University of Tennessee, Knoxville Trace: Tennessee Research and Creative Exchange

Masters Theses Graduate School

8-1977 The eS asonal, Altitudinal, and Vegetational Incidence of Black Bear Scats in the Great Smoky Mountains National Park Susie Jo Kelly Matthews University of Tennessee - Knoxville

Recommended Citation Matthews, Susie Jo Kelly, "The eS asonal, Altitudinal, and Vegetational Incidence of Black Bear Scats in the Great Smoky Mountains National Park. " Master's Thesis, University of Tennessee, 1977. https://trace.tennessee.edu/utk_gradthes/2580

This Thesis is brought to you for free and open access by the Graduate School at Trace: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Masters Theses by an authorized administrator of Trace: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council: I am submitting herewith a thesis written by Susie Jo Kelly Matthews entitled "The eS asonal, Altitudinal, and Vegetational Incidence of Black Bear Scats in the Great Smoky Mountains National Park." I have examined the final electronic copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Master of Science, with a major in Wildlife and Fisheries Science. Michael R. Pelton, Major Professor We have read this thesis and recommend its acceptance: Boyd L. Dearden, Ralph W. Dimmick Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School (Original signatures are on file with official student records.) To the Graduate Council:

I am submitting herewith a thesis written by Susie Jo Kelly Matthews entitled "The Seasonal, !\lti­ tudinal, and Vegetational Incidence of Black Bear Scats in the Great Smoky Mountains National Park.'' recommend that it be accepted in partial fulfillment of theI require­ ments for the degree of Master of Science, a major in and Fisheries Science. with Wildlife

,. < :•<, I-·- ""'�1'""'"i_c..,.h_ ' a_ e--::-l--=R:-" ....._..,t,.,... ) e_,..l_t_ o_ri , a j o r r o e s s o r � 1 F f

have read thesis We this

Accepted for the Council: of�--� Vice Chancellor Graduate Studies and Research THE SEASO::-JAL, ALTITUDINAL, A:.ID 1/EG:ETATION_\L INCIDENCE OF BLACK BEAR SCATS I� T�E

GREAT Sf.IOKY :•!OUNTAL\S NATIONAL PARK

Thesis A Presented for the

�aster Science of Degree

The University of Tennessee, Knoxville

Susie I(el1.y t.To ��e1attlle\•iS August 1977 ACKNOWLEDG��NTS

I am grateful to Dr. Michael R. Pelton, Associate

Professor of Forestry, The University of Tennessee,

Knoxville, who served as Director of Research.

I would like to express appreciation to Dr. Ralph

Dimmick, Associate Professor of Forestry, The University of Tennessee and Dr. Boyd Dearden, Assistant Professor of

Forestry, The University of Tennessee, who served on my co��ittee. Special appreciation goes to those friends and students who helped collect scats and other information from Index Routes.

Special thanks goes to Mike, Natty, and �cGregor for their patience, understanding, and support. Financial support for this study made available was through funds provided by Mcintire-Stennis Project �o. 12 of the Department of Forestry, Wildlife and Fisheries and

Agricultural Experiment Station, The University of Tennessee,

and the r at tur Knoxville G e Smoky Mountains �a al History Association. :\BSTR..A.CT

A study was conducted in tne Great Snaky Moun tains

N ation a Park to determine the variability o� incidence l of black bear scats as a re i l e to the p o p l tio l ab index u a n density. tot a of km were hiked along six dif­ A l 6085.7 ferent routes, each hiked biweekly, from June-October,

1970-1975, resulting in the of The collection 6S7 5cats. o ll of scats was 0.11 scats/km. vera incidence The Elkmont-

Bent ielde 0.21 s c a t / m , Sugarlands Ar� y d s k Index Rou�e 0.13 scats/km, Tremont-Derrick Index Route scats/kre, Knob 0.07 Gregory-Han::1ah Route 06 scats/l<.m, �.foun.tain I::1dex 0. and

Spence-Russell Index s ats / m . Variability 1n Route 0.03 c k the scats/krn over n ne hiking p 2 r i cds was partially the i to the behavioral activities of breeding. attributed Other a�d abundance factors such as rela v the ti e scarcity of summer a:1d fall fco:is, cl mat c corid:itions, and Dre- i i

may have l s o influenced variability. hiberna�ion activities a

The r /km co c: t.o ntunbe of scats lle ted prior ::tnd after the

period was less, in ca ses , except June-October time both Spence-Russell Rcute. Factors possibly fo� the Index less evidence bear presence along this rou:e causing of were the high degree hiker use (40 of percent higher than popular trail) the :arge the next illOS t and number of horseback riders.

iii iv the June-October period. The highest values of scats/km occurred between elevations of 1219-1371 m. Specific vegetation types in the higher elevations were also pre-

£erred over those same types in the lower elevations.

Since certain abundant and available berry crops occur only in the higher elevations, it is assumed that this accounts for the apparent preference.

When the number of scats/km were analyzed con- in junction with a combination factor of altitude-vege:ation, it was found that altitude had at least a percent 95 probability of influencing the location of scats along all routes except the Spence-Russell Index Route.

Numerous factors such as fluctuations in temperature, variations in precipitation, behavioral aspects of bears, etc., �ay influence the incidence of scats. Taking this into consideration when attempting to establish an inde- pendent index to monitor population fluctuations, it was

that certain statistical assumptions concerning found factors that influence the variability of scats incidence would need to be made if the index was to be statistically valid. As tr1ese assu.:nptions could not be made, a less sensitive still effective equation was determined. Used but in conjunction with the of scats collected annually, number

' the equation c:an. oe used as a check papulation fluctua- an

t io�1..s. TABLE OF CONTE�TS

CHAPTER PAGE I. I NTRODU CT O:J. I 1 II. DESCRIPTio:; OF ST'JDY Location and Physiography.-\REA 6 6� Climate ...... I Vegetation ...... 10

I I I. :-.!ATERL\LS .-'1.\JD THODS 20 Description of:-.IE Routes 20 Collection of Scats . JJ Analysis of the Data.

IV. RESULTS .-t'JD DISCUSSIO)J .. Incidence of Scats during Pre- and Post- Summer Seasons. . . . . Incidence of Scats during Sumr.:er Season 2S Distribution of scats �ithin hiking 31 periods ...... Distribution of scats within classes ...... altitudinal.... . Distribution of scats within 41 vegetation e s t t I t t I t I I I I I I I I t t JTP Distribution of scats w n vegetation types ...... ith i altitude- . Backcountry use scat incidence. Expected frequenciesand of scats 36 59 SU?>L\!ARY ,.I • CCNCLUS I .-\\D C\S 62 VI. REC0�0fE\D.-\TIO\S 66

LITERATURE CITED. 68

73 SHEET- INDEX A. SU:.l:.L\R'! TR).I L BL'..C( BL\R STUDY CCLLECT10)JS. .., - ,::y B. CODI\G CLASSIFICAT!C\S.

'7h SC-\TS CCLLECTED . IV c. TOT.:\L D. SC.-US/K�f I

?CR BY _:�L� T I �UDE- I t • I I � t I

,.'I vi

CHAPTER PAGE

E. DESCRIPTION OF I:.JDEX ROUTE LOCATIONS AND VEGETATION TYPES...... 78

F. SCATS COLLECTED BETWEE:.J OCTOBER AND 1 JUNE 82 15 G. At�OVA TABLE FOR SCATS/D! BY HIKING PERIOD . . 85

H. NUMBER OF SCATS COLLECTED WITHIN ALTITUDINAL CLASSES ...... 8 6

I. NU1vffiER OF SCATS COLLECTED WITHIN VEGETATION TYPES ...... 8 7

VITA. 89 LIST OF TABLES

TABLE PAGE

1. Vegetation Types and Their Dominant Species in the Great Smoky Mountains �a tional Park. 12

2 • Alt itudinal Descrip tion of Index Routes in the Great Smoky Mountains National Park ... 25

3. Vegetat ion Types Found Along Index Routes in the Great Smoky Mountains National Park .. 26

4 . Hike r Use of Index Routes Du ring the 1975 Summer Season Determined by Nine All-Day

Counts ...... 5 . Summary of the Number of Scats Co llected Along Index Routes in the Great Smoky Mounta ins

Nat ional Pa rk, 1970-1975 ...... 33 6. Scats/K"Y! Collected Within Altitudinal Classes Along Inde x Routes in the Great Smoky

Mountalns �Jational Park, 1970-1975 •..... 43

7 I o Altit udinal -Vegetation Type Breakdown by Index Route in the Great Smoky Mount ains Nat ional

1970-1975 .. , ...... Park , . . . 49 8 . Coding Used fo r Vegetation and Altitud inal

Classes, . . . c...... , 75

9. Total Numb er of Scats Collected Along Each Index Route and Within Each Ve getation Typ e in the Great Smoky Mountains National Park . 76 lC. Compl eted Jillova Table for Scats/K"Y! by Altitude -Vegetation Type in Great Smoky 1970-1975the . Mount ains National Park, . . . . 77

11. Dates and Numb er of Scats Collected Between 15 October and 1 June...... 82

12. Anova Tab le for Scats/ KM Hiking Period in the G r e at Smoky o unt a insby N a io n a 1 Park ,

19 0-197 . . �! • • . • t . 5 7 5. • ...... 3

vii viii

TABLE PAGE

13. Nu mb er of Scats Colle cted within Altitud inal Classes Along Index Routes in the Great Smoky

.Mountains National Park , 1970-1975 . . . . 86

14. Number of Scats Collect e d Along Index Routes Within Ve getation Types in Respect to Altitude Clas ses in the Great Smo ky Mountains National Park, 1970-1975 ...... 87 LIST OF FIGURES

FIGURE PAGE

1. Map of the Northwestern Qu arter of the Great Smoky Mountains National Pa rk and Index

Routes ...... , ...... 8 .

2 • Vegetation Pattern of the Great Smoky Mountains

(Shanks 1954b) ...... 11

3. Comparison of Scats /K?v1 Va lues for Pre - Summer, Surruner, and Po st-Summer Seasons ov er Six Years (1970-1975) in the Great Smoky

Mountains Nat ional Park . . . . . 29 4. Total Numb er of Scats Collected in the Great Smoky �.fountains National Park, 1970- 1975 . 35 5. Yearly Scats/K� Value s from 1970- 1975 in the Great Smoky Mountains National Park .... 37 6. Scat s /KM Within Vegetation Types in the Great Smoky Mountains National Park , 1970- 1975 ... 39

Relationship Between Scats/KM and the Nine 7. Hiking Periods , June-Oct ober 1970- 1975 , from

the Great Smoky '\fountain s :Jaticnal Park . . . . 40

8 . RelationshiD Bet,�·e en Scats/K:'v! and Altitudinal Classes i n the Great Smoky :V!ountains :�ational

Park, 1970-1 975 ...... 44

9 . Re�ressional Relat ionship Bet '\'>'e en Scats/K:'v! and Alt itude Classes in the Great Smoky Mountains

National Park , 1970- 1975 ...... 45

10. Re lat ionship of Scats / K?-.1 Within Veget ati on Classifications in the Great Smoky Mountains

Natio na l Park, 1970-1975 ...... 51

11. Re lat ionship Between the Summe r Season Scats/K:Vf Value s and the �;umb er of Backcountry (Numb e r of Backpackers Mu ltiplied by the Number of Ni ght s Spent Ou t) Nights for the Same Time Period ...... 57

lX CHAPTER l

INT RODUCT ION

Of primary importance in formulat ing a management plan for wildlife is the necessity to determine popul ation dens ity. Res ear che rs have devoted much time, effort , and resources in developing acc urate methods as sess ing for population numbers .

Current me thods ar e classified into five general groupings . These involve direct counts of an i mal s, capture of marked an imal s, reduction the population size of and subsequently the rate of capture , sele ctive or reduction increase , an d an imal signs an d related ob jects (Ove rton

1971) . Obviously, the mo st desirable method is a total count of a species . Howeve r, only in rare c as s a e is direct count p o s sib e . Methods involving marked animals l are more commonly used. hese methods require considerable T resources set traps, mark animals, reset traps , and to recapture an imals . Ano ther general met�od, involving the reduction of �he populat ion s i z e thus the rate and of capture , is based on the princip le that the rate of

succ essful capture decreases as the po pulat ion decreases;

in this in stance as the ab ove, th e time an d resources �n required ara considerable. The me thod of selective

or increas e, als o kn own as dichotomy method reduct ion the

or 1n compo sition, 1s more complex previous change than the ? methods discussed . The structure o= the population is first modeled into recognizab le classes of an imals and time . A population estimate is derived through use of various mathematical equ ations . In add ition to the mathe­ matical complexit y , the main limitation of this method is that the exact numb ers of animals at a specific time are not usually known . More frequently used is the me asure­ ment of sign or evidence of the presence of the species in quest ion. �,!ajor fact ors that persuade rc;s e a�chers to utilize this last method are lack of funds , equipment, and personnel.

A�imal signs, calls, or other indications of a species' pres ence are often us ed as i n dic a to r of popula­ s tion trends or as indexes to the relative ab undance a of spec ies . Depend ing on the the use of animal species, signs can yield a tot al coun t on one extreme , and on the other, indexes that are very difficult to cal ibrat e (i. e., calls) (Overton 1971) . Overton stated that , at best, th is techn ique can be cons idered a double sample which he de ­ fines as an extensive count of the sign with an extens ive analys is of the relat ionship between the variable counte d and animal . Watt (1968) felt that although me asur ing the an anirr1al popula tion the of signs le ft per by number ani�al has a great many pos sible sources of error, :n some it circumstances to use u l as an independe�t �ay prove be f

on other methcds sa�e check of estimating the populat�ons. 3

As for statistical properties and limitations encountered in using animal signs as an index to populations , Overton

(1971) stated that they are identical to those statistical propert ies and limitations encountered in other types of index counts of animals.

Numer ous papers concerning the use of animal signs as an index to populat ion density have been published .

One of the earliest was by McClure (1939) using dove call­

In g activity. Use of the auditory index has also been ext en ded to species such as ruffed grouse (Pet raborg et al .

1953, Hun gerford 1953) , woodcock (Kozicky et al . 1954),

Gambel quail (Smith and Gal l i:ioli 19 65) , chukar partr idge

(Williams 1961) , turkeys and even squirrel s (Overton 1971) .

Track counts have also been used in the same manner

(Overton 1971) .

The pel let count technique has been used numerous for species . Not only has it been utilized as an index to population level but in some cases it has been calibrated to produce an es timate of the populat ion number ; mo st widely documented are those studies pertaining to deer, elk, and other ungulates (Bennett et al. 1940, Eberhardt and Van Etten 1956, Van Etten and Bennett 19 6 5, Neff 19 68,

Batcheler 1975). The pellet count techn ique has also been used as an index to population trends fer rabbits (Tay lor

Wil liam 1956) and fo r red and fallow goats, and and deer, brushtail opossum (Riney 195�). Conspicuously absent from 4 the literature is mention of the pellet (or feces) count technique and its us e for carnivores.

While the literature contains numerous articles on black bear (Ursus ameri canus), little can be found on practical , yet accurat e me thods of estimating population densities. Davenport (1953) , in his research on agri - cultural depredation by bears, stated that he coul d find no accurate method for estimating population densities of bears . He resorted to estimating the ir numbers from kil l figures, hunter estimates and sign. Spenc er (1961) re- marked that , "No wholly satisfactory me thod has be en devised for censusing black bear.n Some previous tech- niques used direct counts (Barnes and Br ay 1967) , multiple capture (Eriks on and Pe trides 1964) , annual harvest data

(Carpenter 1973) , and radioactive feces tagging (Pel ton and �arcum 1974) . On ly one researcher reporte d us ing black bear sign. Spencer (1 961) used cruise lines ; these were random cruises of known mileages made foot and by on canoe during wh ich all signs (scats , tracks , mark ing posts, and stump workings) we re noted. With the ab ove informa- tion, in conjunction with kill estimates and vi sual ob servat ions, he was ab le to est imate populat ion densities and trends .

Pelton (1972) indicated that bears in their daily move�ents se em to ut ilize trails and als o tend to defecate

. , on the se was sugg este d that hi�ing tra11s trail.s. It by 5 and collect ing scats the data obt ained could be used as an in direct estimation of popul ation density or popula­ tion trends . The present study att empted to determine the feasibility of us ing scats as an index to trends in black bear populations by (1) delineating the seasonal , altit�d inal , and veget at ive distrib ution of scat s of bla ck bears in the Great Smoky Mountains �ational Pa rk

(Park or GS:.INP) , and (2) determining the degree of varia- bility of the inciden ce of scats on Index Routes by date and locat ion . CHAPTER II

DESCRIPT ION OF STUDY AREA

Locat ion and Physiography

The study area is located at the southe rn extreme of the Appalachian Mountains in the Blue Ridge Province.

Within the southe rn sect ion of th is province is the Unaka

Mountain Range which includes the Gr eat Smoky �ountains

Nat ional Park.

The Park encompasses 20 72 km2 and is lo cated :n

sections of southeastern Tennessee and northwestern �o rth

Carolina. On the Tennessee side this includes parts of Blount , Cocke , and Sevier Counties and on the No rth

Caro l ina side it includes sections of Swain and Haywo od

Counties .

Large sect ions of the Park's boundaries are als o

share d with the Cherokee �ational Forest, the Nantahala

National Fo rest , the Pis gah �ational Forest , and the

Cheroke e Ind ian Reservation. The Little Tennessee River and Fontana Lake als o border sections of the Park's

boundary.

Excluding maintenance ro ads , wh ich are closed to

pub lic vehicles , only two maj or paved roads are lo cate d 1n

the Park; S. Highw ay bis ects the running no rth U. 4�1 ?ark to Gat linburg, Tenne sse e, �orth south between and Chero\ee,

0 7

Caro lina and Tennessee Highway which connects Park 73 Headquarters at Sugarlands and Cades Cove (Figure 1) .

The topography of the Park is dominated by the crest of mountains which run northeast to southwest.

For much of its length the altitude does not drop bel ow

1524 m. Elevat ions range from a low of 266.4 m at the

Abr am's Creek-Little Tennes see River junction to a high of 2024.5 mat Clingman 's Dome. The slopes of these rugged mo unt ains tend to be steep er and more on barren the Tennessee side than those on the No rth Carolina side.

Whittaker (1956) estimated that less than 10 percent of the land sur face in the Park has les s than a 10 degree slope.

Narrow V- shaped st ream valleys penet rate th e steep mo untain slopes . These stream valleys dr ain into six maj or rivers ; these are the Big and Little Pigeon Rivers,

Tuckaseegee River, Oconaluftee River , Little River , and Little Tennes see River (King and Stupka 1950) . There are no permanent lakes or ponds in the Park although numero us sp rings exist. Occas ional pending will occur in the cove areas where valleys are flat and broad , out this is only a seasonal phenomenon.

Cl imate

Outside of the tropical areas of the world there are few places that can cla im the veg etat ive that diversity i Park � • ·" !lead afthI ·s '\ '· .,�.-..r · r . '----'L- ...------. ..._____,; y --...-'---'\. J . . ..__.

: ••• '·, \

' ' l ' ' ' \ ,'lj Trails , Trail Shelters �-, / \ X .. Campgrounds Roads .... _ , ... / y / A � Gregory-Hannah Mountain 1 i Spence-Russell 0 1 2 3 32 Tremont-Derrick Knoh I I l l Elkmont-Bent Arm KM 4 Sugarlands 5

Figure 1. J\fap of the northwestern quarter of the Great Smoky Mountains National and Index Routes. Park

00 9 exists in the Great Smoky Mountains . A prime factor of this phenomenon is the cl imate . On an altitude-latitude equivalent the Great Smoky Mount ains climate and vegeta- tion most nearly resembles what one would experience if one would travel from east Tennessee to northeastern Maine and adj acent New Brunswick .

Temperature and precipitat ion ch ange noticeably with an increase in alt itude. Shanks (1954a) found precipita- tion amount s to have increased by percent at the lower SO level of the spruce-fi r forest (1372-1 524 m) when compared to the amount that falls at the base of the mountain range

(approximately 457 m). The me an annual precipitat ion ranges from 127-152 em at the low er elevat ions to 229 ern at elevations over 1829 m (Shanks 195�a, Murless and Stallings

1974) . The driest month of the year is usually September, with the we ttest being July and Au gust. Temperature s can be expected to decrease 1.2if C with eve ry gain of 304m in

-' ' - altitude . Temperature s in the sp ruce-fir forest C 1 " '"1.,I -

1524 m) can be exp ected to be 5.6°-8 .3° C co oler tha� those

at the base of the mountains or in adjace nt valleys . The

average Ju ly temperatures ran ge from 23.6° C in Gatlinburg

at the base of the mountains to 14.7° C at Newfound Gap .

In January the me an temperature in Gatl inbur g is 3.9° C

and at Newfo un d Gap , -5. 2° C. 1 0

Vegetation

The vegetation of the Great Smoky �fountains :.Ja -

tional Park is considered to be one of the mos t divers ifie d

and unique in North America. The Blue Ridge Province has been continually occupied by plant life for perhaps 200 million years while othe r geographical regions of the country have been glaciated, flooded , and/or exposed to various forms of great climat ic and phys ical changes

(Cain 1937) . Thus , the Great Smoky Mountains area acted

as a haven and redistribution center fo r countless spe cies

that had been ob literated in other geographical regions.

Due to its divers ity and subs equent complexity, a thorough

description of the flora of this area has proven to be a

difficult task . The most thorough description of the

vegetat ion of the area was accomplished by Whi ttaker (1956) .

He established vegetat ion patterns , vegetation types, and

their distributional relations.

Shanks (1954b) established a physiographic site-

elevation-moisture category into �hich all dominant tree

1 \ species could be placed (Fi gure .l.) • Shanks ' six categ ories include cove hardwood , northern hardwood , hemlock, sp ruce-

fir, closed oak, and the combinat ion of open oak-pine and

bald s (F igure 2 a n Tab le 1). d Cove hardwood forests are character istic of cove

areas she ltered slopes 1372 m. and up to elevations of De ep, drained are well soi:s prerequisite to this forest type. 11

1970

1820 Spruce-Fir Forests

16 77

1525

,.--_ s 1372 '-'

1067 Ope n Oak and Pine 914 Stands

H H 763 H H H 610 H H

Sheltere d-----(Topographic Pos ition)--- - -Exposed

Figure Vege tation pattern of the Gr eat Smoky 2. '0<:;1}-) (Shanks .1 _, ...__, "-! ,_) ., �cuntains 12 TABLE 1 VEGETATION TYPES A\JD THE IR DOMINANT SPECIES IN THE GREAT S�!OKY :.!OUNTAINS �ATIONAL PARK

Vegetation Type DomL�ant Species

Cove hard�ood Ca�opy Trees

Basswood (Tilia heterophylla) Beech (Fagus �randifolia) Buckeye (Aesculus octandra) Hemlock ( su a canadens1sj Silver bei 1 tHales1a carolina var. monticola) Sugar ITaple (-�er saccfiarum) Yellrni birch {Befula allegneniensis) Yellrni poplar (Lu10aendron tul1p1tera)

Small Trees and Shrubs

temate-leaf dogwood (Cornus alternifolia) AlBear huckleberry (Gaylussac1a urs1na) Hearts-A-Bustin (Euonvuus amer1c��L�) Mountain laurel (Kalm1a lati±olla)-­ Service berry (�melanchier Iaevis) Witch Hazel (Hamrunehs Vi'Ygin1ana)

HeiT�ock Canopy Trees

Beech Hemlock Silver bell Yellow birch Yellow poplar

Small Trees and Shrubs

.Alternate-leaf dogwood Mountain laurel Rhododendron (f\i'lododendron mxirnum) Purple rhododendron (B:_. catahbiense)

:iorthem Canopy Trees hardwood Beech Yelloh Fire cherrybirc.� (�:rtmus pensylvanica) Buckeye 13

(continue d) TABLE 1

Vegetation Type Dominant Species

Small Trees and Shrubs

Striped maple (Acer ensylvanicum) Hydrangea (Hydra""Ti'gea ar orescens) Dog hobble (leucothoe-editorum) Canada blackberry (RUbus canaaensis) Rhododendron Purple rhododendron

Spruce-fir Canopy Trees

Fraser fir C-�ies fraseri) Red spruce (1-lcea rUbens) - Mmmtain ash (Sorbus americana)

Small Trees and Shrubs

Striped maple Rhododendron (Rhododend:ron carolina) High-bush blueberry l Vaccir.luJn SlJ!l.ulatum) 01ountain cranberr1 (V. alnitOTium) Red-berried elder (Si11bucus Plr?ens)

Closed oak CanaDy Trees

Red oak (Querc:tlS rubra) te oak (Q. arb a) 1�hi oak (Q. veiUtina) BlackSour.vood (O'.Aydendron arboreL..nn) Pignut hickory (Carya glabra) Mockernut hickory to11l.entosa) Sweet birch (Betula-lenta)LC.

Small Trees and Shrubs

Floweri.T'lg dog<.vood (Comus florida) Bear huckleberry High-bush blueberry �·lotmtain laure Rhododendron 1 FlaJne azalea (Rhododendron calendulceum) 14 TABLE 1 (continued)

Vegetation Type Dominant Species

Open oak-pine Canopy Trees Virginia pine (Pinus virginiana) Pitch pine (P. r1dgida) Table mou nta1n pine (P. pungens) Scarlet oak (Quercus cocc1nea} Chestnut oak (Q.- pr1nus) hhite oak

Small Trees and Shrubs

Mol..IDtain laurel High-bush blueberry Rhododendron Deerberry (Vaccinium staminelJ11)

Heath balds ;'-.1ormtain laure 1 Rhododendron Flarre azalea Blueberries

Grass balds Blueberries Blackberries (Rubus sp.) Mou;•1tain oatgrass (Danthonia comoressa)

Sources: Shanks 1954b, Whittaker 1956. 15

Eight tree species, six dominant and two subdominant, coexist in the cove hardwood forest with each constitut- ing 10-20 percent of the forest canopy. These species include hemlock (Tsuga canadensis), silver bell (Halesia carolina var. monticola), buckeye (Aesculus octandra), basswood (Tilia heterophylla), sugar maple (Acer saccharum), yellow birch (Betula allegheniensis), yellow poplar

(Liriodendron tulipifera), and beech (Fagus grandifolia).

Whittaker (1956) found that these eight species, in various percentage combinations, make up 80-90 percent of the cove hardwood forest canopy. Below this canopy, a variety of small trees and shrubs make up a broad, but not dense, layer. The herbaceous layer is by far the richest to be found among the six forest types with over a hundred species to be identified during the spring alone (Whittaker

1956, Cain 1937).

Sheltered stream valleys, up to elevations of m, 914 and those areas along the more exposed ridges, up to m, 13 72 are dominated by eastern hemlock. The more important associates of hemlock are silver bell, yellow birch, and occasionally yellow poplar and/or beech (Whittaker 1956).

The hemlock association is dense and dark, providing very little herb stratuin. The acidic soil conditions found under many hemlock fores�s partially explains this pheno1r.enon. The sr.1all tree shrub aye r is oft en and 1 occupied by dense thickets of rhododendron (Rhododendron 16 and/or combinations of nume rous heath species, all of i..,;h ich thrive in acidic soil conditions . Whittaker (1956) noted that at lower elevations stands are mixed with hard­ woods, have less he ath, and a more abundant herbaceous laye r. At higher elevations stands are densest with a thick growth of heaths and little or no herb layer. At the lowest elevations the hemlock forest merges with the cove hardwood forests and hemlock becomes only one of the dominant spe cies.

Located at alt itudes of m and above is t�e 1372 northern hardwood forests , dominat ed by beech and yellow birch. Sometime s referred to as the beech orchards, this forest type is characterized by its open understory with a collection of herbs, sedges and ferns, small trees, and shrubs. Undergrowth is slight if not altogether ab sent .

Whittaker (1956) suggests that the short and stunted ap­ pearance of the dominant tree species as well as the obviously open understory may be attributed to the ex­ treme climatic conditions at this altitude .

The spruce-fir forests are located primarily at altitudes of 1372 m and above . The dominant tre e species are Fraser fir (Abies fraseri) and red spruce (Picea rubens) . Pure stands of Fraser fir may be found above

1829 m. The only ether tree species of maj or importance 1n this ass ociation is mountain ash (Serbus ame ricana) . 17

Depending on the exposure of the slope, undergrowth var1es from a 90 percent coverage of moss and oxalis on north­ facing slopes to percent or less coverage on south­ 60 facing slopes with almost 100 percent coverage by rhodo­ dendron on ridges. While undergrowth is sparse in this forest type, mosses and lichens abound, covering the ground and trees liberally.

The next forest type described is a combinatiJn of

Whittaker's chestnut-oak heath, red oak-chestnut and white oak-chestnut forest types. Shanks (1954b) has called this group of species the closed oak forest. Species of this association are located on intermediate to dry slopes and are dominated primarily by oaks and, prior to the destruc­ tion of the chestnuts by the chestnut blight, by chestnuts and oaks. Some of the dominant tree species today are red oak (Quercus rubra)' white oak cg_. alba) ' chestnut oak

(Q. prinus) , black oak (Q_. velutina) , sourwood (Oxydendron arboreum), pignut hickory (Carla glabra) , ;-r.ockernut hickory

(£. tomentosa), and sweet birch (Betula lenta) . The under story is dominated by heaths but not usually a is continuous stratum. Common to this layer of the forest canopy are mountain laurel (Kalmia latifolia), rhododendron

(Rhododendron maximum), flame azalea (�. calendulaceum), sweet shrub (Calycanthus fertilis), and hydrangea

(Hydrangea ar�orescens) . 18

The last classification, a combination of two types, open oak-pi�e and heat h balds, is composed of those species common to the dry exposed and often rocky ridges and slopes. At lower elevations, Virginia pine (Pinus virginiana) and scarlet oak (Quercus coccinea) are dominant. These forests are comparatively open with only

70 percent tree coverage, 10-40 percent shrub coverage, and 2-10 percent herb coverage. Above the Virginia pine­ scarlet oak grouping (617-975 m) the pitch pine (Pinus ridgida) is dominant. Often found with this species are scarlet oak and chestnut oak. This association is similar to the previous as it is also relatively open, but differs

1n that the shrub layer is denser; the coverage being

40-70 percent. Herb cover is low at 5-2 0 percent coverage.

Above 975 m is the third group of pines. Dominated by table mountain pine (Pinus pungens) tree coverage 70- 80 percent is with scarlet oak, pitch pine and chestnut oak appearing as codominants at lower elevations. Shrub coverage is high at

60-90 percent, dominated by mount ain laurel (Whittaker 1956).

Herb coverage is again low at 5-20 percent.

The most distinctive and con�roversial type of vege­ tation to be found in the mountains are the balds. Found at higher elevations along ridges and mountain tops, balds are either grassy or heath covered. Grassy balds, believed

the result of fire and/or grazing, are dominated to be by mountain oat grass (Danthonia . heath balds compressa; The 19 occur between 1219 m and the uppermost peaks. At lower elevations mount ain laurel is the dominant heath while at higher elevations catawba rhododendron (Rhododendron catawb iense) dominates . CHAPTER III

RI-\TERIALS A\JD �!ETHODS

In 1970, five different hiking routes totaling

130.0 km were established in the northwestern quarter of the Park to collect black bear scats. These rout es con­

sist of trails previousl y established by the �at ional Park

Service. All five routes are located we st of U. S. Highway

441 and north of the Appalachian Trail and the �orth

Carolina state line (Figure 1, page 8) .

Description of Routes

The easternmos t route, Sugarlands , is 17.2 km in

length , ranges in elevat ion from 57 3 n: to 1810 m, and is

characterized by three general vegetation types: spruce ­

fir, open oak-pine and balds , and clos ed oak forests .

This route begins at the Spruce-Fir Nature Trail on the

Clingman 's Dome Ro ad and ends on U. S. Highway 441 (Figure 1

and Appendix E) .

The Elkmont-Bent Arm Route lies west of the Sugar­

lands Route, is 32 .8 km in length, ranges in elevation from

792 m to 1709 m, and is characterized by two general

ve getat ion typ es : closed oak and northe rn hardwoods

fo rests . Th is route begins on the Little Rive r Gravel

Road the Cuetunb er Gap trail head and terninates the at on 20 21

River Gravel Road at the Goshen Pro�g trail head Little (Figure 1, page 8, and Appendix E).

Sharing a pr o ximatel with the Elkmont-Bent p y 4.8 km

Arm Route is the Tremont-Der rick Knob Ro te . It is u 24.8

km in length, ran ges elevation from m to m, 1n 792 1578

1 ' ' and is characterized by closed oak and northern narawooas

forests. The Tremont-Derrick Knob Index ?oute begins at

the Panther Creek trail and terminates at the end of head the Davis Ridge Trail (Figure and Appen�ix E). 1 The Spence-Russell Route is length, 2D.l km in

ranges in elevation from 590 m to m, is char- 1548 and

acterized by five general vegetat on types: hemlock, i

cl osed oak, northern hardwoods, cove hardwoods, and open oak-pine and route begins in the Cades Cove balds. This

Picnic round Creek ��ead, T:l.a�

In�Je:< ..ot:�e. .1 kiil ele'�/a- �'·1CiJntain F �tis 3� lor1g, 1·ange s in

general vege t ation types: closed oak, open oak-pine and

balds, cove hardwoods, and northern hardwoods. The route

be g i n at the G egor y Ridg trail head at the end of Fo rge s r e c�eek I�oad i:-t and terr:.inates t�c �\brar.lS Falls pa.rking aYe2.. 22

Collection of Scats

Between l October and 15 May routes were hiked on an irregular basis. Between 15 May and 1 October all rout es were hiked biweekly, during which time all routes were hiked within a two- to four -day period. Scats collected du ring these periods were used in other studies on food hab its (Be eman 1971) , radioactive tagging of feces (�arcum 1974 , Eagar 19 77) and move�ent studi es

(Eubanks 1976) .

Scats of black bear were collected and records were maintained on the exact date and loc ation of each; any remain ing scats were always brushed from the trail to prevent colle cting the same material at later date. a Information regard ing -weathe r conditions , time of day, condition of the scat , the number (day or over­ and type night) of hikers me t along the and any other routes, pertinent data were recorded on Index Rou te Dat a Sheets

(Appendix A) . Scats were frozen upon return to the lab.

Scats collect ed around trail shelters in proximi ty to

Index Route s not included in the dat a anal y3i s. were

An alysis th e Data of number of scats collected from one rout e cannot The

i r e c om p are d t o the n w!lb e r co 11 e c t e d f l' om an c the r � e d c t 1 y since rout es are of unequal l ngths . For comparative e purposes it was necess ary to convert the scats number of 23 collected to a figure indicating a per unit effort; this was accomplished by transforming all figures into a scats/km value (to convert any scats/km value to km/scat, divide the number one by the scats/km value).

The major portion of the analysis concerned data collected from 1 June through 1 October, a period of time consisting of nine biweekly hikes. Prior to the summer season hiking period of 1 June - 1 October, Index Rou tes were cleared of scats during the 15 May hiking period.

This was to insure that scats collected on 1 June would be no more than two weeks old. Any scats found 1vere included in the pre-suurner season data. The remainder of the analysis utilized data collected from 1 October through

1 June, a series of irregular hikes. Influencing this decision was early work by Pelton (1972), who found that black bear activity in the Park increases appreciably after

May and decreases sharply after September, and the fact that only during the time from 1 June - 1 October were all routes hiked regularly and all infoiillation recorded. Data collected on a regular basis cannot be sta tistically analyzed in conjunction with data collected on an irregular basi3, however, a comparison can be made between the su�ner periods (1 .June - 1 October) and the pre- and post-swnmeT season periods (1 October - 1 June) when each set of data

is described on a scat per kilometer (Appendix F) . basis 24

Locations for each scat collected between 1 June -

1 October were recorded as accurately as possible in terms of distance from a trail head, trail junction, streaD crossing , or other landmark. These locations were then marked on USGS topographic maps to determine the altitude

(to the nearest 12 .2 m contour interval) for each scat.

Using the clas s ification sys tem formulated by

Shank s (195 4b) , in addition to per sonal observation, each scat was categorized into a specific vegetation type.

Individual scat locations could then be ident ified using route name, date of collection (year, month, and day), altitude, an d vegetation type . This information was then coded to facilitate use of a computer for subsequent analysis of variance tes ting (Barr and Goodnight 1976).

Altitudinal classifications were arbit rarily set at

15 2.4 m intervals beginning at 45 7. 2 m and progressing to

16 76+ m. Each class was numbered con secutively, resulting in nine different altitude clas ses (Appendix B). An altitudinal description of (1) individual routes and (2) all routes combined are presented in Table 2; this table includes the dis tanc e each route covers wi thin the re ­ spective altitude cl ass and the percentage of the total length of the route which occupies each altitude clas s.

Vegetative characteristics for each route are similar ly described (Table 3) . Six vegetative classes were us ed : cove hardwoods , closed cak , open oak -pine and TABLE 2

ALT ITUDINAL DESCRIPT ION OF INDEX ROUTES IN TilE GREAT SMOKY MOUNTAINS NAT IONAL PARK

Index Routes Tremont - Gregory- Elkmont - Derrick Spence- Ilannah Overall Altitude Sugarla nds Bent Knob Rus sell Motmtain Total _ Ann 0 0 (m) 11 1i % % KM % Kiv1 % lQv1 KJv1 KM Kiv1

------_, - - - 457- 609 0.5 2.7 ------5.8 15.3 6. 3 4.7

610- 762 1.0 5.5 0.2 0.6 ------5.4 25.5 12.4 32 .6 19.0 14 .4

763- 913 1.3 7.1 4.3 13.7 2.3 9.1 2.6 12.3 8.2 21.6 18.7 13.9

914-1066 2.1 ll. 5 4.8 15.3 3.1 12.2 1.8 8.5 3.4 8.9 15.2 11.3

1067-1218 1. 3 7.1 2.9 9.2 2.9 11.4 3.8 17.9 3. 2 8.4 14 .1 10.5

1219-1371 4.5 24 .6 3.7 11 .8 5. 5 21.7 3. 4 16.0 2.1 5.5 19 .2 14.3

1372--1524 1.3 7.1 4.3 13. 7 9.7 38 .2 3.4 16 .0 2.9 7.6 21. 6 16.1

--- 1525- 1676 3. 2 17. 5 8.4 26 .8 1.9 7.5 0.8 3.8 - -- 14.3 10.6

-- -· - -· - --.. -- 3.1 16 .9 2.8 8.9 ------5.9 4.4 167 7+

10 tfl TABLE 3

VEGl._<;TAT ION TYPES FOUN D ALONG INDEX ROUTES IN TilE GREAT SMOKY MOUNTAINS NATIONAL PARK

Inaex Routes -cregory- Elkmont ­ Tielnon.r.:.:ne rrick Spence - Hannah Vegetation Sugar1ands Bent Ann Knob Rus sell fvbLmtain Total T e KM % )(1\f 1J KM 9o KM % I

Cove 3.0 40.0 4.5 60 .0 7.5 5.6 llanlwood

Closed 3. 4 6.8 17.8 35 .9 14.8 29.8 4.4 8.9 9.2 18. 5 49.6 37.2 Oak

Op en Oak ­ 6.9 22. 5 3. 5 11.4 20 .3 66 .1 30 .7 23.0

Pine .:mel Balds Northern 5.8 48 .5 10.2 31 .3 5.4 16 .6 1.2 3.7 32 .6 24 .4 llard\'.JOOdS ] --- Ilem1 ock 5.3 100.0 --- 5.3 4.0

--- -· -� ------Spruce­ 7.7 100.0 ------7.7 5.8

Fi r Total 18. 0 33 .6 25. 0 21.6 35.2 133.4 km -

h.l (J\ 27 balds, northern hardwoods, hemlock, and spruce-fir . The inclusive vegetation types and the distance that each route covers within each vegetation type are also included.

A combination factor of altitude class and vegeta­ tion type was determined for each scat location. The relat ionship between this combinatio� factor and the resultant scats/km for each was analyzed through the use of a regression procedure (Sakal and Rohlf 1969) . CHAPTE R IV

RE SULTS c\..:\iD DIS CUSS ON I

I. INC IDENCE OF SCATS DURING PRE- .�\J D

POST - SUMMER SEASON S

Index Routes were hiked on a regular basis from 15

May through 1 October and on an irregular basis between

1 Octobe r and May, 1970-19 75. The period time from 15 of

1 June - 1 October will be referred to as the suminer season and the time spans from January - 1 June and from 1

1 October - 1 Jan ua ry will be referred to as pre- and post -summer seasons , respectively.

Scats/km for the pre- and post -summer season we re less than those for the SUIT'u'Tier season except the case in of the Sp ence-Russell Index Route . The pre- and post­ surrJner season values ',.;ere 0. 03 scats/km and 0.06 scats/k�1, respectively for this route (Figure 3) .

There are numerous facto rs which may infl uence the low incidence of scats during the pre-summer season . Bears may be located away from the Index Rout es, and th eir move­ ments may be rest ricted to smaller areas than at othe r times during the year . The availability of natural food s is less during the sp ring and the bears may sti ll be sub ­ sisting off fat stores (rather feeding a d than n thus defecating) unt il later in the season which food at time 28 29

• 2 5 Sugarlands Tremont-Der­ rick Knob Elkmont -Bent Arm - - ·-- Snence-Russell Gregory- -X ­ ·Field • 2 0 Hannah �loun tain

. 1 s ::.:;: � ...... Vl ..., ('j u r.l) .10

• 0 5

Pre ­ Summe r Pos t­ Sur.,_,'ner Summe r

Season

Figure 3. Comparison scats/km values for summe r, sum..rner, and post-sc.mmeofr seasons over six yearspr e- (1970-1975 ) in the Great Smoky :,Jountains :.:atio::1al Park . 30 sourc es are nore plentiful. Assoc iat ed with the above is the possibility that the dormant condition of the gastro ­ intestinal tract after the winter season may deter food intake during the early part of the spring.

Factors that may influence the fluctuation of incidence of scats dur ing the fall should be noted. It is suspect ed that some specific food sour ces (i.e. , mas t) may occur away from the routes during the fall. Beeman

(1975) noted that several of the bears he had radiotracked left the ir normal home ranges during early fall and re ­ turned prior to denning . Decreas ed feeding activ ities prior to denning obviously result in less defecation.

During the denning period no scats are found because no de fecation occurs (Folk et al . 1972) . In his study on the defecation rates of captive brown bears fed ad lib atum

Roth (1977) also found that in winter the defecation rate approaches zero , particularly fo r pre gnant females . Leaf cover in the fall was also found to hinder the location of scats . In years of early leaf fall the number of scats collected appeared to decrease more quickly than in years of later leaf fall indicating the difficulty of locating scats becaus e of leaves.

It appears that some factor or combination of factors are caus ing less evidence of the presence of bears along the Spen ce-Russell Route suggesting �hat bears Index ut ilize this route more freque nt ly during the pre - and 31 post- summe r season than during the summer seaso n . Due to the popularity of the Anthony Creek Trail and the upper section of the Bote Mountain Road (both maj or sections of the Spence-Russell Index Route) hiker intensity is suspected as hav ing some influence on the differences between scats/km for the summer season and those for the pre - and post-summer seasons . A total of 73 overn ight hikers , 36.4 percent of all overnight hike rs encountered on all Index Routes during the suminer season , ',vere 1975 met along the Spence-Rus sell Index Route (Tab le 4) . Marsh

(1973) found that between early Jun e and mid-Se ptember the

An thony Creek Trail was more heavily used than any of the other trails which make up sections of the Index Routes . h'hether the pres ence of larger numbers of people alone causes the bears to avo id this route during the summe r season or whether hikers and horseback riders have trampled the scats into the mu d or whe ther bears have simply quit using this rout e for some other reason (e.g. , dominance of one or more animals over othe rs , climatic con ditions , establishment of othe r game tra ils , etc.) is un known .

Figure 3 illustrates that bears appear to use the route less during the summe r season.

II. INCI DENCE OF SCATS DURI!\G SEASON SU�vL\1E R

Between 1 June and 1 October, 1970-1975 , 697 scats

'-' i

TABLE 4 HIKER USE OF INDEX ROUTES DURING THE 197 5 SU?>1HER SEASON DETERi'vfiNED BY 0fiNE ALL-DAY COUNTS

Number of Hikers 0, Index Route Day OV"emignt Total Number 0 of Grand Total

"" ., Sugar lands 10 36 46 / . ..)

Elkmont - 28 85 113 18.0 Bent Arm

Tremont- 15 73 88 14.0

Derrick K...i1 ob

Spence-Russell 69 174 243 38.8

Gregory- Ha.t""'lna.� >!ountain 27 110 137 21.9 " "" Grand Total 149 478 6 L-1 100.0 TAB LE 5

SUMMARY TilE NUMBER OF SCATS COLLECTED ALONG INDEX ROUTES IN TilE (WGR EAT SMOKY MOUNTAINS NAT I ONAL PARK , 1970-1975

fikin�Pe riods I June June July July 1 Sept . Sept . OCt. Total Total Index Route 1 15 1 15 g.1 Aug.15 1 15 1 Scats Scat s KM KM

Sugar1and s 23 5 21 13 15 29 23 5 877.2 0.18 17.2 17 9 1 5

�[" 32 .8 45 JJ 37 23 25 48 55 33 3 332 1607.2 0.21 ElkmBent o Armnt- 1

24 .8 23 7 7 4 15 12 6 92 1240.0 0.07 TDerrickremont- Knob 7 11

[" Spence - 20.1 4 4 3 0 .) 4 8 3 1 32 964 .8 0.03

Hussell

,- Gregory- 35.1 10 :J 10 10 15 11 7 9 86 1396.5 0 .06 ll<:mn ah 9

-Mountain Total scats 105 56 78 53 58 111 108 72 56 697

130.0 676.5 724.8 729.4 684 .5 659.7 689.6 709.3 654.6 557.3 6085.7 Total km ts/l

vl vi 34

During this five-year period the Elkmont -Bent Arm Index

Route yielded 332 sca ts , Sugarland s 155, Tremont -Derrick

Knob 92, Gregory-Hannah Mountain 86, and Spence-Russell 32.

The overall trend betwee� years shows an increase in the raw number of scats collected between 1970 and 1971 . This incre ase is likely explained by the fact that dur ing 1970 the Gregory-Hannah Mountain Index Route "Y·ias not hiked.

From 1971 thro ugh 1975 the numb er of scats collected steadily declined (Figure 4) suggesting that bear activity along Index Routes was also decreasing .

To enable comparisons to be made among the routes , the ;mmber of scats collected along each rout e 1-1 as con­ verted to a scats/km value . Despite the conversion , the

Elkmont-Bent Arm Index Rout e still produced the highest scats/km at 0.21. Sugarlands produced 0.18 scats /km,

Tremont-Derrick Knob 0.07, Gregory Hannah Mountain 0.06, and Spence-Russell 0.03 scats/km (Tab le S) .

The inclus ive scats /km value for all Index Routes over all years l--ias 0.11 scats/km (or 9.1 km/sca t) . �·larcum

(1974) est imated the dens ity of black bear for 1973 to be

1 bear /2.7 km2 . The scats /km value for 1973 was 0.12 (8.3 km/ scat) . Eagar (1977) determined the density of black bears to be 1 bear /3.0 3 km2 for 1974 and 1 bear/2 .77 km2 for 1975; scats /kn values were 0.09 and 0.08, re sp ect�vely during this period. Study are as during the period when all three estimates derived were the same . Fu ture we re 35

175

150

Ul +..J (1j 100 u V'J

!..I-, 0

� Q)

,..0 '"7 ... � I :J z3

50

70 71 72 73 74 75 Year

Figure Total numb er of scats collected in the Great Smoky4. Mou ntains National Park , 1970-1975. 36 comparative studies from other sections of the black bear' s range in �orth Ame rica would be interesting and useful.

Evaluation of the scats/km on a yearly basis showed that a marked decrease occurred between 1970-1975. The scats/km value for 1975 is 43 percent of the valu e for

1970 . In other words , the numb er of scats/km collected in 1975 was 57 percent smaller than the numb er of scats/km collected in 1970 (Figure 5) .

It appears that the 1973 scats / km figure is greater than it should be according to the overall linear appear- ance (Figure 5) , howeve r it is also possible that the 1972 figure is artificially lower than normally expected. If this 1s the case it is likely explained by the scarcity of mast in 1972, which apparently caused greater mo vements of b ears (�beem an and P e lt on -ol� �-)l! •

LaFollette (1974) also recorded several instances in which bears traveled greater than usual distances when their natural food supply was depleted or nonexistent .

Thus the decrease in incidence of scats along Index Routes is a result of a greater percentage of the bear populat ion moving out of the ir normal summe r and/or fall home ranges

(and in some cases the Park) foraging for mo re substantial food supplies . It was found that scat incidence did vary among vegetation types depend ing on the hiking date . For

exa�ple, incidence in the open oak- pine and balds as type decreased (1 -15 Septembe r) , incidence '! ege tation 1n 0.193

0.165

� 2.. � ...... (/'; 0.13S I \ l \ \ n ' ' 1 v �' • u' ::1 u ' :J

'71 / .) 1-t I :J 70 71

Year Yearly scat s/�r:t '/ alues £ron Figure 5. 19'70-1975 .. Great S;1oky �lo:J:-J.tains \at ional ?arL :n t['e 38 the northern hardwoods type increased, suggesting a shift in feeding, possibly from oak to beech and buckeye ma st

(Figure 6) .

Distribution of Scats within Hiking Periods

In order to analyze the variability in scats/km among hiking periods , an analys is of variance proced�re was performed. The results show that variab ility among scats /km over the nine hiking periods has a probab ility of occurring

75 percent of the time (Appendix G) .

Between June and 15 June (1970-1975) the numb er of 1 scats/km drops from 0.16 to 0.08 (Figure 7) . Since all routes were cleaned two we eks prior to the first hiking , this decline cannot be attributed to scats remaining from earlier months . It 1s felt that although black bear breeding activities were found to peak 1n Jul y, this behavior, which begins in Jun e (Beeman 1975) , may explain the decrease in the numb er of scats collected durirrg late

the onset of the June , perhaps signaling breeding season . Roth (1977) also noted a temporary decrease in the defeca­ tion rate of brown bears from April to Jun e and suggests that the cause may be the apparent lack of appetite in bears , part icularly males, during the breeding season.

The increased incidence of scats betwe en 1 Augus t and .-\u gust is likely explained by r:1aturing lS berry c:Tops .

Beeman (1971) found that blueberries (Vaccinium sp. ) 1 .)�"'0

1.6 Closed Oak • 4 ••• ' •••• Open Oak-Pine and Balds \ --- -- or the rn Hardl•ioods \ :-J -x- x- x- Hemlock 1.4 "\f.. i-\ Spru ce-Fir �\ xxxxxxy.

� \ 1.2 f.� \ 'f. x.XXXX)\ ,.:. \ f. 'f.. X f.. j\ \ y. ' f.. *"' ',( I \ ;<. \ " ' ;<. X: '/.. \ � \ �-� I 1.0 ;< X � � \ ")4. I \ x; I �\ f.. i'/.. 'f. f. I\ :2:: "f. '- l 1 �;i. :::.:: X 1- )( ;.( \ '-.. 'f-. '\. � � t1) 0.8 � X>( 1 \ -1-l " � m � ;< � \ u / \ 'J) f } \ � ;<� 0 � \ i.. i 0.6 .fJ.. . f. 1- \ ..... � � 'f. -t_f f. i ' �·, \\ 0. 4 1- ·f.., ;(.1

I I '''f.� \ I I I . ' 0 2 I \1

0 _ 1. __! 1 2 3 4 5 6 7 8 9 June July Aug. Sept. Oct.

Hiking Periods (Months)

t1�ure 6. Scats/km within vegetation types in the Grea! Smoky Mou nta ins �ational Park, 1970-19 75. 40

• 2 0

• 1 5

. 10

• 0' ,...:::J

... 1 15 1 15 1 15 1 15 1 June July Au g. Sept . Oct .

Hiking Periods (�onths)

Figure Re lationship betwe en scats/km and the nine hiking per7.io ds , June -October 1970-197 5, from the Great Smoky Mountains �ationa1 Park . 41 blackberries (Rubus sp.), and huckleberries (Gaylussacia sp.) accounted for 39 percent of the summe r food volume of black bears (1 July through 30 Au gust) in the Park .

Betw-een 1 September and 15 October black cherry (Prunus serotina) accounted for 60 percent of the food volume while blackberries made up 35 percent and blueberries

13 percent . Pelton (1970) also noted that August is the month that black bears are most actively engaged in foraging �or food in the Park . It is during this late

summer- early fall period that bears gain more we ight than at other times of the year (Beeman and Pelton 1977) .

The continuous decline in incidence of scats/km

from 1 September until 1 October (1970-1975) can prob ab ly be accounted for by movements of bears out of the study area, away from Index Rou tes , oT to lower elevat ions for the purpose of foragin g for fall foods pri or to denning activitie s (Lindzey and Me slow 1976 , Beeman 1975) . Leaf fall , part icularly when it occurs in early autuTim , prob ab ly

intensifies the decline in scats collected.

Distribution of Scats within Altitudinal Classes

In a prel iminary study , Pelton (1972) found that

56 .9 percent of the scats collected along Index Routes

1.;ere fotmd at elevations of 1372 m and above . With 33 .4

percent of the total kilometers ab ove 1372 m and such with

a large percentage of the scats found the higher at 42 elevations , it was felt that altitude might prove to be a maj or factor affecting the distribution of the black bear population in the Park .

Sections of Index Routes between 1372-1524 m were most productive in terms of the overall number of scats collected (Appendix H) . However , when compar ing the scats/km values , the highest value , 8.2 scats/km, fell

in the 1219-1371 m altitude range (Table 6) . The scats/km value s in the different altitude classes were quite variab le (Figure 8) and sugge sted that altitude may affect locat ion of scats. A Chi square test indicates the difference in scats /km among the Index Routes is significant at the 97.5 percent confidence level; altitude classifica­ tion and route classification fo r scat s /km are not independent of each other.

An an alys is of variance with regression was per­ forme d to furthe r explore the effect of altitude on the location of scats . The results of the linear regress ion show that the variable , scats/km, varies in a way that

can be described by a linear equation: Y - -0.8 + 1.16X, where X is altitude and Y is the value of scats/km. The

information in Figure 9 can therefore be us ed, within set confidence levels , in predicting the expected numb er of

scats/km for any altitude level in the study area. This

informaticn may be of va lue the future as a tool for in 43

TABLE 6

SCATS/KM COLLE CTED WITH IN ALTITUDINAL CLASSES ALONG INDEX ROUTES IN THE GREAT SMOKY MOUNTAINS NATI ONAL PARK , 1970-1975

Index Routes

"' ' .A ltitude Elkmont - Tremont- Spence- ureaHannorv-ah Classes (m) Suga:::-lands Bent Arm Derrick Russell 01ountain K,.1 ob

457- 609 *- 0.3

610- 762 3. 0 0.0 0.4 1.9

763- 913 3. 9 3.5 0.4 0.4 5.4

914-1066 4.8 7.9 1.0 1.7 0.9

1067-1218 13.1 11.0 2.1 0.8 0.9

1219-1371 12. 2 17.6 4.7 2.4 1.9

1372-152:1- 13.9 19.3 5.0 3.5 1.7

1525-1676 7.8 7.9 2.6 3.8

1677T 3.5 10.4

*- denotes that the altitude class does not appear along the respective Index Route. 44

8 .5 �

(8 . 2) - 7.5 (7.7)

( 6 . 9) 6. 5 � � (n R !

5. 5 - Key t 0 .-U tit"Llde Class es :

.,. 4. 5 1 4 57- 609 lTI .... _ ::.::: 2 6 10 - 762 m ...... lfl 3 7 ().)- 913 I:'i +J ( 3) 4 9 14-1066 m cti ;---.-;o u 4. 5 1067 -1213 m C./) 3. 5 ( 3.8 ) 6 1219-1 371 m -,� 7 1372 -15 m 'j. ;;,) r- 4 ,...___. 8 15 25-16 6 m 9 16 77+ m

�

1 .... .I. • :> � ( 1. s: I 0 • 5 1- (0.3

1 2 3 5 6 7 3 9 4 Altitude Classes

Figure 8. Re lat ionship between scats/km a�d altitudinal clas ses in :he Great Smoky �ational Park , 1970-1975. Mauntai�s 45

1 - ,...... 1. .J • IJ

I I I I I I 10 .0 I

=

Slope + Y -0.8 1.16 X

-- _.. ,...,.. ...- / / / / 95'\ Confidence /

2. 5 / / / /

1 ,) -r 0 l 6 3

Fi ::1?:cJ. a gure 9. Re gressio�al relatio�ship bet�e en sca:slk 11 c � as s e s r e S ;�·� o c 1-ln a r_J n a m l t it ·d 1 in , - � the G a:: k ;,r �· l t in�· >; at i l. Par}: 19 7 0 1 9 '{ 5 46 predicting intensity of bear use of cert ain areas within the Park.

Distribution of Scats within Vegetation Types

Vegetation patterns are closely related to altitude

1n the Park . Viewing the diversity of vegetation types to be found along the Index Routes and realizing that black bears seem to frequent specific altitude classes , it was assumed that food preferences have an influence on the se asonal distribution of the population and �hus their sign (scats) . For the purposes of this study, food preferences the bear population in the Park were viewed of in terms of vegetation types.

The spruce - fir veget ation type yielded 8.1 scats /km.

- .c "7 The northern hardwoods type accounted .LOT 1 • .) sc ats/kr., closed oak for 5.8 sc ats/km, open oak-pine and balds for

4.3 scats/km , and hemlock for 0.3 scats/km. No sc ats were found in the cove hardwoods vegetation type.

v�ben a r g types, the classified cco din to \'e getat ion overall sc ats/km values indicate that more sc ats occur i� some vegetation types than in others . The spruce -fir vegetation type produced the highest nu mber of scats /km.

It is fel t that the high incidence scats 1n this of veg etation type does re flect the true value of this no t

as a natural feedin g area far bears ; few na � ural type sources sur.1.mer foods are fou:::.d in this type. At high of 47 elevations bears have few choices in moving from one area to ano ther; they, in essen ce , are forced to ut ilize the spruce-fir type in going to and from areas of summe r foods

(berry crops) . The spruce-fir ty?e represents only 5.7 percent of the total kilometers covered by the five Index

Rou tes . The northern hardwoods vegetat ion type accounted for 24.4 percent of the total mileage and 7.3 scats/km were collected along its length. The third most productive vegetation type, closed oak forest, resulted in scats/ 5.8 km . This vegetat ion type represents 37.2 percent of the total distance hiked. The open oak-pine and balds vegeta­ tion type accounted for 23.0 percent of the tota l dist ance hiked and produc ed the fourth highest scats/km (4 .3 scats/ km) . Lastly, the hemlock and cove hardwoods vegetation types , when comb ined , repre sents 9.6 percent of the total mileage , accounted for only 0 . 6 percent of the total scats

collected (Appendix C) •

Due to the effects of altitude on vegetat ion types within the Park , it was necessa ry to delineate the inter­ actions betwe en these two factors (altitude and subsequent vege tat ion type) and the resultant scats/km value .

Distribution of Scats wi�hin Altitude -Vegetation Types

It is obvious that one vegetat ion type can span one or altitude clas ses and vice versa. In order to mo re analy:e the distribution of scats in relat:on to the 48 possible influencing factors of the combination of alti- tude and vegetation types, classifications were estab- lished to describe all pos sible combinat ions of altitude and vegetat ion types (Tab le 7) . An analysis of variance procedure was performed. The results showed that there exis ted a very high probabil ity (at or ab ove the 95 percent confidence level) that the combination of alti tude and its resultant vegetation type did significantly affect the locat ion of bear scats on all routes but one , Spe nce-

Rus s e 11.

Taking each vegetat ion type individually and studying the re sulting pattern produced by the scats/km value and the altitude levels , specific relat ionships are evident

(Figure 10) .

In the cove hardwood vegetat ion type (1) , regardless of the altitude leve l , no scats we re found. Of the 133.4 km of Index Rout es, only 5.6 percent passes through this vegetation type ; it is conceivable that the low incidence of the vegeta t ion type may partially account for the lack of scats collected . Eub anks (1976) also found relat ively low incidence of scats in this vegetat ion type as comp ared to the inc idence in the oak vegetation types. �ast produc ing species in this type include beech, buckeye , and dogwood.

Due to the ccmplexity of the mes ophytic forest, it

is pos. sible ·c :1at areas of cove har(�.,..·ood mav' been

one of the oak forest typ es . 111 classified as error 49

TABLE 7

AL TITUDINAL -VEGETATION TYPE BREAKDOWN BY INDEX ROUTE IN THE GREAT SMOKY MOUNTAINS �AT IONAL PARK , 1970-1975

.Altitude No . Route (m) Vegetat ion Type Scats Scats/fev! KM Sugar lands 457- 609 Closed oak 0.5 0 0 610- 762 Closed oak 1.0 3 3.0 763- 913 Closed oak 1.3 5 3.8 914- 1066 Closed oak 0.5 0 0 914- 1066 Open oak-pine 1.5 10 6.7 balds a� d 1067-1218 Open oak-pine and 1.3 17 13.1 balds 1219 - 1371 Open oak -pine and 4.0 47 ll. 8 balds 1219- 1371 Spruce-fir 0.6 8 13.3 1372-1524 Spruce-fir 1.3 20 15.4 1525-1676 Spruce-fir .),..� 7 23 7.2 1677+ Spruce-fir 3.1 11 3.5

18.3 144 x = 7.1

Elkrnont­ 610- 762 Closed oak 0.2 0 0 Bent 763- 913 Closed oak �::J1 - 3.4 i\.:rm 914- 1066 Closed oak 4.8 38 7.9 1067-1218 Closed oak 2.7 32 11.9 1219-1371 Closed oak 2.6 61 23. 5 1219-1371 Northern hardwood 1.0 4 4.0 1372-1524 Closed oak 2.7 64 77;_._. .I 7 1372-1524 �Jorthen1 hard1.;ood 3.9 19 4.9 1525-1676 Northern hardwood 8.4 66 7.9 • ' 1677+ Northern hardr.v-ood 2.6 29 1.1. .. ?... 33.3 328 X"= 9.84

Tremont­ 763- 913 Closed oak 2.3 1 0.4 Derrick 914-1066 Closed oak 3.2 3 0.9 1067-1218 Closed oak 2.8 6 2.1 Kn.cb 1219-1371 Closed oak 4.5 17 3.8 1219-1371 :\Torthem hardwood 1.0 2 ?-. n',J l3i2-l524 Closed oak 2.2 18 8.2 1372-1524 �Jorthern hardwood 8.4 4,4 1525-1676 Northern hard\,·ood ....1 . 7. ?- � ...0; s 26.1 89 X"= 3. 09 50

TABLE 7 (continued)

Altitude No. Route (m) Vegetation Type Scats Scats/101 KM Spence- 610- 76 2 Hemlock 3.4 2 0.6 ,.., I Russell 610- 762 Cove hardwood '- . .:... 0 0 763- 913 Closed oak 1.4 0 0 76 3- 913 Hemlock 1.3 1 0.8 914-1066 Closed oak 0.7 2 2.9 914-1066 Hemlock 0.7 1 1.4 1067-1218 Closed oak 1.3 0 0 1067-1218 Open oak-p ine and 2.7 3 1.1 balds 1219-1371 Closed oa.\ 1.7 2 1.2 1219-1371 Open oak -pine and 0.7 0 0 balds 1219-1371 Northe rn hardl.;ood 1.1 6 5.5 1372-1524 Closed oak 0.2 0 0 1372-1524 Northern hardwood 3.1 12 3.9 1525-1676 Northern hardwood 0.9 3 3.3 -- - ,.., 21.4 .).:., ::\= 1.48

Gregory - 457- 609 Cove hardv.-ocd 0.7 0 0 HanJ1ah 457- 609 Closed oak 0.9 1 -·1 �1 ain 457- 609 Open oak-pine and 4.1 0.2 �bunt balds 1 610 - 76 2 Cove hardwood 2.3 0 0 610- 762 Closed oak 4.4 15 3.4 610- 76 2 Open oak-pine

LT � � ...J .l. -'1 ' "0 0 ,-i �til' \" 0 ve" 1 a. ..l. \...... �� � ... :.

• 7 ' . )

:'I ·-e

scat s/k� �:�n:n y· e e t a : �ig. c n ur c e C;.lC. s s c:�e 3. lation s nship:1 -: o�e :-- e 3.:: :;� o c L! r1 t :1 �: s g :: l i : i � i o i �-: C 5. �< :·-.. > 1 �- \a:ional Park , 19-G -lJ-3. 52

15.

12 . 5

I Hemlock � 10. 0 Spruce -Fir � 0 i vI Northern Hardwood V-1I 2 ::.::: � ...... � U1 7 . 5 .;J (!j � u U) ffl r1 � 5. 0 !

2. 5 I 'lI I I I I I

I - + N ['<") '-0 ....; ""' '..0 r-- \0 ..... \0 r-- 01 r-- r-- r-- C) 0 ['<") Ln '-0 ,::;_ ....; ri ,...., ,...., I 1 1 I 1 0 ['<") � C) 01 lf1 ,...., •,Q .--1 ....; r-- N \0 ::-- c;. N tl") lf1 .--1 .--1 .--1

Altitude (m)

Figure 10 (cont inue d) 53 classification may contribute to the lack of scats found in the cove hardwood forest type. One other possible reas on for the lack of scats in this type may be histor ical in nature . It is known that the logging and clearing of slopes in the Park reached its peak in the late 1920's.

At this time over 60 percent of the land was at least partial ly cleared (LaFollette 1974) . It may be that the hard mast p�oduc ing species in the cove hardwo ods have not begun to generate enou gh mast to attract bears for any period of time .

The altitudinal re lationships for the closed oak vegetation typ e ( 2) indicate that the scats/km value in­ creased with an increase in altitude . The peak value was

12.8 scats/km; the correspond ing altitude was 1372-1524 m.

This is felt to be particularly noteworthy as this altitude level is also the upper limit of the closed oak vegetation type (Shanks 1954b) .

The open oak-pine and balds vegetation type exhibits

different pattern . Beginnin g at 0.2 scats/km at the lower a elevations , the value increases to 6.1 scats/km, drops to

3.4 scats/km, then rises to 9.4 scats/km and finally drops to 0.8 scats/km . The sharp rise in the scats/km value from

0.2 to 6.1, betwe en alti tude levels of 610-913 m and subsequent drop to scats /km between altitudes 914- 3.4 of 1066 m may be explained by the dominance of Va ccin ium sp . betwe en leve ls of 617-91 3 m (�hi:taker 1956) . alti t�de 54 As bears feed heavily on this species (Be eman 1971) , they

likely frequent the area between 763-913 m more often than

others thus resu lting in the scats/km value of 6.1. The

scats /km value continues to rise to 9.4 scats/km at the corresponding altitude level of 1219-1371 m; at this point

the scats/km value drops drastically to 0.8 scats/km at the

altitude leve l of 1372 m. According to Shanks' diagram

(Figure 2, page 11) , the highest reaches of the open oak­ pine vegetat ion type also term inat es around 1372 m, probab ly explaining the drastic drop in the last scat/km value . The balds , however, start just below 1372 m and continue up to

the high est elevations . As the scats/km value fa lls shar; ly

at the 1372 m alt itude level, it would appear that the re is more to attract bears in the open oak-pine stands than in

the b a ld are a s, poss i b l y food. It may be that prefer bears more open understory and availab ility of oak mast the to denser heath coverage of the heath balds and/or the grassy

cover of the grass balds .

The altitudinal re lat ionships found within the n orth r n hardwoods vegetation type (4) is similar to that e found for the closed oak; the n mb e r of scats/km increased u with each incr ease in altitude . Scat s first occurred within

the 1219-1371 m altitude level and peaked at 11.2 scats /km

in the h i h s t altitude class, 1677+ m. Beeman (1975) g e

suggests that it is not habitat r e fe r e e but the p nc

panhandler behavio·r of some caus es ther.1 1:0 bears that frequent the northern hardwoods vegetation type 1n the summe r; most of the overnight shelters along Index Routes are locat ed in/or on the edge of the northern hardwoods vegetation type. Beeman (1975) also found that those bears in the northern hardwoods vegetat ion type shifted

to the closed oak and cove hardwoods vegetation typ es in the fall .

Within the hemlock vegetat ion type (5) the re lation- ship also indicates an increase in the scats/krn an with increas e in altitu de. The extreme ly lew scats/km values

(Append ix G) show this vegetat ion type to be sec ond least productive . Of the 133.2 km only 12 km or 9 percent of the Index Rout es passed through the hemlock vegetat ion

type . The hemlock type is found primarily in the lower elevations . There are tew food res ources for bears in

this type sinc e the understory consists primarily of

rhododendron.

The spruce- fir vegetation type (6) exhibits an

altogether different pattern when compared to the others .

Scats/km begin to rise between 1219-1371 m and peak at

1524 m. At this point the scats/km va lue begins to

decre ase steadily . This suggests that bears not fre quent do the pure fir stands of the highest altitudes of this

vege�ation type as often as they do the lower altitude

segments. It may be that the particular

food/foods fo und th is altitude level are located only in at 56

the spruce-fir segments and not in the pure fir stands .

It may also be that fewer kilometers of trail exist in the pure fir stands than in the spruce-fir , causing more

emphasis to be placed on these last two segments .

Backcountry Use and Scat Incidence

Recognizing the fact that human impact can and often

does adversely affect some species (Leopold 1966) and the

fact that the overal l number of scats collected per kilometer have decreased from 1970 through 1975 (Figure 4, page 35) , I investigated the relat ionship between the numb er

of scats collected per kilome ter (an indication of popula­

tion trend) and the n��b er of backcountry nights (number of backpackers multiplied by the numb er of night s spent in the backcountry) as report ed by the National Park Service

(Figure 11) . If Figure 11 shows any kind of relationship between the two factors it would appear to be a positive

one for all years except 1973. It is felt, however, that

this conclus ion incorrect. is LaFollette (1974) noted that hike r use of the Park

has increased exp onentially from 8000 in 1970 to 80,000 in

1973. As the popularity of backpacking has also increased

recent years , it would seem logical that a similar pattern in of increase would also be evident Park records . The in determination of backcountry nights, as computed by Pa rk

person:;1 e l, is calculated from the number of camping per:n::_ts 57

.200

- Scats/KM # ,....., Backcountry nights --- # +..l p., tD .17 5 U) 65,000 I tD :::::: ...... 60,000 "J t', '..._./ I './1 \ +..l .150 I 55,000 ...c: I \ bf; ::22� 1 \ . ..., ...... �! \ z V'i I \ +..l \ 50,000 >-. ro I ;.... u \ +..l U) :::::: \ I ..... � ..... 1 ') • J...... ,J \ I 45,000 0 u \ I � u \ ro 40,000 >=Q 14-; 0 .100 35 ,000 :.... tD ...... s ::l 30 ,000 z

.075 70

Year

Figure 11. Relat ionship betvieen summer the season scats/km values and the n��b er of backcountry (n��b er of backpackers multiplied by the numb er of nights spent out ) nights fo r the same time period . 58

issued to overn ight backpackers . It is probab le that

some ne glect to apply for the permit and are thus never

censused. The rat io betwe en those with permits and those without is unknown .

Another reas on for doubting a positive relat ionship between numb er of sca ts/km and number of backcountry nights

is the results (unpublished data) of hiking an ab andoned trail (same biweekly pattern as other Index Routes)

located just to the east of the Spence-Russell Index

Route. The ab andoned route proved to be the most produc­ tive in terms of scats/km, when compa red to the results of all other routes hiked in 1975. As both routes (Spence­

Russell and the ab andoned trail) traverse similar topography and vegetat ion , it appears that some factor may influence the number of scats/km, possibly the degree of human utilization . It should be noted that the AI1 thony Creek

Trail , a maj or portion of the Spence-Russell Index Route ,

is one of the more popular trails in the Park . Marsh

(1973) noted that during the period between early June and

late Ju ly, the An thony Creek Trail had an average of 24 people passing his checkpoint every three hours . Between

the end of July and mid-September this figur e increas ed to

a mean of per three-hour period. Marsh also tabulated 34 similar counts two other tra ils , Elkmont an d Grego ry en Ridge , �hich are maj or segments of two othe r Index Routes ,

Elk:n cnt-Bent and Gregory-Hannah �fountain. Using the Arm 59 same time periods mentioned above , he found Elkmont to produce 5 and 0 persons , respectively, per three-hour period and Gre gory idge to produce 6 and 1 pers ons , R respectively, per three-hour period . Additional data col ­ lected dur ing the 19 7 5 sur.un.er seas on a 1 so shm-:ed that the comb ination of trails which make up the Spence-Rus sell

Index Route was used by a greater number of hikers than any of the other Index Routes (Tab le 4, page 32) . All of the above points to the high degree of popular ity a�d conse ­ quently the high degree of human utiliz ation encountered on the Spence-Rus sell Index Route. It is suspe cted that the large number of hike rs who use the Spence-Russ ell Index

Route may have had some influence on the re sultant scats/km value and thus account for the no s ignificant outc ome of n the analys is of variance procedure . However , the low inci dence of scats on low visitor use trails (e. g. , Gregory ­

Hannah Mountain) cannot be attributed �o hiker impa ct .

Expected Fre quencies of Scats

At the outset of this prel imin ary survey, it was hoped that an independent index could be established to monitor fluctuations in the black bear population of the

Great Smoky Mountains National Park . Due to the compli­ cated nature of the statistical aspects of the problem and the fact that numerous s umptions need to be made concern ­ a s ing the statistical va lidity the many and va ried of 60

influences on the incidence of black bear scats (i.e. ,

temperature , precipitation, behavioral reactions among

individual bears , game trails, etc.) establishing a

precise and accurate index was not possible without a

more in - depth analysis of the ab ove variables .

However, with the data available and with the re -

sults of the statist ical analyses pe rformed , it is possible

to determine the relative expected frequencies , which are

the probabilities of the various outcomes (Snedocor 1950) .

Given a particular alt itude-vegetation class on a given

route and knowing the mean scats/km value for that rout e,

it is possible to predict the numb er of times that

part icular rout e must be hiked in order to collect one or

more scats within a set probab ility. Whe re p = mean numb er

of scats/km for a route (or trail segment) and = numb er � of hikes , the fo llowing equat ion can be set up : M = pN . The calculated value of M is found in a table for relat ive

expected frequencies for individual terms of the Poisson

distribution . Using the me an scats /km value for the

closed oak vegetation type of the Elkmont-Bent Arm Index

Rout e between altitude levels of 106 7-1218 m and assuming

that this section would be hiked 10 times , the equation

would as follows : be

.l\' = p:;

= . 2324 (10) \1

= .1v . .) 2 61

To determine the probability of the occurrence of

a count of Y = 0 events on a distribut ion whe re the mean ,

') - M = the value of ._,.) is found in the appropriate 2. 3, table and is . 1 0 0 5 . In other words , if the closed oak 2 9 vegetation typ e of the Elkmont -Bent Arm Index Route, between the altitude leve ls of 1067-1218 m were hiked 10 times, there is a 90 percent probability that one or more

scats will be found.

Using this equation , it is possible to predict within a set probability whe ther one route and its subse- quent altitude-vegetation type produce the number of scats expected, based on the sample me an (scats/km) from previous hikes. Although this equat ion cannot guarantee to be an accurate monitor of population fluctuations , it is ab le to provide a certain degre e cf probability that a cert ain number of scats can be expected to be collected along particular segments of Index Routes . Any deviations from

the expected frequencies could indicate a fluctuation 1n

the population level. This alone is not an accurat e descript ion of the population trends . With further me asure- ments of extrins ic and intrinsic factors that may affect population density conjunction with the appropriate in statistical analyses, it would be possib le to directly monitor population fluctuations via incidence scats . of CHAPTER V

SUMf.IARY AND CONCLuSIONS

The purpose of this study was to determine the degree of variab ility of incidence of scats by date and locat ion and to determine the feasibility of us ing scats as a reliable index to the population density of black bears in the Great Smoky Mountains National Park .

Seasonal distribution of scats was broken into two categories : summe r season (1 June - 1 October) and post- and pre- su��er season (1 October - , June) . The scats/km values for the and post-swllme r seasons ,,;ere consider- ably lowe r than the summer season scats /km by about 25 percent . The Spen ce-Rus sell Index Route proved to be the only exception y.;here the summer season scats/km value fell below the pre- and post-summer season scats/km values .

During the six years ' time this study cov ers ,

6 0 8 5 . km o f Index Rout e s we r e h ike d b e t ��-e en 1 June and 7 l October, yielding a total of 697 scats. The Elkmont-

Bent Arm Index Ro ut e proved to be the most productive of all the Index Routes , yielding 0.21 scats/km. Next was

Sugarlands with 0.18 scats/km, Tremont-Derrick Knob with

0.07 scats/km, Grego ry- Hannah Mountain with 0.06 scats/km

and S e - R us s e 11 ��,- t h 0 . c a t s I p e r:. c i 0 3 s km . Betwe en 1 June and October route was hiked l each

62 63 approximately nine times or once every two weeks . Definite peaks and depre ssions in the scats/km value occur among the hiking dates . Peaks occurred in early June and from mid­

Au gust through early September. Depressions occurred from mid-June through late July and during late September and early October. It is felt that the decline noted during

June and July may be an indication of the onset of the breeding period for black bears �hile the decline at the end of the summe r season probably indicates the movement of bears away from Index Routes dur ing early fall, foraging activities , and later the onset of winter and the initiation of hibernat ion activities . The productivity, in terms of scats/km, of the time period from early Au gust through early

September was about 50 perc ent greater than other time periods except for early June which was also approximately

SO percent more pro�uctive .

Of the nine altitudinal class es , ranglng from 457-

1677+ m, the class ranging from 1219-1371 m proved to be most productive , yielding 8. 2 scats/km. In fact , those altitudes from 1219-1677+ m produced mo re scats /km than those ranging from 457-1218 m. The lower range of altitudes accounts for 31.6 percent of the total scats col lected while t�e higher altitude range accounts for 68.4 percent of the

scats collected. Looking at the numb er of scats total collected in conj unction with corresponding distances hiked,

turns out over two -thirds of all scats collected it that 64

along Index Routes we re collect ed on less than half of

the total mileage .

Of the six major vegetation types present in the

Park , the spruce-fir vegetation type produced the highest

overall scats /km value at scats/km. It is interesting 8.1 to note that only 6 percent of the total dis tance hike d was in the spruce- fir vegetat ion type. Sorthe rn hardwoods yielded 7.3 scats/km, closed oak scats/km , open oak ­ 5.8 pine and balds 4.3 scats/km and hemlock 0.03 scats/km .

The cove hardwoods yielded no scats.

Realizing that one vegetat ion type can span one or more altitude classes and vice ver sa, an analys is of variance procedure was performed to 2etermine if these t�o

factors coul2 significantly affe ct the number of scats collected along Index Routes . The outcome proved that ther� did indeed exist a probability at or above the percent 95 level of confidence that the comb inat ion of altitude and its resultant vegetation type did sign ificantly affect the

location of the bear scats on all routes excep t the Spence­

Rus sell Index Route.

In order to attempt to explain the results for

Spence-Russell Index Route, one factor was investigated ,

that of hiker intens ity or simply the numb er of backpackers utilizing the routes . During a surve y showed that 1975 17� bad:packers Here met along the rout e during a given time

This number almost oercent larger than the p eriod. is 40 65 next highest number of backpackers encountered along any

of the othe r routes .

Due to the complicated nature of the statistical aspects of the prob lem and the fact that ���erous assump - tions would need to be made concerning the statistical validity of the many influences on the population density

(i.e. , temperature , precipi tat ion, behavioral reactions among individual bears , etc.) establishing such an index

was not feasible without a more in- deoth� analvs/ is of these

factors .

However, with the compiled data, a prediction

equation was established. Where p = mean numb er of scats/km

for a route or trail segment and = number of hikes , then �J

= pN . Using a table for re lative expected frequencies �,! for individual terms of the Poisson distribution to determine the correct value corresponding to it is possible to �! , determine if a given area is producing the nwnb er of expected scats .

It is important thou gh that each Index Route con­ tinues to be hiked , us ing the results from the prediction equation only as a check on the actual results ; nu.mb er of scats/km . Any deviations from expected (calculat ed) value s

could indicate a fluctuation in the population lev el, however, this indicat ion alone should not be considered a

accurate monitor of the population trends without true or fur ther re s earch ani continuation of monitoring o£ the

and its activities . bear population CHAPTER VI

RECQ[ •1MENDAT I O�i S

Add itional re search of associated factors is needed

if an independent index for monitoring population trends via scat collection is to be established. The fol lowing

list of projects might be included in an extended research program.

1. �onitoring of weathe r condi tions particularly among different vegetat ion types and among different

altitude types ; temperature , rainfall and runoff should be stressed.

2. A study of the deteriorat ion rate of black bear scats , on and off trail, to determine if the t�o-we ek

interval between collection periods is teo long for proper identificat ion of scats .

3. Reaction of bears (particularly backcountry bears opposed to panhandlers) to hiker intensity since as percent of the bears trapped and released in areas 95 cf the are never seen again (Beeman 1975) . Fark 4. Determination of the defecation rat e of black bears; with this information it would be possible to

estimate population density using the fol lowing formula:

s(scats)

t C c. i av; ""- ' ) • a.� ( 1'm J<.i. ·) ;- . d(scats/day and an imal)

c:� e 9 6 • ) f f 1 8 66 67

5. Comparison between on -trail and off-trail incidence of scats to determine the degree of variability of incidence and relat ive degre e of trail us e by bears during seasonal movements . LITERATURE CITED CITED LITE Ft"'.TURE

Barnes , G. and Bray . Populat ion chaTacter ­ istiV. cs and actC. E.ivities of196 bla7.ck bear in Yellowstone �at ional Park. Final Rent . Colo. Wildl. Res . Colorado State Un iversit� , pp. Un it . 199 Barr, A. J. and J. H. Goodnight . users guide to the Statistical An alysis Syst1976em. AStu dent Supply Stores, �orth Carolina State University , Rale igh , 260 pp .

Batcheler, Developme nt of a distanc e C. L. cens;1975,Js rrom pellet group . '""_\e[anthodage . �;fo9(·:'4) de :6er11 -652. J. \'i:i.ldl

3ee:nan , Seasonal food haoi::s -;:;"-e L. E. 1971. of blac:�: bear ( Ur su s ar:, e ric anu s) in the Great Smoky "ta of Tenne ssee and �ortli Carolina. UnpublisheYoud .n ins Thesis , The Un iversity of Tenn essee , �.S. pp . Knoxville , 61

Beeman , 75. Popu lation char act eristics , move­ meL.n tsE., and19 activities of the black bear (Ursus ame ricanus) in the Gre at Smoky �- iountains .\ atlonc;l Park. Unpub lished Ph .D. Thesis , The of Tenne ssee , Kn oxville, pp . Un iversity 318 and Pelton . Seasonal foods Beei:·:an , feL.edi E.ng eco:.rlog . yR. of black 1977. in the Smoky i'- fow;.ar�d ­ tains . ith Internat . Con£.be arson Bear Res . and

e � , a s p e , or: t an a ( I n. s s ) :. fan a g K 1 i 11 �I P r e , l � p p .

P, F. English, >1cCain . .�. Benne ttst, udyL� .} of., deer populationsa1 1dby ?� use. pelletl9� ·Dgr . oup ...:0unts . i\-ildl . �!anage . 4(.:1.) :.3ofS3-4 03. J.

a in , . . . r e ina r g u i d e t o t he G re en r i a r - C S Brus.� hy 1 9 3 7 . nat:....ure p tr1 inail , ythe Sr.w ky ,\lounbt ains Na�ional>It Pa. rk . The Un iversitGry eaoft Tennessee , Kn oxville , :'.fioeo ., pp . 29

t e :- 3 . black bear in Virginia Carp e :n , ;. Lo£ 1Ga 0 me7 The Inland FisheriesVirgini, Richma.on d , pp . Corrtr:1 . and 22 Daye:r:port , . .-\ gTicultural depredation L. B. 19 53 by the blacf.: bear 1 n i n i a � . i d . an a e . ( 3 ) : 3 31 - 3 .t 1) • \ � i Y g J 1\. l 1 �'·1 g 1 7

69 70

Eagar, C. 1977. Radioisotope feces tagging as a populD. ation estimator of black bear (Orsus ameri­ canus) density in the Great Smo ky Mountains �at1onal Park . Unpub lished �.S. Thesis, The Univers ity of Tennessee, Knoxville, 89 pp .

Eberhardt , L. and R. C. Van Etten. 1956. Eval uation of the pellet group count as a deer census me thod. J. Wildl . �fanage . 4(4) :398-403.

Erickson , A. W. and G. A. Petrid e s . 1964. Population structure , movements , and mortality of tagged bears in Michigan . In W. Erickson , J. Neller, and -- A. Petrides , ppA.. 46-67 . The black bear in MiG. chigan. Michigan Agr. Expt . Sta . Bul l. 4, 102 pp .

Eubanks , 1976. �!ovements and ac tivities of the black beaL.r in the Great Smoky Mountains National Park . Unpublished M.S. Thesis , The Univers ity of Tennessee , Knoxville, 33 pp .

F o 1 , F . G .?:.. , F o 1 , and J . J . in o r . 19 7 2 . Phy s o 1 o i - k cal . condi , �f . t ions kof thre e speciesyf of bears in winti er g dens . In pp . 107-124, Bears --their biology and management . Proceedings 1st Internat . Con£. on Bear Re s. and Ylanage ., Mer ges , Sw·i :erland. Publ. Series No. 371 pp . t ITJCN 23, Hungerford , K. E. 1953. A ruf fed grouse drumming technioue for Northe rn Idaho conditions . Uncoivunt. Idaho � crest, Wildl. and Range Expt. Sta. Research :� ote 0limeo ., pp . 10 , 3 K.1ng , B. and -� · Stupka. The Great Smoky :�loun­ P.ta ins--t he ir geology and19 50 natural. history. Sci . >fonthly 7 ( 1 ) 1 :31-43. Kozicky, L. , Bancroft , and P. G. Homeye r. 1954. An E.an alysT.is A. singing ground count s, - of woodcock � 1948. ; - 19� z • J • 1'.v l.J.",u -' 1 • : an age . 18(2) .·7 - ::J-9 '�66 - :.. • \l LaFollette , Julie D. 1974. Some aspects of the history of the black bear (Ursus americanus) in the Great Smoky �·l ountains . Unpu5IiSFi.e d :\I.S. Thesis , University of Tenn essee , Knoxville , 149 pp . The

\, . .._ 1 ... 1 . ...,_ o (::' s 1_, 6 � . _.'-'aap t a b"l"1 . 1 t o f � n 1r_0� s ...... a o 1 t � c h Le o:.; l chA.ang e. Q 6 G. Cox, ed. . , y pp . Read:ngs ,_ n iOnIn Jl-rJ.) , ·: onservat ecolc:.::u.v. �reridith � ..� Yo rk , Corn ::.; ew ,) 9 5 p p. �l I .J..

L ind e y G. and E. · lesl oH 1976. i n ter dormancy z , ?. C. � . W in black bears in sou t h� e s tern Washin gton . Wil dl . ��lanage . 40 (3) :408-415. J.

�la rcwn , L. C. An evaluation of radioactive feces tagging as a t e c hn i que for determin ing populat ion densities of the bla ck bear ( Ursus amer ican u s) in the G r e a t Smoky �·fount ains \at 1onaf Park. Unpub ­ l i she d �.S. Thesis , The University of Tenn essee, Knoxvi lle , 95 pp .

�arsh, G. G. 1973. Hikers 1n t h e Great Smoky �ountains Nat ional Park- -their attitudes , characteristi cs , i mpli c a t ion s for management . Unpublished and T h esis, The Un iversity of Tenn essee , �. S. Knoxville , 67 �P · �cClure , H. 1939 . o oin g activity and censusing of the rrt o u rn1n g doYeC. ') l!ildl . \far.age . 3(3) :323-323. J. �� rless, D. and t a ll n s . Hikers gu ide C. S i g 1974 . to Smokies . Sierra C l ub , r a nc i s c o, 375 San F pp . �eff , The pellet group count techn ique for D.bi J.g game1968 tr. end , ce nsu s and distribution ; a review. Wildl. :� !ar.age . 32 (3) :597-614. J.

Ov erton , S. 1971. s timat in g the numb er of l_ n ':·i , E an iii'.als •.d l d 1 i f e o u 1 at ons . R . H . G i 1 e s , r . , c p p i In J cL , . 0 3 - 4 5 6 , i f e I:l a r1 g emen t t e c hn i que s . e �ip p ld4l ife SocietY�\! i l d, 1 Washingtoa n , D. 633 pp . 'l l1 C. , Peltor., 1972 . Use of foot t r a i travellers Creat:1. R. Snloky· >�ct.::J.tains :-�ation a l Parkl to esti=:�.Jin.te �te ::: lack bear act 'i ity . In S. Herrero , e d . , Eears --t he ir bii ology anct management . Paperpps . 36 -�-2 , Proc. of the Internat . o n f on 3ear Res . ancian ci age . , C a a r , 1 e r Ct a . . U �; C Pu S e r i e s o : Lm3' 3 71 pp . l g y ,-\ b I b l . :: . 2 ?e l :on , R. C. !a r c wn . 1974. The notential use of>L radioandi sot opesL. for0 determining dens ii ies of black and other carni v o r es . Predator Sympos ium ,

bears' � ...... J • • � ...... , ...... c n;� \. .. .. �- \_nn . :J o c . \l8 "'T't!.J...lai ...... \ft-n..0a • , _\! 1-S . SOl'la...... , \!o nt-"...... Press) . (ln

P::t:" ::.. bc·rg , v1 . �. ·l \� e llein , :1.Roac, lsideG. and "\�� . E. G1J.::,Ta J. scYl . - a ceYl.Sti.S ' T '] '...,.. ..:'::1 'f ""/""1 • 1. 9S3 � - � � "'l �f� c1 :-u�lJn�-r·--,1in.g]-C'-' • coun.• I �syl·;�: �.,,, ·��., • . sp:-a •ing • met�1oc. I 1 - e d � -- v ·- ::, ·�' ' ..L '. c. ::; '-- • _;._ ', _;, ) rr-,,y" 1-"('1"\.� 2. 2-295, ��I -

Rhiney, T. 1957. The use of faeces cotmts in studies of several free- ranging mammals in New Zealand . N.Z.J. Sci. Technol. 38B(6) :507-S32 .

Roth, H. V. 1977. Defecat ion rates of captive brown bear . Presented at 4th Internat . Conf. on Bear Res . and Manage ., Kalisp ell, Montana (In Pre ss) , 15 pp .

Shan ks , R. E. 1954a. Climates of the Great Smo ky Moun ­ tains . Ecology 35(3) :354-361.

Shanks , R. E. 19S4b. Re ference list of native plants of the Great Smoky Mountains . Botany Department , The University of Tennessee , Mimeo ., pp . 14 Smith, R. H. and S. Galli:ioli. 1965. Predicting hunter success by means of a spring call count of Garnbel quail. J. '�'lildl . �·lanage . 29(4) :806-813.

Snedocor, G. W. 1950. 4th ed. Statistical methods . The Iowa State College Press , Ames , pp . 485 Sakal , R. R. and F. J. Rohlf. 1969. Biometry . Freeman and Company , San Francisco , 776 pp�.. H.

Spencer, H. E. 1961. The black bear and its stat�s Maine . Game Division Bulletin �o . Dept . in Inland Fisheries and Game , Augusta, 4. ofpp . �, !aine , 55 Tay lor, and R. Williams . 1956. The us e of pelletR. H. counts :-1for. est imat ing the density of popula­ tion of the wild rabb it , Oryctolagus cuniculus (L.). N.Z.J. Sci. Technol. 38B(3) :236-256 .

Van Etten, Bennett, Jr . 1965. Some sourceR. s C.of ander rorC. usL. ing pellet group counts for census ing deer. J. Wildl . :-Ianage . 29(4) :723-729.

Watt , K. E. F. 1968. Eco logy and resource management : a quantitat ive approach. McGraw-Hill Book Comp any , New York , pp . 450 Whittaker, R. H. 1956. Vege tation of the Great Smoky �fountains . Ecol. Monog. 26 :1-80.

Williams , H. W. 1961. The influence of phys ical and bio­ logical factors on the ral ly call of the chuk ar partridge (Alectocis graeca) regard to the use - of the call as a cens� s meth odwith. We stern Assoc. State Game an d Fish Ccmm . 41 :117-129 . APPE:JDICES APPEND IX A

SHEET- INDEX TR.I-\IL BLACK STUDY SU1--IMARY SCAT COLLECT IO:JS BEAR

Index Trail Hike r Date ----��-��-______------Trail ------we a the r Total N1oob er o f Bear Scat Co l lected: ------

Scat Exact Number Locat ion Cormnent s

74 APPEND IX B

CODING CLASSIFICAT I ONS

L:'\.B LE 3 CODI :\C; USED FOR \"E GELH ION ASD .-\ L TITUDDiAL CLASSES

Class Code Code Definition

.\1 tit ude ,\ ctua1 Values

1 457- 09 T!1 (1500 -1999 ft ) 6 610- 762 m (2000-2499 ft)

...... 1 763- a�J. j m (2500-2999 ft)

914-1066 El (3000-3"199 .c+.L '- )

5 1057 -1218 Til (3500-3999 f t )

t- � 6 1219-1371 J1l (4000-,t-�99 �F �)

1372-1524 !:t (4500-4999 \.. ) £+-�

1525-1676 m (5000- 5199 f •c I - '- ) s

(\_! 1677+ m (5500+ ft)

Ve getat ion Ty-pe

1 Closed oak

Open oak-pine and balds 3 Northern hard�ood

Heml oc }( Spruce - fir

75 AP PE;,;D IX C

TOTAL SCAT S COLLE CTED

T �\B LE 9

TOTAL OF SCATS COLLECT::D ALO\G EACH I.'�DEX ROUTE.':L0 AN!B ERD WIT H I:: EACH VE GETATION TYPE I� THE GREAT SMOKY 0L�TI O:JAL PARK �!OU�L\DJS

------egetat icr:. Index :�cute '/ Type Scats Scats/!0� # D!

Su gar1ands Closed oak 3.4 Cpen oak-pine 743 6.9 and �alds 10 .

Spruce fir 62 7I 7I 8.1

lk.:.ont- Closed oak 1 17.8 E Bent Ar:n :.:ort.hern hd'A·d . 2 0 15.8 11.8 113 Tremont - C(; rr ick Closed oak 2.4 .8 3. 0 1 ' KI10b :;ort1t.e rn ..,. , .., hd•\',:d . 10.2

Spence -?us sell Cove hardh·ood .),u� n 0' Closed oak "T 4.4 ()'-. J Q 7 r Cpen oal<-.. pine 3 .) , ;:) and balds 0.9 .'Jorthem hd��·d . .., , ' Hemlock -t 5.4 3.9 5.3 ;) . 8 G:regory-Hannah Cove harc�1-:o cc t.a Closed oak !'") 4.5 :. rcUJl L,·1 Cpen oak-pine 36 9.2 balds 20.3 2.8 :-rorthemartd h±�·d. 6 1.2,., � 5.0 Overall Total Cove ha rch,;ood / . .) Closed oak () C):en oak-pine 239 4930.7.6 5.8 2n d balds 133 4.3 .'Jorthern. hd11d . 139 7.3 Hemlock 12.0 0.3 Spruce -fir 62 8.1 ...... - L:u . ·•. APPE�DIX D

A.l'JOVA TAB LE FOR SCATS/KM BY

ALTI TUDE -VEGETATION TYPE

TAB LE 10

COMPLETED }\NOVA TABLE FOR S CA S/K:V1 BY ALTITUDE ­ VEGETATION TYPE IN THE TGR EAT S�OKY MOUNTAINS NAT IONAL PARK , 1970-1975

Source df ?v!S F ss A.rnong classes 8 272. 3486 34. 0435 2.5401

Linear regression 1 226. 9609 226. 9609 35. 0032

Deviat ions from 7 45.3877 6.4840 <1. 00 regre ssion

Within groups 19 254. 6425 13.4022

Total 27 526.9911

77 APPE�DIX E

DESCRIPTION OF INDEX ROUTE LOCAT IONS

AND VE GETAT ION TYPES

Sugar lands

This rout e begins at the Spruce-Fir Nature Trail on the Clingman 's Dome Road. Approx imately 0.8 km from

its starting point is the Mt . Collins trail shelter. The

first section of the Sugarland Route follows the Sugarland

Mountain Trail fo r km at which point it intersects 14 .3 the Huskey Gap Trail. The Index Route turns east at this

junction and follows the east half of Huskey Gap Trail

for km where it terminates on Highway The 2.9 U. S. 441 . Sugarl ands Route is characterized by thre e general vegetation types. The spruce -fir vegetation type is

dominant above m. The open oak-pine and 1372 balds vegetat ion type is found from m to m and the 914 1372 closed oak vegetation type is found from m. 573-914

Elkmont-Bent Arm

Located in the watershed of the East Prong of the

Little River, this rout e begins at the Cucumb er Gap trail

he ad and travels west for 2 km to the Bent Arm manway which

leads southward. The manway intersects Miry Ridge Trail

after km. Continuing southward along Ridge Trail 6 ;,!i ry for the route inters ects the App alachian Trail 4.5 km ,

78 79

(AT) at Buckeye Gap. The Index Route then proceeds east along the AT . Along this sect ion cf the AT , trail

shelters are located at Silers Bald and Doub le Springs

Gap , 4.2 km and 6.9 km east of Buckeye Gap , respectively.

Goshen Prong Trail intersects the AT 7.9 km from Buckeye

Gap. At this intersection the Index Route proceeds for

11.7 km (7. 3 mi) along Goshen Prong Trail and then terminates on the Little River Gravel Road . Two veg eta- tion types , closed oak and northe rn hardwoods , characterize this route. Closed oak fo rests are located pre dominantly at the lower elevations 792-1372 m while the northern hardwoods located at the higher elevations betwe en

1372- 1709 m.

Tremont -Derrick Knob

Starting at the Panther Creek trail head on the

Tremont Road , this route turns southward , after 3. 7 km, at

Jakes Gap and follows Miry Ridge Trail until it terminates at the AT . (This last section along Miry Ridge is also shared with the Elkmont -Bent Arm Index Route.) At this point the route proceeds west along the AT for 2 at 4. k"'TI wh ich point it reaches Derrick Knob . Located at Derrick

Knob is the only trail shelter along this route . the Here route turns north along Davis Ridge and after km 8.5 terminates the Tremont Road just ab ove the originat ing on point of the route. Similar to the Elkmont -De�t route , Arm 80 there are two gene�al vegetation typ es ; closed oak and northern hardwoods . The closed oak fo rests are located betwe en elevations of 792-1372 m while the northern hard­ woods are situated between 1372-1578 m.

Spence-Russell

This route begins in the Cades Cove Picnic Ground at the Anthony Creek trail head. It proceeds for 5.5 km to the Bote �ountain Road where it turns south following the road for km where it intersects the AT crossing 2.6 Spencefield. At the AT the route continues we stward for

3 .7 km where it reaches Rus sell Field. Trail shelters are loc ated at both Spen cefield and Rus sell Field. The rout e turns north at Russell Field and fo llows Leadbetter

Ridge for km until it intersects the Anthony Creek 5.6 Trail. At the junction of the two trails the rout e fallows

Anthony Creek Trail back to the route's originating point.

Five vegetat ion types are found along this route . In the lowest elevations along sheltered streams are the hemlock forests. �loving to higher elevations , but still in sheltered sit uations , cove hardwood forests are found . At the same elevations as the cove hardwoods, but in mo re exposed sites , are the closed oak forests which in turn give way to the open oak-pine and balds vegetation type on most exposed sites along the route . no�thern the The hardwoods vegetation type is found above the m 1372 elevation level. 81

Gregory-Hannah Mountain

The first section of this route begins at the

Gregory Ridge trail head on Forge Creek Road and continues

for 8.0 km (5.0 mi) whe re it terminates at Rich Gap . The rout e continues along the Gregory Bald Trail for 1.1 km to the crest of Gregory Bald. After crossing Gregory Bald , the trail turns north and descends to its termination at

Parson' s Branch Road. The Index Rout e continues along

Hannah Mountain Trail for 12.1 km at which point it inter­ sects Scott Gap on the Rabbit Creek Road . Here the rout e turns east , follows Rabb it Creek Road for 11.3 km , and terminates in the Ab rams Falls parking area at the west end of Cades Cove . The rout e is characterized by four general vegetation types. Cove hardwoods are located in sheltered are as between 522-1372 Closed oak forests, km . found at the same elevat ions as cove hardwoods bu� in more exposed sites , give way to the open oak-pine and balds on the most exposed ridges and gaps . The northern hardwoods

are found ab ove the 1372 m elevation level. APPENDIX F

SCATS COLLE CTE D BETWEEN 15 OCTOBER AND 1 JUNE

TABLE 11

DATES AN'D Nillv1BER OF SCATS COLLECTED BETWEEN 15 OCTOBER A�D 1 JL�E

Index Route Scats Collected Means

Sugarla nds s- 3-/0 2 Post-summer season 0.10 scats/ki-n 5-16-70 1 Pre - sumner season 04 scats/"km 5-30 -70 1 0. 10-18-70 3 10-31-70 1 11-13-70 0 11-1 5-70 1

4- 3-il 0 4-17-71 0 4-31 - il 0 ,., 10-18 -il i. 10-29- il 1

5-28-72 1

10-15-74 4 IT

Tremont -Derrick Knob --4- 4-70 0 Post-surmner season 0.10 scats/k:.-n 5-30-70 2 Pre-swnme r season 05 scats/km 10-18-70 3 0. 11-12-70 .) 12- 9-70 4

4- 3-71 0 4-17-71 0 5-16-71 3 5-29-71 0 10-17-71 10-29-71 5 2

82 83

TAB LE 11 (continued)

Index Route Scats Collected ivfeans

11- 3-71 0 12-15-il 2

5-27-72 3

10-12-74 1 28

Elkmont -Bent

-Ann4- 4-70 0 Post-su�er season 0.05 scats/kw 4-18-70 1 Pre - sumner season 0.07 scats /km 5- 2-70 1 5- 3-70 1 5-16-70 1 5-30-70 12 10-17-70 3

4-17-71 0 4-29-71 0 ,., 5- 1-71 .:.. 1 5-,- 16-71 .... o- 4-71 :::; 10-30-71 2

5-27-72 3 10-28-72 0 TI"

SE_ence-Russell 4- 4 -io 0 Post-summer season 0.04 scats/Nu 4-18 -70 Pre-swnner season 0.06 scats/kra 3-70 61 5-31-70 1 10-22-70 0 11-31-70 0 12- 5-70 1

4 - 3·-71 0 4-18-71 0 5- 2-71 0 0 5-16-71 5-27-71 .) 84

TAB LE 11 (cont inued)

Index Route Scats Collected Means

5-29-71 0 10-16-71 4

3-10 -72 0 5-27-72 3 12-17-7 2 0

10-19-73 1

10-12-74 0 71)

Gregory-Hannah Mountam 4- 4-71 0 Post-sumrr�r season 0.05 scats /k� 4-19-71 0 Pre-STh'1mer season 0. 005 scats/km 5- 1-71 0 5-16-71 1 6- 3-71 10-16-71 01 10-28-71 1

5-27-72 0 10- 2-72 3 11-24-72 0

10-10-73 5

10-12-74 0 11 APPENDIX G

ANOVA TABLE FOR SCATS/KM BY

HIKI:JG PER� OD

TABLE 12

A.t\J OVA TABLE FOR SCATS/Df BY HIKI:JG PERIOD THE GREAT S0!0KY :'

Source df F ss ;,rs Among groups 8 0.0570 0.0071 1.34*

Within groups 44 0.2317 0.0053

� ,.., Total ::>t. 0.2887

* = significant difference at the 75 percent confidence level.

85 APPENDIX H

NUMBER OF SCATS COLLECTED WITHIN

ALTITUDINAL CLASSES

TABLE 13

NUMBER OF SCATS COLLE CTED WITHIN ALTI TUDINAL CLASSE S ALONG INDEX ROUTES IN THE GREAT SMOKY MOUNTAINS NATI ON AL PARK , 1970-1975

Index Routes . Tremont - Gregory- Altitude Elkmont- Derrick Spence- Hannah (m) Sugarla nds Bent ..-\rm Knob Rus sell Mountain

457- 609 --* 2

610- 762 3 0 2 _jr

763- 913 5 15 1 1 44

914-1066 10 38 3 3 3

,.., 1067-1213 17 32 6 3 .)

1219-1371 55 65 26 8 4 1372-1524 18 83 48 12 5

1525-1676 25 66 ;) 3

1677+ 11 29

* = this altitude did not appear the corresponding Index Route . i11

86 APPENDIX I

NUMBER OF SCATS COLLE CTED WITHIN VE GETAT ION TYPES

TABLE 14

NUMBER OF SCATS COLLECTED ALONG INDEX ROUTES WITHIN VEGETATION TYPES IN RE SPECT TO ALTI TUDE CLASSES IN THE GREAT s:,!OKY MOUNTAINS NATIONAL PARK , 1970-1975

Alt itude Class ( T:1) Ve getation Types K?v! Scat s Scats/K?v! it 457- 609 Closed oak 1.4 1 0.7 Cove hardwood 0. 7 0 Open oak-pine 4.1 1 0.2 and balds

,1 ... 610- 762 Cove hardwood '+ • J 0 ' Closed oak 5.6 18 .).., .) � Open oak-pine 5. 4 1.5 and balds 8 Hemlock 3.4 2 0.6

763- 913 Cove hardwood 1.4 0 Closed oak 9. 4 21 2. 2

Open oak-pine ...,I • � 7 44 6.1 and balds Hemlock 1.3 1 0.8

914-1066 Closed oak 10.3 46 4.5 Open oak-pine 2 . 9 10 3.4 and balds Hemlock 0. 1 1.4 7 1067-1218 Closed oak 8. 1 40 4.9 Op en oak-pine 5. 4 21 3.9 and balds

1219-1371 Closed oak 10.1 81 8.0 Open oak-pine 5.1 48 9.4 and balds :J orthern hdwd . 4.0 14 3. Spru ce-fir 0.6 8 13.35

O"" u : 88

TABLE 14 (continued)

A ltitude " Class ( m) Vegetation Types KM :r Scats Scat s/KM

13i2-1524 Closed oak 6. 4 82 12.8 Open oak-pine 1.3 1 0.8 and balds Northern hdwd. 15 . i i2 4.6 Spruce-fir 1.3 20 15.4

1525-1676 �orthern hdwd. 11.0 i4 6.7 Spruce-fir 3. 2 23 .,I • � ?

16i7+ �orthern hdwd . 2.6 29 11.2 Spru ce- fir .)� . � 1 11 3 . 5 VITA

Susie Jo Kelly Matthews was born in Trieste , Italy, on Feb rua ry 19, 1951. She attended elementary schools in

Ludington, �!ichigan and was graduated from Ludington Senior

High School in 1969. In September, 1969 , she entered

Grand Valley State College and received a Bac helor of

Science degree in Environmental Sciences March of 1973. in In September, 1973, she entered the Graduat e School at

The Unive rsity of Tennessee. She received her Mas ter of

Science degree in Wi ldlife and Fisheries Science from the

Department of Forestry, Wildlife , and Fishe ries in August

She is married to ;.richae l R. i>latthel'iS . 19 77 .

89