sign in sign up
<<
Home , Mast (botany), Vaccinium hirsutum

ENERGETICPRODUCTION BY SOFT AND HARDMAST FOODS OF AMERICAN BLACKBEARS INTHE SMOKYMOUNTAINS

ROBERTM. INMAN,1 Department of Forestry,Wildlife, and Fisheries, Universityof Tennessee, Knoxville,TN 37916, USA, email: [email protected] MICHAELR. PELTON,Department of Forestry,Wildlife, and Fisheries,University of Tennessee, Knoxville,TN 37916, USA, email: mpelton@ utk.edu

Abstract: We measuredcaloric productionby 19 species of vegetation used as food by Americanblack bears (Ursus americanus)in GreatSmoky MountainsNational Parkto determinethe significance of productionby mast type, season, and species. Mean annualproduction by all species was 351,209 cal/ha. Hardmast produced74.5% (16.0 billion cal) of total calories availableon the study area;soft mast produced25.5% (5.5 billion cal). Gross energeticcontent of soft and hardmast did not differ (P = 0.488, n = 19). Mid-summerwas the lowest periodof production.Northern red (Quercus rubra)produced 65.7% of calories; squawroot(Conopholis americana) produced 15.8%, and huckleberries(Gaylussacia spp.) produced 5.1 %. A white oak (Q. alba and Q. prinus) mast failureoccurred, and white producedonly 5.1% of calories. Oaks are likely the single most influentialgenera affecting bear ecology in the southernAppalachians. However, availabilityof soft mast likely has a substantialimpact on bear populationsbecause of the timing of production,nutrients available, and its functionas a surrogateduring hard mast failure. Furtherstudy is needed to determinethe effects of soft mast abundanceon age of primiparity,litter interval, recruitment, and density. Once the roles of majorfoods are well understood,appropriate habitat compositions and silvicultural prescriptions may be defined.

Ursus13:57-68 (2002)

Key words: Americanblack bear,Carya, Conopholis,energy, Gaylussacia, mast, Nyssa, Prunus,Quercus, Rubus, Smilax, Smoky Mountains,Ursus americanus, ,Vitis

Nutritionalcondition of Americanblack bears affects nent of black bear habitatsin the southernAppalachian age of primiparity(Jonkel and Cowan 1971, Eiler et al. region. Soft mast composed a greaterpercent than hard 1989), litter interval(Rogers 1976, Eiler et al. 1989), lit- mast in both annualvolume index (45%vs. 14%,Beeman ter size (Beecham 1980, Elowe and Dodge 1989), cub and Pelton 1980) and relative percent density (37% vs. survival (Rogers 1976, Eiler et al. 1989), and yearling 10%,Eagle andPelton 1983) of scatscollected in GSMNP. survival (Jonkel and Cowan 1971; Rogers 1976, 1987). Gypsy moth (Lymantriadispar) infestations in Virginia Nutritionalcondition may also influence fertility (Noyce resulted in complete crop failures, yet bear repro- and Garshelis 1994). Thus, habitatswith adequatefood duction and survival in those areas did not decrease im- sources are essential for effective bearmanagement. mediately after the infestationand subsequentfailure of Because of the importanceof energy storage for the the mastcrop (Kasbohmet al. 1996). The amountof denningperiod, previous southern Appalachian bear stud- soft mast availablein an areainfluences the seasonal and ies were designed to determinethe relationshipbetween annualhome range sizes of bears and their activity pat- fall oak mast availabilityand reproduction. Bear repro- terns (Garshelisand Pelton 1980, Quigley 1982). Bears ductionwas found to be highly correlatedwith hardmast with abundantblack cherry (Prunus serotina) in theirhome indices by Eiler et al. (1989), Pozzanghera(1990), and range displayed delayed movement to areas of abundant McLean (1991). However, Coley (1995) found no sig- oak mast (Garshelisand Pelton 1981). Abundantgrape nificant positive correlationsbetween hard mast indices (Vitisspp.) cropsreduced the impactsof a severe oak mast andbear population size. Coley (1995) analyzed20 years failure on bearreproductive effort (Eiler et al. 1989). of black bear populationdata to determinethe influence Managing bear habitatsrequires identificationof im- of hardmast productionon black bearpopulation dynam- portanthabitat components and determinationof the op- ics in GreatSmoky MountainsNational Park(GSMNP). timal, or at least minimal, habitat mix necessary for Coley (1995) correlatedhard mast indices from GSMNP maintainingpopulations at desiredlevels (Schoen 1990). with variouspopulation estimates for periodsup to 5 years Policies regardingnatural disturbances in nationalparks afterrecorded hard mast data, a broadapproach that should and silviculturalpractices on multiple-uselands influence accountfor the influence of recruitmentin additionto re- food availabilityfor bears. Bear habitatmanagement in production. Because soft mast availabilitywas not docu- the southernAppalachians currently emphasizes the im- mented during the 20-year period that Coley (1995) portanceof the fall hardmast component (Eiler et al. 1989, studied, the variationin populationestimates that could Pelton 1989). A betterunderstanding of the influence of be explainedby soft mast availabilityis unknown. soft mast on bear population dynamics in the southern Soft mast may also be an importantnutritional compo- Appalachiansmay lead to more effective management strategies. In this study, we examined food 1 Presentaddress: Wildlife availability ConservationSociety, 2023 Stadium the calories the of Drive,Suite 1A, Bozeman, MT 59715, USA. by measuring produced by majority 58 Ursus 13:2002 vegetative bear foods in GSMNP (59% annual volume ous studies in the southernAppalachians (Beeman and index of scat; Beeman and Pelton 1980). Our objectives Pelton 1980, Eagle and Pelton 1983, Brody and Pelton were (1) to test for differences in caloric productionper 1988). All were common in GSMNP (Whittaker hectarebetween mast types and among seasons, and (2) 1956, Stupka 1960, Golden 1974). to estimatetotal calories producedon the study area sea- sonally, by mast type, and by species. Sampling Scheme We stratifiedthe study areato select 275 samplepoints by (1) vegetationtype, (2) elevationalrange of each veg- STUDYAREA etation type, (3) aspect, and (4) distance from trails. We GSMNP is located along the borderof Tennessee and used the vegetation classification by MacKenzie (1993; North Carolina (Fig. 1). This study was limited to the Table 1). To stratifyby vegetationtype we overlaidtopo- northwesternquadrant of GSMNP (613 km2). Elevations graphic maps with Landsat Thematic Mapper satellite withinthe studyarea ranged from 270 m to 2,025 m. Soils imagery that was remotely sensed during 1984 are thin and poorly developed with mediumto high acid- (MacKenzie1993). This vegetationdata layer had a pixel ity, low water storagecapacity, and low to moderatefer- resolutionof 90 x 90 m, with each pixel classified as 1 of tility (Soil Survey 1945, 1953). Climate of the area has 14 vegetationtypes. We sampled9 of 14 vegetationtypes: been classified as a warm-temperate rain cove hardwood,xeric oak, mesic oak, mixed mesic hard- (Thorthwaite 1948). Annual precipitationranges from wood, pine, pine-oak, tulip-poplar,northern hardwood, 140 cm at lower elevations to 230 cm at higherelevations and spruce-fir. The 5 unsampledtypes (treeless, grassy (Stephens 1969). GSMNP's microclimaticdiversity rep- bald,grape thicket, heath bald, and water) comprised 2.4% resented site potentials for much of the federally owned of the study area. We did not recordstand age, but it was landin the southernAppalachians, and management goals generally>60 years.Thus, in most vegetationtypes of the Parkare intendedto maintainan absence of human were relativelymature and undisturbed.An exception to alteration. Thus, GSMNP affordeda unique opportunity this was the spruce-fir type, in which over 70% of the to studyhabitat use of wildlife populationsin a relatively Fraserfir (Abiesfraseri) had been killed by the balsam controlledsetting. wooly adelgid (Adelgespiceae) duringthe past 30 years (NationalPark Service [NPS] personnel,Gatlinburg, Ten- nessee, USA, personalcommunication, 1995). We clas- METHODS sified elevation rangesfor each vegetationtype as low or We selected 19 plantspecies for this studythat had been high, and aspects as northeastern(315-134?) or south- identified as primaryfoods for black bears duringprevi- western (135-314?). We distributed30-33 sample plots

j 7 W j

Cherokee NF

Fig. 1. Location of study area in GreatSmoky MountainsNational Park, Tennessee (TN)and NorthCarolina (NC), USA. Muchof surrounding area is national forest (NF). ENERGETICPRODUCTION IN THESMOKY MOUNTAINS * Inman and Pelton 59

Table1. Vegetationtypes as classifiedby MacKenzie(1993), including dominant species, area,and percent of northwest quadrant of Great Smoky Mountains National Park, North Carolina and Tetnnessee, USA. a Vegetation type Dominanttree species Hectares Area (%) Covehardwood Easternhemlock (Tsuga canadensis) 14,710 24% Sweet birch (Betula lenta) Red maple (Acer rubrum) Carolinasilverbell (Halesia carolina) Tulip-poplar(Liriodendron tulipifera) Northernred oak (Quercus rubra) Basswood (Tilia heterophylla) Yellow birch (Betula alleghaniensis) Pine Table-mountainpine (Pinus pungens) 13,476 22% Pitch pine (Pinus rigida) Virginiapine (Pinus virginiana) Scarlet oak (Quercuscoccinea) Xericoak Chestnutoak (Quercusprinus) 13,235 22% Red maple Tulip-poplar Sourwood (Oxydendrumarboreum) Scarlet oak Mixedmesic hardwood Tulip-poplar 8,017 13% Red maple Easternhemlock Chestnutoak Northernhardwood Yellow birch 3,363 6% Americanbeech () Sweet birch Easternhemlock Red maple Northernred oak Red spruce (Picea rubens) Tulip-poplar Tulip-poplar 3,045 5% Red maple Carolinasilverbell Mesicoak Northernred oak 1,770 3% Red maple Chestnutoak Pine-oak Scarlet oak 1,542 3% Table-mountainpine Blackgum (Nyssa sylvatica) Red maple Chestnutoak Spruce-fir Yellow birch 693 1% Red spruce Red maple Fraserfir (Abiesfraseri) a Species within each vegetationtype are orderedaccording to dominance,based on a mean of species basal area >2.0 m2/ha,from MacKenzie (1993).

within each of 9 vegetation types as equally as possible height. We subsequentlyconverted m3 of tree crown vol- the elevation and among aspect combinationswhere that ume andm2 of shrubcoverage into caloric productionper vegetation type was found. If field inspection indicated hectarefor each samplepoint, using informationdescrib- that the had been vegetation incorrectly classified by ing the averagenumber of calories producedper m3or m2 MacKenzie that (1993), plot was discardedand a new plot by each food species. We estimated the timing of that sampled. All sample points were constrainedto be >90 productionas describedbelow. m from the nearesttrail. Distributionand Abundanceof Food Phenology For each species of bear food, 5-45 individualtrees or Species patchesof shrubwere markedalong 4 transectsthat passed At each we used a sampling point squareplot of 0.04 througha variety of watershedsand elevational ranges: ha to measure crown (400 m2) volume (m3) of relevant (1) Forge Creek-Parson's Branch-GregoryBald Trails tree species and percentcover (m2)of relevantshrub and (549-914 m elevation), (2) Lead Cove-Bote Mountain- vine To estimatecrown species. volume we made 3 mea- Schoolhouse Gap-TurkeypenRidge Trails(518-945 m), surementson each tree within the plot: (1) crown width (3) Jake's Creek-Miry Ridge Trails (671-1,372 m), and from east-west, (2) crown width from north-south, and (4) Appalachian-SugarlandMountain Trails (1,524-1,829 crown We used a (3) height. clinometerto estimatecrown m). Individualtrees or shrubsfor each food species were 60 Ursus 13:2002 selected and permanentlymarked so that each species DataAnalysis was representedat variouselevations andin differentwa- We estimatedgross caloricproduction per unit area(m2 tersheds. We surveyed each transectbi-weekly from 20 of shrubcoverage) or volume (m3of tree crown) by each March 1995 to 11 December 1995 and recordedthe phe- species during 1995 as the mean numberof pro- nological conditionof each tree or shrubpatch. Individu- duced per m2 ()or m3() x mean dry weight/ als were recordedas ripe when fruitswere mature,edible, (g) x energeticcontent per g dry weight. To estimate and present (including on the ground). If an individual the temporalavailability of calories, we pooled all phe- tree or shrubpatch was the same condition during con- nological observationsfor a species and assigned each secutive surveys, thatcondition was also assigned for the week a portion of the total caloric production. The as- interveningweek when no survey occurred. If consecu- signed portionof productionwas based on thatparticular tive weeks differed, an entry of no observationwas re- week's percent of the total numberof ripe observations corded for the interveningweek. Thus, all species and recordedfor the species. For example,if a species had 50 individualshad an equal possibility of having been ripe 1 observationsof ripe, and 5 of those occurredin week 1, week more or less than observed. Measures of fruiting then week 1 received 10%of the total caloric production periods were thus imprecise. We defined peak produc- for that species. Dates used to separateseasonal periods tion as the periodwhen >50%of individualshad ripe fruits were spring (1 Apr-15 Jun), summer (16 Jun-15 Sep), available. and fall (16 Sep-15 Dec; Beeman and Pelton 1980, van Manen 1994). FruitProduction We used analyses of varianceto comparecaloric pro- Whenthe majorityof fruitsof an individualtree or shrub duction/habetween mast types and among seasons. Al- patch were ripe, we estimatedtotal fruitproduction from though sample points were located using a stratified thatindividual. To estimatefruit production for trees, we sampling scheme, analyses were based on a randomde- counted the fruit seen within a measuredvolume of tree sign. Differences in mean production/habetween mast crown using a spotting scope (Inman 1997). We took 2 types and among seasons were analyzed using a com- Each samplesat randomlocations from within each tree crown pletely randomdesign with repeatedmeasures. plot hard and averaged the 2 samples to estimate productionfor was sampled for cal/ha by soft mast and by mast, that individual. To estimate fruit productionby shrubs, and during spring, summer, and fall seasons. This re- blocked out the varia- we sampledthe numberof fruits within a 0.5 m2 area of peatedmeasures design effectively the null shrubcoverage. We sampledby randomlytossing a square tion due to vegetationtype. We tested hypotheses hectare did not differ 0.5 m2sampling grid into thepatch of shrubberyand count- that mean caloric productionper and seasons ing all fruits within the grid. Fruits were counted from between mast types (P= 0.05) among (P= the top of the crown to the ground. Two 0.5 m2 0.05). total caloric within the sampleswere takenat each individualpatch, and the mean We also estimated production for each defined sea- of the 2 samples was used as the estimate of production entire study areafor the year 1995, and for each We esti- for thatpatch. son, for both mast types, species. mated species coverage within the study area by FruitWeight and Gross EnergeticContent multiplyingthe mean areasof coverageby each food spe- total areaof each We collected fruits during 1995 and preparedsamples cies within each vegetation type by the GIS for bomb calorimeteryin a ParrOxygen Calorimeter(Parr vegetationtype (fromMcKenzie's [1993] coverage). cal/ha InstrumentCompany, Moline, Illinois, USA). For fruits We estimatedtotal caloric productionby applying to the and with large,indigestible seeds (blackcherry and blackgum) values (above) acreage figures phenological we estimatedthe number or hulls ([Carya spp.]), we removedthe seed or categorizations. Thus, average each food within each hull and analyzed only the fleshy partof the fruit. Spe- of calories producedby species, cies withminute seeds were left intact(huckleberries, blue- vegetationtype, for the entirestudy area through the grow- berries [Vaccinium spp.], blackberries [Rubus spp.], ing season, 1995. 1995 there was a mast failure 2 of grapes). One composite sampleof each species was oven During by species mast north- dried at 60?C until no weight change was detected. The white oaks andan extremelyabundant cropby oak. We scenarios of mast weight of the entiresample was dividedby the numberof ern red developed differing or the levels of oak fruits in the sample to obtain the mean dry weight/indi- productionby increasing reducing to the maximum or minimum vidual fruit for a species. Samples were then measured mast production(cal/m3) 1995. We estimatedtotal annualand for caloricvalue per gram dry weight. We requireda varia- levels recordedin tion in caloric value of <3% over at least 2 tests (Johnson mast type productionfor 2 scenarios: (1) the production of northern and Robel 1968, Golley 1961). of all oak species was set at 30.3 cal/m3(level ENERGETICPRODUCTION INTHE SMOKY MOUNTAINS * Inman and Pelton 61 red oak's abundantcrop during 1995), and pro- Hard mast trees occurredin 8 of 9 vegetation types and duction was increasedto half (15 cal/m3)of the oak pro- accountedfor 97.2% of the tree crown volume, whereas duction (1995 hickory levels representednear total crop soft mast trees occurredin only 5 of 9 vegetation types failure);and (2) we set productionof all oaks at the mean and accountedfor <3%. Black cherryoccurred in only 3 of the 2 lowest producersduring 1995 (mean = 1.6 cal/ vegetationtypes. ,greenbriers (Smilax spp.), m3). The first scenario representeda high mast year for and huckleberrieseach occurredin at least 7 of 9 vegeta- all species and the second scenarioa total mast failure. tion types whereas blackberries,grapes, and squawroot occurredin relatively few. Huckleberriesaccounted for 58.1% of shrub coverage and were the most abundant RESULTS shrub in 7 of 9 vegetation types. Area of coverage by blueberrieswas <15% of huckleberries. Distributionand Abundanceof Food coverage by Species Phenology Oaks occurredin 8 of 9 vegetationtypes and accounted Shrubfruits were ripe earlier (summerand early fall) for 85.7% tree crown volume among species producing thantree fruits (fall; Inman1997). Squawrootwas at peak bear foods (Tables 2, 3, Fig. 2). White oaks were found productionfor 6 weeks (22 May-2 Jul). Peak production in fewer vegetationtypes thanred oaks, but accountedfor by blueberrieswas earlier (Jul) than huckleberries(Jul- more (45.4%) crown volume than did red oaks (40.3%). Aug) with the exception of the high elevation southern

Table 2. Shrub coverage (m2/ha) within 9 vegetation types in the northwest quadrant of Great Smoky Mountains National Park, USA, 1995.

Overstoryvegetation Shrubspeciesa Mean (m2/ha) SD Spruce-fir Thomless blackberry(Vaccinium canadensis) 1,880 2,531 S. mountaincranberry (Vaccinium erythrocarpum) 68 270 N. highbushblueberry () 54 296 Northern hardwood Thornlessblackberry 743 1,914 Allegheny blackberry(Rubus allegheniensis) 358 1,187 N. highbushblueberry 100 326 Greenbriers 4 23 Cove hardwood Huckleberries(Gaylussacia sp.) 1,028 2,466 Greenbriers 332 805 Grapes 168 551 Squawroot 88 259 N. highbushblueberry 17 63 Upland low () 10 55 Deerberry(Vaccinium stamineum) 4 23 Mesic oak Huckleberries 1,383 2,634 Allegheny blackberry 200 779 Greenbriers 312 523 Thomless blackberry 3 18 Upland low blueberry 1 5 Mixed mesic hardwood Huckleberries 2,385 3,351 Greenbriers 461 1,144 Grapes 197 754 Deerberry 4 23 Squawroot 6 24 Tulip-poplar Huckleberries 339 904 Greenbriers 323 613 Grapes 275 1,194 Squawroot 4 23 Xeric oak Huckleberries 1,587 2,337 Upland low blueberry 566 1,372 Greenbriers 385 450 Deerberry 96 247 N. highbushblueberry 4 22 Pine-oak Huckleberries 2,946 3,500 Hairy blueberry(Vaccinium hirsutum) 279 1,064 Greenbriers 235 347 Upland low blueberry 225 855 Deerberry 129 418 Pine Huckleberries 1,434 2,244 Greenbriers 403 1,075 Upland low blueberry 162 464 Grapes 8 44 Squawroot 3 15 a N. = northern,S. = southern 62 Ursus 13:2002

Table3. Treecrown volume (m3/ha) for bear food producingtrees inthe northwestquadrant of GreatSmoky Mountains National Park,USA, 1995. Overstory vegetation Tree speciesa Mean (m3/ha) SD

Spruce-fir Fire cherry (Prunuspennsylvanica) 449 1,958 Northernhardwood American 6,066 2,634 Black cherry (Prunus serotina) 1,704 631 N. red oak 1,417 7,759 Fire cherry 47 255 Cove hardwood N. red oak 14,325 30,669 Chestnutoak 5,563 5,881 Hickories 2,048 7,916 White oak 1,598 4,387 Scarlet oak 1,085 3,675 Mesic oak N. red oak 9,903 12,087 Chestnutoak 3,536 15,143 Black cherry 2,173 11,904 White oak 2,118 8,429 Mixed mesic hardwood N. red oak 11,753 17,390 Hickories 7,984 27,480 Chestnutoak 7,534 12,911 White oak 6,523 26,509 Tulip-poplar Chestnutoak 5,850 18,713 N. red oak 3,519 8,118 Hickories 219 1,200 Xeric oak Chestnutoak 19,645 17,402 N. red oak 4,295 8,856 Scarlet oak 3,207 5,940 White oak 1,646 5,450 Hickories 1,620 2,420 Blackgum 809 2,943 Pine-oak Chestnutoak 14,210 11,371 N. red oak 3,773 6,329 Scarlet oak 2,468 2,383 Hickories 1,107 5,084 White oak 1,066 2,364 Pine N. red oak 4,889 10,982 Chestnutoak 4,366 13,817 Hickories 3,334 7,715 Scarlet oak 3,055 10,378 White oak 2,365 3,565 Blackgum 725 1,806 Black cherry 529 2,992 a N. = northern, S. = southern

35

., 30 - OWhite oak crown volume a)"a 25 --

- o 20 IRed oak crown volume 0o j 15 - 10-

> 5-

00 0 r. 0 uO 0 a .0Y 0 z2

Vegetation type Fig.2. Meanred and white oak crown volume per hectare, by vegetation type, in the northwestquadrant of GreatSmoky Mountains National Park,USA, 1995. ENERGETICPRODUCTION INTHE SMOKY MOUNTAINS * Inman and Pelton 63 mountaincranberry (Vaccinium erythrocarpum), which Gross energeticcontent ranged from 5.27 cal/g for thorn- peakedduring September. Black cherrywas at peak pro- less blackberryto 3.44 cal/g for hairyblueberry (Table 4). duction from late August throughearly October. Oaks The 7 highest values were soft mast species; however, were ripe and available from late August through late mean cal/g of soft (x = 4.47 cal/g) and hard(x = 4.36 cal/ December when we ceased sampling. Peak production g) mast species did not differ (P = 0.488, n = 19). occurredduring the last week of Septemberfor red oaks and early Octoberfor white oaks. FruitCrop Production The numberof tree fruits producedper m3 of crown FruitWeight and Gross EnergeticAnalysis volume ranged from 0.14 for chestnut oak to 80.20 for Although not quantified, we considered acorn sizes black cherry(Table 5). Fruitproduction by red oaks was during 1995 typical of most years, with the exception of 8 times greaterthan thatby white oaks. Shrubfruits pro- white oak (Q. alba) which was less than /2normal size. duced per m2 ranged from 21.5 (greenbriers)to 109.5

Table4. Grossenergetic content and single driedfruit weight of blackbear foods in the northwestquadrant of GreatSmoky Mountains National Park, USA, 1995.

Fruitspeces Saple type Cal/g SD weight(g) na Ihomlessbackbery wholefruit 5.27 0.01 0.17 160 Blackgum seedremoved 4.98 0.01 0.08 142 Firedcny wholefruit 4.94 0.09 0.04 66 Gapes wholeflrit 4.90 0.01 0.09 245 Squawroot enirestalk 4.66 0.03 9.23 8 Alleghy blackbeny wholefruit 4.66 0.04 0.06 11 Hucddebenies wholefirit 4.65 0.07 0.03 83 wholefruit 4.60 0.07 222 35 fickories meatonly 4.53 0.01 0.15 50 Greebriers whoefruit 4.44 0.04 0.09 178 Scarletoak wholefruit 4.26 0.07 0.67 30 UpladlowMuefeny wholefruit 4.24 0.00 0.03 662 Chestntoak whoefiuit 4.19 0.07 2.99 57 Nathenhighush bluebery wholefruit 4.18 0.03 0.06 218 Soure mx tain anbeny wholefriit 4.17 0.07 0.04 57 wholefruit 4.17 0.06 0.04 389 Whiteoak wholefruit 4.05 0.07 0.67 33 Bla dchery seedremoved 3.95 0.01 0.04 100 Hairybluebeny wholefirit 3.43 0.05 5 a Numberof fruitsin composite sample thatwas weighed to estimateindividual fruit weight. Table 5. Fruitsand calories produced per square meter of shrub coverage or cubic meter of tree crown volume in the northwest quadrant of Great Smoky Mountains National Park, USA, 1995.

Species Fruits/m3 Fruits/m2 SD n Cal/m3 Cal/m2 Trees Northernred oak 2.97 5.1 84 30.33 Black cherry 80.17 115.0 90 12.99 Scarletoak 2.69 5.5 86 7.63 Fire cherry 21.64 46.6 48 4.49 Chestnutoak 0.14 0.4 92 1.73 White oak 0.59 1.3 80 1.59 Blackgum 2.76 16.9 78 1.13 Hickories 0.20 1.5 78 0.14 White oaks 0.35 Red oaks 2.83 Shrubsand vines Squawroot 57.33 38.5 6 2,467.33 Thornlessblackberry 67.96 72.1 90 59.46 Grapes 66.36 62.1 14 30.25 Northernhighbush blueberry 70.00 38.7 4 16.66 Huckleberries 109.54 105.8 78 13.23 Upland low blueberry 106.96 155.7 50 13.15 Southernmountain cranberry 74.71 119.6 102 11.84 Allegheny blackberry 32.77 65.6 26 9.46 Greenbriers 21.50 80.7 50 9.08 Hairy blueberry 28.50 31.5 8 5.09 Deerberry 25.50 40.3 8 4.14 64 Ursus 13:2002

(huckleberries;Table 5). Standarddeviations of fruit paring our maximumhard mast scenario with our mast samples were high relativeto means. failure scenario (Fig. 3). In these scenarios, differences in annualproduction levels of hardmast greatly affected CaloricProduction by Areaand Volume total annualproduction, but maximizing annualproduc- Caloricproduction by tree species rangedfrom 0.1 cal/ tion levels of soft mast did not (Fig. 3). m3for hickoriesto 30.3 cal/m3for northernred oak (Table Caloriesproduced per hectarewere greatestduring fall 5). Caloric production by shrub species (excluding (x = 167,600, SD = 455,525, n = 275) and lowest during squawroot)ranged from 4.1 cal/m2for deerberryto 59.5 spring (x = 22,300, SD = 164,950, n = 275, 2 df, F = cal/m2for thornlessblackberry (Table 5). Squawrootwas 284.92, P = 0.0001). Summerproduction was intermedi- by far the most productive shrub species on a per area ate (x = 89,850, SD = 168,200, n = 275). Summerand fall basis. productiondid not differ(P = 0.8896), but springproduc- tion differedfrom both summer(P = 0.0001) and fall (P = EnergeticProductivity 0.0001). Estimates of productionover the entire study Mean productionon the study areawas 351,209 cal/ha area indicated that 12.5% of all calories were available (SE = 49,834). We estimatedthat annual production dur- during spring, 28.2% duringsummer, and 59.3% during ing a year of hardmast failure would be 122,963 cal/ha fall. However,63% of the calories availableduring sum- and 690,344 cal/ha duringa year of high hardmast pro- mer were hardmast, typically considered fall food. There duction. was a periodof low energeticavailability during mid-sum- Caloriesproduced per hectare were greaterfor hardmast mer (Fig. 4). (x = 204,475, SD = 592,775, n = 275) than for soft mast Northernred oak produced65.7% of annualbear food (x = 75,275, SD = 74,550, n = 275, 1 df, F = 32.16, P = calories; squawroot ranked second, producing 15.8% 0.0001). Although productionby mast types differed, (Table6). Although white oaks accountedfor 45.4% of variancewas less for soft mast,indicating less patchydis- the tree crown volume on the studyarea among bear food tribution.Hard mast produced74% of the calorieswithin producers,they producedonly 6.7% of the calories avail- the studyarea, and soft mastproduced 26%. However,in able from trees. Red oaks accountedfor 40.3% of the tree our scenarioof hardmast failure,soft mast accountedfor crownvolume andproduced 91.7% of treecalories. White 72% of all caloric production.We noted a 17-fold differ- oaksproduced only 5.1%of all calories,which was equiva- ence in annual cal/ha contributedby hard mast in com- lent to productionby huckleberries(5.1%). Excluding

750

2 600

4 4c :!5

a 4) u4) *c4) 450 2

2 .aC-3 Cd 0L)

w 300 wCd cO? E.r5

I 150

0 1995 actual Hard mast optimal Hard mast failure Hard mast failure & optimal soft mast

Annual caloric production

Fig. 3. Contributionsof soft and hard mast to annual caloric production per hectare under 4 mast production scenarios in the northwest quadrant of Great Smoky Mountains National Park, USA, 1995. ENERGETICPRODUCTION IN THESMOKY MOUNTAINS * Inmanand Pelton 65 I I I I I _ _ 1.40

1.20 *- Spring ESummer *Fall

0.60 0 ri 0.40

0 . 00 I I I I I I I I I s3 0 -> ? ^ 5 ?? o< '5 rt ? a 3 ? 0 O Q <" 2 S - $!e - - - (1 - Ce - 'U 0 0- . 'U 'U > 0 z P-4 oc;4

Fig. 4. Total calories available/week (1 Apr-15 Dec) of hard and soft mast species used as food by bears in the northwest quadrant of Great Smoky Mountains National Park, USA, 1995.

Table 6. Estimates of total calories produced, percent of study area production, and caloric production per hectare for 19 species of vegetation used as food by black bears in the northwest quadrant of Great Smoky Mountains National Park, USA, 1995.

Species Mast type Total calories % of areaproduction SE Cal/ha SD N. redoak H 14,147,184,151 65.70 648,797,209 230,711 10,581 Squawroot S 3,413,293,121 15.80 551,769,040 55,664 8,998 Huckleberries S 1,093,165,414 5.10 643,004,494 17,827 10,486 Chestnutoak H 877,345,427 4.10 643,004,494 14,308 10,486 Scarletoak H 789,142,618 3.70 602,264,335 12,869 9,822 Thomlessblackberry S 226,337,206 1.10 99,365,195 3,691 1,620 Blackcherry S 217,041,793 1.00 349,561,350 3,539 5,701 Whiteoak H 214,239,731 1.00 638,217,240 3,494 10,408 Greenbriers S 190,784,595 0.89 648,797,209 3,111 10,581 Blueberries(all Vacciniumspp.) S 154,101,237 0.72 2,513 Grapes S 151,090,557 0.70 551,769,040 2,464 8,998 Uplandlow blueberry S 133,703,304 0.62 603,982,343 2,180 9,850 Blackgum S 23,085,653 0.11 466,902,036 376 7,614 Hickories H 22,494,795 0.10 641,391,018 367 10,460 Alleghenyblackberry S 14,723,375 0.07 97,751,627 240 1,594 Northernhighbush blueberry S 11,151,068 0.05 506,272,716 182 8,256 Deerberry S 6,494,296 0.03 540,935,436 106 8,822 Hairyblueberry S 2,191,760 0.01 39,668,758 36 647 Firecherry S 2,103,400 0.01 88,323,839 34 1,440 Southernmountain cranberry S 560,809 0.00 17,834,274 9 291 Total 21,536,133,072 100 351,209 Oaks 261,381 41,296 Berries(Rubus spp., Vaccinium spp., and Gaylussacia spp.) 24,271 41,566 squawroot,huckleberries produced 60% of shrub calo- DISCUSSION ries. All 5 blueberryspecies combinedproduced <1% of We believe our measurementsof energeticavailability all calories, whereas blackberryspecies produced 1.2% fairly representedthe average annuallevel on our study of all calories. Soft mast species rankedas 2 of the top 3 area. Majoralterations likely occur duringyears of com- producers(Table 6). Caloric productionper hectarewas plete hardmast failureor duringextreme weathercondi- 261,381 cal/haby oaks and24,271 cal/haby berries(black- tions such as severe drought. Oaks accountedfor 74% of berries, blueberries, and huckleberries;Table 6, Inman caloriesavailable to bearsduring this study,and have likely 1997). been the most influentialgenera affecting bear ecology in 66 Ursus 13:2002

the recent past. Inherentproductive ability of the red cient sources of energy during late spring and summer. and white oak groupsis not representedwell by this data. We speculate that squawrootabundance affects cub sur- If measurementhad occurred during 1996, white oaks vival and otherimportant population parameters, such as would have produceda largercrop. Hard mast surveys age of primiparityand litter interval. Increasing bear popu- conducted by NPS on the Tennessee side of GSMNP lation size within GSMNP (Coley 1995) could be in part during 1995 resulted in index values of 1.35 mast units due to increasinglyabundant squawroot in a maturingoak for white oaks and 4.05 mast units for red oaks; index forest. values for 1996 were 4.07 mast units for white oaks and Huckleberrywas the thirdmost productivespecies on 1.99 mast units for red oaks (Bill Stiver, National Park ourstudy area, and may be of importanceto the bearpopu- Service Biologist, Great Smoky Mountains National lation as a source of both protein and carbohydrateen- Park, Gatlinburg,Tennessee, USA, personal communi- ergy. The protein found in huckleberriesand other soft cation, 1997) Production by northernred oak during masts is available duringa period when bears are physi- 1995 was the highest recordedlevel in NPS mast survey ologically focused on assimilatingproteins for growthand history (1979-97). muscularmaintenance (Eagle and Pelton 1983, Nelson et Our analyses determinedgross energy ratherthan di- al. 1983, Brody and Pelton 1988). Insects availabledur- gestible energy. The amountof energy availablefor bear ing this period have a high protein content (Eagle and use will be different,depending on the chemical compo- Pelton 1983) and constituted 11%of annualvolume in- sition of the food and its digestibility by black bears. dex of scat in GSMNP (Beeman and Pelton 1980). How- However, we used gross energy as an acceptablerepre- ever, huckleberryand other soft masts constituted37% of sentationbecause trials to determinethe digestibility of scat volume (Beemanand Pelton 1980), and althoughsoft each food species wouldrequire captive facilities and great mast species generally have moderate protein contents expense. (Eagle and Pelton 1983, Inman 1997), increased intake Results of this study strengthenthe hypothesisthat oak can offset deficiencies in nutritionalcontent (Crampton productivityhas been of foremost importancein provid- and Harris1969). Thus huckleberryand other soft masts ing energy for bearsin the Smoky Mountains. However, are likely to be significantcontributors to the annualpro- soft mast was responsiblefor 26%of the energyavailable tein intakeof bears. to bears and may provide even more resources to bears Because soft and hardmast fruits have differentmeth- duringyears of mediumto low hardmast production. Soft ods of seed dispersal,production by soft mast crops may mast may affect population dynamics of bears in the be less variablethan that of hardmast crops. and Smoky Mountainsbecause it has differentphysiological hickory nuts are destroyedby consumption. Hard mast usage than hard mast and provides an abundanceof en- producershave gained selective advantageby occasion- ergy when hardmast is not available. Squawroot,a para- ally producing large crops that saturateseed predators sitic plantthat grows on the roots of oaks, was the second (Silvertown 1980, Sork et al. 1993). Soft mast species, most productive species on the study area, yielding al- on the other hand, have seeds that are scarified and dis- most 16% of available energy. Squawrootgenerates a persed when fleshy fruits are eaten; therefore,consump- surge of carbohydrateenergy (Seibert and Pelton 1994) tion enhances reproductive potential. From an duringlate spring and early summer. After 4-5 months evolutionaryperspective, hard mast producershave se- with no energy intake, bears, especially females with lective advantageby investingin productionof occasional young, are in great need of metabolizableenergy. This large crops whereas soft mast producersgain advantage may be of particularimportance after years of mediumto with relatively stable, high production. Soft mast, there- low hard mast productionand the resultantlack of re- fore, may providea relatively stable source of energy for mainingfat stores. Females with cubs are the most active bears, whereashard mast fluctuates. sex-age group (Garshelis and Pelton 1980, Villarrubia Black bears are large carnivores whose diet consists 1982), and this increasedactivity is likely in response to mostly of vegetation. Consequently,they have a relatively increased nutritional demands in support of lactation. low basal metabolismthat results in low fecundity,long Because squawroot is usually abundantin the habitats generationtime, andaltricial young (McNab 1989). Con- where it does occur,it is a food thatmay be acquiredwith straintson carryingcapacity may be determinedduring relatively little effort. We recorded the presence of the period of lowest energetic availability. Bears have squawrootin 3 of the top 4 vegetation types used by fe- adaptedto low caloric availabilityduring winter with fall male bears in spring (van Manen 1994) in GSMNP. Al- hyperphagiaand winter hibernation,effectively increas- though fall mast crops can be good enough to result in ing the energeticavailability during winter via fat storage high reproduction(Eiler et al. 1989), survival, particu- and inactivity. We found an additionalperiod of rela- larly of cubs, could be affected by the presence of suffi- tively low caloric productionduring summer. More ver- ENERGETICPRODUCTION INTHE SMOKY MOUNTAINS * Inman and Pelton 67 tebratespecies consume fleshy fruits than consume hard blackbear populations inIdaho. International Conference Bear mast (Martin et al. 1951), and inter-specific competi- Researchand Management 4:201-204. and M.R. PELTON. tion may be most intense duringsummer. Nuisance bear BEEMAN,L.E., 1980. Seasonal foods and feeding of black bears in the Mountains. activity in GSMNP typically begins when squawroot ecology Smoky InternationalConference Bear Research and is Management availability decreases, highest during mid-summer 4:141-147. when we found lowest and de- energetic availability, BRODY,A.J., ANDM.R. PELTON. 1988. Seasonal changes in creases with the onset of fall food availability (Singer digestionin black bears. CanadianJournal of Zoology and Bratton 1977, Stiver 1991). In years when fall mast 66:1482-1484. crops are average-to-good, such as 1995, caloric avail- COLEY,A.B. 1995. Populationdynamics of blackbears in Great ability far exceeds what the bearpopulation can consume Smoky Mountains National Park. Thesis, University of (Inman1997). Thus, availabilityof food in fall may limit Tennessee, Knoxville, Tennessee, USA. the populationonly during years of severe mast failure. CRAMPTON,E.W., ANDL.E. HARRIS. 1969. Appliedanimal A thresholdlevel of low fall mast nutrition.W.H. Freeman and Co., San Francisco,California, availabilitymay nega- USA. tively affect the survival and reproductionof bears. EAGLE,T.C., ANDM.R. PELTON.1983. Seasonal nutritionof Both hard and soft masts serve functions necessary black bears in the Great Smoky MountainsNational Park. for bears and are of bear habitat. importantcomponents InternationalConference Bear Research and Management Differences in usefulness of soft and hard masts result 5:94-101. from the timing of productionin conjunctionwith nutri- EILER, J.H., W.G. WATHEN, AND M.R. PELTON. 1989. tional content and bear physiological needs. Inherent Reproductionin blackbears in the southernAppalachian differences in annual fluctuations of productivity may mountains. Journalof Wildlife Management53:353-360. also influence the utility of each mast type for bears. It ELOWE,K.D., ANDW.E. DODGE. 1989. Factorsaffecting black bear success and cub survival. Journal is likely that southernAppalachian bear populations with reproductive of Wildlife Management53:962-968. access to ample carbohydrateenergy and proteinduring GARSHELIS,D.L., ANDM.R. PELTON.1980. Activityof black springand summerhave a lower median of age primipar- bears in Great Smoky MountainsNational Park.Journal of and inter-birthinterval as well as rates of cub ity higher Mammalogy61:8-19. survival, breeding success, and sub-adult survival than , AND . 1981. Movementsof blackbears in the those populations that rely solely upon fat stores from GreatSmoky Mountains National Park. Journal of Wildlife the previous fall. Management45:912-925. GOLDEN,M.S. 1974. Forestvegetation and site relationshipsin the centralportion of the of GreatSmoky Mountains National MANAGEMENTIMPLICATIONS Park. Dissertation, University of Tennessee, Knoxville, Tennessee, USA. Effective habitat managementfor black bears in the GOLLEY, F.B. 1961. Energy values of ecological materials. southernAppalachians requires abundant oak standsthat Ecology 42:581-584. are matureand of both red and white oak composed spe- INMAN,R.M. 1997. Caloric productionof black bear foods in cies This furtherconfirms that oaks (Pelton 1989). study GreatSmoky Mountains National Park. Thesis, University and oak habitatsare of great importancefor bears in the of Tennessee, Knoxville, Tennessee, USA. southernAppalachians. Ourresults also indicatethat the JOHNSON,S.R., ANDR.J. ROBEL. 1968. Caloric values of seeds abundanceof energy found in squawrootand the low en- from four range sites in northeastern Kansas. Ecology ergetic availability during mid-summerhave the poten- 49:956-961. ANDI.M. COWEN. tial to significantly influence bear populationdynamics. JONKEL,C.J., 1971. The black bear in the forest. Wildlife Further study is needed to determine the effects of spruce-fir Monographs27. KASBOHM,J.W., M.R. VAUGHN,AND J.G. KRAUS.1996. Effects squawrootabundance upon survival, recruitment,age of of gypsy moth infestation on black bear reproductionand primiparity,and litter interval. The of mid- relationship survival. Journalof Wildlife Management60:408-416. summerfood abundanceand bear also warrants density MACKENZIE, M.D. 1993. The vegetation of Great further Smoky investigation. Defining the role of soft mast will Mountains National Park: past, present, and future. be a positive step toward effective habitatmanagement Dissertation,University of Tennessee,Knoxville, Tennessee, for bearsof the southernAppalachians. Once the roles of USA. majorfoods arewell understood,appropriate habitat com- MARTIN,A.C., H.S. ZIM,AND A.L. NELSON. 1951. American wildlife positions and silviculturalprescriptions may be defined. and plants: a guide to wildlife food habits. Dover Publications,Inc., New York,New York,USA. MCLEAN,P.K. 1991. The demographic and morphological LITERATURECITED characteristics of black bears in the Smoky Mountains. BEECHAM,J.J. 1980. Some populationcharacteristics of two Thesis, Universityof Tennessee,Knoxville, Tennessee, USA. MCNAB,B.K. 1989. Basal rate of metabolism,body size, and 68 Ursus 13:2002

food habits in the OrderCarnivora. Pages 335-354 in J.L. SILVEnTWN,J.H. 1980. Theevolutionary ecology of mastseeding Gittleman, editor. Carnivore behavior, ecology, and in trees. Biological Journalof the LinneanSociety 14:235- evolution. CornellUniversity Press, Ithaca, New York,USA. 250. NELSON,R.A., G.E. FOLK,E.W. PFEIFFER,J.J. CRAIGHEAD,C.J. SINGER,F.J., AND S.P. BRATTON.1977. Black bear management JONKLE,AND D.L. STEIGER.1983. Behavior, biochemistry, in Great Smoky Mountains National Park. NPS-SER and hibernation in black, grizzly, and polar bears. Research\ResourcesManagement Report No. 13. InternationalConference Bear Research and Management SOILSURVEY. 1945. Sevier County. U.S. Department of 5:284-290. Agriculture, University of Tennessee Agricultural NOYCE,K.V., ANDD.L. GARSHELIS.1994. Body size and blood Experiment Station and Tennessee Valley Authority, characteristicsas indicatorsof condition and reproductive Gatlinburg,Tennessee, USA. performancein black bears. InternationalConference Bear .1953. BlountCounty. U.S. Departmentof Agriculture, Researchand Management9(1):481-496. Universityof TennesseeAgricultural Experiment Station and PELTON,M.R. 1989. The impactsof oak mast on black bears in Tennessee Valley Authority,Knoxville, Tennessee, USA. the southern Appalachians. Pages 7-11 in C.E. McGee, SORK,V.L., J. BRAMBLE,AND 0. SEXTON.1993. Ecology of mast- editor. Proceedingsof the workshop: southernAppalachian fruitingin threespecies of NorthAmerican deciduous oaks. mast management.The Universityof Tennessee,Knoxville, Ecology 74:528-541. Tennessee, USA. STEPHENS,L.A. 1969. A comparisonof climatic elements at POZZANGHERA,S.A. 1990. The reproductivebiology, winter fourelevations in the GreatSmoky Mountains National Park. dormancyand denning physiology of black bears in Great Thesis, Universityof Tennessee,Knoxville, Tennessee, USA. Smoky Mountains National Park. Thesis, University of STIVER,W.H. 1991. Populationdynamics and movement of Tennessee, Knoxville, Tennessee,USA. problem black bears in Great Smoky Mountains National QUIGLEY,H.B. 1982. Activity patterns,movement ecology, and Park. Thesis, Universityof Tennessee,Knoxville, Tennessee, habitatutilization of black bearsin GreatSmoky Mountains USA. NationalPark. Thesis, Universityof Tennessee, Knoxville, STUPKA,A. 1960. GreatSmoky Mountains National Park natural Tennessee, USA. historyhandbook. No. 5. U.S. GovernmentPrinting Office, ROGERS,L.L. 1976. Effects of mast and berrycrop failures on Washington,D.C., USA. survival, growth, and reproductivesuccess of black bears. THORNTHWAITE,C.W. 1948. An approach toward a rational Transactionsof the North American Wildlife and Natural classification of climate. GeographicReview 38:55-94. Resources Conference41:431-438. VANMANEN, F.T. 1994. Black bear habitatuse in GreatSmoky 1987. Effects of food supply and kinship on social Mountains National Park. Dissertation, University of behavior,movements, and populationgrowth of black bears Tennessee, Knoxville, Tennessee, USA. in northeasternMinnesota. Wildlife Monographs97. VILLARRUBIA,C.R. 1982. Movement ecology and habitat SCHOEN,J.W. 1990. Bear habitatmanagement: a review and utilization of black bears in Cherokee National Forest, futureperspective. InternationalConference Bear Research Tennessee. Thesis, University of Tennessee, Knoxville, and Management8:143-154. Tennessee, USA. SEIBERT, S.G., ANDM.R. PELTON. 1994. Nutrient content of WHITTAKER, R.H. 1956. The vegetation of Great Smoky Squawroot, Conopholis americana, and its importanceto Mountains. Ecological Monographs26:1-80. southern Appalachian black bears, Ursus americanus Received: 24 1998. (Carnivora:Ursidae). Brimleyana21:151-156. April Accepted: 18 September2001. Associate Editor: Craig McLaughlin.