DENNINGECOLOGY OF FEMALEAMERICAN BLACK BEARS IN SOUTH CENTRALLOUISIANA

DWAYNEA. HIGHTOWER,'School of Forestry, Wildlife, and Fisheries, State University AgriculturalCenter, Baton Rouge, LA 70803, USA, email: [email protected] ROBERT 0. WAGNER,1School of Forestry, Wildlife, and Fisheries, Louisiana State University AgriculturalCenter, Baton Rouge, LA 70803, USA, email: [email protected] RICHARD M. PACE, 111,2U.S. Geological Survey, Biological Resources Division, Louisiana Cooperative Fish and Wildlife Research Unit, Louisiana State University, Baton Rouge, LA 70803, USA

Abstract: We examinedthe denningecology of female Americanblack bears(Ursus americanusluteolus) in the upper(Inland) and lower (Coastal) AtchafalayaRiver Basin of southcentral Louisiana. We attempted101 den visits of 35 female bearsduring February-March, 1992-2000. We found dens on 28 of 56 (50%) Coastal and 44 of 45 (98%) Inlandattempts. We determinedreproductive status on 62 (61 %) den visits, but our success differedbetween Coastaland Inlandareas (43% and 84%,respectively). The distributionof littersizes differedbetween areas. We did not detect any cub abandonment. Use distributionamong den types (e.g., tree, snag, groundnest, slash, stump, log, and excavated) differed between areas. We found no evidence that the distributionof den security classes differed between parturientand barrenfemales. Using the median location errorof field trialswith test transmitters(206 m) as our criterion,we classified bearsas stationaryduring the denningseason if the distancesof 3 consecutive locations (a location set) from their centroidwere each <206 m. Of 29 dennedbears, 24 had >1 stationarylocation set (classified as denned),and 16 of 117 bearseasons outside of the denningperiod contained >1 stationarylocation set. We classified 10 of 23 Coastalfemales, 6 of 19 Coastalmales, 2 of 4 Inlandfemales, and 1 of 1 Inlandmales of previously unknown status as denned. Based on movement data, many Coastal bears, including adult females, remainedactive or only bedded for shortperiods duringwinter.

Ursus 13:11-17 (2002)

Key words: Americanblack bear,denning, litter size, location constancy,Louisiana, reproduction, Ursus americanus luteolus

Denning by black bears is hypothesizedto be an adap- Denning is not, however, compulsory for all bears in tation to decrease stress during periods of inclement some mild climates. Excludingparturient females, which weather and food shortage (Lindzey and Meslow 1976, must den to give birth,bears may remain active if suffi- Johnson and Pelton 1980a), and is when parturitionoc- cient food is available(Graber 1990). Winter-activebears curs (Alt 1983). After enteringdens, bears cease all in- have been reportedfor manysoutheastern black bear popu- gressive and eliminative functions and enter a lethargic lations (Taylor1971, Hellgrenand Vaughan 1989, Wood- or deep sleep (Nelson et al. 1973). Disturbanceduring ing and Hardisky1992, Weaverand Pelton 1994). These the denningperiod may result in cub abandonmentor re- observationsindicate that bears in milder climates may duce the survivalprobability of adultbears by increasing have winter behavioraloptions not available to bears in energy expenditure when food resources are scarce colder climates, where food may be more limited. Popu- (Linnell et al. 2000). Because managersperceive denned lation-specific knowledge of bear behavior in winter is bears,especially parturientfemales, to be vulnerable,ad- criticalto understandthe seasonal needs of bears and for- equate denning habitatand structureshave been empha- mulateappropriate management strategies. sized as importantin managingblack bear habitat (Johnson To aid in conservationplanning for the threatenedLoui- and Pelton 1981, Weaverand Pelton 1994, Linnell et al. siana black bear (U. a. luteolus; U.S. Fish and Wildlife 2000). Service 1992), we investigatedthe denning habitat,den- Researchersin the southeasternU.S. have reportedblack ning frequency,and reproductiverates of bears in 2 iso- bears using a wide varietyof den structuresincluding trees, lated populations within the Basin, standingsnags, groundnests, logs, stumps,piles of woody Louisiana. Our objectives were to document den struc- debrissuch as logging slash, andexcavated areas beneath tures used, determineif parturientfemales used more se- or human trees,logs, debris, structures(Johnson and Pelton cure dens than other females, document reproductive Smith and 1981, 1985, Hellgren Vaughan1989, Weaver output,and classify wintermovement behavior (e.g., sta- and Pelton 1994). Of these, tree dens are consideredthe tionary or moving). Because of habitat differences be- most secure (Johnson and Pelton 1981). Hence, bears tween study areas (coastal swamps and mesic islands tree dens using are presumedto realize increasedsurvival versus inland bottomlandhardwood forests), known dif- for themselves and their offspring (White et al. 2001). ferencesin movementdynamics (Wagner 1995) andmor- tality (Pace et al. 2000) between and 1 populations, reported Presentaddress: Quantitative Ecological Services, Inc., 200 differencesin den use among Americanblack bear Willow Lake LA popu- Lane, DeRidder, 70634, USA lations in the southeasternU.S., we of 2 Presentaddress: Northeast Fisheries Science Woods comparedaspects Center, between our Hole, MA 02543, USA, email: [email protected] denning ecology study populations. 12 Ursus 13:2002

STUDYAREA deltaic plains (Evans et al. 1983). Most of the land area, We worked with 2 populations of bears located in 2 excluding the salt domes and areas protected by levees, areas of south , referred to as Inland was inundatedthroughout most of the year. The water (approx460 km2in size) and Coastal (approx570 km2in table was generally <0.2 m above the surface. Drier sites size; Fig. 1). The Inland population was located prima- in the swamp and scrub habitatsgenerally were those ar- rily within the MorganzaSpillway, a floodway developed eas surroundingisolated groupsof trees. Detailed habitat by the U.S. Army Corps of Engineersto release floodwa- descriptionswere provided by Nyland and Pace (1997). ters of the River into the AtchafalayaBasin. Forestedhabitats within the floodway were primarilybot- tomlandhardwood stands managed for timberproduction. METHODS Forest managementprescriptions, especially even-aged We capturedbears in leg snares or culvert trapsfollow- management,may have influenced den tree availability. ing the methods detailed in Johnson and Pelton (1980b). Agriculturalfields within the Spillway were adjacentto We trappedfrom April to Septemberduring 1992-99, with the levees and LouisianaHighway 10, which bisected the trappingsessions ranging from 3 to 6 months. Trapping Spillwaynear its northernend. Seasonalflooding occurred effort varied among years and between study areas, with throughoutthe Inlandarea but was not severe during the Coastal receiving more effort during several years. Cap- term of this study. Large areas of slightly higher eleva- turedbears were immobilized with eithera 2:1 mixtureof tion were dry throughoutwinter, and flooded areas were ketamine hydrochlorideand xylazine hydrochloride(4.4 generally <0.5 m deep. mg/kg) or a 1:1 mixture of tiletamine hydrochlorideand The Coastal population was in an area consisting of zolazepam(6.0 mg/kg). Bears were uniquelymarked with small bottomland hardwood stands, baldcypress-water ear tags, lip tattoo, and, beginning in 1997, with a passive tupelo(Taxodium distichum-Nyssa aquatica) swamps, and injectabletransponder (AVID Incorporated,Norco, Cali- coastal marshesthat varied along salinity gradientsrang- fornia, USA). A first premolarwas extractedfor cemen- ing from fresh to saline (U.S. Army Corps of Engineers tum annuli aging (Matson's Laboratories,Incorporated, 1974). Wetlandscrub habitatsoccurred in the transition Milltown, Montana,USA), and radiocollarswith leather zone between swamp and marsh. Bottomlandhardwood spacing materialwere attachedto bears >2 years old (Ad- stands were restrictedto the borders of sugarcanefields vanced TelemetrySystems, Incorporated,Isanti, Minne- protectedfrom flooding by levees, and along spoil areas sota, USA; Telemetry Systems, Incorporated,Mequon, nearwaterways. Coastalhabitats were highly interspersed Wisconsin, USA; Telonics, Incorporated,Mesa, Arizona, and most containedopen water, sloughs, bayous, and ca- USA). nals influencedby tidalfluctuations (Chabreck 1970). The We attemptedden visits for adult and subadultfemale Coastal area also contained 4 salt domes, characterized bears during Februaryand March from 1992 to 2000 to by upland cover types, rising >30 m above surrounding replaceradiocollars, document den location and structure, and determine reproductive status. Conditions permit- ting, bearswere immobilized in dens and the young sexed and weighed. We listened for nursing cubs at den sites we were not able to enter. To assess cub survival, se- lected litters after 1995 were radiotagged following the protocol of Higgins et al. (1997). Females were assigned to reproductiveclasses at den visits or from observations made after den emergence: solitary adult female, solitary subadultfemale, adult female with cub (born in current winter),subadult female with cub, adultfemale with year- ling, and subadultor adult female of unknown condition. We tested for differences between areasin the proportion of den visits resulting in den observationsusing a X2test and for differences between areas in female reproductive conditionusing Fisher'sexact test. In these analyses,each den attempt represents the total number of field excur- 50 0 50 100 150 Kilometers sions we made attemptingto assess the denning status of a female bear during 1 winter (bear-winter). Because sizes were small, data from adult and subadult 1. in the River Basin in south sample Fig. Study areas Atchafalaya females were across withinareas for all central Louisiana (1992-2000). pooled years analy- LOUISIANABEAR DENS * Hightoweret al. 13 ses of den data. We assumed multiple observations of sets outside the denningperiod. individualsacross years were independent. A Type I er- ror rate of 5% was used for all statisticaltests. We classified den types according to structure(e.g., RESULTS groundden excavatedbeneath standing structure, ground nest, hollow log, slash pile, snag, stump,tree) and tested Den Visits for differences in den types used between areas using We attemptedden visits for 77% of bear-wintersmoni- Fisher's exact test. We also subjectively grouped struc- tored (Table 1). We did not attemptto visit bears whose tures into classes based on relative security from distur- telemetrylocations in the weeks priorto the periodof den bance similar to McDonald and Fuller (1998). Tree and inspection indicatedmovement (19 adult and 1 subadult snag dens with elevated entranceswere consideredmost Coastal females, and 1 adult Inlandfemale), or on bears secure, groundnests least secure, and the remainderwere located on inaccessible privatelands (5 adult Coastal fe- classified as moderate. We groupedbears of known re- males and 1 adult Inlandfemale). We observed 72 dens productivecondition into parturientor barrenfemales and of 19 Coastal and 16 Inland female bears. We did not tested for differences in den security class within areas enter 17 dens because of safety concerns (i.e., anesthetic using Fisher's exact test. could not be safely administeredto bear, bear appeared Number,age (cub, yearling), and sex of offspring ob- preparedto flee den, or den tree unsuitablefor climbing). served at the den site were recorded. Because sample Overall, 56% of attemptedden visits were successful. sizes were small, we pooled littersizes acrossyears within A total of 15 dens were observed after the bear left the areas for analysis. We assumed litter sizes were not dif- den site. Coastal females were more likely to move on ferent among years and that multiple litters born to the our approachthan Inland females (0.70 vs. 0.11; X2=34.8, same female were independent. We used Fisher's exact 1 df, P < 0.01), and we were less likely to find Coastal test as evidence for differences in litter-size distribution dens once bears fled (n = 44, P = 0.04). We believe that (e.g., 1, 2, and 3 cubs) between areas. Coastal bears fled when researcherswere fartheraway Because we were sometimes uncertain if bears that than did Inland bears, resulting in only 28% of Coastal moved away during our approachwere denned and be- dens being observed if the bear fled versus 80% for In- cause some dens were inaccessible, we assessed denning land bears. Those factors resultedin den observationin remotely. We used aerial location constancy to classify only 50% of Coastal attemptscompared to 98% of In- the winter(November-April) movement of bears,includ- land. ing males, from 1992 to 1995 when den visits were un- The reproductivecondition of bears leaving observed successful or not attempted. Locations were obtained dens included 6 with cubs (all Coastal), 6 with yearlings approximatelyweekly over the periodconsidered. After (3 Coastal, 3 Inland), 1 solitary (Coastal), and 2 of un- 1995, bears were only located monthly,thus locations af- knownreproductive condition (1 Coastal,1 Inland). Four- ter 1995 were excluded from analyses. We characterized teen of 15 (93%) vacated dens were groundnests, and 1 aerialtracking precision as the mediandeviation (distance was in a stump. Using aerial or ground telemetry, we error)obtained when locating test transmitters,placed at verified the returnof females with cubs to the areaof the known locations, in 1994. Bears were considered sta- den following disturbanceand detected no permanentcub tionary during the denning season if the distances of 3 abandonment. Of 17 radiotagged cubs representing5 consecutive locations (a location set) from theircentroid Coastaland 4 Inlandlitters, all were determinedto be with were each less thanthe mediandistance error on test trans- the birthing female until den emergence. Several cub mitters (206 m; n = 70). We evaluatedthe techniqueby radiotagswere recoveredat the den site after the female applying it to location sets of denned bears and location emerged from the den, but we did not find any evidence Table1. Denvisits, den observations,and den season reproductivedeterminations of subadultand adult female black bears in the upper(Inland) and lower(Coastal) Atchafalaya River Basin (1992-2000).

Bear den-years Den inspections Bears observed Reproductivecondition Study area Age class monitored attempted in dens Dens observed determined

Coastal Adult 75 51 17 27 24 Subadult 6 5 0 1 0 Subtotal 81 56 17 28 24 Inland Adult 41 37+1V 33 37 33 Subadult 6 6+1" 7 7 5 Subtotal 47 43+2 40 44 38 Total 128 101 57 72 62 a One unmarkedbear was discovered in the den by a landowner;bear was subsequentlyinspected by projectpersonnel. 14 Ursus 13:2002

of mortality. All others were determinedto be with the when we first observedthem. In contrast,the majorityof female afterden emergence. bearsin otherden types seemed oblivious to ourpresence We determinedthe reproductivecondition of bears in and did not move until we approachedto about 3 m. 86%of dens observedin both areas. Because of the lower We inspectedden sites of 19 and 14 parturientfemales likelihood of observing Coastal dens, overall success in and 6 and 24 barrenfemales in the Coastal and Inland determiningreproductive condition when den inspection areas,respectively (Table 3). We detectedno differences was attemptedwas lower in the Coastal areathan the In- amongproportions of den securityclasses between partu- land area (43% vs. 84%). Groupingbear visits that were rient and barrenfor Inlandfemales (n = 38, P = 0.64) or not attemptedbecause of movement prior to the den in- Coastalfemales (n = 25, P = 0.83). spectionperiod with unsuccessfulden visits, Coastalbears accountedfor 52 of 62 (84%) bear-wintersof unknown LitterSize reproductivestatus. We determinedlitter size of 14 Coastal and 11 Inland Seventeen of 38 (45%) Inlanddens where the females litters. The distributionof litter sizes differed between reproductivecondition was determinedwere occupied by areas(n = 25, P = 0.05). Mean littersizes for the Coastal solitary females, 15 (39%) by females with cubs, and 6 andInland areas were 2.4 (SE = 0.17) and 1.5 (SE = 0.16), (16%) by females with yearlings. The reproductivecon- respectively. Observedfrequencies of 1-, 2-, and 3-cub dition of females in observedCoastal dens differedfrom litters were 1, 6, and 7 for Coastal and 4, 6, and 1 for Inland(n = 62, P < 0.01), with only 1 of 24 (4%) females Inland. solitary,17 (71%)with cubs, and 6 (25%) with yearlings. LocationConstancy DenTypes Werecorded 2,591 locationsof 56 male andfemale bears Den types were not observed in the same proportions from 1992 to 1995, forming 2,228 3-location sets, dis- betweenareas (n = 71, P = 0.02). Most dens in the Coastal tributedacross 193 bear-seasons (76 den season, Nov- area were ground nests, in contrastto Inland where the Apr; 117 summer-fall, May-Oct). Three bear-winters numberof treedens with elevatedentrances (tree-elevated) with <3 locations were excluded from the analysis. In and ground nests were about equal (Table 2). Dens in of and debris from recenttim- Table3. Den securityclasses for dens used by parturient piles logs woody resulting and barrensubadult and adult female black bears in the or were recordedin both but ber harvest clearing areas, upper(Inland) and lower(Coastal) Atchafalaya River Basin 80% were Inland. We did not estimate the number of (1992-2000). structuresavailable in eitherarea. How- potentialdenning Den securityclass ever, because 93% of dens that were observed when the Reproductive sow fled on our approachwere groundnests, we believe Study area condition High Moderate Low that the majorityof unsuccessful den visits were a result Coastal Parturient 3 6 10a of bears dens. This was consis- Barren 1 1 4 abandoningground-level Inland Parturient 5 6 3 tent with our observationsof dennedbear response to our Barren 6 10 8 a presence. Bears in groundnests were generallyaware of Includes 1 female whose reproductivestatus was determinedafter our presence and usually in a standingor sitting position den emergence. Table2. Percentof subadultand adult female black bear den types in the upper(Inland) and lower (Coastal) Atchafalaya River Basin(1992-2000), Tensas River Basin in northeastLouisiana (1988-91; Weaver and Pelton 1994), Great Dismal Swamp (GDS) in Virginiaand North Carolina (1984-87; Hellgren and Vaughn 1989), Osceola (1983-85) and Ocala (1985-88) National Forests in Florida(Wooding and Hardisky1992), White River National Wildlife Refuge in Arkansas(1993-95; Oli et al. 1997),and Big Islandand MontgomeryIsland in Arkansas(1992-94; White et al. 2001). Percentswere rounded to the nearestwhole number; samplesizes (N)areindicated in bottomrow. Study area

Den type Coastal Inland Tensas GDS Florida White River ArkansasIslands

Excavated 5 2 0 48 0 0 0 Groundnest 57 28 14 44 100 2 8 Hollow log 5 0 0 0 0 0 0 Slash pile 7 19 27 0 0 8 75 0 Standingsnag 0 12 0 0 0 0 Hollow stump 11 2 0 4 0 0 0 Tree, elevated entrance 14 30 59 4 0 90 16 0 Tree, groundentrance 1 7 0 0 0 0 N 28 43 22 23 14 51 12 LOUISIANABEAR DENS * Hightoweret al. 15 testing the technique,we assumedbears were not station- caution should be used in interpretingthis result. Our ary duringthe summermonths. Accuracyof location sets small sample sizes resultedin low power to detect differ- to classify bear movement statusboth duringand outside ences. the denningperiod as stationary(bears with observed dens) The high proportionof groundnests we observedin the or moving (summerlocations) was 87%, overall. Thir- Coastalarea was consistentwith reports from other coastal teen percent of summer-fall bear-seasons were plain populationsin northcentral Florida (Wooding and misclassified as stationaryand 17%of bears observed in Hardisky1992) and the GreatDismal Swamp of Virginia dens were misclassifiedas moving. Misclassifieddenned and North Carolina(Hellgren and Vaughan1989). Also, bears included 1 Coastal female with cubs, 1 barren all bearsobserved in the MississippiAlluvial Valley popu- Coastal female, 2 Inland females with yearlings, and 1 lations on Big Island and MontgomeryIsland in Arkan- Inlandfemale of unknownreproductive condition. sas used groundnests (White et al. 2001). Based on location constancy,we classified 27 of 82 (62 An absenceof tree dens was thoughtto have influenced female and 20 male) bear-winterswhere denning status den types used in the Great Dismal Swamp and on Big was previouslyunknown. Twelve females (10 Coastal,2 Islandand Montgomery Island in Arkansas(Hellgren and Inland)were classified as stationaryfor at least 1 location Vaughan 1989, White et al. 2001). Unlike the Coastal set (i.e., approximately3 weeks) and 15 (13 Coastal, 2 forests, most Inlandforested lands known to be occupied Inland)were classified as moving. Seven males (6 Coastal, by bears were managed for commercial hardwoodpro- 1 Inland)were classified as stationaryfor at least 1 loca- duction,potentially reducing the numberof availabletree- tion set and 13 Coastal males were classified as moving. elevated or high security dens. However, Inland bears We believe wintermovement determinations based on our used and successfully gave birthin low and moderatese- decision rule conservativelyestimate denning because in curitydens. Concealment,rather than energetics, may be test comparisonswe were more likely to classify station- the most importantfactor for black bearden site selection ary bears as moving than the opposite. Increasing the in southeasterncoastal plain populations (Hellgren and maximumdistance above the mediandeviation increased Vaughan1989), as most low and moderatesecurity dens the misclassificationof summer-fall bear-seasonsto un- were located in thick cover. realisticlevels. Treedens arenot necessaryfor successful reproduction in Louisiana'smild climate if flooding and humandistur- bance are minimized. Our data were insufficient to test DISCUSSION for differencesin cub survivalamong den securityclasses High rates of tree den use have been reportedin some or den types. However,in Minnesotaand Massachusetts, southeasternU.S. black bearpopulations including White areas with more severe winter weather, McDonald and River National Wildlife Refuge, Arkansas(Smith 1985, Fuller (1998) were not able to detect any differences in Oli et al. 1997), GreatSmoky MountainsNational Park, cub survival rates among dens of 3 thermal efficiency Tennessee (Johnson and Pelton 1981), and the Tensas classes. Their results suggest that cub survivalis not re- River Basin of Louisiana (Weaverand Pelton 1994). It ducedby use of open dens at groundlevel. We agree with has been hypothesizedthat tree or snag dens with elevated Smith et al. (1994) thatbears adapt to theirenvironments entrancesare importantto easternblack bearpopulations andmay use whatevermaterials are availablefor den con- because they provide a well-insulatedcavity (Johnsonet struction. al. and more 1978) protectionfrom disturbance(Johnson Denned Coastal and Inland bears behaved differently and Pelton 1981, Alt and Gruttadauria1984). Oli et al. when approached;Coastal bears were more likely to flee (1997) suggested that using tree dens may affordparturi- upon approachwhen occupying low security dens than ent females moreenergy for activitiesassociated with cub Inland bears. Flushing of denned bears by researchers the rearingby reducing amount of energy expended for has been reportedin most studies of black bear denning maintaining body temperature. If these assertions are ecology in the Southeast(Hamilton and Marchinton 1980, then we correct, would expect bears,especially parturient Hellgren and Vaughan1989, Weaver and Pelton 1994). females, to use tree-elevatedor high security dens when Linnell et al. (2000) hypothesized that denned bears in possible. Because we did not measure den availability, milder climates lowered their threshold to disturbance we could not directly address den site selection. How- because the energetic costs of den abandonmentwere the of ever, proportions high (tree-elevated),low (ground lower. Our observationsin the Coastal area supportthis nest), and moderate(all others) den securityclasses used assertion. However, Inlandbears were more tolerantof by parturientfemales in ourstudy areas did not differfrom intrusionin nearly the same climate. The Coastal terrain barrenfemales. These results that suggest securityclass was more difficult to navigate and preventedus from ap- of dens used was of independent birthingstatus, but some proaching den sites as stealthily as was possible in the 16 Ursus 13:2002

Inland area. Our approachthrough thick cover around apparentdifference in litter size distributionbetween In- Coastaldens was noisy andprobably contributed most to land and Coastalshould be interpretedwith caution,even den abandonment. thoughthe significancelevel of the statisticaltest met our Bear movement away from den sites raised our con- a priori level. We were unable to determinethe repro- cern aboutpotential cub abandonment.Many Coastallit- ductive condition of the female in 50% of Coastal den ters were examinedin the den afterthe female fled, andin visits, and a small increasein the proportionof Coastal 1- many cases she was not observed during the visit. Al- cub litterswould have changedour results. though some females with cubs changed den sites after our visit, we did not detect any cub abandonmentas a result of our activities. MANAGEMENTRECOMMENDATIONS Coastal and Inland bears also differed in extent and Availability of suitable dens does not appearto be a probability of winter movement. Location constancy limiting factorin the 2 southcentral Louisiana black bear analyses indicatedmany Coastalbears and a small num- populations,and den trees are not requiredfor successful ber of Inlandbears continued to move throughoutthe den- reproduction.However, we do recommendprotection of ning period or only bedded for periods of <3 weeks. potentialden trees within the Inlandarea because a large Additionally,some Coastalbears classified as stationary portionof the Inlandpopulation occupies a spillway de- duringthe den season by our location constancydecision signed to divert waters from the . No rule were stationary only briefly during November or flooding occurredwithin the spillway during this study, Decemberand began moving beforeour den visits in Feb- but if waters were released into the spillway during the ruary. Distances between weekly winter locations were denningseason, all littersin dens at groundlevel could be also greaterfor Coastalbears (R.O. Wagner and R.M. Pace, lost. Candidateden trees are defined in the U.S. Fish and unpublisheddata). Winter-activeblack bears have been Wildlife Service listing rule for the Louisianablack bear reportedin many areas with mild climates (Hellgrenand (U.S. Fish and Wildlife Service 1992) as baldcypressor Vaughan 1989, Graber 1990, Weaver and Pelton 1994, watertupelo >91 cm diameterat breastheight with a vis- Oli et al. 1997). Graber(1990) suggestedwinter dormancy ible cavity, occurringin or along rivers, lakes, streams, may be elective for non-parturientfemales in mild cli- bayous, sloughs, or otherwater bodies. We observedbear mates where sufficient food is available throughoutthe dens in many tree species including oaks (Quercussp.), winter. We did not measure winter food availability in Americanelm (Ulmusamericana), sweetgum (Liquidam- our study areas,but we did observe fresh scats near win- bar styraciflua),and water hickory (Carya aquatica). We ter beddingsites of manyCoastal and Inland females with recommendmodification of this definition to include all yearlings,as was also reportedfor bearsin the TensasRiver tree species meeting the size and cavity criteria,regard- Basin (Weaverand Pelton 1994). Some winterfoods were less of location relative to waterbodies. still availableto bearsduring February and may have been We also recommendcontinued monitoring of both bear sufficientto allow bearsto remainactive andforage while populations. Parametersnecessary to model the dynam- maintaininga positive energy budget. ics of either of the populations,including cub and year- We only detected 1 barrenCoastal female throughden ling survivaland interbirth interval, are unknown. Studies season checks, suggesting most unsuccessful den visits designed to addressthese questions,as well as determine were of barrenfemales or females with yearlings. How- the cause of low reproductiveoutput of Inlandfemales, ever,we concedethat females with 1-cublitters could have are needed. carried the cub away from the nest on our approach (Ericksonand Martin 1960). We do not believe this oc- curredin the majorityof unsuccessfulCoastal den visits, ACKNOWLEDGMENTS but we did record 1 instance where a female moved, un- Our work was funded by: U.S. Forest Service, South- detected, away from a ground nest at our approachand ernForest Experiment Station, Southern Hardwoods Labo- was observedwith 1 cub the following summer. ratory;Louisiana Cooperative Fish andWildlife Research We detected litters at the majorityof observed Coastal Unit; Louisiana State University Agricultural Center; dens and they appearedlarger, on average,than those ob- LouisianaDepartment of Wildlifeand Fisheries (LDWF); servedInland. We cannotexplain the apparentdifference U.S. Fish andWildlife Service; Riverbend Station, Entergy in litter-size distributionbetween the Inlandand Coastal Corporation;and the National Council of the Paper In- populations. The age of females in the 2 studyareas were dustryfor Air andStream Improvement, Incorporated. We similar and most bears in the Inlandarea appearedto be are gratefulto many LDWF personnelwho providedas- healthy,based on a subjective assessment of weight and sistance, especially S. Shively, G. Linscombe, and N. body condition during den inspection. We believe the Kinler. We thankall the studentsand technicianson the LOUISIANABEAR DENS * Hightoweret al. 17

LSU bearcrew for theirefforts. In southLouisiana, bears Wildlife Society Bulletin 28:400-413. live almost solely on privateland, and the gracious con- MCDONALD, J.E., Jr., ANDT.K. FULLER. 1998. Testing sent and supportof many landowners,hunters, and farm- assumptions in bear research: Using statistical power to estimate effects of ers of Iberia,Point and St. Parishesmade analysis den type on black bear cub Coupee, Mary survival. Ursus 10:405-411. our work We thank reviewers J. Clark and D. possible. NELSON,R.A., H.W. WAHNER,J.D. JONES,R.D. ELLEFSON,AND Immell for comments on this manuscript. P.E. ZOLLMAN.1973. Metabolism of bears before, during, and after winter sleep. American Journal of Physiology 224:491-496 LITERATURECITED NYLAND,P.D., ANDR.M. PACE,III. 1997. Utility of National ALT,G.L. 1983. Timing of parturitionof black bears (Ursus WetlandsInventory data for black bear habitatassessment americanus) in northeastern Pennsylvania. Journal of in CoastalLouisiana. ProceedingsSoutheastern Association Mammalogy64:305-307. of Fish and Wildlife Agencies 51:297-310. , AND J.M. GRUTrADAURIA. 1984. Reuse of black bear OLI, M.K., H.A. JACOBSON,AND B.D. LEOPOLD. 1997. Denning dens in northeastern Pennsylvania. Journal of Wildlife ecology of black bears in the White River NationalWildlife Management48:236-239. Refuge, Arkansas.Journal of WildlifeManagement 61:700- CHABRECK,R.H. 1970. Marshzones and vegetativetypes in the 706. Louisiana coastal marshes. Dissertation, Louisiana State PACE,R.M., III, D.A. ANDERSON,AND S. SHIVELY.2000. Sources University,Baton Rouge, Louisiana,USA. and patterns of mortality in Louisiana black bears. ERICKSON,A.W., AND P. MARTIN. 1960. Black bear carries cubs Southeastern Association Fish and Wildlife Agencies from den. Journalof Mammalogy41:408. 54:365-373. EVANS, D.L., P.Y. BURNS,N.E. LINNARTZ,AND C.J. ROBINSON. SMITH, M.E., J.L. HECHTEL,AND E.H. FOLLMANN. 1994. Black 1983. Forest habitatregions of Louisiana. LouisianaState bear denning ecology in interior Alaska. International University, School of Forestry, Wildlife, and Fisheries Conference on Bear Research and Management9(1):513- Resource Report 1, Baton Rouge, Louisiana,USA. 522. GRABER,D.M. 1990. Winterbehavior of black bearsthe Sierra SMITH,T.R. 1985. Ecology of black bears in a bottomland Nevada, California. International Conference on Bear hardwood forest in Arkansas. Dissertation, University of Researchand Management8:269-272. Tennessee, Knoxville, Tennessee, USA. HAMILTON,R.J., ANDR.L. MARCHINTON.1980. Denning and TAYLOR,E.F. 1971. A radio-telemetrystudy of the black bear relatedactivities of black bears in the coastal plain of North (Euarctosamericanus) with notes on its history and present Carolina. InternationalConference on Bear Research and statusin Louisiana.Thesis, Louisiana State University, Baton Management4:121-126. Rouge, Louisiana,USA. HELLGREN,E.C., AND M.R. VAUGHAN. 1989. Denning ecology U.S. ARMY CORPS OF ENGINEERS. 1974. Preliminary draft of black bearsin a southeasternwetland. Journalof Wildlife environmental statement for the Management53:347-353. Floodway,LA. U.S. ArmyCorps of Engineers,, HIGGINS, J.C., M.R. VAUGHAN, D. MARTIN,AND A.D. VASHON. Louisiana,USA. 1997. Expandableradio collar for black bear cubs. Forum U.S. FISHAND WILDLIFE SERVICE. 1992. Listing the Louisiana on Wildlife Telemetry, Jamestown, North Dakota, USA. blackbear as a threatenedsubspecies and designation of other NorthernPrairie Wildlife ResearchCenter Home Page http:/ free-living bears of the species U. americanusas threatened /www.npwrc.org/resource/tools/telemetry/telemetry.htm, due to similarityof appearance.Federal Register 57(4):588- Accessed 14 October 1997. 595. JOHNSON, K.G., D.O. JOHNSON, AND M.R. PELTON. 1978. WAGNER,R.O. 1995. Movementpatterns of blackbears in south Simulation of winter heat loss for a black bear in a closed centralLouisiana. Thesis, LouisianaState University, Baton tree den. Proceedingsof the EasternBlack Bear Workshop Rouge, Louisiana,USA. 4:155-156. WEAVER, K.M., ANDM.R. PELTON.1994. Denning ecology of , ANDM.R. PELTON.1980a. Prebaiting and snaring black bears in the Tensas River Basin of Louisiana. for black bears. techniques Wildlife Society Bulletin 8:46- InternationalConference on BearResearch and Management 54. 9(1):427-433. . 1980b. ,AND Environmentalrelationships and WHITE,T.H., J.L. BOWMAN,H.A. JACOBSON,B.D. LEOPOLD,AND the denning period of black bears in Tennessee. Journalof W.P. SMITH.2001. Forest managementand female black 61:653-660. Mammology bear denning. Journalof Wildlife Management65:34-40. ,AND . 1981. Selection and availabilityof dens WOODING,J.B., ANDT.S. HARDISKY.1992. Denning by black for black bears in Tennessee. Journal of Wildlife bears in northcentral Florida. Journal of Mammalogy Management45:111-119. 73:895-898. LINDZEY,F.G., ANDE.C. MESLOW.1976. Characteristicsof black bear dens on Long Island, Washington. NorthwestScience 50:236-242. Received: 28 May 2001. LINNELL,J.D.C., J.E. SWENSON,R. ANDERSEN,AND B. BARNES. Accepted: 16 October 2001. 2000. How vulnerable are denning bears to disturbance? Associate Editor: Immell.