but closer to riparian corridors, than did females inMichigan or Mississippi. Females in Michigan used locations with

206 Can. J. Zool. Vol. 95, 2017

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

NDVI values 14%–15% less than females in Missouri or Mississippi,which had similar values.

Selection of home rangesVariation in resource selection of females in Michigan was best

explained (Table 2) by the functional NDVI (i.e., vegetation produc-tivity) model, suggesting that females selected tertiary and mainroads and locations with greater vegetation productivity, butavoided riparian corridors (Table 3; Fig. 2a). Females in Michiganshowed functional selection of tertiary roads between about 300and 500 m and NDVI of about 7000–8000. Variation in resourceselection of females in Missouri was best explained by the func-tional riparian corridors model, suggesting that females selectedriparian corridors, tertiary roads, and locations with greater veg-etation productivity, but avoided main roads (Fig. 2b). Females inMissouri showed functional selection of tertiary roads betweenabout 300 and 2000 m and riparian corridors between about 500and 6500 m. Variation in resource selection of females in Missis-sippi was best explained the by functional NDVI model, suggest-ing that females selected tertiary and main roads and locationswith greater vegetation productivity, but avoided riparian corri-dors (Fig. 2c). Females in Mississippi showed functional selectionof tertiary roads between about 1 500 and 4 500 m, riparian corri-dors between about 5 000 and 10 000 m, and NDVI of about 7 000and 10 000. Greater coefficient values indicated the resource se-lection behaviors of females in Missouri were more pronouncedthan females in Michigan or Mississippi. Females in Mississippiand Michigan had analogous resource avoidance behaviors, butthe resource selection behaviors of females in Michigan weremore pronounced.

Selection within home rangesFemales across populations had similar resource selection be-

haviors to those they used to establish home ranges across thelandscape. But coefficient values suggested that female resourceselection behaviors were less pronounced within home rangesthan those used to establish home ranges across the landscape,with the exception of main roads and riparian corridors in Missis-sippi. Although the functional NDVI model best explained varia-tion of resource selection across populations (Table 2), femalesused locations that avoided or were indifferent to areas withgreater vegetation productivity (Table 3; Figs. 2d, 2e, 2f). Femalesin Michigan showed functional selection of riparian corridors be-tween about 7 000 and 11 000 m and females in Missouri showedfunctional selection of NDVI of about 4000–7000.

DiscussionReproductive-aged female bear resource selection was best ex-

plained by a functional response to vegetation productivity acrossnearly all populations and spatial scales, supporting our first hy-pothesis. Similar to previous bear studies (Milakovic et al. 2012;Fowler 2014), we presume vegetation productivity correlated withfood resources reproductive females needed to maintain nutri-tional condition for maximizing their reproductive success (Dahleand Swenson 2003; Ayers et al. 2013). Marked support for a func-tional response to vegetation productivity emphasizes femaleswere adjusting their behavior based on variation in vegetationproductivity across the landscape, and to a lesser extent withintheir home ranges. Compared with females in Missouri and Mis-sissippi, females in Michigan had the broadest functional selec-tion of vegetation productivity, which were also the least NDVIvalues across populations, despite similar or greater availabilityacross all landscapes. We suspect this trend in functional behaviorwas due to greater intraspecific competition for areas with pro-ductive vegetation (i.e., food and cover), which increased despoticbehavior (Beckmann and Berger 2003) from Mississippi to Michi-gan. For example, the low-density population in Mississippi wouldhave minimal intraspecific competition for those females andT

able

1.M

ean

and

stan

dard

devi

atio

n(S

D)

ofva

riab

les

rela

ted

tolo

cati

ons

ofre

pro

duct

ive

fem

ale

blac

kbe

ar(U

rsus

amer

ican

us)

inM

ich

igan

,M

isso

uri

,an

dM

issi

ssip

pi,

USA

,du

rin

gsp

rin

g–su

mm

erof

2009

–201

3.

Mic

hig

anM

isso

uri

Mis

siss

ipp

i

Var

iabl

eD

efin

itio

nR

epre

sen

tati

onEs

tim

ate

Mea

nSD

Mea

nSD

Mea

nSD

Bea

rde

nsi

ty(n

o.of

bear

s/10

0k

m2)

Mic

hig

an:J

.L.B

elan

t,u

np

ubl

ish

edda

ta;M

isso

uri

:Wil

ton

etal

.20

14b;

Mis

siss

ipp

i:R

.Ru

mm

ell,

MD

WFP

,per

son

alco

mm

un

icat

ion

Res

ourc

eco

mp

etit

ion

Pop

ula

tion

den

sity

14–1

9—

1.7

—<

1—

Dis

tan

ceto

nea

rest

tert

iary

road

(m)

Loca

lcou

nty

orci

tyro

ads,

off-

road

veh

icle

trai

ls,a

nd

pri

vate

road

s(U

nit

edSt

ates

Bu

reau

ofth

eC

ensu

s20

11)

Mor

tali

tyri

skor

mov

emen

tco

rrid

or

Den

sity

(km

/km

2)

1.69

—1.

28—

0.31

—U

sed

loca

tion

s49

638

651

131

637

0232

51LS

avai

labi

lity

653

518

823

767

5476

5637

HR

avai

labi

lity

625

414

611

325

3799

3256

Dis

tan

ceto

nea

rest

mai

nro

ad(m

)Pr

imar

yan

dse

con

dary

(mai

n)r

oads

are

gen

eral

lydi

vide

d,lim

ited

-acc

ess

hig

hw

ays

orm

ain

arte

ries

and

hav

eon

eor

mor

ela

nes

oftr

affi

cin

each

dire

ctio

n,a

nd

typi

cally

hav

ein

ters

ecti

ons

wit

hm

any

oth

erro

ads

(Un

ited

Stat

esBu

reau

ofth

eC

ensu

s20

11)

Mor

tali

tyri

skD

ensi

ty(k

m/k

m2)

0.03

—0.

18—

0.14

—U

sed

loca

tion

s21

4213

2376

8926

8617

1612

93LS

avai

labi

lity

5808

2948

1914

2156

4792

3585

HR

avai

labi

lity

5914

3058

4172

1799

6460

2685

Dis

tan

ceto

nea

rest

rip

aria

nco

rrid

or(m

)

Stre

ams

and

rive

rs,b

raid

edst

ream

s,ca

nal

s,di

tch

es,a

nd

pon

dan

dla

ke

edge

s(U

nit

edSt

ates

Geo

logi

calS

urv

ey20

14)

Mov

emen

tco

rrid

ors

orar

eas

offo

odre

sou

rces

Den

sity

(km

/km

2)

1.17

—0.

08—

0.33

—U

sed

loca

tion

s58

3330

9740

6417

8463

6626

83LS

avai

labi

lity

3260

2125

9429

5961

4542

3709

HR

avai

labi

lity

2271

1349

7792

2690

1814

1301

ND

VI

16-d

ayco

mp

osit

en

orm

aliz

eddi

ffer

ence

vege

tati

onin

dex

Food

reso

urc

esU

sed

loca

tion

s64

4998

574

7850

375

7610

49LS

avai

labi

lity

6160

1240

6143

1925

5169

2248

HR

avai

labi

lity

6522

1007

7583

530

7595

1228

No

te:M

DW

FP,M

issi

ssip

piD

epar

tmen

tof

Wil

dlif

e,Fi

sher

ies,

and

Park

s;LS

,lan

dsca

pe

scal

e;H

R,h

ome

ran

ge.

Duquette et al. 207

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

allowed them to select choice patches of food and cover (Beyeret al. 2013) as we observed at the landscape scale. Although fe-males in Missouri also exhibited a threshold of vegetation pro-ductivity selection, the prominence of riparian corridor use byMissouri females was probably due to the steep terrain of thestudy area and anthropogenic risk (e.g., vehicles; Brody and Pelton1989) on hilltops, which filtered animals nearer to streams orrivers (Garneau et al. 2008; Waller et al. 2013). Areas near ripariancorridors may have also provided suitable patches of food (Lyonset al. 2003) or cover in that landscape.

The shift in selection to avoidance of vegetation productivityfrom the landscape to home-range scale further suggests thatintraspecific competition influenced female resource selection(van Beest et al. 2014). This shift supported our second hypothesisthat females would reduce their selection for profitable resourcesas population density (i.e., intraspecific competition) increased.The negative relationship between profitable resource access andintraspecific competition was verified by greater selection of veg-etation productivity by lower bear density populations in Mis-souri and Mississippi compared with Michigan. With greater beardensity, females in Michigan would have maximized their life-time reproductive success by occupying areas with reduced mor-tality risk to themselves or cubs (LeCount 1987; Garrison et al.2007), even if those habitats had poorer food resources (Ben-Davidet al. 2004; McDonald and Fuller 2005). But females in Michiganlikely offset these constraints by establishing suitable homeranges across the landscape using prior knowledge of food distri-bution (McDonald and Fuller 2005). This strategy can allow fe-males to minimize their space use while providing flexibility tomove among patches of food and avoid risks (Mitchell and Powell2007), as indicated by the functional selection of vegetation pro-ductivity by each population.

With the exception of vegetation productivity, females in eachpopulation established home ranges across the landscape basedon similar resource selection patterns as those within homeranges. Although consistency in resource selection behaviors sug-gests that these resources were perceived similarly at each spatialscale, the strength of these behaviors were greater at the land-scape scale than within the home range, supporting our third

hypothesis. Similar to vegetation productivity, broad-scale knowl-edge of variation in resources (McDonald and Fuller 2005) andmortality risks was likely more beneficial for females than at-tempting to monitor and respond to smaller scale changes in theirhome ranges (Ordiz et al. 2011, 2012). For example, females inMississippi increased their selection of main roads and avoidanceof riparian areas from the landscape to home-range scale, plausi-bly because of a 35% increase and 40% decrease in the mean dis-tances of those resources at the home-range scale, respectively.The scale-dependent behaviors of female bears in our study cor-roborate those of other large mammals (Rettie and Messier 2000;Latif et al. 2011; Beyer et al. 2013) and similarly suggest femalesfunctionally adjusted their behavior to the spatial scale at whicheach factor most limited their reproductive success (Rettie andMessier 2000).

The breadth of resource selection trade-offs was greatest forfemales in Missouri, followed by Michigan and Mississippi, notsupporting our fourth hypothesis that females would exhibitgreater behavioral trade-offs from Mississippi to Michigan. Varia-tion in resource selection across populations appeared to reflectresource availability in each landscape, rather than bear popula-tion density (Bojarska and Selva 2012). Although population-levelvariation in selection of vegetation productivity appeared to beinfluenced by bear density, females likely maximized their re-productive success by secondarily adapting to other landscape-specific resources or mortality risks. Brown bear (Ursus arctos L.,1758) life-history strategies were similarly shown to be primarilyinfluenced by variation in environmental pressures across a largeportion of their range (Ferguson and McLoughlin 2000). Differ-ences in resources and mortality risks among populations mayexplain why females in Missouri had greater behavioral trade-offsthan in Michigan, despite bear density being 8.2–11.2 times greater inMichigan than in Missouri.

Within populations, female trade-offs in resources and mortalityrisks were likely related to landscape-specific variation in those fac-tors. Tertiary road avoidance increased as bear population densityincreased from Mississippi to Michigan. This trend in avoidancewas closely associated with landscape availability, as tertiary roaddensity was 5.5 and 4.0 times greater in Michigan and Missouri,

Table 2. Model rankings of generalized linear mixed-effect models assessing resource selection of reproductive female black bear (Ursusamericanus; n = 29) across the study landscape and within home ranges in Michigan, Missouri, and Mississippi, USA, during spring–summer of2009–2013.

Landscape Home range

State Model K AICc �AICc

AICc

weight LL Model K AICc �AICc

AICc

weight LL

Michigan (n = 7) NDVI 8 14 233.0 0.0 1.0 –7 108.5 NDVI 8 12 522.6 0.0 1.0 –6 253.3Riparian corridor 8 20 368.1 6 135.1 0.0 –10 176.0 Riparian corridor 8 19 403.0 6 880.4 0.0 –9 693.5Main roads 8 20 838.6 6 605.6 0.0 –10 411.3 Main roads 8 19 395.8 6 873.2 0.0 –9 689.9Tertiary roads 8 20 882.9 6 649.9 0.0 –10 433.5 Tertiary roads 8 19 391.6 6 869.0 0.0 –9 687.8Naïve 7 20 949.1 6 716.1 0.0 –10 467.6 Naïve 7 19 489.0 6 966.5 0.0 –9 737.5Null 3 42 886.8 28 653.8 0.0 –21 440.4 Null 3 42 864.4 30 341.8 0.0 –21 429.2

Missouri (n = 11) Riparian corridor 8 18 270.1 0.0 1.0 –9 127.1 NDVI 8 12 522.6 0.0 1.0 –6 253.3NDVI 8 18 365.8 95.7 0.0 –9 174.9 Tertiary roads 8 19 391.6 6 869.0 0.0 –9 687.8Main roads 8 18 870.7 600.6 0.0 –9 427.3 Main roads 8 19 395.8 6 873.3 0.0 –9 689.9Tertiary roads 8 25 841.9 7 571.8 0.0 –12 912.9 Riparian corridor 8 19 403.0 6 880.4 0.0 –9 693.5Naïve 7 26 346.3 8 076.2 0.0 –13 166.2 Naïve 7 19 489.0 6 966.5 0.0 –9 737.5Null 3 95 957.0 77 686.9 0.0 –47 975.5 Null 3 42 864.4 30 341.8 0.0 –21 429.2

Mississippi (n = 11) NDVI 8 43 575.4 0.0 1.0 –21 779.7 NDVI 8 30 313.8 0.0 1.0 –15 148.9Riparian corridor 8 44 157.3 581.9 0.0 –22 070.7 Riparian corridor 8 38 815.4 8 501.6 0.0 –19 399.7Main roads 8 49 918.9 6 343.5 0.0 –24 951.5 Main roads 8 38 997.6 8 683.9 0.0 –19 490.8Tertiary roads 8 51 588.6 8 013.2 0.0 –25 786.3 Tertiary roads 8 40 040.5 9 726.8 0.0 –20 012.3Naïve 7 60 549.4 16 974.0 0.0 –30 267.7 Naïve 7 40 520.6 10 206.9 0.0 –20 253.3Null 3 128 997.9 85 422.5 0.0 –64 496.0 Null 3 95 957.0 65 643.2 0.0 –47 975.5

Note: Models included individual bear and year as random effects on the intercept. Models presented Akaike’s information criterion corrected for small sample size(AICc), difference in AIC relative to minimum AIC (�AICc), AIC weight of the model relative to other models (AICc weight), number of parameters (K), and log-likelihood(LL). The model NDVI is the normalized difference vegetation index.

208 Can. J. Zool. Vol. 95, 2017

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

respectively, than in Mississippi. By adjusting avoidance behaviorto landscape-specific tertiary road density, females in each popu-lation could have maximized accessed resources (e.g., food) be-tween tertiary roads with human mortality risk (Ciarniello et al.2007; Reynolds-Hogland and Mitchell 2007) and interior habitatswith greater risk of conflicts with male bears (Garrison et al. 2007).Our results were corroborated by Stillfried et al. (2015) who con-ducted an independent analysis of female black bear resourceselection in the Michigan study area. Similarly, avoidance of pri-mary roads by females in Missouri was likely due to main roadsbeing 1.3–6.0 times denser than in Mississippi and Michigan, re-spectively. The functional selection of riparian corridors betweenabout 1000 and 6000 m by females in Missouri could represent asimilar scenario as in Michigan, whereby females used areas (e.g.,hillsides) between main roads with human risks (e.g., vehicles)and riparian corridors where conspecifics tend to concentrate(Lyons et al. 2003). Given the choice, use of areas near ripariancorridors were likely safer than areas near main roads for femalesin Missouri. In contrast, landscape terrain in Michigan and Missis-sippi was similarly flat and had low-density human developmentthat could have allowed females to use habitats closer to mainroads with less human disturbance. These landscape characteristicscould have allowed females to access food and cover between mainroads with human risk (Jonkel and Cowan 1971; Reynolds-Hoglandand Mitchell 2007; Roever et al. 2008) and riparian areas with poten-tially greater conspecific risk (Garneau et al. 2008; Waller et al.2013) or thick vegetation that impeded movement.

ConclusionsWe demonstrated variation in spatially dependent resource se-

lection across black bear populations that allowed females to pro-cure resources while reducing apparent mortality risks. Femalesin our study adjusted their behavior to establish home rangeswith a threshold of vegetation productivity, under the confines ofinterspecific competition and mortality risk within their popula-tion. Although this behavior decreased within home ranges, fe-males maintained behavioral trade-offs among resources andmortality risks across two spatial scales. This behavior empha-sizes the importance of considering the functional behavior ofpopulations at multiple spatial scales (Rettie and Messier 2000) toprovide the best knowledge to guide population management.Similar functional trade-offs have been reported for other largemammals (Godvik et al. 2009; Matthiopoulos et al. 2011; Moreauet al. 2012), emphasizing that behavioral flexibility allows animalsto meet their life-history requirements under spatially variableintraspecific competition and environmental conditions. As blackbear populations in Missouri and Mississippi increase, greatercompetition for food and space will likely greater constrain beartrade-offs between resources and mortality risks, as we observedbetween bear density and vegetation productivity at the land-scape scale. Increased competition from conspecifics may in-crease female use of areas with greater mortality risks (e.g., foodnearer to tertiary roads), which could limit population growth

Table 3. Generalized linear mixed-effect modelsbest explaining resource selection of reproductivefemale black bear (Ursus americanus; n = 30) acrossthe study landscape or within home ranges inMichigan, Missouri, and Mississippi, USA, duringspring–summer of 2009–2013.

Landscape Home range

Michigan�0 –4.223 –3.501SE 0.344 0.358P <0.001 <0.001�tertiary roads –5.286 –2.568SE 0.249 0.213P <0.001 <0.001�main roads –3.717 –3.275SE 0.072 0.078P <0.001 <0.001�riparian 3.232 2.919SE 0.073 0.072P <0.001 <0.001�NDVI 1.766 –0.065SE 0.102 0.071P <0.001 <0.001VarNDVI 7.901 6.226SD 2.811 2.495Varbear 0.539 0.294SD 0.734 0.543Varyear 0.049 0.170SD 0.221 0.412

Missouri�0 –0.511 –2.979SE –0.170 0.283P 0.865 <0.001�tertiary roads –4.316 –2.872SE 1.143 0.162P <0.001 <0.001�main roads 8.959 2.021SE 0.602 0.029P <0.001 <0.001�riparian –9.321 –2.226SE 0.493 0.030P <0.001 <0.001�NDVI 8.378 –0.104SE 0.569 0.074P <0.001 0.160VarNDVI 6.204 7.281SD 2.560 2.698Varbear 0.869 0.054SD 0.932 0.233Varyear 0.441 0.176SD 0.664 0.419

Mississippi�0 –2.241 –0.981SE 0.330 0.299P <0.001 0.001�tertiary roads –0.215 0.038SE 0.015 0.016P <0.001 0.019�main roads –2.525 –4.477SE 0.031 0.053P <0.001 <0.001�riparian 1.583 4.180SE 0.024 0.051P <0.001 <0.001�NDVI 2.668 –0.142SE 0.048 0.029P <0.001 <0.001

Table 3 (concluded).

Landscape Home range

VarNDVI 5.367 3.154SD 2.317 1.776Varbear 1.104 0.387SD 1.051 0.622Varyear 0.029 0.307SD 0.169 0.554

Note: Models presented with standardized parameterestimates (�), standard errors (SE) or standard deviations(SD), probability values (P), and the variance (Var) of thefunctional response parameter (normalized differencevegetation index (NDVI) or riparian), individual bear, andyear included as random effects on the intercept.

Duquette et al. 209

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

(Simek et al. 2015). We suggest that maintenance and establish-ment of movement corridors linking high-quality food and coverwithin and across bear populations is important for maintainingor restoring demographic and genetic connectivity of bears (Dixonet al. 2006; Wilton et al. 2014a; Simek et al. 2012, 2015). In additionto yielding important information for guiding black bear manage-ment, our study underscores how multiple factors can influencebroad variation in spatially dependent functional resource selec-tion of a species across large portions of its range. Advances in

computational abilities should allow greater exploration of mod-els incorporating multiple functional relationships and improvedinformation for management.

AcknowledgementsThis project was supported by the Federal Aid in Wildlife Res-

toration Act under Pittman–Robertson. We thank the MichiganDepartment of Natural Resources; Mississippi Department of Wild-life, Fisheries, and Parks; Missouri Department of Conservation;

Fig. 2. Coefficient relationships of the best generalized linear mixed-effect model explaining the resource selection of reproductive-age femaleblack bears (Ursus americanus) across the study landscape or within home ranges in Michigan (a, d), Missouri (b, e), and Mississippi (c, f), USA, duringspring–summer of 2009–2013. Solid thick line is main roads, solid thin line is tertiary roads, long-dashed line is riparian corridors, and dotted line isthe normalized difference vegetation index (NDVI). Gray shading around lines represent ±1 standard error of the mean of each resource.

210 Can. J. Zool. Vol. 95, 2017

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

Safari Club International Foundation; and Safari Club International–Michigan Involvement Committee for financial support. Wethank the Mississippi State University Department of Wildlife,Fisheries, and Aquaculture; Carnivore Ecology Laboratory; andForest and Wildlife Research Center for logistical support. Muchgratitude go to participating landowners for land access andC. Albright, S. Edwards, D. O’Brien, P. Lederle, B. Roell, B. Young,and numerous technicians for field and technical support. Wethank the Institute for Wildlife Studies for providing financialassistance for the primary author during the preparation of thispaper. We thank A. Bridges for valuable comments on an earlierdraft of this paper.

ReferencesAyers, C.R., Belant, J.L., and Millspaugh, J.J. 2013. Directness of resource use

metrics affects predictions of bear body fat gain. Polar Biol. 36: 169–176.doi:10.1007/s00300-012-1247-7.

Basille, M., Van Moorter, B., Herfindal, I., Martin, J., Linnell, J.D.C., Odden, J.,Andersen, R., and Gaillard, J.M. 2013. Selecting habitat to survive: the impactof road density on survival in a large carnivore. PLoS ONE, 8: e65493. doi:10.1371/journal.pone.0065493. PMID:23874381.

Bates, D., Maechler, M., and Bolker, B. 2011. lme4: Linear mixed-effects modelsusing S4 classes. R package version 0.999375-42. Available from http://CRAN.R-project.org/package=lme4 [accessed 2 June 2015].

Beckmann, J.P., and Berger, J. 2003. Using black bears to test ideal-free distribu-tion models experimentally. J. Mammal. 84: 594–606. doi:10.1644/1545-1542(2003)084<0594:UBBTTI>2.0.CO;2.

Belant, J.L., Griffith, B., Zhang, Y., Follmann, E.H., and Adams, L.G. 2010.Population-level resource selection by sympatric brown bears and Americanblack bears in Alaska. Polar Biol. 33: 31–40. doi:10.1007/s00300-009-0682-6.

Belant, J.L., Etter, D.R., Mayhew, S.L., Visser, L.G., and Friedrich, P.D. 2011. Im-proving large scale mark–recapture estimates for American black bear pop-ulations. Ursus, 22: 9–23. doi:10.2192/URSUS-D-10-00011.1.

Ben-David, M., Titus, K., and Beier, L.R. 2004. Consumption of salmon by Alaskanbrown bears: a trade-off between nutritional requirements and the risk ofinfanticide? Oecologia, 138: 465–474. doi:10.1007/s00442-003-1442-x. PMID:14673639.

Beyer, H.L. 2012. Geospatial Modelling Environment. Version 0.7.1.0 [computerprogram]. Available from http://www.spatialecology.com/gme [accessed 2 June2016].

Beyer, H.L., Ung, R., Murray, D.L., and Fortin, M.J. 2013. Functional responses,seasonal variation and thresholds in behavioural responses of moose to roaddensity. J. Appl. Ecol. 50: 286–294. doi:10.1111/1365-2664.12042.

Bojarska, K., and Selva, N. 2012. Spatial patterns in brown bear Ursus arctos diet:the role of geographical and environmental factors. Mammal Rev. 42: 120–143. doi:10.1111/j.1365-2907.2011.00192.x.

Brody, A.J., and Pelton, M.R. 1989. Effects of roads on black bear movements inwestern North Carolina. Wildl. Soc. Bull. 17: 5–10.

Burnham, K.P., and Anderson, D.R. 2002. Model selection and multimodelinference: a practical information–theoretic approach. Springer, New York.

Ciarniello, L.M., Boyce, M.S., Heard, D.C., and Seip, D.R. 2007. Components ofgrizzly bear habitat selection: density, habitats, roads, and mortality risk.J. Wildl. Manage. 71: 1446–1457. doi:10.2193/2006-229.

Creel, S., Winnie, J.A., Jr., Christianson, D., and Liley, S. 2008. Time and space ingeneral models of antipredator response: tests with wolves and elk. Anim.Behav. 76: 1139–1146. doi:10.1016/j.anbehav.2008.07.006.

Curators of the University of Missouri–Columbia. 2011. Missouri Spatial DataInformation Service. Available from http://msdis.missouri.edu/ [accessed2 June 2015].

Czetwertynski, S.M., Boyce, M.S., and Schmiegelow, F.K. 2007. Effects of huntingon demographic parameters of American black bears. Ursus, 18: 1–18. doi:10.2192/1537-6176(2007)18[1:EOHODP]2.0.CO;2.

Dahle, B., and Swenson, J.E. 2003. Seasonal range size in relation to reproductivestrategies in brown bears Ursus arctos. J. Anim. Ecol. 72: 660–667. doi:10.1046/j.1365-2656.2003.00737.x.

Dixon, J.D., Oli, M.K., Wooten, M.C., Eason, T.H., McCown, J.W., and Paetkau, D.2006. Effectiveness of a regional corridor in connecting two Florida blackbear populations. Conserv. Biol. 20: 155–162. doi:10.1111/j.1523-1739.2005.00292.x. PMID:16909668.

Elowe, K.D., and Dodge, W.E. 1989. Factors affecting black bear reproductivesuccess and cub survival. J. Wildl. Manage. 53: 962–968. doi:10.2307/3809596.

Environmental Systems Research Institute (ESRI), Inc. 2011. ArcGIS. Version 10.0[computer program]. ESRI, Inc., Redlands, Calif.

Ferguson, S.H., and McLoughlin, P.D. 2000. Effect of energy availability, season-ality, and geographic range on brown bear life history. Ecography, 23: 193–200. doi:10.1111/j.1600-0587.2000.tb00275.x.

Fowler, N.L. 2014. Expression of facultative den behavior in Ursidae. MS thesis,Mississippi State University, Mississippi State.

Fry, J.A., Coan, M., Homer, C., Meyer, D., and Wickham, J. 2011. Completion of

the 2006 National Land Cover Database for the conterminous United States.Photogramm. Eng. Remote Sens. 77: 858–864.

Garneau, D.E., Boudreau, T., Keech, M., and Post, E. 2008. Habitat use by blackbears in relation to conspecifics and competitors. Mamm. Biol. 73: 48–57.doi:10.1016/j.mambio.2006.11.004.

Garrison, E.P., McCown, J.W., and Oli, M.K. 2007. Reproductive ecology and cubsurvival of Florida black bears. J. Wildl. Manage. 71: 720–727. doi:10.2193/2005-689.

Gillies, C.S., Hebblewhite, M., Nielsen, S.E., Krawchuk, M.A., Aldridge, C.L.,Frair, J.L., Saher, D.J., Stevens, C.E., and Jerde, C.L. 2006. Application of ran-dom effects to the study of resource selection by animals. J. Anim. Ecol. 75:887–898. doi:10.1111/j.1365-2656.2006.01106.x. PMID:17009752.

Godvik, I.M.R., Loe, L.E., Vik, J.O., Veiberg, V., Langvatn, R., and Mysterud, A.2009. Temporal scales, trade-offs, and functional responses in red deer hab-itat selection. Ecology, 90: 699–710. doi:10.1890/08-0576.1. PMID:19341140.

Hiller, T.L., and Belant, J.L. 2015. Sexual size dimorphism mediates effects ofresource dispersion on American black bear space use. J. Zool. (Lond.), 295:200–207. doi:10.1111/jzo.12234.

Hiller, T.L., Belant, J.L., Beringer, J., and Tyre, A.J. 2015. Resource selection byrecolonizing American black bears in a fragmented forest landscape. Ursus,26: 116–128. doi:10.2192/URSUS-D-15-00023.1.

Johnson, C.J., and Seip, D.R. 2008. Relationship between resource selection,distribution, and abundance: a test with implications to theory and conser-vation. Popul. Ecol. 50: 145–157. doi:10.1007/s10144-008-0078-4.

Johnson, K.G., and Pelton, M.R. 1980. Prebaiting and snaring techniques forblack bears. Wildl. Soc. Bull. 8: 46–54.

Jonkel, C.J., and Cowan, I.M. 1971. The black bear in the spruce–fir forest. Wildl.Monogr. 27: 3–57.

Kiffner, C., Kioko, J., Kissui, B., Painter, C., Serota, M., White, C., and Yager, P.2014. Interspecific variation in large mammal responses to human observersalong a conservation gradient with variable hunting pressure. Anim. Conserv.17: 603–612. doi:10.1111/acv.12131.

Kreeger, T.J., and Arnemo, J.M. 2007. Handbook of wildlife chemical immobili-zation, 3rd ed. Wildlife Pharmaceuticals. Sunquest, Shanghai, People’s Re-public of China.

Latif, Q.S., Heath, S.K., and Rotenberry, J.T. 2011. An ‘ecological trap’ for yellowwarbler nest microhabitat selection. Oikos, 120: 1139–1150. doi:10.1111/j.1600-0706.2010.18835.x.

LeCount, A.L. 1987. Causes of black bear cub mortality. Int. Conf. Bear Res.Manage. 7: 75–82.

Lyons, A.L., Gaines, W.L., and Servheen, C. 2003. Black bear resource selection inthe northeast Cascades, Washington. Biol. Conserv. 113: 55–62. doi:10.1016/S0006-3207(02)00349-X.

Mace, R.D., Waller, J.S., Manley, T.L., Lyon, L.J., and Zuuring, H. 1996. Relationshipsamong grizzly bears, roads and habitat in the Swan Mountains Montana. J. Appl.Ecol. 33: 1395–1404. doi:10.2307/2404779.

Manly, B.F.J., McDonald, L.L., Thomas, D.L., McDonald, T.L., and Erickson, W.P.2002. Resource selection by animals. 2nd ed. Kluwer Academic Publishers,Boston, Mass.

Martin, J., Basille, M., Van Moorter, B., Kindberg, J., Allainé, D., and Swenson, J.E.2010. Coping with human disturbance: spatial and temporal tactics of thebrown bear (Ursus arctos). Can. J. Zool. 88(9): 875–883. doi:10.1139/Z10-053.

Matthiopoulos, J., Hebblewhite, M., Aarts, G., and Fieberg, J. 2011. Generalizedfunctional responses for species distributions. Ecology, 92: 583–589. doi:10.1890/10-0751.1. PMID:21608467.

McDonald, J.E., Jr., and Fuller, T.K. 2005. Effects of spring acorn availability onblack bear diet, milk composition, and cub survival. J. Mammal. 86: 1022–1028. doi:10.1644/1545-1542(2005)86[1022:EOSAAO]2.0.CO;2.

McLoughlin, P.D., Ferguson, S.H., and Messier, F. 2000. Intraspecific variation inhome range overlap with habitat quality: a comparison among brown bearpopulations. Evol. Ecol. 14: 39–60. doi:10.1023/A:1011019031766.

McLoughlin, P.D., Morris, D.W., Fortin, D., Vander Wal, E., and Contasti, A.L.2010. Considering ecological dynamics in resource selection functions. J. Anim.Ecol. 79: 4–12. doi:10.1111/j.1365-2656.2009.01613.x. PMID:19732211.

Milakovic, B., Parker, K.L., Gustine, D.D., Lay, R.J., Walker, A.B.D., andGillingham, M.P. 2012. Seasonal habitat use and selection by grizzly bears innorthern British Columbia. J. Wildl. Manage. 76: 170–180. doi:10.1002/jwmg.235.

Milner, J.M., Nilsen, E.B., and Andreassen, H.P. 2007. Demographic side effects ofselective hunting in ungulates and carnivores. Conserv. Biol. 21: 36–47. doi:10.1111/j.1523-1739.2006.00591.x. PMID:17298509.

Mississippi Automated Resource Information System. 2014. Statewide datadownload. Available from http://www.maris.state.ms.us/HTM/DownloadData/Statewide.html [accessed 2 June 2015].

Mitchell, M.S., and Powell, R.A. 2007. Optimal use of resources structures homeranges and spatial distribution of black bears. Anim. Behav. 74: 219–230.doi:10.1016/j.anbehav.2006.11.017.

Moreau, G., Fortin, D., Couturier, S., and Duchesne, T. 2012. Multi-level func-tional responses for wildlife conservation: the case of threatened caribou inmanaged boreal forests. J. Appl. Ecol. 49: 611–620. doi:10.1111/j.1365-2664.2012.02134.x.

Naiman, R.J., and Rogers, K.H. 1997. Large animals and system-level characteris-tics in river corridors. BioScience, 47: 521–529. doi:10.2307/1313120.

Duquette et al. 211

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.

National Oceanic and Atmospheric Administration (NOAA). 2014. National Cli-matic Data Center. Available from http://cdo.ncdc.noaa.gov/qclcd/ [accessed2 June 2015].

Ordiz, A., Støen, O.G., Delibes, M., and Swenson, J.E. 2011. Predators or prey?Spatio-temporal discrimination of human-derived risks by brown bears.Oecologia, 166: 59–67. PMID:21298447.

Ordiz, A., Støen, O.G., Sæbø, S., Kindberg, J., Delibes, M., and Swenson, J.E. 2012.Do bears know they are being hunted? Biol. Conserv. 152: 21–28. doi:10.1016/j.biocon.2012.04.006.

Palomares, F., and Caro, T.M. 1999. Interspecific killing among mammaliancarnivores. Am. Nat. 153: 492–508. doi:10.2307/2463664.

Puckett, E.E., Kristensen, T.V., Wilton, C.M., Lyda, S.B., Noyce, K.V.,Holahan, P.M., Leslie, D.M., Jr., Beringer, J., Belant, J.L., White, D., Jr., andEggert, L.S. 2014. Influence of drift and admixture on population structure ofAmerican black bears (Ursus americanus) in the Central Interior Highlands,USA, 50 years after translocation. Mol. Ecol. 23: 2414–2427. PMID:24712442.

Quinn, G.P., and Keough, M.J. 2002. Experimental design and data analysis forbiologists. Cambridge University Press, New York.

R Core Team. 2013. R: a language and environment for statistical computing.R Foundation for Statistical Computing, Vienna, Austria. Available fromhttps://www.r-project.org/.

Raeker, G., Fleming, J., Morris, M., Moser, K., and Treiman, T. 2010. Missouri’sforest resource assessment and strategy, seeking a sustainable future forMissouri’s forest resources. Missouri Department of Conservation and USForest Service, Northern Research Station, Newtown Square, Pa.

Reagan, S.R., Ertel, J.M., Stinson, P., Yakupzack, P., and Anderson, D. 2002. Apassively triggered foot snare design for American black bears to reducedisturbance by non-target animals. Ursus, 13: 317–320.

Rettie, W.J., and Messier, F. 2000. Hierarchical habitat selection by woodlandcaribou: its relationship to limiting factors. Ecography, 23: 466–478. doi:10.1111/j.1600-0587.2000.tb00303.x.

Reynolds-Hogland, M.J., and Mitchell, M.S. 2007. Effects of roads on habitatquality for bears in the southern Appalachian: a long-term study. J. Mammal.88: 1050–1061. doi:10.1644/06-MAMM-A-072R1.1.

Roever, C.L., Boyce, M.S., and Stenhouse, G.B. 2008. Grizzly bears and forestry. I:Road vegetation and placement as an attractant to grizzly bears. For. Ecol.Manage. 256: 1253–1261. doi:10.1016/j.foreco.2008.06.040.

Scheick, B.K., and McCown, W. 2014. Geographic distribution of American blackbears in North America. Ursus, 25: 24–33. doi:10.2192/URSUS-D-12-00020.1.

Shepherd, B., and Whittington, J. 2006. Response of wolves to corridor resto-ration and human use management. Ecol. Soc. 11: 1. doi:10.5751/ES-01813-110201.

Sikes, R.S., Gannon, W.L., and Animal Care and Use Committee of the AmericanSociety of Mammalogists. 2011. Guidelines of the American Society of Mam-malogists for the use of wild mammals in research. J. Mammal. 92: 235–253.doi:10.1644/10-MAMM-F-355.1.

Simek, S.L., Belant, J.L., Young, B.W., Shropshire, C., and Leopold, B.D. 2012.History and status of the American black bear in Mississippi. Ursus, 23:159–167. doi:10.2192/URSUS-D-11-00031.1.

Simek, S.L., Belant, J.L., Fan, Z., Young, B., Leopold, B.D., Fleming, J., andWaller, B. 2015. Source populations and roads affect American black bearrecolonization. Eur. J. Wildl. Res. 61: 583–590. doi:10.1007/s10344-015-0933-5.

Steyaert, S.M.J.G., Reusch, C., Brunberg, S., Swenson, J.E., Hackländer, K., andZedrosser, A. 2013. Infanticide as a male reproductive strategy has a nutritiverisk effect in brown bears. Biol. Lett. 9: 20130624. doi:10.1098/rsbl.2013.0624.PMID:24088563.

Stillfried, M., Belant, J.L., Svoboda, N.J., Beyer, D.E., Jr., and Kramer-Schadt, S.2015. When top predators become prey: black bears alter movement behav-ior in response to hunting pressure. Behav. Processes, 120: 30–39. doi:10.1016/j.beproc.2015.08.003. PMID:26277059.

Thomas, D.L., and Taylor, E.J. 2006. Study designs and tests for comparing re-source use and availability II. J. Wildl. Manage. 70: 324–336. doi:10.2193/0022-541X(2006)70[324:SDATFC]2.0.CO;2.

United States Bureau of the Census. 2011. Topologically Integrated GeographicEncoding and Referencing (TIGER) system files. Available from https://www.census.gov/geo/maps-data/data/tiger.html [accessed 2 June 2015].

United States Geological Survey. 2013. 250 m MODIS normalized difference veg-etation index. Available from https://earthexplorer.usgs.gov/ [accessed 2 June2015].

United States Geological Survey. 2014. National hydrography data. Availablefrom https://nhd.usgs.gov/ [accessed 2 June 2015].

van Beest, F.M., Uzal, A., Vander Wal, E., Laforge, M.P., Contasti, A.L., Colville, D.,and McLoughlin, P.D. 2014. Increasing density leads to generalization in bothcoarse-grained habitat selection and fine-grained resource selection in alarge mammal. J. Anim. Ecol. 83: 147–156. doi:10.1111/1365-2656.12115. PMID:23931034.

Waller, B.W., Belant, J.L., Leopold, B.D., Evans, D.L., Young, B.W., and Simek, S.L.2013. Scale-dependent den-site selection by American black bears in Missis-sippi. Mamm. Study, 38: 211–216. doi:10.3106/041.038.0309.

Waller, B.W., Belant, J.L., Leopold, B.D., Young, B.W., Simek, S.L., and Evans, D.L.2014. Influence of landscape attributes on American black bear den-site se-lection in Mississippi. Mamm. Study, 39: 115–119. doi:10.3106/041.039.0207.

Watts, H.E., and Holekamp, K.E. 2009. Ecological determinants of survival andreproduction in the spotted hyena. J. Mammal. 90: 461–471. doi:10.1644/08-MAMM-A-136.1.

Wilton, C.W., Belant, J.L., and Beringer, J. 2014a. Distribution of American blackbear occurrences and human–bear incidents in Missouri. Ursus, 25: 53–60.doi:10.2192/URSUS-D-13-00017.1.

Wilton, C.M., Puckett, E.E., Beringer, J., Gardner, B., Eggert, L.S., and Belant, J.L.2014b. Trap array configuration influences estimates and precision of blackbear density and abundance. PLoS ONE, 9: e111257. doi:10.1371/journal.pone.0111257. PMID:25350557.

Zar, J.H. 1999. Biostatistical analysis. 4th ed. Prentice Hall, Inc., Upper SaddleRiver, N.J.

Zedrosser, A., Dahle, B., and Swenson, J.E. 2006. Population density and foodconditions determine adult female body size in brown bears. J. Mammal. 87:510–518. doi:10.1644/05-MAMM-A-218R1.1.

212 Can. J. Zool. Vol. 95, 2017

Published by NRC Research Press

Can

. J. Z

ool.

Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

MIS

SISS

IPPI

ST

AT

E U

NIV

LIB

on

03/0

6/17

For

pers

onal

use

onl

y.