W&M ScholarWorks
Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects
1982
Reproductive success of the common tern (Sterna hirundo) and black skimmer (Rynchops niger) in different habitats in Virginia
Daniel Carey Smith College of William & Mary - Arts & Sciences
Follow this and additional works at: https://scholarworks.wm.edu/etd
Part of the Ecology and Evolutionary Biology Commons, and the Zoology Commons
Recommended Citation Smith, Daniel Carey, "Reproductive success of the common tern (Sterna hirundo) and black skimmer (Rynchops niger) in different habitats in Virginia" (1982). Dissertations, Theses, and Masters Projects. Paper 1539625169. https://dx.doi.org/doi:10.21220/s2-6ptf-ps17
This Thesis is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. REPRODUCTIVE SUCCESS OF THE COMMON TERN (STERNA HIRUNDO) AND
BLACK SKIMMER (RYNCHOPS NIGER) IN DIFFERENT HABITATS IN VIRGINIA
A Thesis
Presented to
The Faculty of the Department of Biology
The College of William and Mary in Virginia
In Partial Fulfillment
Of the Requirements for the Degree of
Master of Arts
by
Daniel Carey Smith
1982 APPROVAL SHEET
This thesis is submitted in partial fulfillment of
the requirements for the degree of
Master of Arts
Daniel Carey Smith, Author
Approved, April 1982
Mitchell A. Byrd
Ruth A. Beck
Gregory M. Capelli
Stewart A. Ware
R. Michael Erwin, U.S. Fish and Wildlife Service TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS...... ii
LIST OF TABLES...... iii
LIST OF FIGURES ...... iv
ABSTRACT...... v
INTRODUCTION...... 2
LITERATURE REVIEW ...... 5
STUDY AREA AND METHODS...... 17
RESULTS...... 26
DISCUSSION...... 46
APPENDIX A ...... 53
LITERATURE CITED...... 54
VITA...... 60 ACKNOWLEDGMENTS
I would like to express my utmost gratitude to Dr. Mitchell A.
Byrd, who first interested me in seabird research, for providing much needed moral and logistical support throughout this study. I also thank
Mrs. Ruth Beck, Dr. Gregory Capelli and Dr. Stewart Ware for their
ctitical reading of the manuscript. Invaluable assistance with the
data analysis was provided by David Reed of the William and Mary
Computer Center and Ken Williams of the U.S. Fish and Wildlife Service.
Bob Bower and Jewel Thomas of the William and Mary Biology Department
also lent their technical expertise to the project.
The Nature Conservancy allowed me to conduct research on the barrier islands of the Virginia Coast Reserve. The Eastern Shore Lab of the Virginia Institute of Marine Science provided storage facilities
and technical assistance in 1980, and NASA's Wallops Island Flight
Center, in cooperation with the Chincoteague National Wildlife Refuge, provided housing in 1981. I also owe many thanks to the members of my
family and to Fenton Day for assisting with the field work.
Finally, this thesis could not have been completed without the continual support and guidance of Dr. R. Michael Erwin of the U.S. Fish and Wildlife Service. He provided funding during both years of the study through the Non-game Section of the Patuxent Wildlife Research
Center and advised me during every aspect of the project. I am forever indebted to him for his supervision, patience and friendship.
ii LIST OF TABLES
Table Page
1. Colony locations in Virginia-Maryland and New Jersey...... 21
2. Mean temperature and rainfall data...... 29 •'i
3. Common tern nest and egg fate data...... 33
4. Common tern clutch sizes and hatching success ...... 34
5. Common tern chick fate d a t a ...... 35
6. Common tern fledging success...... 36
7. Black skimmer nest and egg fate data...... 37 v 8. Black skimmer clutch sizes and hatching success ...... 38
9. Black skimmer chick fate d a t a ...... 39
10. Black skimmer fledging success...... 40
11. Common tern conspecific nest spacing...... 41
12. Common tern heterospecific nest spacing ...... 42
13. Black skimmer conspecific nest spacing...... 44
14. Black skimmer heterospecific nest spacing ...... 45
iii LIST OF FIGURES
Figure Page
1. Colony locations on Eastern Shore of Virginia-Maryland . . . 22
2. Horsehead Tump and Cedar Island salt marsh island colonies . 23
3. Metomkin Island barrier island beach colonies...... 24
4. Smith Island barrier island beach colony ...... 25
5. Common tern and black skimmer initial egg-laying dates . . . 27
iv ABSTRACT
The reproductive success of the common tern (Sterna hirundo) and black skimmer (Rynchops niger) was compared between barrier island beach and salt marsh island habitats. Parameters monitored to determine relative productivity included egg-laying dates, clutch size, hatching success, fledging success and nearest neighbor distance. Both
species began egg-laying later in 1981 than in 1980, with no differences found between habitats. Common terns had significantly greater clutch sizes and hatching success rates on beach habitat than in the marsh and in 1980 than in 1981. Common tern fledging success showed no difference between habitats but was significantly greater for both habitats in 1980 than in 1981. Large scale colony washout and losses to predation in 1981 limited the sample size and may have negated any differences which existed by habitat. Black skimmer productivity data were not tested due to small sample sizes from the marsh habitat, but showed a reversal in trend from 1980 to 1981. Beach colonies of skimmers had greater success in all parameters in 1980, but were not as great as the marsh success rates in 1981. Common tern conspecific nest spacing was greater.on the beach than in the marsh and greater in 1981 than in 1980. Black skimmer conspecific nest spacing was greater on the beach also, but did not differ by year.
Heterospecific nest spacing for both species was greater than the corresponding conspecific distance on the beach, was greater in distance on the beach and was greater in 1981 than in 1980.
v REPRODUCTIVE SUCCESS OF THE COMMON TERN (STERNA HIRUNDO) AND
BLACK SKIMMER (RYNCHOPS NIGER) IN DIFFERENT HABITATS IN VIRGINIA INTRODUCTION
Common terns (Sterna hirundo) and black skimmers (Rynchops niger)
have traditionally nested on barrier island beaches on the mid-Atlantic
coast of the U.S. (Bailey 1913, Bent 1921, Austin 1929, Stone 1937,
Palmer 1941, Erwin and Korschgen 1979). Recently, however, human
encroachment has caused habitat shifts by terns and skimmers into
alternative habitat types in some areas (Buckley and Buckley 1975, 1977,
Burger 1977, Burger and Lesser 1977, Erwin 1980). Habitats now being
used for breeding include salt marsh islands, dredge deposition sites,
and man-made structures (Erwin 1979b). Recent coastal surveys (Erwin
1979b, Erwin and Korschgen 1979, Parnell and Soots 1980, Buckley and
Buckley in press) have shown that the use of these alternative habitat
types by traditional beach-nesters is greatest in states with the
greatest degree of coastal zone development. An example of differential habitat utilization between areas of similar coastal structure can be
seen in Virginia and New Jersey. In Virginia, where 85 per cent of the
coastline is protected by private and Federal agencies, over 80 per cent
of the beach-nesting seabirds are found on the barrier islands. In
contrast, New Jersey has only 25 per cent of its coastline protected
from unrestricted recreation, and only 10 per cent of the same species
are found on the barrier islands, the remainder having moved into salt marsh areas.
These dramatic shifts in breeding habitat use mean that each
species must adapt to changes in the physical makeup of the habitat at
2 3 the colony and nest site as well as to changes in the biotic environment.
The major factor affecting the relative fitness of the individual in different habitats include the risk of flooding of eggs and young
(Greenhalgh 1974, Montevecchi 1978, Burger 1979a, Burger and Lesser
1979), the physical stability of the sand, rock and marsh substrates
(McNicholl 1975, Southern 1977), predation (Buckley and Buckley 1972,
Burger 1979, Burger and Lesser 1978) and competition for nest sites
(Crowell and Crowell 1946, Nisbet 1973, Erwin 1980). These measures of habitat quality must be weighed against the effects of density of breeding birds within a particular habitat. Fretwell (1968) described habitat suitability as a density-dependent gradient.which varies from high quality with high density to low quality with lower density.
Potential breeding birds might not settle in the habitat of highest quality if the density of the birds already utilizing that habitat would cause their success to be less than the average success of the local population. They might instead choose to colonize habitats of lower suitability because of the lower density.
For ground-nesting seabirds such as the common tern and black skimmer, the shift from open, dry sandy beaches to more vegetated wet marshes of lower elevation represents a severe change (Erwin et al.
1981). The common tern has demonstrated the ability to adapt somewhat to the new habitat by building larger nests (Borodulina 1966, Burger
1977) and exhibiting nest repair behavior (Burger 1979). The black skimmer may not be as flexible in its habitat utilization, being more restricted to beach habitats throughout its range. The ability of
these species to cope with shifts in breeding habitat will ultimately be reflected in their reproductive success (Erwin et al. 1981). Low reproductive success in seabirds has been associated with declining 4 colony size and the abandonment of colony sites (Crowell and Crowell
1946, Morris and Hunter 1976).
Although the shift of adult common terns and black skimmers into new habitats has been documented, only preliminary work has been done comparing their relative reproductive success (Storey 1978,
Burger and Lesser 1979, Smith et al. 1981). Before habitat management decisions can be made, it is necessary to evaluate the importance of alternative habitats from the perspective of productivity. In order to evaluate the relative reproductive success of the common tern and black skimmer between barrier island beaches and salt marsh islands, mixed breeding colonies of the two species were monitored on the
Eastern Shore of Virginia and Maryland in 1980 and 1981. Parameters of success which were recorded included colony phenology, clutch size, hatching success, fledging success and environmental factors such as the incidence of storm washouts and predation. LITERATURE REVIEW
Habitat selection
Selection of breeding sites that allow for the successful production and survival of offspring is essential to individual and
species survival. An important consideration in the study of a species
reproductive biology then, must be the mode of selection and utilization
of breeding habitat, together with the inherent ecological factors which
influence reproductive success. Although seabirds depend on the marine
environment for their food supply (Lack 1967), even the most highly pelagic forms must return to land to breed and it is there that they are exposed to selective factors such as predation and competition for
food and nest sites. Several physical and biotic factors combine to form
the characteristics of the nesting habitat chosen by seabirds, including microclimate, microtopography, substrate, nest-building material, food resources, and the presence of conspecifics or other species (Buckley and
Buckley 1980).
Habitat selection is based on both ultimate and proximate factors
(Lack 1954). Ultimate factors are those which govern the survival and reproductive success of a species, such as food availability, morphology, and protection offered by the habitat from predation and
adverse environmental conditions (Immelman 1972 in Buckley and Buckley
1980). Proximate factors are those psychological preferences of a
species for a particular food, nesting site or other aspect of the habitat which is not necessarily in itself essential for existence
5 6
(Lack 1954). Hilden (1965) called these the characteristic stimuli of species-specific habitats. To understand why species select particular habitat types, we must learn which elements of the habitat are perceived by the species as relevant (Klopfer and Hailman 1965).
Several studies have attempted to quantify the relevant cues by which seabirds choose particular habitat types. Buckley and Buckley
(1972) identified four topographic features of royal tern colonies which appear to be consistent over a large geographic area. These include general accessibility and excellent visibilty of surroundings, extensive areas of adjacent shallows, the absence of quadruped predators and location at an inlet between bay and ocean. Burger and Lesser (1978) investigated 34 common tern salt marsh colonies and 225 unoccupied salt marsh islands along the New Jersey coastline. Variables which seemed to determine the occurrence of tern colonies included island size, the presence of open water, vegetation type, distance to the nearest island, and distance to the shore. Using these criteria, it was determined that only 3 of the 225 unoccupied islands were suitable for colonization by terns, and only those 3 islands were used in subsequent years. For the black skimmer, the most important requirement for choosing a nesting location is the presence of common terns as nesting associates, the normally beach-nesting skimmers accepting a wide range of habitat types in order to nest with terns (Gochfeld 1977). The passive skimmers apparently achieve protection from predators due to the aggressive anti-predator behavior of the terns (Erwin 1979a). Skimmer nests also had less vegetation cover than tern nests or random points, but could not be significantly differentiated from other available sites by differences in shell cover, substrate texture or soil moisture content. Site tenacity (Austin 1949) is the tendency for birds to return to their natal site to breed after imprinting on particular habitat features, and has also been more widely interpreted as the repeated return to previous breeding sites (McNicholl 1975). Site tenacity is selectively advantageous in reducing susceptibility to predation and other adverse factors by familiarizing a bird with its surroundings and allowing it to return to the site of previously successful breeding attempts. If attachment to a site continues even after adverse environmental factors decrease productivity, however, the condition may become detrimental. Austin (1949) gives an example of a common tern colony persisting despite rat predation and flooding which caused complete failure. Common terns and black skimmers also exhibit group adherence, or the tendency for individuals to remain together even when relocating to new colony sites (Austin 1949). Nest site tenacity will be the most well developed trend in stable habitat with group adherence aiding in the selection and colonization of new colony sites, possibly in new habitats (McNicholl 1975).
Predation
An ultimate factor in the selection of a breeding site is the protection that the habitat provides from predators. Using the characteristics of behavioral and ecological adaptations to predation the degree of development of young, and the utilization of feeding zones, Lack (1968) divided the seabird group into two categories.
The first group breeds in large colonies at relatively safe sites, does not defend its nests against predators, and feeds offshore. This group includes the orders Sphenisciformes, Procellariiformes,
Pelecaniformes, and the Alcid Charadriiformes. The second group nests 7 in more accessible sites, has cryptically colored eggs and young, aggressively defends its nest sites, and feeds inshore. This group includes the Laridae, Rynchopidae, and the Stercorariid
Charadriiformes. Coloniality and the choice of inaccessible nesting sites are adaptations which have arisen at least in part due to selective pressures from both aerial and terrestrial predators (Buckley and Buckley 1980). Ground-nesting species such as the common tern and black skimmer are highly vulnerable to predation and as a result exhibit the highest degree of crypsis, spacing of nests, and anti predator behavior. Both the common and arctic terns on the Farne Islands remove egg shells, disperse their, defecations, space;out their nests and make vigorous attacks on intruders in an effort to increase the concealment of eggs and young (Cullen 1960). In contrast, sandwich terns at the same location showed a different strategy, nesting in dense colonies and remaining on the nest at the intrusion of an aerial predator such as a black-headed gull. Nesting in dense colonies provides an advantage to those individuals with nests in the center of the group in that it is more likely for a predator to first encounter an edge nest (Kruuk 1964). Royal terns nest as densely as possible and have adopted hexagonal packing of nests as the most efficient configuration
(Buckley and Buckley 1977). This strategy most probably resulted from their habit of selecting colony sites on small, bare islands which are free from mammalian predators but which offer only limited area for nesting above the mean high water mark. Egg predation by laughing gulls on the periphery of their colonies also may have contributed to the adoption of this behavior (Buckley and Buckley 1972).
Anti-predator adaptations in gulls seem to be directed only toward diurnal predators (Southern and Southern 1979), with most species 8 showing no response to nocturnal predation except for flight or colony site abandonment. The only adaptation for avoiding mammalian nocturnal predators may be the choice of breeding sites which are inaccessible to them, however, black-crowned night herons (Morris and Hunter 1976) and great horned owls (Nisbet 1975a) are both known nocturnal predators of common tern colonies also. Habitat features may also provide shelter to terrestrial predators from the aggressive colony defense of seabirds.
Lemmetyinen (1971) found differences in the effectiveness of common tern colony defense between the treeless coastal North Sea and islets in the Finnish Archipelago covered with trees and shrubs. Mink were able to prey successfully on nests and chicks while under cover of vegetation but could not withstand aerial attacks in open areas.
The presence of herring gulls in the salt marsh strongly influences the colony site establishment of other species. In attempting to select gull-free islands, terns and skimmers may often be left with fewer and poorer quality sites (Erwin et al. 1981). Although constant overwashing prevents the establishment of terrestrial predators, avian predators such as the herring gull, laughing gull, black-crowned night heron, fish crow and great horned owl may cause great damage to salt marsh island colonies (Buckley and Buckley 1972, Burger 1974, Nisbet 1975a, Hunter and
Morris 1976, Burger and Lesser 1979). In some cases, gull predation and competition for nest sites may cause complete colony abandonment by terns and skimmers (Crowell and Crowell 1946, Erwin 1980). Burger
(unpubl. data) found colony destruction levels of 50 per cent due to predation in New Jersey black skimmer colonies and suggests that skimmer colonies are more subject to colony abandonment due to predation than tidal flooding. This is based on the differences in predictability of the two factors, with a high level of probability that predation levels will remain constant from one year to the next. Predation by gulls in tern colonies located on marsh islands in New Jersey is most severe in the densest areas of the colony, the opposite of what has been found with mammalian predators (Storey 1978). This is due to the ineffectiveness of terns in driving gulls out of the colony once they have landed and to the small distances between,nests, aiding in the quick destruction of several eggs or chicks. Predation rates in New
Jersey have been calculated at 0 to 83 per cent, with islands containing herring gulls suffering the most damage (Burger and Lesser 1977).
Competition
Competition for breeding sites, both intra- and interspecifically, may determine species distribution in a particular habitat (Partridge
1978). In areas of low population density, only low level competition for a particular optimal habitat type will exist, with the birds showing a stenotypic tendency, or the restriction to a single habitat type. This is usually true at the edges of a species geographic distribution
(Hilden 1965). In areas of high population density, however, competition for available optimal habitat will be severe and many individuals may be forced to utilize less preferred or suboptimal habitat. Fretwell (1968) described habitat suitability as a density-dependent gradient which varies from high quality with a high density of breeding birds to low quality with lower density. Potential breeding pairs must weigh the chance of lower than average reproductive success due to high density in the preferred habitat to the unpredictability of nesting in a new habitat with low density. Favorable conditions have allowed the expansion of herring gulls into many areas where the density of breeding individuals is much lower than at traditional sites (Drury and Nisbet 10
1972). This species has undergone great population increase and range expansion in the last 50 years (Drury 1973) and in doing so has exploited a wide variety of habitat types. Although formerly coastal island nesters restricted to Maine in the U.S. (Bent 1921), they now nest as far south as South Carolina and have been found nesting on highway median strips and abandoned building rooftops in New York
(Buckley and Buckley 1980), on dredge spoil islands, osprey nesting platforms, and salt marsh islands in New Jersey where they are competing with laughing gulls and common terns (Burger and Shisler 1978).
Mixed-species breeding colonies may show interspecific competition for available nesting sites where the inhabitants have overlapping habitat requirements. In this case, each species will exhibit greater stenotopy toward particular species-specific conditions. These preferences for different substrate and microtopographic features will allow the partitioning of a particular site and the compatible coexistence of the nesting species. This situation exists for common terns and black skimmers, which nest in large mixed colonies on the mid-Atlantic coast of the U.S. Terns arrive before skimmers and choose nest sites which are usually adjacent to vegetation (Gochfeld 1977).
Skimmers then fill in the vacant areas in the colony which are free of vegetation. By the time laying occurs, there may be several subgroups of skimmers surrounded by terns. This association between species seems to strongly outweigh any disadvantages in competing for nest sites since large expanses of seeming identical habitat are left unoccupied.
Black skimmers derive benefit from nesting with common terns in the form of protection from predators (Erwin 1979a). Skimmers use only passive injury-feigning distraction displays at the intrusion of ground predators and rarely chase aerial predators. Terns mob and dive-attack 11
ground predators and aggressively chase aerial predators away from the colony site. Nesting in close proximity to the aggressive terns is not harmful to the skimmers, since reproductive success is no different between conspecific and heterospecific nearest neighbors. Differences
in microhabitat preference and foraging strategy preclude competition
for nest sites or food information sharing hypotheses as reasons for the association.
Habitat shifts
Common terns and black skimmers have traditionally nested on sand
or pebble barrier island beaches on the mid-Atlantic coast of the U.S.
(Bailey 1913, Bent 1921, Austin 1929, Stone 1938, Erwin and Korschgen
1979) although common terns have also utilized salt marsh areas to
some extent. Common terns are found in England on sand and rock beaches
and in some salt marsh areas (Marples and Marples 1934, Greenhalgh 1974,
Cramp et al. 1974), in Finland on both inland and marine habitats
(Lemmetyinen 1973) and in the USSR on grass-covered or bare islands,
silt or sand spits, and on rice paddies far from any large expanse of water (Borodulina 1966).
Increasing human disturbance and development of traditional barrier beach breeding sites has recently caused major habitat shifts by common terns, black skimmers and other seabirds into alternative and possibly suboptimal habitat types (Buckley and Buckley 1975, 1977,
Burger 1977, Burger and Lesser 1977, Erwin 1980). These shifts may
present many problems in adjusting to new ecological conditions.
Changes in predation pressure, competition for nest sites, substrate
type and other factors may affect the individual fitness and productivity
of the colonizing species. For seabirds moving from open, dry and 12 sparsely vegetated sandy beaches to low-lying, wet and more thickly vegetated salt marsh islands, the shift would seem to represent a severe change (Erwin et al. 1981).
After witnessing habitat shifts by common terns in New Jersey,
Storey (1978) attempted to determine if common terns have adopted a nesting strategy which results in successful marsh nesting. Nest sites were examined to determine if they were more safe from flooding than unused sites, how they compared to nests of beach-nesting common terns, and how they compared to nests of the strictly marsh-nesting Forster’s tern. Over a two-year period, an average of 96 per cent of the tern nests in the marsh were located on mats of dead Spartina grass. This was also seen by Burger and Lesser (1978) who found that common terns on salt marsh islands nested on dead eelgrass windrow (Zostera) 80 per cent of the time even though it accounted for only 4.6 per cent of the island surface, the majority of each island being covered by live
Spartina alterniflora. In the Storey study, terns chose the greatest ground height available and were more prone to nest on the ground when at greater heights. When nesting on wrack or windrow, terns prefer large, thick mat to small, thin mat. Areas of lower than average vegetation density were also preferred. In contrast, salt marsh tern colonies in Lancashire, England had a predominance of nests on the ground or on piles of hardened cattle manure, widely spaced to avoid being trampled by livestock (Greenhalgh 1974).
Similarities exist between marsh and beach-nesting common terns in their preference for ground height and vegetation density, although vegetation density increases with ground height on beaches while decreasing with ground height in low-lying marshes (Storey 1978). The rim height of marsh nests is significantly higher than those on beaches, 13
presumably as another prevention against washouts. Marsh nests also have
a more extensive lining, but lining is also important in beach nests as
a buffer against heat and wind (Marples and Marples 1934). The lining in
sand nests heats more slowly than marsh nests but also reaches a higher
temperature and retains heat longer (Storey 1978). Marsh nests must be
incubated for longer periods of time to account for reduced temperature
of the substrate, but are far safer from overheating during periods of
disturbance when the eggs are unattended. Common terns showed some
differences in nest site selection from Forster's terns, which are
traditional marsh-nesters. Forster’s terns nest on the leeward side of
the islands so that their nests are not exposed to the worst wave action.
They also choose lower areas of the marsh with greater vegetation
density, but utilize larger mats of wrack for nest placement. In
adapting to the marsh environment, Forster’s terns have evolved other
strategies for dealing with tidal flooding. This species begins its breeding season earlier than common terns and, in cases of washout,
relays sooner and generally has a full complement of three eggs in the
second clutch.
Among herring gulls, common terns and black skimmers, colonies on barrier island beaches are larger and more stable than those on marsh
islands with differences most likely due to a limitation of nesting
substrate in the marsh (Erwin et al. 1981). Marsh-nesting laughing
gulls and Forster's terns have similar rates of colony site change as
those of the beach-adapted species in marsh habitats. These traditional
marsh-nesting species have also evolved certain strategies for
achieving maximum reproductive success in the marsh habitat, but it
appears that the herring gull and common tern also possess enough
flexibility to adjust to this new environment, as evidenced by their 14 adaptation to building larger nests (Borodulina 1966, Burger 1977) and exhibiting nest repair behavior after tidal flooding (Burger 1979).
Tidal flooding has been found to be the major source of nest and egg-destruction in colonies in many areas. An average of 40 per cent of the common tern clutches on the Volga delta are destroyed every year by tidal floods and complete washouts due to storms and high water have occurred. The major source of productivity reduction at colonies in
England was found to be tidal flooding (Greenhalgh 1974). In the U.S., tidal flooding was found to be the cause of low reproductive success on small salt marsh islands inhabited by common terns (Storey 1978,
Burger and Lesser 1979). Tides which destroyed 70 to 100 per cent of laughing gull nests in New Jersey occurred at least once in six out of ten breeding seasons between 1965 and 1974 (Montevecchi 1975). Storey
(1978) gives a review of tidal damage to breeding colonies of seabirds.
Reproductive success
Common terns begin egg-laying at three years of age and are maximally productive at four years of age (Austin 1938). The rate of egg deposition is variable, usually with an interval of two days between eggs,(Palmer 1941). Clutches normally consist of three eggs (Bent 1921,
Palmer 1941), although a range of one to five eggs has been found.
Records for common tern colony average clutch sizes include 1.92 eggs/ clutch for approximately 2000 nests at Great Gull Island (Cooper et al.
1970), 1.5-2.5 eggs/clutch at 36 colonies in New York (Duffy 1977),
2.27-2.96 eggs/clutch in New England colonies (Nisbet and Drury 1972), a mean of 2.9 eggs/clutch in southern Finland (Lemmetyinen 1973) and
2.5 eggs/clutch on salt marsh colonies in New Jersey (Burger and Lesser
1979). Young colonies of terns without an established nucleus of 15 sexually efficient individuals between four and ten years of age may have smaller clutch sizes than would be expected (Cooper et al. 1970).
Common terns begin incubation after the first egg is laid, hatching normally occurring after 21 days (Palmer 1941). Variation in the duration of incubation is caused by different levels of disturbance, and may range from 20 to 30 days (Austin 1932). Replacement clutches may be attempted if the first eggs are destroyed (Bent 1921, Palmer
1941) with a large percentage of relaying indicating low initial success of a colony.
Hatching success of eggs is a good indication of relative nesting success (Cooper et al. 1970). Great Gull Island experienced only 16.5 per cent success from almost 2000 eggs in 1966 (Cooper et al. 1970), but increased to an 88 per cent rate of hatching in a later year
(LeCroy and LeCroy 1974). Rates of 95-98 per cent hatching were found in New England (Nisbet and Drury 1972) and hatching success in southern
Finland averaged 80 per cent over a four year period (Lemmetyinen 1973).
Hatching success varied from 64-78 per cent in an English salt marsh from 1971 to 1973.
Precocial common tern chicks fledge at approximately 21 days of age and the rate of fledging success has been used as a measure of reproductive success in several studies (see Erwin and Custer in press for a review of productivity measures and suggestions for uniform reporting). Storey (1978) predicts that common tern chicks have an equal chance of survival in the marsh as on the beach after finding that 10 of 12 marsh colonies produced chicks in 1976 and 1977, while only 6 of 10 beach colonies were productive. Other reports of success in salt marsh areas include 0-0.96 young/nest produced on 11 different islands in New Jersey (Burger and Lesser 1979) and 45.6 per cent 16 fledging success of terns in Lancashire, England, where the limiting factors were found to be tidal flooding and predation of eggs and young by gulls. Records of common tern fledging success at beach colonies include 0-2.10 chicks/nest fledged in New England (Nisbet and Drury
1972), 0.91-1.65 fledglings/nest in Ontario (Morris et al. 1976),
0.91-1.65 fledglings/pair in southern Finland (Lemmetyinen 1973), and
59-88 per cent survival of hatched chicks at Coquet Island, England
(Langham 1972). Other reports include 42 per cent fledging success from 17,500 eggs in New England (Austin 1929), 0.4-0.7 chicks/nest fledged in New York (LeCroy and Collins 1972), and 0.14 chicks/nest fledged in Canada (Switzer et al. 1971).
The breeding ecology of the black skimmer has only recently been thoroughly documented (Erwin 1977) and few reports of its reproductive success are found in the literature. Skimmers begin egg-laying during the third week in May in Virginia and most commonly have clutches of four eggs (Erwin 1977) although a range of 1-5 eggs has been reported
(Bent 1921). The mean clutch size for Virginia colonies in 1973 and
1974 was 3.55 eggs (Erwin 1977). In New York, clutch sizes of seven colonies ranged from 1.8 to 3.4 eggs/clutch (Duffy 1977).
Black skimmers begin incubation after the first egg is laid and hatching normally takes place after 23 days (Erwin 1977). As with common terns, relaying may occur if initial clutches are unsuccessful
(Bent 1921). Erwin (1977) found 78.6 per cent hatching success from
95 nests over two years in Virginia, with no difference in success between years. Fledging success of those chicks equaled only 0.40 and
0.37 young/nest during the two years, with 10 of the 11 fledged young being the first hatched in the brood. It was hypothesized that food limited production in those colonies. STUDY AREA AND METHODS
During 1980 and 1981, I monitored eight mixed colonies of common
terns and black skimmers at five different locations, including three barrier island beach sites on the Eastern Shore of Virginia and two salt
marsh islands in Chincoteague Bay, Maryland (Table 1, Figure 1). In a
cooperative study, ten common tern colonies on six different salt marsh
islands in' New Jersey Were studied by Dr. R. Michael Erwin of the U.S.
Fish and Wildlife Service during the same two year period (Table 1)
(Smith et al. 1981). In the Virginia-Maryland study area (hereafter
designated VA-MD), the salt marsh island colonies studied were Horsehead
Tump in both years and the northeasternmost tump in the Cedar Islands
group in 1981 (Figure 2). These islands are small (less than 0.5 ha) and
flat, vegetated predominantly by Spartina alterniflora. Other congeners
nesting on the islands included Forster's terns (Sterna forsteri) and
royal terns (Sterna maxima). Other islands in the Cedar Islands group, which is approximately 1 km south of Horsehead Tump, supported a small
number of nesting herring gulls (Larus argentatus) and great black- backed gulls (Larus marinus).
Two colony sites, designated North and South, were monitored on
Metomkin Island each year (Figure 3). The exact locations of the two
colonies shifted between years, but each may have consisted of many of
the same breeding pairs during the two seasons. Metomkin Island, 35 km
south of the Chincoteague Bay colonies, is part of the Virginia barrier
island chain, consisting of narrow, sandy beaches protecting the
17 18
mainland from the Atlantic - Ocean together with extensive Spartina
salt marsh and tidal mud flats. The island is approximately 7.5 km in
length and varies from 50 to 200 m in width. Tern and skimmer nests were usually found adjacent to clumps of sea rocket (Cakile edentula)
or beach grass (Ammophila breviligulata). The island also supports breeding colonies of least terns (Sterna albifrons), gull-billed terns
(Gelochelidon nilotica), royal terns, herring gulls, great black- backed gulls, and laughing gulls (Larus atricilla).
Smith Island is another member of the Virginia barrier island chain, lying 55 km south of Metomkin Island and consisting of similar beach habitat (Figure 4). One mixed colony of terns and skimmers was monitored on the north end of Smith Island in 1980. The island also
supports breeding colonies of least terns and gull-billed terns.
Aerial surveys of the VA-MD coastline were made at the beginning of each season to verify colony locations and approximate species numbers. Brief ground visits were then made to each prospective study
site to confirm the suitability for research and the onset of egg- laying. I attempted to select colonies containing between 40 and 100 nests of each species. Colonies chosen for study were visited every
5-8 days until fledging of all chicks under observation. Colony visits were limited to 1 hr/day in an attempt to minimize disturbance. No visits were made during periods of excessive heat, rain or high wind.
Individual nests were marked on the first visit with numbered wooden or plastic stakes and then monitored during incubation to determine egg-laying dates, clutch sizes and hatching success. In 1980, all nests
in each colony were staked, but in 1981 only several small aggregations of tern and skimmer nests were marked, in an effort to further reduce
the level of disturbance. 19
Previous studies (Nisbet and Drury 1972a, Erwin 1975) have
demonstrated the benefit of enclosures in calculating seabird
reproductive success without significant detriment to the colony being
studied. Prior to peak hatching in this study, 30 cm high, 2.5 cm mesh wire enclosures were erected in each colony to restrict the wandering
of chicks and simplify the recapture of banded chicks throughout the
study period. In 1980, one enclosure was erected in each colony,
encircling 25-39 nests of both species. These single enclosures measured 20-25 m in diameter and contained both edge and center nests
from within the colony. In 1981, to provide a more representative sample
from the colony structure (Erwin and Custer in press), 3-6 smaller (10-
15 m in diameter) enclosures were erected in each colony, each
containing an average of five nests of each species which had been
staked at the initiation of egg-laying. One single-species enclosure
of ten nests was also set up for each species in colonies where
sufficient numbers of birds were present. .
Common tern chicks were banded immediatedly after hatching whenever possible with aluminum U.S. Fish and Wildlife Service bands.
Black skimmer nestlings were not banded until the second visit after
hatching due to the potential for band loss caused by the large size of
the band. Chicks were checked weekly to determine their fate and were
considered fledged if not found after 21 days for terns and 23 days for
skimmers when known to have been present on the previous visit. At the
conclusion of each breeding season, nearest neighbor distances were
taken in each enclosure between both conspecific and heterospecific
nests. All nest stakes and wire enclosures were then removed from each
colony site. Weekly temperature and rainfall means were taken from
U.S. Department of Agriculture weekly weather and crop bulletins, using the Norfolk monitoring station for reference.
Clutch size, hatching success and fledging success were all
analyzed using a log linear model which more normally distributes the
discrete data points by equating the natural log of the expected
frequencies with a linear model on a log scale (FUNCAT procedure of the
Statistical Analysis System, Barr et al. 1979). Both conspecific and heterospecific nearest neighbor distances were measured at the end of
the fledging period in each colony and analyzed for significant
differences by habitat, colony and year. An analysis of variance using
a general linear model (GLM procedure of the Statistical Analysis
System, Barr et al. 1979) was employed. 21
Table 1. Location of common tern and black skimmer mixed colonies studied in Virginia-Maryland and New Jersey. Years Colony studied Location Coordinates
VA-MD 1980 1981
North Metomkin Isl. X X Accomac County, VA 37°45'N 75 °35’W South Metomkin Isl. X X Accomac County, VA 37°451N 75 °35' W Smith Island X Northampton Co., VA 37°10,N 75 °501W Horsehead Tump XX Worcester Co., MD 38°001N 75°15’W Cedar Island X Worcester Co., MD 38°00'N 75 °151W
NJ
Boomerang Island X X Atlantic County, NJ 39°30'N 74°25’W Wading Thoro X X Atlantic County, NJ 39°30'N 74°25'W Obes North XX Atlantic County, NJ 39°30'N 74 °25 ’ W Obes South X Atlantic County, NJ 39°30'n 74 °25'W Boot Island X Atlantic County, NJ 39°30'N 74°25’W Main Marsh Thoro X Atlantic County, NJ 39°30fN 74°25' W Parker Island X Atlantic County, NJ 39°30IN 74°25TW 22
Figure 1. Map of Virginia-Maryland Eastern Shore showing study colony locations (from Erwin 1980, with permission). CHINCOTEAGUE BAY colonies ^ _ MARYLAND_------—
CHESAPEAKE BAY
METOMKIN ISLAND co Ionie s
RGINIA
ATLANTIC OCEAN
SMITH ISLAND colony
10 km 23
Figure 2. Map of Chincoteague Bay showing location of Horsehead Tump and Cedar Island colonies (from U.S. Geological Survey 7.5' Topographic Series, Boxiron Quadrangle, 1963). 38°02—
HORSEHEAD TUMP 1 980 and 1981 colony
CHINCOTEAGUE BAY
1981 colony
CEDAR ISLANDS
O
km ATLANTI C OCEAN 24
Figure 3. Map of Metomkin Island showing location of colonies studied (from U.S. Geological Survey 7.3’ Topographic Series, Metomkin Inlet Quadrangle, 1966-67). 37°45'““
ME TOMKI N ISL AND
North colony 1980
North, colony 1981
METOMKIN BAY
South colony 1981
ATLANTIC OCEAN
South colony 1980
1 km 25
Figure 4. Map of Smith Island showing location of 1980 study colony (from U.S. Geological Survey 7.5r Topographic Series, Ship Shoal Inlet Quadrangle, 1967). 37°10/
1980 colony
SMI TH ISLAND
ATLANTIC OCEAN
1 km RESULTS
Arrival and egg-laying dates
Common terns arrive at their Virginia colony sites in late April and early May, going through courtship and territory establishment for approximately two weeks. Black skimmers arrive approximately one week later, filling in open areas of the ternery with aggregations of nests.
Aerial and ground censuses of the Eastern Shore in both years found relatively consistent numbers of terns and greater numbers of skimmers in 1981 than had been present in 1980. On the two barrier islands which were monitored during the study, common terns increased from 1200 to
1700 individuals on Metomkin Island but decreased from 1200 to 960 birds on Smith Island (B. Williams, Williamsburg, VA, pers. comm.).
Approximately 160 common terns were present on the marsh island
Horsehead Tump in both years. Black skimmers increased from 2000 to
4100 and from 1000 to 1600 individuals on Metomkin and Smith Islands, respectively, from 1980 to 1981. Ten additional pairs of skimmers nested on Horsehead Tump in 1981, increasing the number there to 15 breeding pairs.
Clutch initiation and median peak laying dates were compared between habitats, colonies and years, with the only major differences being detected by year for both species (Figure 5). Common tern egg- laying began between 10-16 May in 1980 and from 18-26 May in 1981. Peak laying dates for terns varied from 20-23 May in 1980 to 28 May-4 June in
1981. The duration of the common tern egg-laying period was
26 27
Figure 5. Common tern and black skimmer initial egg-laying dates for 1980 and 1981. COMMON TERNS
□ 1980 n=172 median date:20-21 MAY
m i 1981 n= 53 median date : 28~29 MAY
BLACK SKIMMERS
L-J 1980 n= 93 median date:24-25 MAY
1981 n= 41 median date: 7—8 JUNE
J I L 10 14 18 22 26 30 1 5 9 13 MAY JUNE 28 consistent between years, ranging from 14 to 20 days.
Black skimmer clutch initiation and median peak laying dates also showed a wide disparity between years, with egg-laying beginning between
12-18 May in 1980 to 20 May-1 June in 1981, a difference of 12-14 days.
Peak laying dates ranged from 20-25 May in 1980 to 22 May-8 June in
1981, a difference of 8-14 days. The duration of egg-laying was from
12-16 days per colony during both years. Weekly temperature and rainfall means showed no substantial difference between years for the two month tern and skimmer arrival and egg-laying stage of late April to early
June (Table 2).
In 1980, common tern laying dates were recorded in Rhode Island,
New Jersey, and North Carolina, in addition to VA-MD (Smith et al.
1981). A trend of later clutch initiation with increasing latitude was found, with the earliest dates occurring in the southern regions of
Virginia, on 10-11 May, North Carolina, on 12-13 May, followed by Rhode
Island and New Jersey on 20-21 and 24-25 May, respectively. An analysis of median peak laying dates for the four states also showed that among- regional differences were significantly greater than within-regional differences.
Late season clutch initiation attempts were recorded in the VA-MD study area during 1980 and 1981 and showed a low percentage of nests as compared to numbers of peak nests in both years (Table 3 for terns,
Table 7/ for skimmers). The highest incidences of relaying by common terns were seen in the Horsehead Tump marsh colony during both years.
For skimmers, the only major late season nesting attempt came on
Metomkin Island in 1981, following a washout of the South Metomkin colony in early July, when a large (greater than 200 pair) concentration of skimmers begin renesting at a location between the Table 2. Mean temperature and rainfall data recorded in Norfolk, during the 1980 and 1981 common tern and black skimmer egg-laying period. Date Temperature (°.F) Rainfall (in)
1980 1981 1980 1981 1980 1981
4/22 4/21 56 59 0.3 0.2 4/29 4/28 62 57 1.1 0.9 5/06 5/05 61 64 1.2 0.6 5/13 5/12 65 58 0.1 0.1 5/20 5/19 68 67 0.6 0.6 5/28 5/27 71 62 4.4 0.5 6/03 6/02 72 76 0.1 1.0 6/10 6/09 76 77 0.3 2.4
Total 8.1 6.3 Mean 66.4 65.0 1.0 0.8 former North and South Metomkin study sites. Late season clutch initiation by common terns was recorded for four mid-Atlantic states in
1980 (Smith et al. 1981) and was found to be very limited in most colonies. Attempts to correlate the number of late clutches with both peak colony size and the success of peak clutches in Virginia and
New Jersey showed no apparent relationship.
Colony washout and predation
All VA-MD study colonies were relatively successful in 1980, with no washouts by storm tides and no abandonment by adult birds during any stage of the breeding season. A brief period of apparent gull predation late in the fledgling period resulted in the death of 18 common tern chicks and the disappearance of 13 others from the North Metomkin enclosure. In 1981, the Cedar Island colony was completely abandoned between 15 and 21 June after continued predation on adult birds, apparently by great horned owls (Bubo virginianus). Horsehead Tump suffered a similar method of predation late in the season, resulting in the death of approximately 25 common and Forsterfs tern fledglings.
On June 30, 1981, tropical storm "Brett" coincided with a new moon high tide, overwashing many of the Virginia barrier islands completely (B. Truitt, Nature Conservancy, pers. comm.). The South
Metomkin colony was completely washed out, with all nests and nest contents either missing or destroyed. All species breeding on the southern end of the island abandoned their colony sites, with only black skimmers attempting to renest at a different location on the island. Four enclosures of five nests each of terns and skimmers were destroyed, along with two single-species enclosures of ten nests each.
The North Metomkin breeding colony was not affected by the storm tide 31
because of its location on a raised shell bed. The four enclosures in
that colony (three mixed and one skimmer) were subjected to a high
degree of predation during the remainder of the season, however, again
apparently by gulls. A total of 15 tern and 16 skimmer chicks
disappeared from the colony during the season.
, Reproductive success
Clutch size, hatching success and fledging success were all
recorded to determine the relative reproductive success of common terns
and black skimmers between the traditional beach and newly colonized
salt marsh habitats. Due to the failure of two colonies in 1981 and
the low productivity of the remaining colonies, it was also necessary
to test the differences between years. For common terns, a total of 99
beach nests and 75 marsh nests were staked and enclosed over the two
year period. An additional 157 marsh nests were monitored in New Jersey
and are included in the analysis where noted. Of the common tern nests,
25 beach nests at South Metomkin were washed out after hatching, 25
marsh nests on Cedar Island were abandoned before hatching, and 8
beach nests on Smith Island were eliminated from the analysis because
of uncertainty as to the identification of some chicks from that colony.
For black skimmers, a total of 100 beach nests and 18 marsh nests were
monitored during the study. Of these, 30 beach nests on South Metomkin
were lost to washout prior to hatching. The small number of marsh
nests over the two years precluded the testing for significant
differences between habitats for skimmers, due to the requirements of
the SAS FUNCAT model used in the analysis (Ken Williams, U.S. Fish and
Wildlife Service, pers. comm.).
Common tern clutch size and hatching success were found to be 32 significantly greater in 1980 than in 1981 (Tables 3 and 4).
Differences were significant for data from the VA-MD region alone and after combining with data from NJ. Common tern fledging success showed no difference between habitats but was found to be significantly greater in 1980 than in 1981 (Tables 5 and 6). Although not tested for significance, black skimmer productivity figures seemed to reverse trends from 1980 to 1981. The barrier beach colonies showed greater clutch size, hatching success and fledging success in 1980, but the marsh colony was more successful for every variable in 1981 ( Tables 7,
8,9 and 10).
Nearest neighbor distances
Common tern conspecific nest spacing was found to be significantly greater on the beach than in the marsh, both in the VA-MD region alone and after addition of data from the NJ marsh colonies (Table 11). This difference between habitats was also reflected by a significant variation in tern conspecific spacing by colony (Appendix A). Distance to the nearest conspecific was found to be significantly greater in
1981 than in 1980 for both habitat types (Table 11).
Tern heterospecific spacing, or the distance from common tern nests to the nearest black skimmer nest, was found to be significantly greater than the corresponding conspecific distance on the beach but not in the marsh (Table 12). The difference in heterospecific spacing by habitat showed the same trend as conspecific spacing, with distances between nests being significantly greater on the beach than in the
( marsh. This trend was also consistent by colony and year, the beach colonies having greater mean heterospecific nest distances and 1981 showing increased spacing over 1980 (Table 12). 33
Table 3. Common tern nest and egg fate data. Peaka Clutch^ Latec No. No. No. No. No. hatched/ Colony ; nests size nests eggs hatched missing addled no. eggs N s - o tz: K P rt O 3 P 3 ho rt Ln 3 CTv hO P ho I-1 CTV CTv P O P Ln o 00 Ln 1+ /—s ✓ ^— - s o (D O 3 S' H* H* 3 3 CO CTv o CTv ho I-* M I-1 hO hO p p p p L0 hO •^1 H- -- LOOO O .. -o •• , ^—\ ^^ n CO I-* rt P1 P P 00 H* OO 00 CTV P CU MM rt rt CTv ho 3 P 00 Ln c M vo . ••• O L0 H- LO "Ooo . /—s /—N P4 3 rt) o S3 O P CU i-3 3 3 Ln 01 hO 3 o CTV p I-*VOvo I-1 CTv O i—1 3* • 00 o P . CTV ^—N <■—-N N—'' XJ 1 bd P rt O Ln ho rt ho CTV ho CO P H1 CTv CTv vo f—1 O LO LO 1— P O 1—* P LO ho 1—* 00 LO | /—s v> ' _ S PI-1 P I-1 rt)P M 3 CD CO ho p4 P4 ho hO o4 Ov ho o rt P VOho CO 3 3 o o '-O Ln \ l 1-* OO o Ov .• 'O /—s _ 00 (-> I-* rt VO \ — o P4 O O O O o ho hO a o hO Ln Ln Ln 3 rt o o I-1 00 LO O OV (-* ho p LO • o LO 'O Ln LO "vj ..•• H- H- s "—J ✓—\ ' _ K s rt> rt> rt rt O PP H- H- P4 hO p rt ho O ho o 3 3 Ln 3 o o o o o 3 3 M 00 CO Ln Ln Ln CTv OO Lo .• •oj • P VO P . H- LO ✓—s /—s s ' s _ - o- 3 00 P Pu 01 t-1 P CU Ln 3 o o I-1 Ln Ln p p CTv o -o O o CTv o o p /—x v S P ho ho ho P4 rt> ho H-* h-» 01 ho S3 3 ho o rt> cu H 3 Ln -P p H--H- o o Ln oo ho Ln M ho o .. •• p VO I-1 00 p "J /-S ^—\ X) 4 ✓ O' p a4 to a> P 3 rt p O rt c o O 00 O ho M O o O - MO CTv I-1 ho ctv — -p /—s /-S ' s-' N-/ N-' s-' N -- K M H* P P P ro H rt P4 O 01 01 a4 3 ho h-* ov ho 3 Ln CO O o -P ho I—1M VO 1—J o o ho Ln t—1 ho p Ln p LO -P -• >—■ CU 1-3 o H1 P rt rt oo P4 P ho ho av rt) ho vo VO ho bd VO P LO Ln p ho OV vo P VOOO o j—l "J VO •..•• ^—s v-" '—•" VO P O COn hO O 3 hO K ro Ln p M 1—* Ln o Ln P ho ho M ho CSV CTVVO "4 ..•• 'O P VO O 00 • LO /—s 'w' ) P O P n> 3 3 rt> 3 P n> 3 3 o CO rt 01 P H-h p rt) M O CO rt> CU o 3 rt 3 3 rt O rt P h-1 3 3 CO 3 3 cu 0 H* 3 25 o ro P o t-» o 3 H* 3 p 3 3 rt rt) 01 01 a> X) "C r* 3 r* DClutch size for enclosed nests are given by mean ± 1 standard deviation. cNumber of new nests found approximately one month after peak laying, with percentage relative to peak nest count given in parentheses. ^Colony failures at Cedar Island and South Metomkin in 1981 are incorporated into data. 34
Table 4. Common tern clutch sizes and hatching success by habitat and year. Clutch sizea
Habitat n 1980 n 1981 : n : Combined years
Beach 59 2.73 40 2.12 99 2.48
Marsh 25 2.40 50 1.94 75 2.09
Combined 84 2.64 90 2.03 habitats
Habitat effect: X2 = 7.59, P< 0.0059
Year effect: X2 = 37.02, P< 0.0001
Hatching success3- (No . hatched/no. eggs)
Habitat n 1980 n 1981 n Combined years
Beach 59 0.83 40 0.74 99 0.79
Marsh 25 0.67 50 0.24 75 0.38
Combined 84 0.79 90 0.49 habitats
Habitat effect: X2 = 47.57, P< 0.0001
Year effect: X2 = 33.72, P< 0.0001
aAnalysis by FUNCAT procedure of the Statistical Analysis System, Barr et al. 1979. 35
Table 5. Common tern chick fate data. P S bd (-3 cn cn PP O CO 3 eggs no.cn Cn no. nests dead missing fledged PJ batched CO nests 2! 1—1 Colony O PP O C C o P o O VO c C O 3 o O VO o P P r t c r cr* tl CP e H 00 o ' U4 H H- p M 00 cn O P r t r t cn P r t r t I-1 r t r t P r t r t r t o O P 4 P* MO O ft) H P* P4 OO P P 4 P 4 P 4 P P r t r t P 4 r t r t P4 ^<5 P P P M K S P PP M S s I—1 P CO p P I-1 M P P P p Hi CP H* r t r t CP h-1 r t r t P s W P O o K W P OO H* p ft) i-3 P 3 3 P P 1-3 P 3 3 M p P P CP 7? i-{ P C CP I*4 7? C cn O 3 H* H* cn o 3 H* H* H P 4 P 4 to P P P4 p 4 to P p ft) CO
P No. fledged/ No. fledged/ No, No. No, No,Enclosed r t •sj VO Cn -fP to to to l—1 to Cn to 1-* h-» to Cn VO O O Cn Cn Cn Cn Cn VO Cn 00 ■o -P- o P (P P M P I—1 H* Ot VO tO Ot P ' C O P W P Ui P -£> VO -P ' Cn O -P O si H O) P CP P p CP CO C O V O h-> to 00 N) U W to o O C O O Cn O I O C n VO 00 O C o O Cn p rt P 4 s ft) rt O ts3 Cn t— * C o •P" I- 1 t—1 to H* I—1 N) VO O '-J -P - Cn to to to M 1 OO CO I- O I H* P H« P H* h-1 4> to to M VO to ■O .O' U) 1 1 h-1 00 -P" OO co CO 0O OO M P HO O o o o o o o O M O h-» t—1 I-1 P • . * •«• • • . . • • • • -P- VO to h-* to o o co 00 C n 00 00 ■sj o H* O VO o at o o o CO O -P> o C O a t -P* OP O H •P O H P rt P CP O o o O o o o o o o o o o o • m ••• • • * • ••. •• H* I-1 CO H* o i-* o o 1-* C O C n C O a t at CO P 00 oo O 00 o o o ■sj C O - o C O V O -ps oo
CP ____ P rt P
36
Table 6. Common tern fledging success by habitat and year. Fledging success (No. fledged/no. nests)
Habitat n 1980 n 1981 n Combined years
Beach 59 1.54 40 0.16 99 •0.99
Marsh 25 0.80 50 0.20 75 2.09
Combined 84 1.36 90 0.18 habitats
Fledging success3 (No. fledged/no. eggs)
Habitat n 1980 n 1981 n Combined years
Beach 59 0.57 40 0.08 99 0.38
Marsh 25 0.33 50 0.10 75 0.18
Combined 84 0.51 90 0.09 habitats
Habitat effect: X2 = 0.54, P< 0.4627
Year effect: X2 =110.29, P< 0.0001
aAnalysis by FUNCAT procedure of the Statistical Analysis System, Barr et al. 1979. 37
Table 7. Black skimmer nest and egg fate data. Peaka Clutch13 Latec No. No. No. No. No. hatched/ Colony nests size nests eggs hatched missing addled no. eggs h-> o 00 VO 1 > no rt* 52! P 3! p JN O rt 3 3L NO a to f-* CO (— 4 X—N ^ x x— s X X - - 00 O P K 3 3 P H* H- M NO o rt OO 3 P p t-1 oo 3 Cn Cn VD c o C O c o h-» O H- . a x— \ - X-*N ^ \ - - cn cn P P O o oa o H* cn cn 3 M 3 I-1 a to - a0 o . -IN 3 M -P- -P- C O COCOCO h-> rt rt rt v X NaX \ - - - P 3 O P 3 a Cn 00 00 -o 4> -o 00 00 o o 33 H* 3 o Cn I—1 cn Cn to cn cn cn CO g o i-* Cn Cn Cn -o O O Cn ^ s l O "•j . o o o o o ooo CO -o -o oo W Nl 4C O 4C WNl H - 14- X— •. > X N-X X X N - - o n no 3 H 3P 3 c n t—1 to td 3I-1 3 O t—1 t—1 i—1 h-* 3 3 P rt V C D X - - 0 0 O O J N V O Cn o O N O s 3 3 3 3 P Cn Cn Cn -O H Cn 1 Cn t o -O o o Cn l—* c o c o 3 3 P rt rt rt co -o C O - o X-N X— \ - X-Nx - ' M CO OO t—1 CTN cn P o to a . . NO & ON -O NO -P" C n O 3 NO n o — -X 'O 52! 3 Cn ♦ « rt x— s '— x /—S. s 1 + X n o 1 f _ OO O NO PPOO rz: 3 3H* H* l-t o 4 N o O cn o O O rt rt 3 PP CO H- JN 3 rt • CO • v r s x-N ' X '— X v - _ — Cn p o 3 o c nN O c n C n 3 P 3 a 3 3 P C n 3 3 H* CO c o 4 > • -o co OOO o o O-f> Cn I—* H- . C O o _ - X . s ~ ~ 3 3 H O rt P I—1 3 3 - to to W rt p CO ' S—X- ' ' '—/ ' \ X—\ \ x s , s _ - _ — 3 O a 4 I-1 s 3 3 3 3P n H* ON H-1 cn co cn t— O O O CD I-1 OO 3 3 rt I—1 CnI—1 4> H* Cn Ch I-* CO I - 1 CO CO t-* Cn to !—1 -O to c n c n C O J > 4 > C n c n C n vD I—1 vD I—1 CD O I—1 Cn 1—» CO CO Cn v _ , a rt H O rt 3
' ' n _ _ ON Cn O a ■^J to o Cn ON h-1 I-* P 4 I-1 3 3 o !-* - o o O • . a- CO to v _ /— s X-\ x C D X \ ' - _ 3! ro cn "-J cn P o n oCn C O C O to . 3 3 3 O oo to Cn " 0 I-1 0 0 o o C o C O C n t—* co t—* - O C O C n - ' V , 2 P O 3 3 3 a 3 3 3 p 3 rt 3 3 c r 3 3 3 3 O 3 a 3 rt rt O Mi rt 3 3 O rt 3 I-1 3 3 H- 3 3 H* rt I-1 O O H’ 3 3 P 3 * rt < 3 3 3 3 3 3 3 rt V# X OP H- t-h N n P 3 3 3 M CO O M 3 CO 3 3 3 rt rt, O t-1 33 3 1-* o a 3 3 3 rt O cr H* C 3 3 3 1-* 3 3 3 3 3 a. H« 3 O O 3 a 4 H* a 3 rf a 4 a 3 O rt O 3 < H- v# *<5 vj cr rt 3 O 2! fD r« €3 3 Ho 3 3 O 3 3 rt fD CO rt CO t-h 3 3 a 3 3 H* o fD OP t—* O3 3 3 rt 3* 3 s-ti rt H* 3 3 3 X fD 3 3 fD 3 fD 3 P I-* 3 3 H* P 3 3 o 3 rt 3 H- 3 rt 3 rt < rt O 33 3 M 3 X) X "<5 »* X OP X OP P ft> P 3 H- CO ro 3 o fD ft) CO rt 3 P rt H* P 3 P r{ 3 to CD to P rt O 3 <1 » no XI OP a Colonyft) o cr failures at Cedar Island and South Metomkin are incorporated into data. 38
Table 8. Black skimmer clutch sizes and hatching success by habitat and year. Clutch size
Habitat n 1980 n 1981 n Combined years
Beach 45 3.49 55 2.54 100 3.11
Marsh 5 3.00 13 3.54 18 3.27
Combined 50 3.36 68 2.87 habitats
Hatching success; (No. hatched/no. eggs)
Habitat n 1980 n 1981 n Combined years
Beach 45 0.77 55 0.25 100 0.56
Marsh 5 0.33 13 0.41 18 0.37
Combined 50 0.66 68 0.30 habitats 39
Table 9. Black skimmer chick fate data. s w H CO CO CC cn a (-* CO P PJ cn cn 2! pT fD O C C O o o VO p P o 3 o O rj P r t cr* cr* ti p n 00 c r c r l-{ H* P P CO O P r t r t CO r t r t h-1 r t r t p r t r t r t c r c r I- 1 O O (D c r c r o O CD P* c r c r P r t r t c r r t r t c r P fD fD 3 3 P P CD M 3 3 I - 1 I - 1 P fD (D h-1 M p P CD fD CD r t r t CD I - 1 r t r t 3 W OO 3 dd P OO CD (D 1-3 3 3 P CD HP 3 3 l-l P P S 4 ? r PP £ CD fc4 p r Hy.CO O 3 H* H- CO O 3 H- H- c r T3 P P c r c r td P P
h -1 h-1 o H 1 Ln M Lo No -p I-* (— 11-* oo o OO Ln LO O Ln On On On On On On
i-1 I- 1 n o Ln M O0 M 00 NO -P Oo -P P NO VO h-* VO O I-1 Ln t - 1 On -P ■^J O
M Ln I - 1 h-1 ■P h-1 f-1 h-> NO CTV O '-J o 1 -O DO vo NO -^j OO -P
Ln t—* 1—i -P I- 1 f-1 I—1 VO 00 Ov Ov CTv 1 Ov OO NO OO o Ln
00 OO t—1 1—* OO oo CO 00 00 1 0O OO o O vo I-1
O o O o O o O O o f—* f—1 O Ln h-1 -L o NO -P O - p I-* NO vo 00 f - 1 CN ■P Ov o 00 O VO o OO o Oo
o o O o o o o O o o o o O • «•• •' • • • • • • • (—*NO NO o DO o h-1 f—*Oo M OO 0o NO 00 h-1 NO Ln NO o I-1 Oo f- 1 O0 o Ov OO CD t-t! OQ l-» CTQ fD CO Cu CTQ fD CD 40
Table 10. Black skimmer fledging success by habitat and year. Fledging success: (No. fledged/no. nests)
Habitat n 1980 n 1981 n Combined years
Beach 45 1.09 55 0.14 100 0.71
Marsh 5 0.40 13 0.76 18 0.58
Combined 50 0.92 68 0.35 Habitats
Fledging success (No. fledged/no. eggs)
Habitat n 1980 n 1981 n Combined years
Beach 45 0.31 55 0.05 100 0.21
Marsh 5 0.13 13 0.22 18 0.18
Combined 50 0.27 68 0.11 Habitats 41
Table 11. Common tern conspecific nest spacing. VA-MD VA-MD and NJ
Mean Mean Habitat n dist.(cm) Prob. Habitat n dist.(cm) Prob
Beach 89 305.7 F = 77.96 Beach 89 305.7 F = 73 06 Marsh 67 114.1 P < 0.0001 Marsh 295 174.2 P < 0 0001
Mean Mean Year n dist.(cm) Prob Year n dist.(cm) Prob
Beach:
1980 59 270.8 t = 2.1675 no beach habitat in NJ 1981 30 374.2 P < 0.0365
Marsh:
1980 not tested separately 1980 137 149.9 t = -3 9560 1981 1981 158 195.4 P 0 0001
Combined habitats:
1980 101 208.2 F =5.34 no beach habitat in NJ 1981 55 251.4 P < 0.0254
Colony n=17, range of mean distances=457.7 cm to 91.6 cm, F = 5.34, P < 0.0001 Table 12. Common tern heterospecific nest spacing. I. Difference between nearest conspecific and nearest heterospecific distance.
Nearest Mean Habitat neighbor n dist.(cm) Probability
Beach Consp. 89 305.7 t = 5.1483 Heterosp. 67 520.5 P < 0.0001
Marsh Consp. 74 114.1 t =-1.1861 Heterosp. 18 104.1 P * 0.2519
II. Heterospecific spacing differences.
Mean Habitat n dist.(cm) Probability
Beach 74 520.5 F = 31.41 Marsh 18 104.1 P < 0.0001
Year
F = 4.37, P < 0.0395
Colony
F = 2,78, P < 0.0451 Despite the limited sample size of black skimmer marsh nests, a significant difference was found in conspecific nest spacing by habitat, with beach nests being more separated (Table 13). No difference in : nearest conspecific distance was found by colony or between years. As was found for common terns, skimmer heterospecific spacing was significantly greater than conspecific spacing on the beach but not in the marsh (Table 14). Distance to the nearest heterospecific nest varied significantly by habitat and year, but showed no difference between colonies. 44
Table 13. Black skimmer conspecific nest spacing. Mean Habitat n dist.(cm) Probability
Beach 75 237.6 F = 7.20 Marsh 13 181.4 P <0.0088
Mean Year n dist. (cm) Probability
1980 55 244.7 F = 2.14 1981 33 203.6 P < 0.1478
Colony
F = 2.58, P < 0.0582
« Table 14. Black skimmer heterospecific nest spacing. \
I. Difference between nearest conspecific and nearest heterospecific distance.
Nearest Mean Habitat neighbor n dist.(cm) Probability
Beach Consp. 88 237.6 t = 5.8795 Heterosp. 75 440.4 P < 0.0001
Marsh Consp. 13 181.4 t =-0.9478 Heterosp. 13 153.8 P < 0.3619
II. Heterospecific spacing differences.
Mean Habitat n dist.(cm) Probability
Beach 75 440.4 F = 18.14 Marsh 13 153.8 P < 0.0001
Year
F = 28.31, P < 0.0001
Colony
F = 2.40, P < 0.0727 DISCUSSION
Egg-laying dates
Factors influencing the onset of egg-laying in colonial waterbirds are probably numerous, complex and species specific. The age or experience of birds may affect both the timing,of nesting and the clutch size of common terns (Nisbet 1975). The number of birds present may also v be correlated with both the timing of breeding and colony synchrony, with larger colonies breeding earlier and being more synchronous in their laying dates (Darling 1938). Conflicting evidence exists for the importance of temperature as a cue for the initiation of breeding.
Schreiber (1980) suggests that cold winter temperatures inhibit early nesting by Florida brown pelicans (Pelecanus occidentalis), while
Rodgers (1980) discounts temperature as a factor in the timing of egg- laying by little blue herons (Florida caerulea) in Florida. A positive correlation was found between common tern egg-laying dates and air temperature in Finland, with the earliest laying dates occurring in years of highest temperature (Lemmetyinen 1973b). On the mid-Atlantic coast of the U.S., no relationship exists between a critical minimum temperature and median peak egg-laying dates by common terns, although the onset of laying is coincident in regional difference to the among- regional temperature differences seen between Virginia-North Carolina and
Rhode Island-New Jersey (Smith et al. 1981). Evidence for wide fluctuation in egg-laying dates is found in the records of Ontario common terns initiating egg-laying in 1972 as early as birds from
46 47
North Carolina in 1980 (Morris et al. 1976, Smith et al. 1981). In my study, temperature and rainfall means do not appear to differ enough between years to have caused the delayed clutch initiation seen in 1981 by VA-MD common terns and black skimmers. On the beach, the smaller numbers of terns within the study colonies may have resulted in decreased social stimulation, delaying the onset of egg-laying as predicted by Darling (1938).
Renesting by both species was limited to only a small percentage of the peak laying attempts in each year. The 1980 season was relatively abbreviated and successful, with no need for large scale relaying. In
1981, the delay in the onset of laying resulted in a late season, precluding the possibility for common tern renesting after the colony failures at South Metomkin and Cedar Island. When considering all marsh colonies from VA-MD and NJ together over the two year period, common terns showed three times the percentage relaying at marsh colonies as was found at the beach colonies, 18.8 per cent to 6 per cent. A high percentage of relaying is associated with large scale failure of the first nesting attempt and would indicate lower success of marsh nests.
Colony washout and predation
This study provided a view of potential productivity by seabirds under two different regimes of environmental pressure. In 1980, there were no colony washouts and large scale predation was suffered at only one site, while in 1981, dramatic incidents of both storm tide washout and predation caused the complete failure of two colonies and limited the production of the remaining two. The predation of adult birds on
Cedar Island early in 1981 and the brief predation of fledglings on
Horsehead Tump later that season were very similar in method, with the 48
heads and wings of the prey severed but left with the carcass. This
method of predation is characteristic of great horned owls (Austin 1948)
and seems the most likely explanation with the proximity of a known
population of great horned owls on wooded Assateague Island only 1 km to
the east of the colony (M. A. Byrd, College of William and Mary, pers.
comm.). Nisbet (1975a) found similar predation in a Massachusetts common
tern colony, although the death of chicks was consistent throughout the
season and caused only night desertion by adults. The predation of adult
birds on Cedar Island caused complete abandonment by all breeding
species, while the later pxedation on fledglings at Horsehead Tump did
not affect the remaining birds.
The large scale death and disappearance of both common tern and
black skimmer chicks from the North Metomkin colonies in 1980 and 1981
seems most likely attributable to the presence of both herring gull and
laughing gull colonies within 100 m of the site in both years. Both of
these species are known egg and chick predators (Hatch 1970, Buckley and
Buckley 1972, Burger and Lesser 1979), and both species were observed
being chased from the colony by both terns and skimmers on several
occasions. No terrestrial predators have been reported on Metomkin Island
(Dueser et al. 1979), but another known avian predator of eggs and
chicks, the black crowned night heron (Nycticorax nycticorax)(Hunter and
Morris 1976), does nest in the Spartina salt marsh behind the island.
Reproductive success
Even with the relatively poor success of all VA-MD colonies in 1981 negating the calculation of significant differences for fledging success between habitats for common terns, differences in both clutch size and hatching success seem to indicate that the beach habitat provides greater 49
potential for success than the newly colonized marsh habitat. Lower
clutch size might also indicate a higher proportion of young or
inexperienced breeders in the marsh population, however. A long-term banding and recapture study of terns and skimmers would indicate the age structure of the population, as well as the origin of individuals.
Marsh-nesting birds may be moving into that habitat from the higher density barrier beaches in VA-MD or emigrating from the marsh populations of New York and New Jersey to the north.
Although black skimmer productivity data were not analyzed for
significant differences in success by habitat, the reversal in success rates between years by habitats is noteworthy. The small sample size of marsh nests should not be considered representative of the population, however. Combined with the lack of washouts and predation within
enclosures on Horsehead Tump in 1981, this small sample gave the appearance of increased productivity in the marsh.
Continued monitoring of the comparative success of birds between habitats is needed to more accurately quantify any differences which exist, so that intelligent management decisions can be made. The need for long-term study is evident in the different degrees of predation pressure and colony washout witnessed between 1980 and 1981. How limiting are
these factors and which of the two years most accurately represents the levels of productivity which can be expected over several years? The success of the two species should be compared to the strictly marsh- nesting Forster1s tern, to identify differences in adaptation and determine where the common terns and black skimmers are failing in the reproductive cycle, in clutch size, hatching success, fledging success or some other factor. The success of the terns and skimmers in the marsh could also be compared to the success achieved on other alternative habitat types such as dredge spoil islands and man-made structures.
Nearest neighbor distances
The significant difference in common tern and black skimmer conspecific nest spacing by habitat may be due in part to the structure of the two habitat types. Barrier island beaches are expansive in VA, allowing breeding pairs to space themselves at an optimal distance: for social interaction and predator avoidance. Nesting on salt marsh islands, however, is limited to accumulated mat of Spartina wrack, or more commonly in MD, Zostera windrow. This drift material forms a borde around each island and usually provides the only relief from the thick
Spartina vegetation cover. Common terns in New Jersey nest on eelgrass windrow 80 per cent of the time even though that substrate accounts for only 5.per cent of each island on the average (Burger and Lesser 1978).
Restriction to this substrate would seem to minimize the nearest neighbor distance, as was found in my study, and likewise to limit the number of pairs which can occupy any particular island.
The significant difference in common tern conspecific nest spacing by year may be partly a result of the decrease in the number of adults present in the beach colonies in 1981 from 1980, density decreasing with population size (Fretwell 1968). The increase in spacing seen between years for common tern marsh nests cannot be related to numbers, as they remained constant between years, but may instead be due to a decrease in the amount of nesting substrate available in 1981. New
Jersey marsh islands had substantially lesser amounts of wrack deposition in 1981 than had been present in 1980 (R. M. Erwin, U.S.
Fish and Wildlife Service, pers. comm.). No measure of percentage wrack cover was recorded in MD, however. The significant difference in common 51 tern conspecific nest spacing by colony can be attributed to the difference in habitats, with the five beach colonies being ranked in the top seven of seventeen total colonies.
‘ Erwin (1979a) found ....a difference in nest density in mixed colonies of common terns and black skimmers on Fisherman Island, VA in 1973, hat no difference in success according to nearest neighbor species. Black skimmers were found to have a smaller heterospecific distance to common terns than conspecific distance in New Jersey salt marshes (Burger and
Lesser 1978) but this was accounted for by the small number of skimmer nesting pairs. Lack of substantial numbers of skimmers may also account for the failure to find a, difference between conspecific and heterospecific distance in the marsh during this study, although the / limited availability of nesting substrate must also be considered. The
The significant difference between conspecific and heterospecific . distance on the beach is expected in VA, again because of the seemingly unlimited amount of potential breeding habitat on the barrier islands.
Heterospecific spacing considered separately for each species followed the same trend as seen for conspecific spacing, differing by habitat and by year. The reason for these differences should also be the same, related to the restriction of usable habitat in the marsh, and for terns, the presence of smaller colonies in 1981. The effect of the reduced tern-skimmer distance in the marsh needs to be investigated to determine if levels of aggression are remaining consistent with those reported on beach habitat (Erwin 1979a). With the skimmers associating with the terns in the marsh as they do on the beach, it appears that the decrease in spacing does not outweigh the benefit which the skimmers receive in protection from predation.
The differences in spacing patterns by both species indicate that adapting to the marsh habitat is requiring the adjustment of certain behaviors. Although other changes have occurred which indicate the flexibility of terns, including nest repair behavior after tidal flooding (Burger 1979), the results of this study indicate that the level of success possible on the barrier beach habitat is not being achieved. The colonization of the MD salt marsh islands would not seem to be of selective advantage with the close proximity of the extensive beach habitat in VA, but it may be that the density of the barrier island population has forced some individuals to shift into this
’’inferior11 habitat type. Although the pristine VA barrier islands serve as a refuge on the mid-Atlantic coast of the U.S., where most of this traditional seabird nesting habitat has been destroyed, the survival of populations in other areas depends on their ability to maintain sufficient levels of productivity in alternative habitat types. In years such as
1981, when the barrier islands are subjected to tidal flooding and high levels of predation, salt marsh islands in protected bays which are removed from large colonies of gulls and potential terrestrial predators may in fact offer a better alternative. 53
Appendix A. Common tern conspecific nest spacing by colony.
VA-MD and NJ
Colony Year Mean dist.(cm) n Grouping'
North Metomkin 1981 457.7 15 A Wading Thoro 1981 359.3 23 B Smith Island 1980 311.1 18 BC South Metomkin 1981 290.8 15 BCD South Metomkin 1980 270.3 17 CD Parker Island 1981 245.0 18 CDE North Metomkin 1980 241.0 24 CDEF Wading Thoro 1980 214.7 16 , DEF North Obes 1980 190.0 21 EF Main Marsh Thoro 1981 183.0 28 EFG South Obes 1980 176.4 19 EFG Boomerang Isl. 1981 166.7 28 FGH North Obes 1981 160.6 36 FGH Boomerang Isl. 1980 142.4 19 FGH Horsehead Tump 1980 120.1 42 GH Horsehead Tump 1981 104.1 25 GH Boot Island 1980 91.6 18 H
aColonies with the same letter are not significantly different from each other; tested with the Duncan multiple range test. LITERATURE CITED
Austin, O.L. 1929. Contributions to the knowledge of the Cape Cod Sterninae. Bull. Northeastern Bird Banding Assoc. 5: 123-140.
_____ . 1932. The status of the Cape Cod terns in 1932. Bird-Banding 3: 143-156.
_____ . 1938. Some results of adult tern trapping in the Cape Cod colonies. Bird-Banding 9: 12-25.
. 1940. Some aspects of individual distribution within Cape Cod tern colonies. Bird-Banding 11: 155-169.
. 1948. The status of the Cape Cod terns in 1947. Bull. Mass. Audubon Soc. 32: 137-138.
. 1949. Site tenacity, a behavior trait of the common tern (Sterna hirundo). Bird-Banding 20: 1-39.
_____ . 1951. Group adherence in the common tern. Bird-Banding 22: 1-15.
Bailey, H.H. 1913. The Birds of Virginia. Lynchburg, VA. J. P. Bell Co., Inc•
Barr, A.J., J.H. Goodnight, J.P. Sail, W.H. Blair, and D.M. Chilko. 1979. SAS Userfs Guide. SAS Institute, Inc. Raleigh, N.C. pp. 221- 263.
Bent, A.C. 1921. Life histories of North American gulls and terns. U.S. Nat. Hist. Mus. Bull. 113: 337 pp.
Borodulina, T.L. 1966. Biological and economic importance of gulls and terns of southern USSR water bodies. Ecology-morphology of birds and mammals. Israel program for scientific translations, Jerusalem.
Buckley, F.G. and P.A. Buckley. 1972. The breeding ecology of royal terns (Sterna thalasseus maxima maxima). Ibis 114: 344-359.
_____ • 1980. Habitat selection in marine birds, pp. 69-112 in Behavior of Marine Animals Vol. 4 (J. Burger, B. 011a, H. Winn, eds.) Plenum Press, New York.
54 55
Buckley, P.A. and F.G. Buckley. 1975. The significance of dredge spoil islands to colonially nesting waterbirds in certain national parks, in Proc. Conf. Management of Dredge Islands in North Carolina Estuaries (J. Parnell and R. Soots, eds.) pp. 35-45, Univ. North Carolina Sea Grant Publication, UNC-SG-75-01, 142 pp.
_____ . 1977. Human encroachment on barrier island beaches of the northeastern United States and its impact on coastal birds, in: Coastal Recreation Resources in an Urbanizing Environment: A Monograph (J.H. Noyes and E.H. Zube, eds.) pp. 68-76, Univ. Mass., Amherst.
_____ . Population and colony site trends of Long Island waterbirds for five years in the mid-1970's. Trans. Linn. Soc. N.Y. in press.
Burger, J. 1977. Nesting behavior of herring gulls: invasion into Spartina salt marsh areas of New Jersey. Condor 79: 162-169.
____. 1979. Nest repair behavior in birds nesting in salt marshes. J. Comp. Physiol. Psych. 11: 189-199.
_____ . 1980. Nesting adaptation of the herring gull (Larus argentatus) to salt marshes and storm tides. Biol, of Behav. 5: 147-162.
Burger, J. and P. Lesser. 1978. Determinants of colony site selection in common terns (Sterna hirundo). Proc. 1977 Conf. Col. Waterbird Grp. 1: 118-127.
_____ . 1978. Selection of colony sites and nest sites by common terns (Sterna hirundo) in Ocean County, New Jersey. Ibis 120: 433-449.
_____ . 1979. Breeding behavior and success in salt marsh common tern colonies. Bird-Banding 50: 322-337.
Burger, J. and J. Shisler. 1978. Nest site selection and competitive interactions of herring and laughing gulls in New Jersey. Auk 95: 252-266.
Cooper, D., H. Hays, and C. Pessino. 1970. Breeding of the common and roseate terns on Great Gull Island. Proc. Linn. Soc. N.Y. 71: 83-104.
Cramp, S.W., W.R.P. Bourne, and D. Saunders. 1974. The Seabirds of Britian and Ireland. New York: Taplinger Publ. Co.
Crowell, E.M. and S. Crowell. 1946. Displacement of terns by herring gulls at the Weepickett Islands. Bird-Banding 17: 1-10.
Cullen, J.M. 1960. Some adaptations in the nesting behaviour of terns. Proc. 12th Internat. Ornithol. Congr., Helsinki, pp. 153-157.
Darling, F.F. 1938. Bird flocks and the breeding cycle. Cambridge Univ. Press, Cambridge, England, 124 pp. 56
Drury, W.H. 1965. Gulls vs. terns; clash of coastal nesters. Mass. Aububon-summer.
. 1973-1974. Population changes in New England seabirds. Bird- Banding 44: 267-313; 45: 1-15.
. 1979. Population dynamics in northern marine birds, pp. 123-139, in Conservation of Marine Birds in Northern North America (Bartonek and Nettleship, eds.) U.S. Fish and Wildlife Serv. Rept. No. 11.
Drury, W.H. and I.C.T. Nisbet. 1972. The importance of movements in the biology of herring gulls in New England, pp. 172-212, in Population Ecology of Migratory Birds, U.S. Fish and Wildlife Serv. Rept. No. 2.
Dueser, R.D. 1979. Mammals on the Virginia barrier islands. J. Mammal. 60: 425-429.
Duffy, D. 1977. Populations of terns and skimmers on Long Island Sound and eastern Long Island, 1972-1975. Proc. Linn. Soc. N.Y. 73: 1-48.
Erwin, R.M. 1975. Comparative foraging and breeding strategies among three seabird species, the royal tern, common tern and black skimmer. PhD. thesis, Univ. MD. Ill pp.
. 1977. Black skimmer breeding behavior and ecology. Auk 94: 709-717.
_____ . 1979a. Species interaction in a mixed colony of common terns and black skimmers. Anim. Behav. 27: 1054-1062.
_____ . 1979b. Coastal waterbird colonies: Cape Elizabeth, Maine to Virginia. U.S. Fish and Wildlife Serv., Office of Biol. Serv., U.S. Dept, of Interior FWS/0BS-79/10, 212 pp.
_____ . 1980. Breeding habitat use by colonially-nesting seabirds in two mid-Atlantic regions under different regimes of human disturbance. Biol. Conserv. 18: 39-51.
Erwin, R.M. and T. Custer. 1982. Measuring reproductive success in waterbirds: some biases and alternatives. Colonial Waterbirds 5: in press.
Erwin, R.M. and C.E. Korschgen. 1979. Coastal waterbird colonies: Maine to Virginia, 1977. An atlas showing colony locations and species composition. U.S. Fish and Wildlife Serv., Biological Services Program, FWS/OBS-79/08.
Erwin, R.M., J. Galli, and J. Burger. 1981. Colony site dynamics and habitat use in Atlantic coast seabirds. Auk 98: 550-561. 58
Marples, G. and A. Marples. 1934. Sea Terns or Sea Swallows. Country Life, Ltd. London.
McNicholl, M.K. 1975. Larid site tenacity and group adherence in relation to habitat. Auk 92: 98-104.
Montevecchi, W.A. 1975. Behavioral and ecological factors influencing the reproductive success of a tidal salt marsh colony of laughing gulls. PhD. thesis, Rutgers Univ., New Jersey.
Morris, R.D. and R.A. Hunter. 1976. Factors influencing desertion of colony sites by common terns (Sterna hirundo). Can. Field-Nat. 90: 137-143.
Morris, R.D., R.A. Hunter, and J.F. McElman. 1976. Factors affecting the reproductive success of common tern (Sterna hirundo) colonies on the lower Great Lakes during the summer of 1972. Can. J. Zool. 54: 1850-1863.
Nisbet, I.C.T. 1973. Terns in Massachusetts: Present numbers and historical changes. Bird-Banding 44: 27-55,
_____ . 1975a. Selective effects of predation in a tern colony. Condor 77: 221-226.
_____ . 1975b. Asynchronous hatching in common and roseate terns. Ibis 117: 374-379.
. 1976. Returns from terns. Mass. Audubon Newsletter 15: 3-5.
Nisbet, I.C.T. and W.H. Drury. 1972. Measuring breeding success in common and roseate terns. Bird-Banding 43: 97-106.
Palmer, R.S. 1941. A behavior study of the common tern (Sterna hirundo hirundo L.). Proc. Boston Soc. Nat. Hist. 42: 1-119.
Parnell, J. and R. Soots. 1980. Atlas of colonial waterbirds of North Carolina. Univ. N.C. Sea Grant Publ. UNC-SG-80-06.
Partridge, L. 1978. Habitat selection, in Behavioural Ecology, an Evolutionary Approach (J.R. Krebs and N.B. Davies, eds.) pp. 351- 376, Sinauer Associates Inc., Sunderland, Mass.
Rodgers, J.A. 1980. Breeding ecology of the little blue heron on the west coast of Florida. Condor 82: 164-169.
Schrieber, R.W. 1980. Nesting chronology of the eastern brown pelican. Auk 97: 491-508.
Smith, D.C., R.M. Erwin, T.W. Custer, and J.0. Fussell. 1981. Common tern colonies along the mid-Atlantic coast: I. Nesting chronology. Colonial Waterbirds 4: 160-165. 59
— Southern, W.E. 1977. Colony selection and colony site tenacity in ring-billed gulls at a stable colony. Auk 94: 469-478.
Southern, W.E. and L.K. 1979. Absence of nocturnal predator defense mechanisms in breeding gulls. Proc. 1978 Conf. Colonial Waterbird Group 2: 157-162.
Stone, W. 1937. Bird studies at Old Cape May, Vol. 2. Philadelphia, Delaware Valley Ornithol. Club.
Storey, A. 1978. Adaptations in common terns and Forster’s terns for nesting in the salt marsh. PhD. thesis, Rutgers Univ., New Jersey.
Switzer, B. 1971. DDE levels in eggs and reproductive success in common terns in Alberta. Can J. Zool. 49: 418-439.
Wolk, R.G. 1959. Some reproductive behavior patterns of the black skimmer. PhD. thesis, Cornell Univ., New York. VITA
Daniel Carey Smith
Born in Jacksonville, Florida on December 8, 1955. Graduated from
McLean High School, McLean, Virginia in June of 1974. Awarded the
Bachelor of Science degree in Biology from the Virginia Polytechnic
Institute and State University, Blacksburg, Virginia in March of 1979 after completing a Cooperative Education program with the U.S. Fish and
Wildlife Service at the Patuxent Wildlife Research Center, Laurel,
Maryland. Began graduate study toward the Master of Arts degree in
Biology at the College of William and Mary in Virginia, Williamsburg,
Virginia in August of 1979 as a graduate teaching assistant. Research completed in association with the Migratory Bird and Habitat Research
Laboratory of the U.S. Fish and Wildlife Service, Laurel, Maryland.