Warning This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards, as per the Policy on Communications and Federal Identity.

Skip booklet index and go to page content

COSEWIC Update Status Report on the Roseate Tern in Canada 1999



Foraging habitat of breeding Roseate Terns

Roseate Terns generally forage in shallow areas close to shore, near shoals and tide rips (Safina 1990).  At some colonies, Roseate Terns travel up to 30 km round trip to find food (Heinemann 1992).  Common Terns forage under a wider range of habitat conditions, and are less restricted by physical oceanography (Safina 1990).  As a result, Roseate Terns prey on a limited number of fish species, whereas Common Terns have a more diverse diet (Richards and Schew 1989, Safina et al. 1990).  The specialized nature of Roseate Tern foraging habitat may partially explain why this species is both less abundant and less widely-distributed than Common Terns (Safina 1990, Nisbet and Spendelow 1998).  Furthermore, because Roseates in given colonies prey primarily on only one or two species, they are vulnerable to environmental perturbations affecting these fish (Safina et al. 1988, 1990).

Table 3.  Canadian Roseate Tern breeding sites occupied since the early 1980s.  Bold entries are from the original status report (Kirkham and Nettleship 1985)
Breeding SiteYearDetailsSource
1. Magdalen Islands
    *=breeding confirmed
1972-19831-5 prKirkham and Nettleship 1985
 19872 adults*Shaffer and Robert 1996
 19886 adults*Shaffer and Robert 1996
 19893 adults*Shaffer and Robert 1996
 19905 adultsFradette 1990
 19914 adultsShaffer and LaPorte 1996
 19922 adults*Shaffer and LaPorte 1996
 19937 adults*Shaffer and LaPorte 1996
 19942 adultsShaffer and LaPorte 1996
 19954 adultsShaffer and LaPorte 1996
 19963 adultsShaffer and LaPorte 1996
 19972 adults*F. Shaffer pers. comm.
2. Machias Seal Island19821 prKirkham and Nettleship 1985
 1983-5absentKirkham and Nettleship 1985
 19881 prKennedy 1988
 1992absentanonymous 1992
 199314 adultsP. Dooley unpubl. Data
 19941 nest, second suspectedK. Amey pers. comm.
 19952 nests susp.K. Amey pers. comm.
 19962 nests susp.K. Amey pers. comm.
 19975 adultsJ. Hudson pers. comm.
3. Peter’s Island19811-2 adultsKirkham and Nettleship 1985
 19821 prKirkham and Nettleship 1985
 19881-2 prsJ. Cohrs pers. comm.
 19951 prBoates and Sam 1996
 1996absentD. Currie pers. comm.
 1997absentD. Currie pers. comm.
4. Holmes Island19911 adultD’Eon 1991
 19931 adultBoates et al. 1993
 1994-7absentD’Eon 1994-1997
5. Tusket Island198315-20 prsKirkham and Nettleship 1985
 1997absentD’Eon 1997
6. Ile Chesapeake19962 prs, questionableD’Eon 1996
 1997absentD’Eon 1997
7. The Brothers198255-60 prsKirkham and Nettleship 1985
 198436+ prsT. D’Eon unpubl. data
 199120 prsD’Eon 1991
 199223 prsD’Eon 1992
 199330 prsD’Eon 1993
 199434 prsD’Eon 1994
 199533 prsD’Eon 1995
 199648 prsD’Eon 1996
 199754 prsD’Eon 1997
8. Mud Island19822-3 adultsKirkham and Nettleship 1985
 1991absentD’Eon 1991
 1995absentBoates and Sam 1996
9. Westhaver Island19828 prsKirkham and Nettleship 1985
 1997adults foraging nearbyD. Currie pers. comm.
10. Grassy Island199340 adultsBoates et al. 1993
 199415-20 prsDickie 1994
 199530 prsBoates and Sam 1996
 199712 prsP. Mills pers. comm.
11. Wedge Island19856 prsKirkham and Nettleship 1985
 1995absentBoates and Sam 1996
12. Sambro Island19826 adultsKirkham and Nettleship 1985
 1995absentBoates and Sam 1996
13. Fisherman’s Beach19827 prsKirkham and Nettleship 1985
 1983absentKirkham and Nettleship 1985
14. Inner West Bird Island**1987probableErskine 1992
 19978 prsWhittam 1997
15. Thrumcap Island19822 prsKirkham and Nettleship 1985
 1995absentBoates and Sam 1996
16. Spit off Fisherman’s Harbour*19975 prsWhittam 1997
17. Country Island198725 prsField notes, Erskine 1992
 199530 prsBoates and Sam 1996
 199645 prsWhittam 1997
 19971 prWhittam 1997
18. Unnamed Island off Charlos Cove*19974 prsWhittam 1997
19. Hog Island1987several pairs, suspected nestingField notes, Erskine 1992
20. Cooks Island1987several pairs, suspected nestingField notes, Erskine 1992
 1997absentNSDNR unpubl. data (aerial survey)
21. Sable Island198510-20 prsKirkham and Nettleship 1985
 19934 prsZ. Lucas pers. comm.
 19943 prsZ. Lucas pers. comm.
 19952 prZ. Lucas pers. comm.
 19971 prZ. Lucas pers. comm

*Roseate Terns at these sites were believed to have immigrated from Country Island in 1997 (Whittam 1997).

**Erskine (1992) noted Roseate Terns “probably” breeding in square 20TNE56.  The Bird Islands are in this square; however,
original field notes for this square (unavailable at the time of writing) should be consulted to confirm this record.

The northeastern Roseate Tern population is divided into two groups based on differences in oceanography and diet.  These are: a) the “warm-water” group, that nests mostly between Long Island, NY, and Cape Cod, MA, and feeds primarily on American sand lance (Ammodytes americanus) and lesser numbers of bluefish (Pomatomus saltatrix), mackerel (Scomber scombrus) and anchovies (Anchoa spp.); and b) the “cold-water” group that nests in the upper Gulf of Maine and southeastern Canada and feeds primarily on Atlantic herring (Clupea harengus harengus) and white hake (Urophycis tenuis, Nisbet and Spendelow 1998).

While there has been no thorough study of prey selection by Roseate Terns in Canada, some information has been gathered.  On Country Island, Roseate Terns prey primarily on equal numbers of sand lance and white hake, although butterfish (Peprilus triacanthus) and lumpfish (Cyclopterus lumpus) are also taken (Whittam and Leonard unpubl. data).  Sand lance are commonly taken on Sable Island (I. McLaren pers. comm.).  On The Brothers, Roseate Terns have been observed feeding on Atlantic silversides (Menidia menidia), butterfish and Atlantic herring (D’Eon 1994, 1996).


Staging and wintering habitat

After fledging in early August, juvenile Roseate Terns from the northeastern population disperse with their parents to staging areas located from Long Island to Nantucket and Cape Cod, and in the Gulf of Maine (e.g. Stratton Island; Shealer and Kress 1994).  At these sites, terns feed offshore on abundant sand lance and return to roost at night (Shealer and Kress 1994).

Roseate Terns migrate south in late August and early September.  They arrive in South America by October, where they have been recovered along the north coast from western Colombia to eastern Brazil, between 11° and 18° S (Nisbet 1984, Hays et al., 1997).  A large concentration of about 10,000 terns, including up to 3,000 Roseate Terns, was discovered in 1997 at Mangue Secco, Bahia, Brazil (Hays et al. in press).  This concentration contained banded Roseate Terns from every major breeding colony in the U.S. (Hays et al. in press).  Almost nothing is known about the winter ecology and behaviour of Roseate Terns from the northeastern population (Nisbet and Spendelow 1998).


Breeding habitat

Roseate Terns nest in colonies almost exclusively on small islands, frequently vegetated with beach grass and herbaceous plants (Nisbet 1981).  In northeastern North America Roseate Terns always nest in association with Common or Arctic Terns. Terns require colony sites that are relatively free from predators, and will abandon a colony after a season of heavy predation (Nisbet 1981).  Roseate Terns breeding in North America are limited by the number of available predator-free (or predator-controlled) colony sites that are also in close proximity to good foraging sites.

Within a colony, Roseate Terns nest at sites that provide more cover than nest sites of Arctic or Common Terns (Burger and Gochfeld 1988, Ramos and del Nevo 1995, Whittam 1997, Whittam unpubl. data).  This cover is usually in the form of dense vegetation or strewn rocks, boards, or driftwood (Nisbet 1981, Spendelow 1982). Roseate Terns will nest in boxes, half-buried tires, or other shelters provided by humans (Spendelow 1982).  Reproductive success has been found to be greater under artificial shelters than in natural sites (Spendelow 1996).  Table 4 provides information on the type of nesting habitat used by Roseate Terns at major Canadian breeding colonies. Similar descriptions of nest sites at U.S. colonies can be found in Nisbet (1981, 1989).

Table 4.  Nesting habitat used by Roseate Terns in Canada
The Brothers IslandsTwo 1/3 ha rocky islands, 600 m apart, vegetated with low grasses.60% concealed near or under shelter such as lobster crates, nest boxes, tires, dory bottom, plywood; vegetation, 20% in the openD’Eon 1997
Grassy Island1/3 ha rocky island surrounded by narrow cobble beachConcealed under nest shelters and under patches of herbaceaous plants . Center covered in grasses, plants, especially seaside sedges and herbaceous plants.P. Mills pers. Comm.
Country Island19 ha rocky island with organic herbaceous plants, especially sedges, herbaceous plants and small copses of white spruce.Mostly concealed under patches. Some under rocks, 1 under nest box in 1997Whittam 1997
Sable Island40 km long, 1.5 km wide island. Sand dunes covered in marram grass and herbaceous plants.Normally in clumps of marram grass.Kirkham and Nettleship 1985


The effect of gulls on Roseate Tern breeding habitat

Many tern colonies have been abandoned this century due to the presence of gulls (Crowell and Crowell 1946, Kress 1983, Howes and Montevecchi 1993).  In Canada, Roseate Terns have almost completely abandoned both Sable Island and Country Island, the latter due almost certainly to gull predation (Whittam 1997).  In the absence of predation, these two islands are high-quality breeding sites, primarily because they are located far from the mainland (160 km and 5 km, respectively) and are thus less susceptible to terrestrial predators and human disturbance (Lock et al., 1993, Whittam 1997).

In general, Roseate Terns in Canada currently breed on small sites close to the mainland.  For example, The Brothers Islands are only 600 m from the shore. Furthermore, evidence suggests that Roseate Terns that abandoned Country Island in 1997 moved to three sites, two of which were either attached to, or very close to, the mainland (Fisherman’s Harbour and Charlos Cove, respectively).  In the U.S., gulls have forced terns to move to nearshore sites which now form the bulk of Roseate Tern breeding colonies (U.S. Fish and Wildlife Service 1989, Nisbet and Spendelow 1998). 

It would be valuable to determine whether the gull population in eastern Canada is increasing or decreasing. In 1987 there were an estimated 30,000 pairs of Great Black-backed Gulls and 28,000 pairs of Herring Gulls nesting in Nova Scotia (Lock 1987).  (Lock et al., 1993, p. 16) stated that “Gull populations are not expected to increase greatly in the future, but neither are they expected to decrease greatly over the next few decades.” Unfortunately there has been no census of gull numbers in Atlantic Canada over the last 10 years.  The only recent information is from Gros Morne National Park in Newfoundland, where gulls on Stearing Island increased from 0 pairs in 1976 to 615 pairs in 1992 (Howes and Montevecchi 1993).  The population dropped to 282 pairs in 1993, however, probably due to the fisheries moratorium imposed in 1992 (Deichmann 1993). 


Protection of breeding habitat

The presence of Herring and Great Black-backed Gulls at Roseate Tern breeding colonies is the most serious threat to this species in Canada. Simply acquiring habitat and assigning protected status will not solve this problem, although it should expedite the solution.  Colonies that are deemed essential to the survival of this species in Canada need to be monitored yearly and kept free of breeding gulls.  Gulls currently nest on three of the six Canadian Roseate Tern colony sites (Table 2).  In the U.S., 97% of the Roseate Tern population nests at sites that are managed and protected by biologists or wardens.  Management includes destroying gull nests, harassing adult gulls, and removing persistent predators (gulls and other species; reviewed in Nisbet and Spendelow 1998).  Predator management is clearly necessary for the survival of this species (Nisbet 1981, 1989).


Essential breeding colony sites

Below is a list of the essential colony sites (i.e. “core colonies”, defined in Lock et al. 1993) for Roseate Terns in Canada, along with a description of current ownership, degree of protection, and threats to terns.  A summary of current knowledge on tern reproductive success at each colony is also given.  Numbers in brackets refer to each site’s position in Fig. 1.  General habitat descriptions for each site are in Table 4.


The BrothersIslands, Nova Scotia (7)

These islands were recently acquired by the Province of Nova Scotia (S. Boates pers. comm.).  Both islands are censused 2-3 times each breeding season by T. D’Eon, who has recorded the number of nests since 1991 (Table 3), and clutch sizes of nests since 1996 (see Table 5).  The colony appears to be productive because Roseate Tern fledglings were observed in 1996 and 1997 (D’Eon 1996, 1997).  Productivity at this colony was estimated to be at least 0.62 fledglings per nest in 1997 (data from D’Eon 1997, calculated by Whittam 1997).  It should be noted, however, that mean annual clutch size of Roseate Terns at this site (Table 5) is considerably lower than that of Roseate Terns at the Country Island colony (Table 6), and at most U.S. colonies (i.e. 1.5 to 1.8 eggs; Nisbet 1981).

Table 5.  Number of nests (n) and average clutch size for Roseate Terns nesting on The Brothers Islands in 1996 and 1997 (D’Eon 1996, 1997)
YearnClutch size (± 1SE)
1996481 .29 ± 0.07
1997541.30 ± 0.06

Table 6.  Summary of breeding data for Roseate Terns on CountryIslandin 1996 (reprinted from Whittam 1997)
Breeding CharacteristicData

1-egg nests

2-egg nests



Mean clutch size

Modal clutch initiation date


June 10

Mean incubation period (days)23.4

Hatching success (eggs hatched/nest)   

Eggs hatched (%)

Eggs depredated (%)

Unhatched eggs (%)


47 (63)

18 (24)

10 (13)

Nests where at least one egg hatched (%)29 (64)
Fledging success (fledglings/hatchlings)0.07
Productivity (fledglings/nest)0.08

D’Eon also maintains nest shelters, and destroys nests and eggs of Herring and Great Black-backed Gulls on The Brothers.  A Department of Natural Resources (DNR) “Tern Colony” sign has been placed on North Brother Island to discourage human disturbance.  In 1994 a finfish aquaculture site was set up 60 m from the northeast site of North Brother Island (Boates and Sam 1996).  There was concern that this operation would attract gulls to the area and threaten the nearby nesting terns, or that associated human disturbance would disrupt the terns’ breeding behaviour.  As of the 1997 breeding season, it appears that there has been no negative effect on the terns (Boates and Sam 1996, T. D’Eon pers. comm.).  If the operation expands, however, it may threaten the suitability of this site for Roseate Terns.


Grassy Island, Nova Scotia(10)

This island is owned by the Province of Nova Scotia. Roseate Terns were discovered breeding at this site in 1993, but they have not been closely monitored. Productivity of this colony needs to be ascertained, as do factors potentially affecting productivity (i.e. predation).  Gulls are not known to nest on Grassy Island, but they do nest on many nearby islands (P. Mills pers. comm.).  In 1994, nest shelters were placed to encourage tern breeding, and a DNR “Tern Colony” sign was erected to discourage human disturbance.  Both Grassy Island and The Brothers, including a 250 m circle of water surrounding the islands, will eventually be designated Nova Scotia Wildlife Management Areas (S. Boates pers. comm.).


Country Island, Nova Scotia(17)

This island is owned by the Department of Fisheries and Oceans and is under the jurisdiction of the Canadian Coast Guard. Environment Canada is considering the establishment of a Migratory Bird Sanctuary at this site.  If the island is surplussed by the Department of Fisheries and Oceans Environment Canada would seek to obtain title to the property.  National Wildlife Area status may eventually be given, which will afford protection to both terns and their habitat (Boyne 1998).

As mentioned earlier, Roseate Terns are known to have nested on Country Island in relatively high numbers in 1987, 1995 and 1996 (Table 3), but they produced almost zero chicks in 1996 (Table 6).  Only one pair nested in 1997, and they abandoned their egg after 10 days of incubation (Whittam 1997). 

The Canadian Wildlife Service (CWS) has designed a Tern Restoration Plan for Country Island with the support of the Canadian Roseate Tern Recovery Team, the Nova Scotia DNR, and the Canadian Coast Guard.  A two-year pilot study beginning in April of 1998 will examine the viability of non-lethal gull and corvid control to re-establish Roseate and other terns at this site (Boyne 1998).  Common Eiders (Somateria mollissima) and Leach’s Storm Petrels (Oceanodroma leucorhoa) also nest on Country Island (pers. obs.), and are expected to benefit from predator management (Boyne 1998).  Currently 60-90 pairs of Herring Gulls, 20 pairs of Great Black-backed Gulls, 2 pairs of American Crows, and 1 pair of Northern Ravens nest on Country Island (Whittam 1997).

While predation is the most immediate threat to this colony, an indirect threat is posed by the construction of the Sable Offshore Energy Project (SOEP) pipeline.  This pipeline will transport natural gas from underwater oil fields near Sable Island to Country Harbour, Nova Scotia.  The pipeline will fall within 5-6 km of Country Island, and concerns were raised that its construction could disrupt local prey fish and affect tern foraging behaviour. Furthermore, the associated increase in ship traffic, noise, and possibly pollution, could affect tern breeding behaviour (Whittam and Leonard 1996).  As a result, construction of the portion of the pipeline that lies within 20 km of Country Island has been restricted to the non-breeding season (September - April) (Fournier et al., 1997).


Sable Island, Nova Scotia(21)

This island is administered by the Department of Fisheries and Oceans and is under the jurisdiction of the Canadian Coast Guard.  It was designated a Migratory Bird Sanctuary by CWS to protect nesting migratory bird populations, especially terns and Ipswich Sparrows (Passerculus sandwichensis princeps).  Access to the island is restricted by Coast Guard regulations (Lock et al. 1993).

There is a long history of Roseate Terns breeding on Sable, and 10-20 pairs were reported breeding in 1985 (Kirkham and Nettleship 1985).  Only one or two pairs currently breed on Sable, however, and nothing is known about their breeding success. Approximately 2570 Common Terns, and 286 Arctic Terns, nested in 20 colonies on Sable Island in 1995 (Z. Lucas unpubl. data).

About 2800 pairs of Herring Gulls and 1200 pairs of Great Black-backed Gulls were known to nest on the island in the early 1970s (Lock 1973).  Z. Lucas censused gulls on Sable in 1997, but the results of this census are not yet available.  Lucas also monitored potential effects of gull predation on tern reproductive success on Sable Island in 1997.  She found that predation occurred at only one of five colonies studied (Z. Lucas unpubl. data).  While she did not monitor the colony where Roseate Terns nest (“East Light”), casual observations indicate that gull predation is probably not a problem at this colony (Z. Lucas pers. comm.).

SOEP has the potential to disrupt terns and other birds breeding on Sable Island. According to current understanding, only occasional visits to Sable will be made as a result of SOEP, and these visits will be subject to a code of practice (Fournier et al., 1997). Furthermore, SOEP must conduct an environmental effects monitoring program for at least five years to ascertain any effects on Sable Island birds (Fournier et al., 1997).

The developing “Sable Island Conservancy” is a non-profit partnership formed between private industries, government, and local nature groups to preserve the integrity of Sable Island’s flora and fauna.  If federal funding is provided, this organization could help manage and conserve terns on Sable Island (I. McLaren pers. comm.).

Two ancillary sites where Roseate Terns have nested in small numbers for at least five years include:

MachiasSeal Island(2), which is a Migratory Bird Sanctuary.  It supports 67% of all terns breeding in the Bay of Fundy (Lock et al. 1994), and 1-2 pairs of Roseates have bred at this site since 1979 (Kirkham and Nettleship 1985, K. Amey pers. comm.). Nothing is known about the productivity of these birds.  A resident warden keeps gulls from nesting on the island and controls human access.

MagdalenIslands(1). The three colonies where Roseate Terns currently breed are of mixed ownership. Île aux Cochons is privately owned, Deuxieme îlot is crown land and the ownership of îlot C is unknown (F. Shaffer pers. comm.).  Predation by Great Black-backed Gulls is a problem, but no predator management has been planned (F. Shaffer pers. comm.).  Fox predation is also a serious concern, and since 1994 electric fences have been placed around two major tern colonies to prevent foxes from entering (F. Shaffer pers. comm.).