InternationalScholarlyResearchNetwork ISRNZoology Volume2012,ArticleID197356,8pages doi:10.5402/2012/197356 Research Article Swift Foxes and Ideal Free Distribution: Relative Influence of Vegetation and Rodent Prey Base on Swift Fox Survival, Density, and Home Range Size CraigM.Thompson1,2andEricM.Gese1,3 1DepartmentofWildlandResources,UtahStateUniversity,Logan,UT84322,USA 2PacificSouthwestResearchStation,ForestService,U.S.DepartmentofAgriculture,Fresno,CA93710,USA 3NationalWildlifeResearchCenter,WildlifeServices,U.S.DepartmentofAgriculture,Logan,UT84322,USA CorrespondenceshouldbeaddressedtoEricM.Gese,[email protected] Received23April2012;Accepted16May2012 AcademicEditors:A.ArslanandA.Ramirez-Bautista Copyright©2012C.M.ThompsonandE.M.Gese. This is an open access article distributed under the Creative Commons AttributionLicense,whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkis properlycited. Swiftfoxes(Vulpesvelox)areanendemicmesocarnivoreofNorthAmericasubjecttoresourceandpredation-basedpressures. Whileswiftfoxdemographicshavebeendocumented,thereislittleinformationontheimportanceoftop-downversusbottom-up pressuresortheeffectoflandscapeheterogeneity.Usingaconsumableresource-basedidealfreedistributionmodelasaconceptual framework, we isolated the effects of resource-based habitat selection on fox population ecology. We hypothesized if swiftfox ecologyispredominantlyresourcedependant,distribution,survival,andspaceusewouldmatchpredictionsmadeunderideal freedistributiontheory.Wemonitoredsurvivalandhomerangeuseof47swiftfoxesinsoutheasternColoradofrom2001to2004. Annualhomerangesizewas15.4km2,andseasonalhomerangesizewas10.1km2.Attheindividuallevel,annualhomerange sizewasunrelatedtosurvival.Estimatesoffoxdensityrangedfrom0.03to0.18foxes/km2.Seasonalsurvivalrateswere0.73and 1.0anddidnotdifferseasonally.Foxesconformedtothepredictionsoftheidealfreedistributionmodelduringwinter,indicating foxesarefoodstressedandtheirbehaviorgovernedbyresourceacquisition.Duringtherestoftheyear,behaviorwasnotresource drivenandwasgovernedbysecurityfromintraguildpredation. 1.Introduction thetenetsoftheidealfreedistribution,high-qualityhabitat should be occupied to a certain threshold at which point Swift foxes (Vulpes velox) are a mesocarnivore endemic to competition or social structure forces animals into poorer- theGreatPlainsregionofNorthAmerica.Historically,swift qualityareas.Inthiscase,survivalandreproductivesuccess foxesoccupiedtheextensiveshortgrassprairieregionsfrom wouldbesimilaracrosslandscapesthoughdensityandhome central Canada into New Mexico and Texas and from the range size would vary. However, most applications of the RockyMountainseastintoIowa[1,2].Today,theyarefound ideal free distribution to predator-prey interactions have in a variety of landscapes, ranging from shrub steppe to assumed predation risk is constant between habitats [8, 9]. agriculturalto ranchland tonative shortgrassprairie [3,4]. Predationriskisrarelyconstantacrosshabitatsandpatches Whiletheyarecapableofexploitingawiderangeofhabitats, [10],andextensiveworkhasindicatedthatunderthethreat reportedpopulationparameterssuchasdensityandsurvival of predation, prey species preferentially forage in “safer” vary widely indicating variation in habitat quality. There habitats [10–13]. The inclusion of variable predation risk is little information on landscape structure or disturbance into ideal free distribution models has indicated that prey thresholds that control swift fox abundance or lead to densityismorestronglyinfluencedbythe“riskiness”ofthe exclusion[5]. habitatthanbyresourceavailability[8]. Theidealfreedistribution[6]hasbeenusedtostudythe Swift fox population parameters in different landscapes influenceofpredationonpreydistribution[7].Accordingto have been well documented [14]; however, there is little Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 3. DATES COVERED 2012 2. REPORT TYPE 00-00-2012 to 00-00-2012 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER Swift Foxes and Ideal Free Distribution: Relative Influence of Vegetation 5b. GRANT NUMBER and Rodent Prey Base on Swift Fox Survival, Density, and Home Range Size 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Utah State University,1Department of Wildland REPORT NUMBER Resources,Logan,UT,84322 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES International Scholarly Research Network , ISRN Zoology Volume 2012, Article ID 197356, 8 pages,Government or Federal Purpose Rights License 14. ABSTRACT Swift foxes (Vulpes velox) are an endemic mesocarnivore of North America subject to resource and predation-based pressures. While swift fox demographics have been documented, there is little information on the importance of top-down versus bottom-up pressures or the effect of landscape heterogeneity. Using a consumable resource-based ideal free distributionmodel as a conceptual framework, we isolated the effects of resource-based habitat selection on fox population ecology. We hypothesized if swift fox ecology is predominantly resource dependant, distribution, survival, and space use would match predictions made under ideal free distribution theory.We monitored survival and home range use of 47 swift foxes in southeastern Colorado from 2001 to 2004. Annual home range size was 15.4 km2, and seasonal home range size was 10.1 km2. At the individual level, annual home range size was unrelated to survival. Estimates of fox density ranged from 0.03 to 0.18 foxes/km2. Seasonal survival rates were 0.73 and 1.0 and did not differ seasonally. Foxes conformed to the predictions of the ideal free distribution model during winter, indicating foxes are food stressed and their behavior governed by resource acquisition. During the rest of the year, behavior was not resource driven and was governed by security from intraguild predation. 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 9 unclassified unclassified unclassified Report (SAR) Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 2 ISRNZoology information available addressing the differences between saltbrush (Atriplex canescens) and greasewood (Sarcobatus reported values. In order to begin addressing the effects vermiculatus),aswellaspricklypearcactus(Opuntiaphaea- of landscape structure on swift fox distribution and demo- cantha), tree cholla (Cylindropuntia imbricata), and yucca graphics,weexploredtheeffectsoflandscapeheterogeneity (Yucca glauca). The remaining landscape was dominated on population parameters likely to reflect habitat quality, by pinyon-juniper woodland (Pinus edulis and Juniperus such as population density, home range size, and survival. monosperma). Elevation varied between 1,310 and 1,740m, For this paper, we assessed the relative quality of different average temperatures ranged from 1◦C in January to 23◦C landscapesasfox habitatbasedon vegetation structureand in July, and precipitation averaged 30cm but can fluctuate preybaseusingtheidealfreedistributiontheoryasaconcep- widely [16]. Monthly precipitation was the highest in July tual framework. While the ideal free distribution is limited with an average of 4.3cm of rain, though the 35% of the duetoitsresource-basedapproach,itprovidesaconceptual annualprecipitationthatfellduringthecool-season(March– starting point for investigating the relative importance of May) had a proportionally greater impact on productivity resourceavailabilityonanativemesocarnivore. [17]. Under the ideal free distribution theory we made the The maneuver site was purchased by the US Army in following series of predictions. (1) At the regional level, 1982 to facilitate large-scale mechanized infantry training estimates of swift fox density will vary across landscapes activities. Prior to this, both the maneuver site and the while survival rates will not. (2) At the individual level, surrounding area had been grazed by domestic livestock swiftfoxsurvivalrateswillbeunrelatedtohomerangesize. since the late 1800s [18]. Livestock were excluded in 1983 (3) Swift fox home range size will be negatively related to and training began in 1985 [17]. On the maneuver site, rodentpreybase.(4)Swiftfoxhomerangesizewillberelated training intensity varied widely simply due to the logistics to vegetative structural elements that enhance rodent prey of maneuvering large, mechanized infantry units. As a availability. The strength of these relationships or variation result, remnant areas subject to very little disturbance were from these predictions can indicate the relative importance scattered throughout the base. At the same time, areas ofresource-basedinfluencesonswiftfoxecology.Toaddress surrounding the base continued to be grazed by domestic these predictions, we evaluated swift fox spatial ecology at livestock. This natural heterogeneity provided the basis 4spatial-temporalscalesinsoutheasternColoradobetween for our identification of areas with variable vegetation 2001 and 2004. Scales included combinations of individual structure. versus population parameters and seasonal versus annual responses.Bycomparingfielddatatothepredictionsmade 2.2.StudyDesign. Weestablished6studysitesinareassub- under the ideal free distribution model, we were able to jected to 3 land use regimes: livestock grazing, mechanized investigate the relative influence of top-down (predation) militarytraining,andunused.Unusedsiteswereperipheral versusbottom-up(resource/vegetation)factorsonswiftfox portionsofthemilitarybasethatwerenotusedformilitary distributionandrangeuse.Thestudywasconductedonand training yet were subject to fire suppression and livestock around the US Army Pin˜on Canyon Maneuver Site where exclusion. Sites were named according to local landmarks recent, dramatic shifts in the disturbance regime due to or historical owners: Private (PRV), Biernacki’s (BTS), militarypurchase,aswellasdiscreteboundaries,resultedina Pronghorn (PRN), Red Rocks (RRK), Bent (BNT), and locallyheterogeneousenvironment.Thespatialscaleofour Comanche(COM).Datacollectionwasbasedon3seasons: research was small enough to assume that prior to military winter(breeding/gestation:15December–14April),summer ownership, the landscape had been fairly homogonous and (pup-rearing: 15 April–14 August), and fall (dispersal: 15 swift fox population parameters were spatially consistent. August–14December)[18–20].Between2001and2004,we Theabruptshiftinownership,thediscreteboundaries,and conducted fox trapping surveys, radio telemetry of collared thepatternsoflanduseintheareacoalescedintoanatural foxes, small mammal trapping, and vegetation surveys on experimentontheeffectsoflandscapestructureonanative eachsiteduringeachseasoneveryyear. mesocarnivore. 2.3. Swift Fox Capture and Radiotelemetry. We captured 2.MaterialsandMethods swift foxes using double-door box traps (Tomahawk Live Trap Company, Tomahawk WI) baited with chicken [20, 2.1. Study Site. We conducted research on and around the 21]. Traps were placed 500m apart along a 10km transect 1,040km2 Pin˜on Canyon Maneuver Site (PCMS) located bisecting each site resulting in 21 traps per site. Each trap in Las Animas County, Colorado (Figure1). North of the was oriented and covered with brush to provide protection PCMS the study area extended into the US Forest Service, fromexposure.Trapsweresetinthelateafternoon,checked ComancheNationalGrassland.Thestudyareaalsoextended earlythefollowingmorning,andleftclosedthroughoutthe southwardontoprivateranchlands.Theregionwasclassified day. Each transect was trapped for 4 consecutive nights 3 assemiaridgrasslandsteppe,withapproximately60%ofthe times per year. For recollaring or targeting animals, a trap- PCMS categorized as shortgrass prairie dominated by blue enclosuresystemwasusedatdensites[22].Capturedfoxes grama (Bouteloua gracilis), western wheatgrass (Agropyron were handled without anesthesia and weighed, sexed, and smithii), and galleta (Hilaria jamesii) [15]. Shrublands aged through tooth wear (adult, juvenile). Foxes were con- interspersedthroughoutthestudyareaincludedfour-winged sideredjuvenileuntilthepup-rearingseasonfollowingtheir ISRNZoology 3 Comanche National Grassland 6 Otero Co. 5 Las Animas Co. 4 wy 350 3 H U.S. Pin˜on Canyon Maneuver Site 2 1 (km) N 5 0 5 10 15 20 25 (1) Private (grazed) (4) Red rocks (unused) (2) Biernacki’s (military activity) (5) Bent (unused) (3) Pronghorn (military activity) (6) Comanche (grazed) Figure1:SixstudysitesonandaroundthePin˜onCanyonManeuverSiteinsoutheasternColorado.Locationsof6transectsareindicated, aswellastheassociateddominantlanduseforthatsite. birth(15April).Foxeswereear-taggedandcollaredwith30– unsuitable and all home ranges contain elements of each. 50gradiotransmitters(AdvancedTelemetrySystems,Isanti, Therefore, kernel range estimators are more appropriate MN). than minimum convex polygon methods due to the fact Welocatedfoxesaminimumof3timesperweek,twice that there are areas within ranges that are avoided by duringnighttimehourswhenanimalswereactivelyhunting, residents. andonceduringdaylighthourstolocatedensites.Locations Site-level survival rates, both seasonal and annual, were were considered independent when separated by at least 4 estimated using the known fate model in Program MARK hours[23],morethansufficienttimeforaswiftfoxtocross [27]. Due to our emphasis on established breeding territo- its home range. Nocturnal locations were estimated using ries, only adult animals were used. Individuals not located triangulationof2-3bearingswithin5minutesandseparated during a season were censored for that season. Individual by at least 40◦. Triangulation was done using Program fox survival was calculated as the proportion of either LocateII(Pacer,Truro,NovaScotia).Diurnallocationswere season or year that the fox survived [20]. Proportions were collected visually by approaching the animal until either a weighted by the number of days they could be monitored dencouldbeidentifiedortheanimalwasseen. after capture. Periods when the animal was not found were censored due to our inability to determine whether the animal was present and missed or had temporarily left the 2.4.SwiftFoxPopulationParameters. Swiftfoxhomeranges area. were calculated at 4 spatiotemporal scales; annual and sea- Weestimatedthenumberoffoxesineachsiteusingthe sonalrangesatbothsiteandindividuallevels.Allcalculated robustmodelinProgramMARK.Seasonalsurvivalestimates ranges were based on a minimum of 30 locations. Previous weretakenfromatelemetry-basedknownfatemodeldueto area-observation analyses [18] indicated 31 locations were thegreaterprecision.Dispersalrateswereestimatedbasedon the minimum needed to estimate an individual, unbiased the number of radioed animals that abandoned an existing swift fox home range. All ranges used in analyses were home range. Juvenile animals were considered dispersers if based on the 95% fixed kernel isopleth [24] using a least- they left the vicinity of their natal territory, whether or not squarescross-validationsmoothingparameter.Least-squares theysurvivedtoestablisharange.Populationestimateswere cross-validation allows the smoothing parameter (h) to convertedintodensityestimatesbycalculatingthe“effective be chosen that minimizes the squared distance between trapping area” associated with each transect [18, 28]. The the fitted surface and the target surface by varying the radius of the average seasonal 95% kernel home range for bandwidth [25]. Core use areas (50% isopleth) were also each site was used to buffer the transect in ArcView GIS calculatedforcomparisonbetweensites.Allrangeswerecal- [18,28].Theresultingpolygonwasconsideredthe“effective culatedusingtheAnimalMovementsextensionofArcView trapping area” for that site [18, 28]. Density estimates were GIS 3.3 [26]. In this environment there are no “habitat consolidated into seasonal averages as well as an overall patches”; quality is a gradient blending from suitable into estimateforeachsite. 4 ISRNZoology 2.5. Landscape Variables. For each of the 6 study sites, we wasusedtoidentifywhichlandscapevariablesbestpredicted evaluatedvegetationstructurebasedon50×70msampling foxhomerangesize(SASv.9.2).Independentvariableswith grids randomly located <1km of the trapping transect. P<0.1wereincludedintheregressionmodelsandtolerance Four grids were sampled each season, and new grids were valuesweresetat0.1tocontrolformulticollinearity[33,34]. selectedeachsubsequentseason.Eachgridconsistedofseven 50m line transects oriented north-south and spaced 10m 3.Results apart. On each line transect, vegetation type and height were evaluated by dropping a measuring pin every 1m A total of 47 (24M:23F) individual animals survived long and recording the type and height of the tallest vegetation enough and met the minimum number of locations for encountered [29]. For each grid, point measurements were calculating a seasonal home range at least once (Table1). combinedintoestimatesofpercentbasalcover,percentbare Forthese47animals,wecalculated69temporallyindepen- ground,percentlitter(deadmaterial),andmeanshruband dent annual (34 male and 35 female) and 136 temporally grass height. Shrub density was calculated by counting all independentseasonal(68maleand68female)homeranges. woody plants >20cm high within the grid. Grid estimates Overall the average annual home range size of swift foxes were combined into seasonal and annual averages for each was 15.4 ± 9.2 (SD) km2, and the average seasonal home study site. Standard deviations of grid estimates for each range size was 10.1 ± 6.7km2. The average core use area study site were used to represent the homogeneity of (50%kernelisopleth)forswiftfoxeswas1.2±1.1km2.The vegetationcharacteristicsacrosseachstudysite. averageseasonalhomerangeandcoreuseareaformalefoxes Following vegetation sampling, 35 Sherman live traps were12.0±6.9km2 and1.5±1.3km2,respectively.Female were placed throughout the grid with 10m spacing to homerangesandcoreuseareasweresmaller,averaging8.1± evaluate swift fox prey base. Analysis of 659 swift fox scats 5.8km2 and 0.9±0.8km2, respectively. Estimated seasonal collectedonthePCMSbetween1997and1998foundsmall homerangesinthebreedingseasonwere14.2±8.2km2 for mammalsconstitutedthebulkofthefoxes’dietthroughout male foxes and 7.3±4.0km2 for female foxes. During the theyear[19].Insectswereanephemeralfoodresourceonly pup-rearingseason,averagehomerangeswere9.5±5.6km2 during the summer, but rodents were utilized year-round for male foxes and 6.2±3.6km2 for female foxes. During andthereforeweconcentratedpreysamplingontherodent the dispersal season, ranges averaged 12.7 ± 86.3km2 and prey base. Traps were baited with equine sweet feed (corn, 11.7 ± 8.0km2 for males and females, respectively. Home oats, and molasses). Trapping grids were run for 4 nights, rangesizesandsamplesizesvariedamongthesixsites(Table checkedandclosedeachmorning,andreseteachafternoon. 1). CapturedrodentsweremarkedwithSharpiepensonthetail and abdomen allowing for identification of recaptures over 3.1. Prediction 1: At the Regional Level, Swift Fox Densities the 4-day trapping period. Species richness was calculated Will Vary Across Landscapes While Survival Will Not. Swift as the number of species captures. Diversity was estimated fox population density estimates (Table2) differed between using the Shannon-Weaver index [30]. Prey biomass was sites(F =7.48,P<0.001).Estimatesrangedfrom0.03±0.05 estimated using the species average weight [31] multiplied foxes/km2 on a grazed site (PRV) to 0.18±0.10 foxes/km2 bythenumberofindividualanimalscaptured. on an unused site (BNT). Seasonal survival rates ranged ThePearsoncorrelationcoefficientswereusedtoreduce between 0.73 on a military site to 1.0 on a grazed site and the number of variables prior to analysis. Final vegetation ratesdidnotdiffersignificantlybetweenseasons(F = 0.56, variables selected were mean grass height, mean shrub P =0.58).TheBentsitewasexcludedfromthisanalysisdue height,percentbasalcover,shrubdensity,standarddeviation tosmallsamplesizeandpoorsurvivalestimates. ofbasalcover,standarddeviationshrubheight,andstandard deviation shrub density. Standard deviation variables were included to represent landscape heterogeneity. Final prey 3.2. Prediction 2: At the Individual Level, Swift Fox Survival base variables selected included capture rates of the three WillBeUnrelatedtoHomeRangeSize. Attheindividuallevel, most common species (Ord’s kangaroo rat (Dipodomys annualhomerangesizewasunrelatedtosurvival(R2=0.03, ordii), northern grasshopper mouse (Onychomys leuco- F = 2.68, P = 0.11). Individual-level results do reflect gaster), and deer mouse (Peromyscus maniculatus)), total somepseudoreplicationasmorethanoneannualrangewas captures of all species, mean per capita biomass, species includedfor15of47animals. richness, and species diversity. The standard deviation of richness and diversity was included as predictor variable 3.3.Prediction3:SwiftFoxHomeRangeSizeWillBeNegatively to reflect community heterogeneity. Per capita biomass Related to Rodent Prey Base. Between 2001 and 2004, 185 was included because as grassland converts to shrubland, small mammal trapping grids were sampled for a total a community shift may sometimes occur where relatively of 740 grid nights. Individual species capture rates were rare, large bodied rodents are replaced by smaller-bodied uncorrelatedwitheachother,buttheywerecorrelatedwith specieswithfewerantipredatordefenses[32].Allvegetation rodent community indices such as diversity or richness. and prey base variables were log-transformed to improve The Pearson correlation coefficients for these comparisons normality. rangedfrom0.07to0.77(Table3).Thesecorrelationsstem We used ANOVA to determine whether density or fromthefactthatthreespecies(Ord’skangaroorat,North- survival varied between sites. Stepwise multiple regression erngrasshoppermouse,anddeermouse)accountedforover ISRNZoology 5 Table1:Seasonalhomerangesandcoreuseareasformaleandfemaleswiftfoxesin6studysiteslocatedonandaroundthePin˜onCanyon ManeuverSite,Colorado,2001–2004.Parenthesesindicatesamplesizesoffoxesthatwereidenticalbetweenhomerangeandcoreuse.Total rangesarepresentedas±SE. Homerange(95%) Coreuse(50%) Site Sex Dispersal Breeding Pup-rearing Dispersal Breeding Pup-rearing M 10.46(9) 11.03(2) 6.86(5) 1.56 1.12 0.86 PRV F 7.55(6) 7.32(5) 4.89(11) 0.75 0.64 0.72 M 16.28(5) 16.69(10) 14.04(7) 1.99 2.00 2.11 BTS F 12.34(4) 9.03(7) 10.39(6) 1.39 1.16 0.92 M 15.94(5) 14.71(3) 11.09(5) 1.58 1.54 1.29 PRN F 17.33(6) 6.65(4) 6.33(6) 1.88 0.87 0.60 M 13.14(1) 8.98(1) 6.35(3) 0.78 0.56 0.75 RRK F 18.03(1) 6.69(1) 6.32(3) 1.82 0.62 0.68 M — 31.79(1)1 — — 4.131 — BNT F — — — — — — M 7.65(3) 6.27(4) 4.93(4) 1.03 1.19 0.60 COM F 5.64(3) 2.82(2) 2.27(3) 0.49 0.40 0.24 M 12.63±2.0 12.23±1.6 8.36±1.0 1.47±0.3 1.53±0.3 1.04±0.2 Total F 11.40±1.8 7.54±1.5 6.48±0.9 1.19±0.2 1.01±0.4 0.66±0.1 1Throughoutthestudy,onlyoneanimalremainedand/orsurvivedontheBentsitethroughoutafullseason.Thisanimalwasbelievedatransient,thoughit mettherequirementsforcalculatingaseasonalhomerange. Table2:Estimatesofpopulationdensity(±SE)andsurvivalrates(±SE)foradultswiftfoxeson6sitesinsoutheasternColorado,2001–2004. Seasonalsurvivalrates Site Density Annualsurvival Breeding Puprearing Dispersal PRV 0.18(0.10) 0.54 0.81(0.09) 0.83(0.08) 0.81(0.09) COM 0.04(0.05) 0.92 0.92(0.08) 1.0(0.0) 1.0(0.0) BTS 0.11(0.08) 0.54 0.84(0.08) 0.78(0.10) 0.82(0.08) PRN 0.09(0.06) 0.54 0.73(0.11) 0.94(0.06) 0.79(0.09) RRK 0.05(0.03) 0.50 0.80(0.18) 0.83(0.15) 0.75(0.22) BNT1 0.03(0.05) — — — — 1Throughoutthestudy,onlyoneanimalremainedand/orsurvivedontheBentsitethroughoutafullseason.Asaresultwewereunabletoestimatesurvival ratesforthatsite. 75% of all captures and therefore dominated community 3.4.Prediction4:SwiftFoxHomeRangeSizeWillBeRelated indices.Despitethesecorrelations,allvariableswereusedin to Vegetative Structural Elements That Enhance Rodent Prey the analysis due to biological importance and interpretive Availability. Vegetationstructuralmeasurementsweretaken capacity. As stated above, regression tolerance values were on the same 185 grids described above. A number of the adjustedtoreducetheeffectsofthesecorrelations. measurementswerecorrelated,withthePearsoncorrelation Acrossallspatial-temporalscales,capturerateofNorth- values ranging from 0.06 to 0.90 (Table3). In particular, ern grasshopper mice was the variable most consistently percent basal cover and the standard deviation of shrub associated with swift fox home range size (Table4). Deer height were highly correlated (P < 0.05) with most other mousecaptureratewasthesecondmostcommonvariable, variables. As described above, all variables were included emerging once as the primary and twice as a secondary in the analysis due to biological importance and regression explanatoryvariable.Whileregressionmodelsweregenerally tolerancevalueswereadjustedaccordingly. statisticallysignificant,theyhadlowexplanatorypowerwith Onlyonevegetativevariable(shrubdensity)wasconsis- R2<0.20.Oneexceptionwastherelationshipbetweenmean tently related to home range size across all spatiotemporal population-levelbreedingseasonhomerangesandNorthern scales(Table4).Ingeneralthisrelationshipexplainedalow grasshoppermicecaptures(R2=0.467,P=0.02).Inclusion proportion of the variance in home range size with the ofcommunityrichnessraisedtheR2to0.65butraisedtheP exception of the population level, breeding season analysis valueto0.08.Coreuseareas,bothannualandseasonal,were (R2 = 0.39, P = 0.04). Core use areas, both annual and unrelated to prey base variables; controlling for sex had no seasonal,werenotsignificantlyrelatedtolandscapevariables. effectontheanalyses. Controllingforsexhadnoeffectontheanalyses. 6 ISRNZoology Table3:ThePearsoncorrelationcoefficientsforswiftfoxpreybaseandvegetationstructurevariableson185samplinggridsinsoutheastern Colorado, 2001–2004.Capture rates indicate the total number of individuals captured for selected species.Total captures, biomass, and communityindicesincludeallspecies.Valuesexceeding±0.34aresignificantattheP=0.05level. Ord’s N.grasshopper Percapita SD SD Preybasevariables Deermouse Totalcaptures Richness Diversity kangaroorat mouse biomass richness diversity Ord’skangaroorat — 0.26 0.46 0.57 0.53 0.55 0.62 0.55 0.73 N.grasshopper — 0.18 0.50 0.37 0.50 0.30 0.48 −0.18 mouse Deermouse — 0.46 0.32 0.44 0.50 0.53 0.41 Totalcaptures — 0.59 0.99 0.71 0.72 0.48 Percapitabiomass — 0.58 0.41 0.46 0.32 Richness — 0.83 0.73 0.56 Diversity — 0.72 0.82 SDrichness — 0.67 SDdiversity — Vegetationstructure Meangrass Meanshrub SDbasal SDshrub %basalarea Shrubs/100m2 SDshrubs/100m2 variables height height area height %basalcover — 0.43 0.46 0.30 0.11 0.41 0.23 Meangrassheight — 0.76 0.39 0.31 0.20 0.20 Meanshrubheight — 0.65 0.26 0.88 0.52 Shrubs/100m2 — 0.06 0.39 0.90 SDbasalarea — 0.26 0.15 SDshrubheight — 0.35 SDshrubs/100m2 — Table4:Resultsofstepwiseregressionanalysisof(a)vegetationstructurevariablesand(b)preybasevariablesagainsthomerangesizefor swiftfoxesinsoutheasternColorado,2001–2004.Italicsindicateanegativerelationship. (a) Partial Spatialscale Temporalscale Primaryvariable Secondaryvariable PartialR2 ModelR2 ModelP R2 Annual Shrubdensity 0.044 None 0.044 0.085 Dispersal SD(%basalarea) 0.121 None 0.121 0.040 Individual Pup-rearing SD(meanshrubht) 0.142 None 0.142 0.028 Breeding SD(meanshrubht) 0.284 None 0.284 0.002 Annual Shrubdensity 0.172 Meangrassht 0.129 0.301 0.025 Dispersal (novariablesenteredthemodel) Population Pup-rearing (novariablesenteredthemodel) Breeding Shrubdensity 0.393 None 0.393 0.039 (b) Spatial Primary Partial Temporalscale Secondaryvariable PartialR2 ModelR2 ModelP scale variable R2 Annual N.grasshoppermousecaptures 0.086 Deermousecaptures 0.052 0.137 0.056 Dispersal N.grasshoppermousecaptures 0.143 Deermousecaptures 0.138 0.281 0.025 Individual Pup-rearing Meanpercapitabiomass 0.107 None 0.107 0.068 Breeding deermousecaptures 0.181 None 0.181 0.043 Annual N.grasshoppermousecapture 0.105 Totalcaptures 0.264 0.369 0.001 Dispersal (novariablesenteredthemodel) Population Pup-rearing (novariablesenteredthemodel) Breeding N.grasshoppermousecaptures 0.467 Communityrichness 0.179 0.646 0.078 ISRNZoology 7 4.Discussion [38]. Other studies have indicated many of the primary prey items of swift foxes are found in areas of increased Currently,thereareanumberofgapsinourunderstanding landscapeheterogeneity[30,39].Therefore,whilelowshrub of swift fox ecology, including the influence of landscape densityareasmaybehigh-quality,“safe”habitatduringmost variation on local density and demographics. Fundamental of the year, swift foxes in these areas may need to expand to this understanding is the need to quantify the influence their ranges into riskier habitat during winter months to of resource limitation versus predation pressure to better meet resource requirements despite the increased risk of understandtheprocessesinfluencingswiftfoxdensity.While predation. At the same time, areas of moderate to high theymaybecapableofexploitingawiderangeoflandscapes, shrub density may provide for energetic requirements year swiftfoxesappeartobesensitivetothresholdsinvegetation round, but the increased risk of predation makes them structure,preyavailability,orpredatordensitybeyondwhich less desirable during spring, summer, and fall months. Our they are excluded. While the ideal free distribution model resultsindicatethatthehighest-qualityswiftfoxhabitatsare is limited in its applicability to mesocarnivores, it is a heterogeneouslandscapeswhereresidentanimalscanspend convenient point at which to start investigating the relative themajorityoftheyearinareasoflowriskbuthaveaccess importanceofbottom-upforces. toriskier,resource-richpatchesduringwintermonths.Over Our results indicated swift foxes conform to 2 of the 4 the past century, reductions in the complexity of grassland predictionswemadeundertheidealfreedistributiontheory disturbance regimes due to fire suppression and livestock year-round but conform to the remaining 2 only during husbandry have led to a homogenization of the landscape, the winter breeding season. As the ideal free distribution reducing swift foxes’ ability to effectively balance resource is a resource-based model, it follows that swift foxes are needsandpredationrisk. only resource (food) limited during the winter months. Duringtheremainingtwo-thirdsoftheyearfoxhomerange Acknowledgments placementreflectedotherpressures,mostlikelyavoidanceof intraguildpredationbycoyotes[5,35].Thispatternappears Funding and logistical assistance was provided by the to reflect what has been termed an “adaptive compromise” US Army, Directorate of Environmental Compliance and [11]; during periods of food stress swift foxes may trade Management, Fort Carson, Colorado, through the US Fish securityforenhancedresourceavailability.Similarbehavior and Wildlife Service, Colorado Assistance Office, Denver, has been well documented in aquatic systems and with Colorado, and the US Geological Survey, Utah Cooperative invertebrates,birds,andmammals;see[10]forareview. Fish and Wildlife Research Unit at Utah State University, Theideathatswiftfoxesmakethisadaptivecompromise Logan, Utah. Additional support was provided by the US during winter months is supported by two additional lines Department of Agriculture, Wildlife Services, National of evidence. First, several studies have reported that winter Wildlife Research Center at Utah State University, Logan, monthsarethemostresourcepoorforswiftfoxes[5,14,36]. Utah. The authors thank T. Warren, B. Rosenlund, M. In our study, 68% of small mammal captures occurred Klavetter, G. Belew, R. Bunn, and D. Sharps for logistical duringsummermonths(April15–August15).Twenty-three assistance and E. Joyce, J. White, E. Cleere, M. Watkins, percentofcapturesoccurredinwintermonthsand9%infall D. Degeranno, A. Larkins, C. Roemer, D. Fletcher, W. months.Thescarcityofmammalianpreyinthefallisoffset Ulrey, S. Schopman, C. Gazal, A. Knipps, J. Garner, and by increases in insect abundance; insects made up nearly C. Briggs for field assistance. They thank M. Conner, M. 80%ofthedietofswiftfoxesduringfallmonths[19].While Ernst, J. MacMahon, and F. Provenza for reviews of the mammalian prey becomes important again during winter paper.ResearchprotocolswereapprovedbytheInstitutional months[19],asignificantportionofthiscouldbescavenged Animal Care and Use Committees at the National Wildlife carrion [37]. These results support the idea that during ResearchCenterandUtahStateUniversity. wintermonthsswiftfoxesmayaltertheirbehaviorinsearch ofadditionalfood.Second,bothourresultsandaprevious References study[18]suggestthatthepredationrateofadultswiftfoxes by coyotes on the PCMS is the highest in winter months. [1] J. M. Scott-Brown, S. Herrero, and J. Reynolds, “Swift fox,” Whiletheresultswerenotstatisticallysignificant,thepattern in Wild Furbearer Management and Conservation in North isconsistentover5years.Thissupportstheideathatduring America,M.Novak,J.A.Baker,M.E.Obbard,andB.Malloch, wintermonthsfoxesaremoreexposedtopredationrisk. Eds., pp. 433–441, Ontario Trappers Association North Bay, Swift foxes have the potential to persist in a range Ontario,Canada,1987. of environments, from agricultural to prairie remnants to [2] FaunaWest,AnEcologicalandTaxonomicReviewoftheSwift Fox (Vulpes velox) with Special Reference to Montana, Fau- ranchland [3, 4]. However, survival, density, and home naWestWildlifeConsultants,Boulder,Mont,USA,1991. rangeestimatesvarywidely[5].Interpretationofourresults [3] V.L.JacksonandJ.R.Choate,“Densanddensitesoftheswift suggeststhatswiftfoxbehaviorisacomplexbalanceofsea- fox,Vulpesvelox,”SouthwesternNaturalist,vol.45,no.2,pp. sonalfoodavailabilityandavoidanceofintraguildpredation. 212–220,2000. Earlierworkindicatedthatswiftfoxdensitywassignificantly [4] R.S.Matlack, P.S. Gipson,andD. W. Kaufman,“The swift negatively related to grass height [35], supporting the idea fox in rangeland and cropland in western Kansas: relative thatswiftfoxesavoidareasofdensevegetationwheretheyare abundance,mortality,andbodysize,”SouthwesternNaturalist, moresusceptibletocoyotepredationduetovisualobscurity vol.45,no.2,pp.221–225,2000. 8 ISRNZoology [5] A.Moehrenschlager,B.L.Cypher,K.Ralls,R.List,andM.A. box-trap enclosure: kit foxes as a test case,” Wildlife Society Sovada,“Swiftandkitfoxes,”inBiologyandConservationof Bulletin,vol.31,no.3,pp.630–633,2003. WildCanids,D.W.MacdonaldandC.Sillero-Zubiri,Eds.,pp. [23] R.K.SwihartandN.A.Slade,“Influenceofsamplinginterval 185–198,OxfordUniversity,Oxford,UK,2004. onestimatesofhome-rangesize.,”JournalofWildlifeManage- [6] S. D. Fretwell and H. L. Lucas, “On territorial behavior and ment,vol.49,no.4,pp.1019–1025,1985. other factors influencing habitat distribution in birds—I. [24] B.J.Worton,“Kernelmethodsforestimatingtheutilization Theoretical development,” Acta Biotheoretica, vol. 19, no. 1, distributioninhome-rangestudies,”Ecology,vol.70,no.1, pp.16–36,1969. pp.164–168,1989. [7] M. R.Heithaus, “Habitat selection by predators and prey in [25] G. Hemson, P. Johnson, A. South, R. Kenward, R. Ripley, communitieswithasymmetricalintraguildpredation,”Oikos, and D. Mcdonald, “Are kernels the mustard? Data from vol.92,no.3,pp.542–554,2001. global positioning system (GPS) collars suggests problems [8] D.M.HugieandL.M.Dill,“Fishandgame:agametheoretic for kernel home-range analyses with least-squares cross- approachtohabitatselectionbypredatorsandprey,”Journal validation,”JournalofAnimalEcology,vol.74,no.3,pp.455– ofFishBiology,vol.45,supplementA,pp.151–169,1994. 463,2005. [9] B. Luttbeg and A. Sih, “Predator and prey habitat selection [26] P.N.HoogeandB.Eichenlaub,AnimalMovementExtension games:theeffectsofhowpreybalanceforagingandpredation toArcView,Version1.1,AlaskaBiologicalScienceCenter,U.S. risk,”IsraelJournalofZoology,vol.50,no.2-3,pp.233–254, GeologicalSurvey,Anchorage,Alaska,USA,1997. 2004. [27] G. C. White and K. P. Burnham, “Program MARK: survival [10] S.L.LimaandL.M.Dill,“Behavioraldecisionsmadeunder estimationfrompopulationsofmarkedanimals,”BirdStudy, the risk of predation: a review and prospectus,” Canadian vol.46,supplement,pp.S120–S139,1999. JournalofZoology,vol.68,no.4,pp.619–640,1990. [28] E.R.Schauster,E.M.Gese,andA.M.Kitchen,“Anevaluation [11] A.Sih,“Optimalbehavior:canforagersbalancetwoconflict- of survey methods for monitoring swift fox abundance,” ing demands?” Science, vol. 210, no. 4473, pp. 1041–1043, WildlifeSocietyBulletin,vol.30,no.2,pp.464–477,2002. 1980. [29] M. R. Dale, Spatial Pattern Analysis in Plant Ecology, Cam- [12] S. E. Hampton, “Habitat overlap of enemies: temporal pat- bridgeUniversity,Cambridge,UK,1999. ternsandtheroleofspatialcomplexity,”Oecologia,vol.138, [30] P. J. Morin, Community Ecology, Blackwell Science, Malden, no.3,pp.475–484,2004. Mass,USA,1999. [13] B. P. Kotler, J. S. Brown, A. Bouskila, S. Mukherjee, and [31] J.O.Whitaker,NationalAudubonSocietyFieldGuidetoNorth T. Goldberg, “Foraging games between gerbils and their American Mammals, Alfred A. Knopf, New York, NY, USA, predators: seasonal changes in schedules of activity and 1994. apprehension,”IsraelJournalofZoology,vol.50,no.2-3,pp. [32] E. P. White, S. K. Ernest, and K. M. Thibault, “Trade-offs 255–271,2004. in community properties through time in a desert rodent [14] M.A.SovadaandL.Carbyn,EcologyandConservationofSwift community,”AmericanNaturalist,vol.164,no.5,pp.670–676, FoxesinaChangingWorld,CanadianPlainsResearchCenter, 2004. UniversityofRegina,Saskatchewan,Canada,2003. [33] B.G.TabachnickandL.S.Fidell,UsingMultivariateStatistics, [15] R.B.Shaw,S.L.Anderson,K.A.Schulz,andV.E.Diersing, HarperCollinsCollege,NewYork,NY,USA,3rdedition,1996. PlantCommunities,EcologicalChecklist,andSpeciesListforthe [34] K.R.Crooks,“Relativesensitivitiesofmammaliancarnivores U.S.ArmyPinonCanyonManeuverSite,Colorado,Colorado tohabitatfragmentation,”ConservationBiology,vol.16,no.2, State University Science Series 37, Fort Collins, Colo, USA, pp.488–502,2002. 1989. [35] C.M.ThompsonandE.M.Gese,“Foodwebsandintraguild [16] R. B. Shaw and V. E. Diersing, “Tracked vehicle impacts on predation: community interactions of a native mesocarni- vegetation at the Pin˜on Canyon Maneuver Site, Colorado,” vore,”Ecology,vol.88,no.2,pp.334–346,2007. JournalofEnvironmentalQuality,vol.19,no.2,pp.234–243, [36] T. L. Olson and F. G. Lindzey, “Swift fox (Vulpes velox) 1990. home-range dispersion patterns in southeastern Wyoming,” [17] D. G. Milchunas, K. A. Schulz, and R. B. Shaw, “Plant CanadianJournalofZoology,vol.80,no.11,pp.2024–2029, communityresponsestodisturbancebymechanizedmilitary 2002. maneuvers,”JournalofEnvironmentalQuality,vol.28,no.5, [37] D.M.ZumbaughandJ.R.Choate,“Winterfoodhabitsofthe pp.1533–1547,1999. swiftfoxonthecentralhighplains,”PrairieNaturalist,vol.17, [18] E.R.Schauster,E.M.Gese,andA.M.Kitchen,“Population pp.41–47,1985. ecologyofswiftfoxes(Vulpesvelox)insoutheasternColorado,” [38] J. F. Kamler, W. B. Ballard, E. B. Fish, P. R. Lemons, K. CanadianJournalofZoology,vol.80,no.2,pp.307–319,2002. Mote, and C. C. Perchellet, “Habitat use, home ranges, and [19] A. M. Kitchen, E. M. Gese, and E. R. Schauster, “Resource survivalofswiftfoxesinafragmentedlandscape:conservation partitioningbetweencoyotesandswiftfoxes:space,time,and implications,”JournalofMammalogy,vol.84,no.3,pp.989– diet,”CanadianJournalofZoology,vol.77,no.10,pp.1645– 995,2003. 1656,1999. [39] W.S.LonglandandM.V.Price,“Directobservationsofowls [20] S. M. Karki, E. M. Gese, and M. L. Klavetter, “Effects of andheteromyidrodents:canpredationriskexplainmicrohab- coyote population reduction on swift fox demographics in itatuse?”Ecology,vol.72,no.6,pp.2261–2273,1991. southeasternColorado,”JournalofWildlifeManagement,vol. 71,no.8,pp.2707–2718,2007. [21] D.F.Covell,Ecologyoftheswiftfox(Vulpesvelox)insoutheast- ernColorado[M.S.thesis],UniversityofWisconsin,Madison, Wis,USA,1992. [22] A. J. Kozlowski, T. J. Bennett, E. M. Gese, and W. M. Arjo,“Livecaptureofdenningmammalsusinganimproved