Date First Listed--November 2, 1987

Historic Range: Tropical and temperate coasts of Atlantic Basin and East Africa.

Family: Laridae

This Status (i.e. endangered) Likely To Occur In: CT, MA, ME, NC, NJ, NY, RI, VA, Canada-(Newf., N.S, Que.)

Population To Which This Status Applies: U.S.A. (Atlantic Coast south to NC), Canada (Newf., N.S, Que.), Bermuda

Description: The roseate tern is about 4O centimeters in length, with light-gray wings and back. Its first three or four primaries are black and so is its cap. The rest of the body is white, with a rosy tinge on the chest and belly during the breeding season. The tail is deeply forked, and the outermost streamers extend beyond the folded wings when perched. During the breeding season the basal three-fourths of the otherwise entirely black bill and legs turn orange-red.

Habitat: Roseate terns breed primarily on small offshore islands, rocks, cays, and islets. Rarely do they breed on large islands. They have been reported nesting near vegetation or jagged rock, on open sandy beaches, close to the waterline on narrow ledges of emerging rocks, or among coral rubble.

Population and Range: The largest numbers of breeding Roseate Terns (Sterna dougallii ) nest in tropical and subtropical areas of the Indian and North Atlantic Oceans, but temperate zone breeding populations can be found in North America, Europe, South Africa, and Western Australia. North temperate zone breeding populations on both sides of the North Atlantic Ocean have declined considerably since the 1950s.

Currently about 6,000-6,500 Roseate Terns breed in an area from the south shore of Long Island, New York, north to Nova Scotia, Canada. The Northeast U.S. breeding population was declared "Endangered" by the U.S. Fish and Wildlife Service in December 1987; the Caribbean breeding population, which includes the birds nesting in Florida, was declared "Threatened." There is an unequal sex-ratio in the Northeast U.S. breeding population with more females than males, but the causes of this unequal sex-ratio and its effects on overall reproductive success are not fully understood.

During the late 1980s and early 1990s, about 90% of the Northeast U.S. breeding population nested south and west of Cape Cod, Massachusetts, with about 3,000- 3,300 adults (about 50% of the regional breeding population) at Bird Island, Massachusetts. About 2,500-2,800 Roseate Terns nested at Great Gull Island, New York, about 280-360 nested at Falkner Island, Connecticut, and about 300 nested on the south shore of Long Island, mostly at Cedar Beach and Shinnecock Bay. A former tern colony site taken over by gulls at Ram Island, Massachusetts was restored in the early 1990s and had about 240 nesting Roseate Terns in 1994.

Nesting Habits: Roseate Terns in northeastern North America almost always nest in colonies with Common Terns (S. hirundo). A Roseate Tern can be identified by its lighter gray back and wings, longer outer tail feathers, and, at the beginning of the breeding season, by its black bill. As the breeding season progresses, the bill becomes pink at the base. While they must compete with Common Terns for both food (small fish caught by plunge-diving) and nesting sites, Roseate Terns benefit from the aggressive "colony-site defense" behavior of the Common Terns.

In contrast with Common Terns which usually nest in open or exposed sites, Roseate Terns usually hide their nests under some sort of protective cover such as rocks, vegetation, or washed-up debris. Different types of habitats are used at each of the colony sites of the 6 largest northeastern colonies. At Falkner Island, the birds using protected nest sites created by the research staff usually have higher reproductive success than those using less-protected naturally occurring sites.

Roseate Terns begin arriving on the breeding areas at the end of April, and the earliest eggs now typically are laid during the third or fourth week of May. From 1981-1992, the terns began nesting later each year for as yet unknown reasons. Roseate Terns usually will continue to lay new or replacement clutches until the middle of July, but sometimes will lay eggs as late as early August. Except for renesting, young birds usually nest later than the more experienced individuals. A complete clutch usually contains two eggs laid 2 or 3 days apart; young birds and birds renesting following a failure may lay only a single egg. Clutches of 3 or more eggs, and most 2-egg clutches laid less than 2 days apart, usually are the result of 2 or more females laying eggs at a single nest-site.

For Roseate Terns, incubation usually takes 23-24 days (2 days longer than for Common Terns), and chicks usually fledge 25-28 days after hatching. Roseate Tern chicks often leave the nest site after only 3-7 days to find new hiding places. Young Roseate Terns usually do not return to their nests or hiding sites once they fledge (i.e., become capable of sustained flight), and they may leave the colony with one or both parents a few days later.

Productivity varies both geographically and temporally, but except at colony sites where there is avian or mammalian predation, most first-hatched A-chicks usually survive to fledgling. Most geographical and temporal variation in productivity, therefore, results from differences in B-chick survival.

Some of the variation in productivity may be related to how far adults must travel to find small fish to bring back to the colony site to feed their young. During the first few days after they hatch, the chicks in the New York-Massachusetts area are fed almost exclusively American Sand Lance (Ammodytes americanus ). Studies done at Bird Island and at Falkner Island indicate that some adults may travel more than 20 km from their colony site to foraging areas.

Although some adults and fledglings from all study colony sites in the New York- Massachusetts region have been seen as far north as Stratton Island on the coast of Maine, most usually begin moving in late July to staging areas either at the eastern end of Long Island, New York or Cape Cod, Massachusetts, before starting migration a few weeks later to the north coast of South America. Most recoveries of banded birds in November-December have come from Guyana, South America, but we do not know for certain where the adults and young go from January-April. It has been suggested that they go out to sea and become pelagic, and plans are being made to search for them over the mid-Atlantic Ridge.

Growth, Development, and Dispersion:  Immatures do not return north to the breeding grounds (and we suspect that they remain at the wintering areas) for their first summer as 1-year-olds. A few birds may return and nest as 2-year-olds, but most do not begin to breed until they are at least 3 or 4 years old. Usually about 20% of the fledglings survive the typical "minimum" 3-year maturation period. Roughly 88% of the young birds that survive the maturation period are philopatric and return to their natal colony for their first breeding attempt; the other 12% disperse to another colony site.

In 1987-88, about 2/3 of the new recruits to the third largest Roseate Tern colony at Falkner Island were birds raised at other colony sites. Therefore, although the Falkner Island breeding population appears to be "stable", the data suggest that it is not self-supporting, and that it is being maintained by immigration of new recruits from the two largest colonies at Bird Island and Great Gull Island.

An analysis of data collected from 1988-1992 at the 4 largest colonies in the New York-Massachusetts region suggests that roughly 75% of adults survive each winter and return to the same colony the next spring, about 7% survive and move to a different study site, and about 18% fail to return. The overall average survival rate of about 82% is low in comparison to the survival rates of other species of marine birds. The oldest known Roseate Tern reached 21 years old, but we suspect that a few birds may survive into their mid or late 20s. Older adults show more colony-site fidelity than young birds, and an examination of some factors thought to influence movements of adults indicates that movement probabilities are more closely associated with the identity of the destination colony site than with either the identity of the site of origin or the distance between colony sites.