BREEDING HABITAT USE BY COLONIALLY NESTING WATERBIRDS IN TWO MID-ATLANTIC US REGIONS UNDER DIFFERENT REGIMES OF HUMAN DISTURBANCE

Transcription

1 Biological Con~ervalion 18 (1980) 3951 BREEDING HABITAT USE BY COLONIALLY NESTING WATERBIRDS IN TWO MID-ATLANTIC US REGIONS UNDER DIFFERENT REGIMES OF HUMAN DISTURBANCE R. MICHAEL ERWIN* Massachusetts Cooperative Wildli]e Research Unit, Department of Forestry & Wildli~b, University of Massachusetts, Amherst, Massachusetts 01003, USA ABSTRACT More than 80% of the beach-nesting seabirds (common tern, least tern, black skimmer, and herring gull) in coastal Virginia nest on natural barrier island beaches, while in New Jersey the vast majority nest on dredge deposition material or natural marsh islands. This contrast probably results j~'om the differences m human disturbance in the two regions. Although 75 % of all oceanji'ont in New Jersey allows unrestricted reereation, about 85 % of the Virginia beaches are "protected" under the ownership of several conservation agencies. Attendant with changes in habitat utilisation in New Jersey, competitive interaetions have apparently intensified with herring gulls usurping tern and laughing gull nest sites. Other implications are discussed. INTRODUCTION Historically coastal regions have witnessed intense man-induced perturbations of the natural system, such as real-estate and recreation development, while shipping ports and more recently oil-related activities have significantly altered fish and wildlife habitats and populations (Cramp et al., 1974). Despite the interest in coastal environments, little effort has been made to assess the impact of man's activities on habitat utilisation by either marine or terrestrial faunas. Only recently has any attempt been made to document habitat use by one of the dominant (in size) and most conspicuous faunal components, colonial waterbirds (Soots & Parnell, 1975a, b; Buckley & Buckley, 1975, 1977). In this paper, I examine breeding habitat use by four species of'beach-nesting" * Present address: US Fish & Wildlife Service, Migratory Bird and Habitat Research Laboratory, Laurel, Maryland 20811, USA. 39 Biol. Conserv /80/ /$02'25 Applied Science Publishers Ltd, England, 1980 Printed in Great Britain

2 R. MICHAEL ERWIN seabirds in two areas of the mid-atlantic coast with contrasting regimes of human disturbance. Common terns Sterna hirundo, least terns S. albifrons, black skimmers Rynchops niger, and herring gulls Larus argentatus are abundant breeders from Long Island, New York, to Virginia. Historically, these species nested primarily on the open beaches and dunes of islands along the eastern US coast (Bailey, 1913; Bent, 1921; Stone, 1937; Kadlec & Drury, 1968; Bull, 1974). In recent years, the disturbances along the oceanfront caused by spreading urbanisation in areas such as southern New Jersey and Long Island, New York, have presumably caused habitat shifts by some colonial waterbirds (Buckley & Buckley, 1975, 1977; Burger, 1977; Burger & Lesser, 1977). These recent studies, however, suffer from a lack of comparable data from periods before human encroachment. Records of breeding distributions from the 19th and early 20th century are generally anecdotal with little attempt at comprehensive coastal surveys. Because of incomplete data from earlier, "pre-treatment', periods, this paper compares current habitat use patterns in southern New Jersey, a highly-urbanised region, with the eastern shore of Virginia, probably the most pristine barrier island region along the Atlantic coast (Hennessey, 1976). Evidence for changes in interspecific interactions resulting from habitat shifts is also presented. METHODS Rationale Unequivocal demonstration of man's effects on bird populations requires carefully controlled field and/or laboratory experimentation. Such conditions in the field are often difficult or impossible to attain (Colwell & Fuentes, 1975). Although less powerful, the 'natural experiment" approach offers an alternative research strategy (Connell, 1975). By comparing the responses of species from two or more areas under different regimes of human activity, reasonably conclusive results can be obtained. Of course, it must be assumed that geographically separated breeding populations are genetically homogeneous and that the only differences between areas are in the variables being tested. Both of the above approaches are superior to the correlative approach where local changes in populations coincident with some human disturbance factor(s) are suggested as constituting an effect-cause relationship (Allen, 1937; Miller, 1943; Howard, 1968: Bull, 1974). Stbt(ll' UFCH Southern New Jersey and the eastern shore of Virginia are separated by only a short (ca. 90 km) expanse of oceanfront in Maryland and Delaware. Both areas are very similar physiognomically, with long, narrow barrier islands protecting vast areas of salt marsh. An intracoastal waterway parallels the shoreline in both areas. Recreational and development pressures in the two areas were compared, using three primary criteria:(l) the extent of permanent road access along the oceanfront;

3 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE 41 (2) the percentage of the beachfront with either public or private bathing access; and (3) the degree of recreational vehicle access to beaches. State road maps and US Geological Survey maps were used to determine criteria (i) and (2). Information concerning off-road recreational vehicle (ORV) use policy was obtained from state coastal zone agencies and state and Federal park personnel. Property ownership of the oceanfront was categorised into Federal, state, conservation group, or private/municipal (non-conservation group) based on map measurements and documentation from state agencies. Ownership is used as one measure of the degree of protection from human encroachment. Censuses All colonially-nesting waterbirds were censused during early June 1976 and 1977 along the northeast coast as part of a US Fish & Wildlife Service project (Erwin & Korschgen, in press). One team of experienced ornithologists in each state conducted ground and aerial censuses during the incubation period of the nesting cycle. Census methods, times, and reporting procedures were standardised to ensure comparability. At each location, estimates of adults and/or nest counts were made and the habitat type recorded for each species. The habitat categories examined were: (1) barrier island, (2) natural marsh island, (3) mainland, or (4) man-modified sites, either from landfill or dredge deposition. The latter sites result from maintenance dredging in channels and inlets, with the 'spoil' deposited either on adjacent marsh areas or in open water areas, creating new islands. Because of the physical similarity of the two regions, l have assumed that the relative proportions of habitat types 1-3 above are roughly equal. To compare the relative abundance of dredge deposition sites in the two regions, reference was made to current (1976) US Army Corps of Engineers nautical charts from the Philadelphia, Pennsylvania and Norfolk, Virginia district offices. Measurements of intracoastal waterways were made and all designated deposition sites were counted. The relative density of sites as number per unit length of waterway was then calculated. Although this paper is concerned primarily with four 'beach-nesting' seabird species, the habitat use of all wading bird (i.e., herons, egrets, and allies) species in the two regions is also summarised. Only 1977 habitat data are presented here because of incompleteness in the 1976 data set. Competition Rapid population increases and southwestward expansion of the herring gull's breeding range have apparently had a significant effect on the habitat use of other seabird species (Drury, 1973, 1974; Nisbet, 197 I, 1973). This effect was examined in two ways: (1) Because the gulls frequently occur with common terns on dredge deposition sites, 1 compared substrate use by the terns in the presence or absence of gulls. All sites from Long Island, New York, to Virginia were examined where both "high" substrate types (dredge sand, herbaceous cover) and qow" (marsh grass, drift

4 R. MICHAEL ERWIN material) types were available; (2) At colonies where herring gulls increased from 1976 to 1977, the fate of associated species was recorded. Only those cases were included where colony nesting sites appeared to be limiting (i.e. small islands) and where the nesting area of the gulls either overlapped or abutted the nesting area of the other species. RESULTS Habitat use Seabirds nest almost exclusively (81 ~o) on barrier island beaches in Virginia while, in New Jersey, less than 10 ~o are located on these "traditional" sites (Table 1). Least terns, black skimmers, and herring gulls concentrate on the dredge deposition/landfill sites as 'alternatives" to barrier beaches in New Jersey. Common TABLE 1 DISTRIBUTION OF NESTING PAIRS OF SEABIRDS AND WADING BIRDS AMONG FOUR HABITAT TYPES IN NEW JERSEY (NJ) AND VIRGINIA (VA) 1977 Island Habitat Species Barrier Marsh Dredge/fill Mainland Total pairs Common tern NJ 295 (7) 3445 (77) 302 (7) 420 (9) VA 3159 (91) 130 (3) 198 (6) Least tern NJ 489 (27) (54) 325 (18) VA 165 (100) Black skimmer NJ 335 (35) 64 (7) 550 (58) VA 2581 (98) 59 (2) Herring gull NJ 47 (1) 2365 (40) 3492 (59) 9 ( < 1) I VA 1266 (49) 201 (8) 1125 (43) Seabirds combined NJ 1166 (9) 5874 (45) 5316 (41) 754 (5) VA 7171 (81) 390 (4) 1323 (15) X 2 = 11909, P < 0-001, df = 3 b Wading birds combined NJ 3225 (37) 1830 (21) 3656 (42) 3 (< 1) I VA 2719 (44) 1234 (20) 2242 (36) X 2 = 75"5, P < 0-001, df = 2 b a Data: Top line gives breeding pair total with percent of total in parentheses. Bottom line shows the number of colonies in each habitat type. b Combined seabirds and wading birds, each comparing New Jersey with Virginia.

5 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE 43 terns, however, seldom nest on such sites, choosing instead to nest on low Spartina alterniflora marsh islands. In contrast to seabirds, the distribution of wading birds among habitats is very similar in Virginia and New Jersey even though a Chi-square test resulted in a significant difference (Table 1). Part of the marked regional differences in barrier beach use by seabirds may be explained by examining property ownership of the beachfront in the two areas (Table 2 and Figs. 1 and 2). The four seabird species show strong selection for Federal land and avoidance of municipal-private land in New Jersey. No such differences occur in Virginia. More than 90 ~o of all seabirds nesting on Federal property in New Jersey were concentrated at Brigantine National Wildlife Refuge, essentially the only totally-undisturbed barrier beach remaining in coastal New TABLE 2 DISTRIBUTION OF BEACHFRONT LAND AND NESTING SEABIRDS AMONG FOUR LAND OWNERSHIP CATEGORIES IN NEW JERSEY AND VIRGINIA, 1977 c Barrier beach ownership State Federal ~ Conservation b State Municipal or private New Jersey Land Seabird I (2) (6) Contingency coetficient, C = 0.43, P < 0.001, 2d.f. Virginia Land Seabird (2) (12) (2) (3) C = 0.03, P > 0.99, 3d.f. Combined lands of US Fish & Wildlife and National Park Services, US Coast Guard, and National Aeronautic & Space Administration. b Nature Conservancy (Virginia Coast Reserve). " Data presented as percentage figures. Number of colonies shown in parentheses. Jersey. Along the eastern shore of Virginia, beachfront ownership has little bearing on the relative "attractiveness" of potential colony sites. Of the 20 barrier islands, only one has permanent human habitation (Wallops Island, National Aeronautics and Space Administration). Even during the summer period, Cedar Island is the only other island with human habitation and vehicular use. All of the remaining islands are nearly pristine, with only occasional fishing and clamming parties. More direct measures of human disturbance are illustrated in Table 3. The vast majority of oceanfront in New Jersey is accessible by permanent roadways. Further, bathing and related recreational activities are permitted on virtually all of New Jersey's beaches, while, in Virginia, only about 5 7o of the beachfront receives any significant recreational pressure. Measurements of dredge deposition sites showed that the availability of dredge locations for colony sites is very similar in southern New Jersey (0.51/km) and

6 R. MICHAEL ERWIN 40o30, PA. NEW!~ JERSEY ~ Brigantine NWR Fig. I. Atlantic City KEY FEDERAL "~ [] PRIVATE ' [] STATE 0 20km Cape May : Sc~le- " ' Land ownership of oceanfront, Sandy Hook to Cape May Point, New Jersey. 'Private" includes municipal, corporate, or individual ownership.

7 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE 45 qas A(Wa Iio ps) Island Bay Y ~> Fisherman Island NWR C, O KEY [] FEDERAL ['--I PRIVATE [] STATE [] NATURE CON S E RVANCh 0 10 km J Scale I 37"00" Fig. 2. Land ownership of oceanfront, eastern shore of Virginia.

8 R. MICHAEL ERWIN TABLE 3 HUMAN ACTIVITY PRESSURE ALONG THE OCEAN BEACHFRONTS OF NEW JERSEY AND VIRGINIA Use New Jersey Virginia Paved road access (parallel to beaches) 82 ~ 5 ~o Beaches under vehicle restriction 13 ~o 79 Bathing beaches (public and private) 90 ~o 14 ~o a Total beachfront 190 km 136 km Only five of the 22km beachfront is subject to intense recreational use. The remainder is private with infrequent use. coastal Virginia (0.45/km). Unfortunately, no data exist on the vegetative conditions of the sites, which can influence colony site suitability (Soots & Parnell, 1975a, b; Parnell et al., 1978). Competition The presence of herring gulls appears to have a significant effect on the nesting substrate choice of common terns (Table 4). The earlier nesting gulls apparently usurp the higher sites, leaving only the peripheral marsh-drift area for terns. TABLE 4 NESTING SUBSTRATE USE BY COMMON TERNS ON DREDGE DEPOSITION ISLANDS a IN THE PRESENCE AND ABSENCE OF HERRING GULLS Substrate used Gulls High sites Low sites present (dredge sand~upland (marsh grass or wrack) grass) Yes 1 7 No 13 2 Fisher exact probability value, P = Islands include all those from Long Island, New York to Virginia where both nesting substrate types are available. Where nesting area appeared to be limiting, numerical increases of nesting herring gulls have a negative effect in five of six colonies of associated terns, skimmers, and/or laughing gulls Larus atricilla (Table 5). Only recently has the latter species come into close contact with the opportunistic herring gull in marsh habitats (Burger, 1977; Burger & Shisler, 1978). DISCUSSION Habitat use Natural barrier islands seem to provide 'preferred' nesting habitat for the four seabird species examined. Under relatively 'natural' conditions (in Virginia), more than 90 ~o of all terns and skimmers select barrier beach colony sites while about

9 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE ~o of all herring gulls do so (Kane & Farrar, 1977). The extremely low seabird use (9~o of total) of barrier islands along the disturbed New Jersey coast strongly implicates human intrusion as the cause. Without further studies, however, the specific types of disturbance (pets, vehicles, pedestrians) cannot be weighed relative to their detriment. Significantly, wading birds utilise barrier islands in similar proportions in the two regions. Because they nest in dense shrubs (often including poison ivy, Rhus radicans) on the bayside of barrier islands (Erwin & Korschgen, in press), they are essentially immune from human disturbances suffered by the beach-inhabiting species (Erwin, in press). TABLE 5 CHANGES IN BREEDING ABUNDANCE OF HERRING GULLS AND ASSOCIATED SPECIES AT SIX SMALL ISLANDS IN NEW JERSEY, DELAWARE, MARYLAND, AND VIRGINIA Year Island Species Big Reedy HG LG Outward Tump HG LG 45 0 Robbins Marsh HG LG Metomkin HG LG 80 0 West Sloop HG CT 87 0 BS 13 0 Clam (Main) HG LG " Species codes: Herring gull (HG), laughing gull (LG), common tern (CT), and black skimmer (BS). Where humans with vehicles prevail on barrier islands, dredge deposition sites apparently provide refugia for ground-nesting seabirds. If sandy and/or shell-laden 'spoil' has recently been deposited, dredge sites may even resemble barrier beaches both in substrate composition and vegetation (Buckley, 1978). Since repeated deposition increases the elevation of dredge islands, birds nesting on such sites do not suffer the risk of flooding common to natural marsh and even barrier island sites (Soots & Parnell, 1975a, b; Burger & Lesser, 1977; Buckley, 1978, in press). Further, the low insular nature and small size of many deposition sites (Buckley, 1978) reduces the probability that a resident mammal population could be maintained (Burger & Lesser, 1977). However, rats Rattus norvegicus and red foxes VulpesJulva have been found in colonies in all four habitat types on Long Island, New York and New Jersey (F. G. Buckley, pets. comm.). The relative impact of ground predators in the various habitats requires investigation.

10 R. MICHAEL ERWIN The suitability of dredge deposition sites to seabirds and wading birds is highly time-dependent (Soots & Parnell, 1975b). Recently-deposited 'spoil' attracts black skimmers and least and common terns, but successional encroachment of vegetation favours herring gulls and later, in the shrub-tree stage, wading birds. Other factors influencing dredge island site suitability are 'spoil' sediment quality and containment (i.e. whether the island is diked or not). Diked islands with predominantly silt-clay sediment deposition do not attract most species of waterbirds (Parnell et al., 1978). Provision of sand and/or shell is a prerequisite for attracting large numbers of black skimmers, least terns, and several other tern species (Buckley, in press; Parnell et al., 1978). The significance of dredge deposition sites transcends that of safety from human disturbance and flooding. With an archipelago of man-made sites available in areas where natural beach habitat is rapidly deteriorating, regional populations may avoid the dangers of concentrating into a small number of refugia, where an 'island dilemma' (Diamond, 1975) may develop. Already, in New Jersey, virtually the entire black skimmer population is concentrated in only three colonies. Under these circumstances, a local catastrophe (e.g. oil spill, depleted food supply) could drive a regional population to the brink of extinction. Management of vegetation and predators on dredge islands could ensure that a number of alternatives is available each nesting season. Natural marsh islands are used extensively by herring gulls and common terns only in New Jersey. The ability of herring gulls to adapt to new nest site conditions has already been documented (Burger, 1977). It is surprising that common terns concentrate on marshes rather than dredge islands as alternatives to barrier beaches. In earlier years, with beach development just beginning in New Jersey, dredge islands provided 'acceptable nesting grounds for the terns in lieu of the beaches' (Stone, 1937). On Long Island, New York (Buckley & Buckley, 1975) and Cape Hatteras, North'Carolina (Soots & Parnell, 1975b), the majority of common terns at present nest on dredge sites where adjacent beaches are heavily disturbed. Perhaps, in New Jersey, herring gulls have usurped the more attractive (i.e. early succession) dredge deposition sites, limiting the choice of alternatives for the terns (see section below). Some marsh nesting had been documented as early as the 19th century in New Jersey (Buckley, in press), but its extent was unknown. Mainland sites are relatively unimportant for all colonial waterbirds except least terns. Numerous mainland colonies of least terns have been well-documented in New England (Nisbet, 1973) and in other East and Gulf Coast areas as well (Bent, 1921). Effects of nesting habitat shifts Selection pressures should be expected to change with any major shift in nesting habitat use. Adjustments to new abiotic (water levels, substrates) and biotic (competition, predation) conditions in the environment are essential.

11 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE 49 The rapid numerical increase of herring gulls from New Jersey south (Burger, 1977; Erwin, in press) and their invasion into new habitats (Burger, 1977) may have wide-ranging impacts on associated waterbirds. Gull nesting appears to have a major effect on both the habitat and substrate choice of common terns. Further, with increasing use of Spartina marshes, herring gulls could displace laughing gulls in many areas (Burger, 1977; Burger & Shisler, 1978) similar to the displacement that occurred earlier in Massachusetts (Nisbet, 1971). Soon, other marsh-dwelling ground-nesters such as the Forster's tern Sternajorsteri and clapper rail Rallus Iongirostris may have to compete for nest sites with the larger, aggressive gull. Not only do herring gulls outcompete smaller, later-nesting species for nest sites, but they are also predators on the eggs and chicks of terns (Hatch, 1970). Burger & Lesser (1977) found predation rates on common tern nests were much higher on islands with nesting herring gulls than those without gulls. Thus, the expansion of the herring gull population from New Jersey south (Erwin, in press) presents a double-edged threat to the prospects of marsh-nesting waterbirds. In addition to the habitat reduction caused by human disturbance, the ~gull problem" confronting associated species also seems to be largely man-induced. Herring gulls feed heavily upon human refuse during both the breeding and winter periods (Spaans~ 1971; Cramp et al., 1974). Major nest microhabitat changes have accompanied the shift into new colony habitats. Nesting in Spartina marsh increases risks of flooding (Burger & Lesser, 1977) and subjects the nest to temperature and moisture regimes quite different from those on sandy beaches. Burger (1977) has documented the adaptive behavioural response of herring gulls to nesting in wet marsh habitats in New Jersey. Whether common terns are sufficiently adaptable to reproduce as successfully in marsh habitats as on barrier beaches is unknown (A. Storey, pers. comm.). Studies of relative reproductive success in different habitats are needed. CONCLUSIONS Natural barrier islands provide optimal nesting habitat for many colonially nesting waterbirds, especially beach-nesting terns, black skimmers, and gulls. In heavilydisturbed areas such as New Jersey, barrier islands have been developed, greatly reducing the extent of beach habitat. Consequently, birds have shifted colony sites to dredge deposition sites, or to a lesser extent, marsh islands. Dredge islands may provide good alternatives to barrier islands since they are often higher in elevation and are often free of mammalian predators (including man). However, increasing numbers of the large, dominant herring gull and its ability to invade both marsh and dredge island habitats appear to affect nesting common terns and laughing gulls adversely. Management of dredge deposition islands could ensure wide availability of nesting habitat for many colonial waterbirds.

12 R. MICHAEL ERWlN ACKNOWLEDGEMENTS Support for the project was provided by the Office Of Biological Services, US Fish & Wildlife Service contract no to the Massachusetts Cooperative Wildlife Research Unit, University of Massachusetts. Many of the ideas germinated from the studies of F. G. and P. A. Buckley, J. Parnell, and R. Soots, Jr. I am grateful to F. G. and P. A. Buckley and M. A. Howe for their comments on an earlier draft. REFERENCES ALLEN, R. P. (1937). Black-crowned night heron colonies on Long Island. Proc. Linn. Sot'. New York, 49, BAILEY, H. H. (1913). The birds oj' Virginia. Lynchburg, J. P. Bell Co., Inc. BENT, A. C. (1921). Life histories of North American gulls and terns. U.S. ham. Mus. Bull., BUCKLEr, F. G. (1978). Use of dredged material islands by colonial seabirds and wading birds in New Jersey. U.S. Army Corps of Engineers Tech. Rep. D BUCKLEY, F. G. (in press). Colony site selection by colonial waterbirds in coastal New Jersey. In Proc conj~ colonial waterbird group, ed. by W. Southern. Dekalb, Northern Illinois Umv. BUCKLEY, P. A. & BUCKLEY, F. G. (1975). The significance of dredge spoil islands to colonially nesting waterbirds in certain National Parks. In Proc. conj. management ~ dredge islands in North Carolina estuaries, ed. by J. Parnell & R. Soots, Raleigh, North Carolina State Univ. Sea Grant Publ. UNC-SG BUCKLEr, P. A. & BUCKLEr, F. G. (1977). Human encroachment on barrier beaches of the northeastern U.S. and its impact on coastal birds. I n A ~Tmposium on coastal recreation resources in an urbanizing environment, ed. by J. Noyes, Amherst, Massachusetts Coop. Ext. Serv. Monogr. Univ. Massachusetts. BULL, J. (1974). Birds oj'new York State. Garden City, New York, Doubleday. BURGER, J. (1977). Nesting behavior of herring gulls: invasion into Spartina saltmarsh areas of New Jersey. Condor, 79, BURGER, J. & LESSER, F. (1977). Determinants of colony site selection in common terns (Sterna hirundo). In Proc. 1977conJi colonialwaterbirdgroup, ed. by W. Southern, Dekalb, Northern Illinois University. BURGER, J. & SrtlSLER, J. (1978). Nest site selection and competitive interactions of herring and laughing Gulls in New Jersey. Auk, 95, COLWELL, R. & FUENTES, E. (1975). Experimental studies of the niche. A. Rev. Ecol. & Syst., 6, ed. by R. Johnston & P. Frank, Palo Alto, California, Annual Reviews, Inc. CONrqELL, J. (1975). Some mechanisms producing structure in natural communities. In Ecology and evolution of communities, ed. by M. Cody & J. Diamond, Cambridge, Massachusetts, Belknap. CRAMP, S., BOURrqE, W. & SAUrqDERS, D. (1974). The seabirds oj Britain and Ireland. New York, Taplinger. DIAMOrqD, J. (1975). The island dilemma: lessons of modern biogeographic studies for the design of natural reserves. Biol. Conserv., 7, DRURV, W. ( ). Population changes in New England seabirds. Bird-Banding, 44, ; 45, ERWIN, R. M. (in press). Coastal waterbird colonies: Cape Elizabeth, Maine to Virginia. U.S. Fish & Wildl. Serv., O/rice o/ Biol. Serv., FWS/OBS 79/10. ERWlN, R. M. & KORSCHGEN, C. E. (in press). Coastal waterbird colonies: Maine to Virginia, An atlas showing colony locations and species composition. U.S. Fish & Wildl. Serv., O/.ht'e o/biol. Serv., FWS/OBS 79/08. HATCH, J. (1970). Predators and piracy by gulls at a ternery in Maine. Auk, 87, HENNESSEY, G. (1976), Barrier island preservation: the Virginia coast reserve system. In Barrier islands and beaches, ed. by J. Clark, Washington, D.C., The Conservation Foundation. HOWARD, D. (1968). Do New England terns have a future? Bull. Mass. Audubon Sot'., 53, 2-9.

13 COLONIAL NESTING WATERBIRDS AND HUMAN DISTURBANCE 5 KADLEC, J. & DRURY, W. (1968). Structure of the New England herring gull population. Ecology, 49, KANE, R. & FA RRAR, R. (1977) coastal colonial bird survey of New Jersey. New Jers.,4 udubon Oces. Pap., No MILLER, R. (1943). The great blue heron. Cassinia, 33, NISBET, I. C. T. (1971). The laughing gull in the northeast. Am. Birds, 25, NISBET, i. C. T. (1973). Terns in Massachusetts: present numbers and historical changes. Bird-Banding, 44, 2%55. PARNELL, J., DUMONO, D. & NEEDHAM, R. (1978). A comparison of plant succession and bird utilization on diked and undiked dredged material islands in North Carolina estuaries. U.S. Army Engineer Waterways Exp. Stn Tech. Rep., D Vicksburg, Miss. SOOTS, R., JR & P^RrqELL, J. (1975a). Introduction to the nature of dredge islands and their wildlife in North Carolina and recommendations for management, in Proc. conj] management of dredge island~ in North Carolina estuaries, ed. by J. Parnell & R. Soots, Raleigh, North Carolina State Univ. Sea Grant Publ., UNC-SG SOOTS, R., JR & P^R~ELL, J. (1975b). Ecological succession of breeding birds in relation to plant succession on dredge islands in North Carolina estuaries. North Carolina State Univ. Sea Grant Publ., UNC SP^ANS, A. ( 1971 ). On the feeding ecology of the herring gull Larus argentatus Pont. in the northern part of the Netherlands. Ardea, 59, STOrqE, W. (1937). Bird studies at Old Cape May. Delaware Valley Ornithologists' Club, 2 vols.