Avian Population Density in the Maritime Forest of Two South Carolina Barrier Islands

HABITAT STUDY

Avian population density in the maritime forest of two South Carolina barrier islands

An investigation of the factors that influence speciesdiversity on these islands

W. David Chamberlain

INTRODUCTION giniana), CabbagePalmetto (Sabal pal- METHODS metto) and Loblolly Pine (Pinus taeda). HEBARRIER ISLAND system of the The understory vegetationis dominated MODIFIEDSTRIP census method southeastern United States is under by Wax Myrtle (Myrica cerifera), (Emlen 1971) was employed, increasing developmental and recrea- Yaupon (Ilex vornitoria) and Cabbage with three transects laid out on each tional pressure. Little is known of the Palmetto (Table 1). island. All transects traversed the width faunal communities associated with the The topography of both islands is of the island from salt marsh to dune maritime forest on most of these is- characterizedby alternatingdune ridges field. Each transect was 400 feet { 122 m) lands. In South Carolina research on and high ground with drains and depres- wide and crosseda variety of habitats. island avian communities has been lim- sions. These low areas often have stand- Transects were designed to measure ited to qualitative statements produced ing water. This pattern of alternate xeric areas of heavily disturbed, lightly dis- by general field studies and Christmas and mesic sitesallows the development turbed and undisturbed habitat. All Bird Counts. This study was designedto of several vegetation communities transects were sampled biweekly from document avian density in maritime for- which vary in extent on each island. March-September 1979 between 0600 estsand to predict the impact of recreational and developmentalpressures on barrier island habitat. BIJLL 8,•g

STUDY AREAS ".[

ISLAND ATAWERE GATHERED fromCap- er's Island, 15 miles north of Charleston, Charleston County, South CAPERSISLAND Carolina, and Kiawah Island, 21 miles south of Charleston. These islands were selected because of their differences in size, accessibilityand vegetation. Cap- er's Island is a small state-owned wildlife sanctuary with approximately 830 acres of high land (Warner 1976). Ac- ß CM,•RLESTOIiI H,•RBOR cessible only by small boat, Caper's Island enjoys light recreational use limited to nature observation and occa- sional primitive camping. On the other hand, Kiawah Island, with approximately 3300 acres of high land, (Warner 1976)is privately owned, and undergo- ing intensedevelopment as a resort and second home community. It is accessi- ISLAND ble by bridge, and there is heavy recreational and residential use over one half of the island.

Maritime forest communities on both NORIll fOl$1'O RIVfR Kiawah and Caper's islands are dominated by Laurel Oak (Quercus General Stlldy area map/courtesy of Pete Laurie, Sollth Carolina Wildlife and Marine !aurifolia), Live Oak (Quercus vir- ResourcesDepartmentß

142 American Birds, March 1982 Table 1. Vegetationdensity of undisturbedMaritime Forest in two South Carolina barrier islands

Height Canopy Cover Canopy trees/ Understorytrees/ Shrubsand saphngs/ Community (fi) % 100sq.m. I00 sq.m. 16 sq.m.t Kiawah Island Loblolly Pine•abbage palmetto 70-80' 60% 5.0 5.5 I 1.0 Transoct 3 Laurel Oak---Magnolia 30-40' 65-80% 6.8 9.0 4.0 Cabbage Palmetto 20-30' 50-60% 7.2 5.1) .5 Maritime Shrub 10-20' 20-50% -- 13.5 6.2 Loblolly Pine•weetgum--Live Oak 40-80' 60-70% 5.0 10.5 1.0 Laurel Oak--pine--Cabbage Palmetto 30-70' 60% 14.8 9.5 3.5

Caper's Island Laurel Oak---magnolia 30-40' 65-80% 3.8 3.2 .5 Transect 3 Loblolly Pine--Live Oak 60-75' 60% 2.2 3.8 1.5 Cabbage Palmetto-' 20-30' 50-60% I 1.5 27.8 10.0 Laurel Oak--pine 40-70' 60% 4.2 l0 5 .2 •A saplingis a tree species<6.6 cm 13in.} dbh. -'Thepresence of ChineseTallow Tree drasticallyincreased the densityof this community. [Source: Gaddy, unpublishedms.; Sharitz. 1975] and 1000 hours. The data presented are viduals/km: (Table 2). Analysis of the confined to observations made in results by island and transect reveals maritime forest communities. Overfly- considerable variation particularly in ing marine and estuarine species were average density. not included. Similarly, swallows and Although the study lasted less than a similar specieswere omitted becauseof year, seasonal variations both in indi- the difficulty of accurate counting im- viduals and in densities were noted. In- posed by the dense vegetation. Care dividuals seen monthly per transect are was also taken to avoid the error im- given for both islands. Patterns as- posed by "driving" species along the sociated with the departure of winter transect. Individuals observed were residentsand the influx of springand fall mapped indicating position on each migrantsare more apparent on Caper's transect and distance from the center Island (Table 3). Further examination of Caper's I., •}'ansect 1, Maritime shrub. line. Additional notations were made seasonal distribution revealed the for singing males, copulating pairs or land Transect 3 traversed undisturbed greated number of species of all active nests. maritime forest. transects were permanent residents. Transects were laid out on each island Because continuing disturbance was Migratory species were approximately to take advantageof varying vegetation a factor on both islands. botanical data evenly distributed throughout the communities and levels of disturbance. were gathered only for undisturbed transects (Table 4). On Kiawah Island, Transec! I was transects.Sample plots were selectedin Habitat relationships strongly in- through highly altered maritime forest. representative communities along each fluenced both the number of species Portions of this transect included de- undisturbed transect. Plots were 400m' veloped areas of single family housing in area with 10m-' areas sampled for while the majority was along a golf coursefairway. All portionsi•fTransect trees,and4m-' forshrubs andsaplings. 1 were heavily disturbed. Transect 2 on RESULTS Kiawah Island ran along a lightly used dirt road through maritime forest and into an old pasture that was reverting to •3054•VER THEindividuals SAMPLE •fiodrepresenting atotal 77of •'•,";•'•{•'•' •:•' ' "•' •½•' the surroundingforest type. Transect 3, ran through undisturbedmaritime forest rangefor individualspecies ranged from _ and included the margin of a small <1 individual/kmeoraea. • to 66/km aeni,vz. On brackish pond. KiawahIsland theaverage density was On Caper's Island, Transect I ran a 2• individuals/km:, while the average KiawahI., 3'ansect1, Heavilydisturbed short distance through maritime forest on Caper's Island was 237 indi- maritimeforest. and included the margins of two large brackish impoundments. These pond Table 2. Number of bird species,individuals and averagedensity by transect margins were dominated by maritime shrub speciessuch as Wax Myrtle, Sea Area in Acres Average Densit)' Myrtle (Baccharis halimifolia), Yaupon Transect (Hectaresl Total Species Total Individuals lndividaals/km 2 and Chinese Tallow Tree (Saplure Caper's I. I 33.6 (13.6) 51 548 287 2 31.8 02.8) 30 305 170 sebiferum). Disturbance on Transect I 3 30.3 (12.2) 35 438 254 was light and limited primarily to hikers and campers. Transect 2 was located Kiawah 1. I 48.7 (19.7) 38 309 112 entirely in maritime forest along a 2 42.4 (17.1) 49 742 309 lightly travelled dirt road. Caper's Is- 3 34.9 (14.0) 47 712 361

Volume 36, Number 2 143 Table 3. Seasonal abundance of individuals on two coastal South Carolina Islands these, 15 were permanent residents. On Caper's Island 14 species were noted Transect March April May June July August September with seven being permanent residents. Caper's 1. I 114 84 63 69 96 62 69 Significantdifferences in the number of 2 34 45 44 57 33 34 58 3 77 48 43 32 56 122 68 singingindividuals were noted between the islands; the average numbers of Total 225 177 150 158 185 218 195 singingspecies encountered each sam- Kiawah I. I 46 65 49 55 37 35 41 ple day were 15.6 species for Kiawah, 2 85 90 131 128 92 107 II0 6.6 for Caper's. 3 109 143 84 98 84 103 91 Common species were also omitted Total 240 298 264 281 213 245 242 from the list of singing males for both islands. Two such species were the recorded on each transect and the total guilds and their number of representa- Great Crested Flycatcher (Myiarchus number of individuals. Species diver- tives empha.size the importanceof crinitus) which was present but gave sity was clearly determined by the total habitat. As would be expected, only a short alarm call, and the Carolina diversity of available habitats. This fac- transects with seral stages displayed Chickadee (Parus carolinensis) which tor was particularly significant on larger numbers of ground-seed and did likewise. The Solitary Vireo (Vireo Kiawah Island where the presenceof an ground-insect guild representatives. •olitarius) recorded on Kiawah is a win- abandoned pasture reverting to mari- The impact of habitat alteration noted ter resident which apparently begins time forest accounted for a number of on Kiawah Transect i is again demon- singingprior to its springdeparture. speciesabsent from the other transects. strated by the paucity of ground-seed Table 4. Speciestotals by transectson two On Caper's Island the influenceof seral and ground-insectrepresentatives. The South Carolina barrier islands vegetation was also important on lack of foliage-nectarrepresentatives on Transect 1 where maritime shrub two of the Caper's Island transects and Perre. Sum. Wint. growth is present along the pond mar- Res. Res. Res. Trans. gins. Kiawahl. Disturbance factors noted on each Transect I 23 9 5 I transect did not appear to significantly Transect 2 32 9 6 3 Transect 3 27 12 6 2 affect bird populations with the excep- tion of Kiawah Transect 1. Dramatic Caper's 1. impacts related both to continuingdis- Transoct I 30 10 7 6 turbance and the wholesale removal of Transoct 2 16 7 5 3 Transect 3 17 9 6 4 understory vegetation were immedi- ately apparent (Table 5). •Permanent resident species including non- breedingspecies. are present year round. Summer The structureof populationswas also residentsare those occurring and breeding during examined employing feeding guilds the summer. Winter residents are those species (Salt 1953, Manuwal 1968). The feeding presentfrom fall to early spring.Transient species are thoseappearing only duringmigration. !

DISCUSSION

HELIMITED duration of thisstudy resulted in insufficient data for con- Caper's !., Transect 2, Maritime forest. All clusive statistical treatment and pointed pfiotoslWDC. out the need for additional studies in timber-searching representatives on these habitats, especially in winter Kiawah Isl•d appear to be related to months when large numbers of visitors the overall size (Table 6). move into the area. Passefine density Kiawah !., Transect2, Old pasture reverting A tot• of 25 speciesof singingmales could well be greatest during this season. to maritime forest. was recorded on Kiawah Isl•d •d of Density figures for the heavily dis-

Table 5. Habitat utilization of birds on two South Carolina barrier beach islands

AreraRe tultnber Disturbance Arerage ntunber individuals Transect Habitat Factor •ype Speciesper stintpie per sa.tple Ca•r's I. I pondmargins, maritime shrub, light natureobservation, camping 13.4 39 maritime forest 2 maritime forest, dia road light foot traffic 8.3 22 3 maritime forest undisturbed 8.6 3 I

Kiawah I. I disturbed maritime forest heavy residential developmenl. 11. I 22 golf course 2 abandonedpaslure, dirt road, medium roadtraffic f8.5 53 maritime forest 3 pond margin, maritime forest undisturbed 16.2 51

144 American Birds, March 1982 Table 6. Feedingguild distribution on two South Carohna barrier islands

Ground- Foliage- Foliage- Ground- Air- Timber Timber Foliage- Transect seed seed nectar insect insect drilling searching insect Mist Caper's I. I 12 I 1 7 3 2 I 18 6 2 6 2 1 1 1 I 15 4 3 7 2 4 3 2 1 13 3

Klawah I. I 5 2 1 2 3 4 -- 17 4 2 10 I 1 6 2 3 -- 21 6 3 9 1 1 4 3 3 -- 21 5 turbed transect on Kiawah Island are transects, but the Hairy Woodpecker Heritage Conservation and Recreation well below all other transects while (Picoides. villosus), Eastern Phoebe Service and the Heritage Trust Divi- other differences in individual species (Sayornis phoebe) Acadian Flycatcher sion, South Carolina Wildlife and densities are related to habitat. On (Empidonax virescens), Eastern Wood Marine ResourcesDepartment. I thank Klawah Island Transect 2 for example, Pewee (Contopus virens), Carolina Thomas Kolhsaat, Director, S.C. Hen- the heavy density of White-eyed Vireos Chickadee, and Tufted Titmouse (Parus tage Trust Division, The Kiawah Island (Vtreo griseus) and Painted Buntings bicolor) were not recorded. While sev- Company for its kind cooperation, De- (Passerina ciris) is associated with the eral of these speciesmay not have had nnis M. Forsythe and E. Burnham old field-seralcomponent. Similarly on preferred habitat available, the absence Chamberlain for editorial comments Capeifs Island the large number of Red- of the Carolina Chickadee and Tufted winged Blackbirds (Agelaius phoeni- Titmouse remains a mystery. Both is- LITERATURE CITED ceus) on Transect 1 reflects the impor- lands failed to host a number of common tance of pond margins to this species. winter residents,but this was probably CHAMBERLAIN, W.D. and E B Capeifs Island also displayed an overall owing to the brevity of the study. The CHAMBERLAIN. 1975. Avifauna of low density along Transect 2. This House Sparrow (Passer domesticus) Kiawah Island in: Environmental In- ventory of Kiawah Island. Environmen- hghtly disturbed transect through the and Starling(Sturnus vulgaris) were the tal Research Center, Inc. Columbia, maritimeforest did not appear to differ only abundant mainland species that S.C. •n habitat quality. This unexplained were not observed. COCKFIELD, B.A., J.B. TORMEY and scarcity can perhaps be attributed to While this studyfocused primarily on D.M. FORSYTHE. 1980. Barrier I Maritime Forest, WB-PS. Am. Birds patchy distribution. The unduly high passerines, some interesting informa- 34:29. density level of the Common Grackle tion on large predatory specieswas also EMLEN, J.T. 1971. Population densities (Qmscalus quiscula) on Caper's Tran- gathered. The low density figures for of birds derived from transect counts sect 3 is an artifact created by a single hawks, owls, and vultures possibly Auk 88:323-342. large foraging flock encountered only reflectsa lack of preferred habitat in the ß 1978. Density anomalies and regula- tory mechanisms in land bird popula- once. case of the Red-tailed Hawk (Buteo tions on the Florida Peninsula. The It •s clear that there are differences in jamaicensis) and American Kestrel Amer. Naturalist 112:265-286. species diversity between individual (Falco sparverius), the size of the island KARR, J.R. and R.R. ROTH. 1971. Vege- barrier islands which are directly re- and consequentrange restriction in the tation structure and avian density in several New World areas. The Amer lated to the complexity of vegetation. case of vultures; and the difficulty in Naturalist 105:945. Perhaps more important are the differ- locating prey in the case of owls. Hawks MacARTHUR, R.H. and E.R. PIANKA ences in avian density. While this too and owls are both apparently limited by 1966. On optimal use of a patchy envi- can be attributed to island size, to some the density of vegetation through most ronment. The Amer. Naturahst 100: 603 -609. degreeit can perhapsbe assignedto the of the forest. Avian density was also MANUWAL, D.A. 1968. Breeding bird patchy bird distribution. Patchy distri- clearly related to disturbance. This fac- populations in the coniferous forest of bution has been noted in tropical areas tor was particularly evident on Kiawah western Montana. M.S. Thesis Umv of extensive homogeneous habitat Island where the heavily disturbed Mont. lZ/6p. (MacArthur and Pianka 1966) and could Transect 1 showed 20% fewer species SALT, G.W. 1953. An ecological analysis of three California avifaunas. Condor well apply to maritime forest. Addition- and a 66% reduction in average density. 55:258-273. ally, tropical species in homogeneous At present there exists little informa- SHARITZ, R.R. 1975. Forest communities habitats have been found to have larger tion on South Carolina mainland of Kiawah Island in: Environmental In- home ranges (Karr and Roth 1971). At habitats which allows a density com- ventory of Kiawah Island. Environmen- present home ranges have not been parison. Densities obtained by Emlen tal Research Center, Inc. Columbia, S.C. studied on barrier islands, but this con- (1978) for northern Florida (305 indi- TORMEY, J.B., B.A. COCKFIELD and cept may aid in explaining the absence viduals/km:) are perhaps similar to the •' D.M. FORSYTHE. 1980. Barrier I of several common speciesand the role regional mainland. Indications are that Golf Course Subdivision. WB-PS. Am of large predatory species. maritime forest densities are substan- Birds 34:40. WARNER, L. 1976. The status of the bar- Each island lacked sixteen species tially below the adjacent mainland. rier islands of the southeastern coast recorded on the other. On Kiawah all Open Space Institute and Natural Re- sixteen species were observed off the sources Defense Council. New York transectsor have been recorded previ- ACKNOWLEDGEMENTS ously (Chamberlain and Chamberlain 1975). On Caper's Island, none of the This study was supported by a grant ---204 Palm Street, sixteen species was observed off the from the U.S. Department of Interior, Mount Pleasant, SC 29464

Volume 36, Number 2 145