Food Habits

Main Foods Taken

Small fish up to 150 mm long, frequently and sometimes predominantly crustaceans or insects, occasionally squid; rarely other invertebrates. In breeding area, takes almost exclusively live prey. In wintering area, however, takes dead fish discarded by fishermen or used for bait to trap terns on beach (Morris et al. 1982, Blokpoel et al. 1984, P. Trull).

Microhabitat For Foraging

In breeding area, open waters within about 20 km of breeding sites, where fish available within 50 cm of surface—e.g., shallow coastal waters, bays, inlets, shoals, tide-rips, drift lines, saltmarsh creeks, lakes, ponds, or rivers. On Atlantic Coast, usually within 1 km of shore; sometimes recorded feeding over predatory fish in open water (Duffy 1986, Safina 1990). Very few records of marine birds taking freshwater fish in North America (in contrast to Europe; Becker et al. 1993a). Many defend feeding territories and feed within 100 m of shore (Nisbet 1983d). In Maine, even birds nesting on islands 5–15 km offshore thought to feed mainly inshore (Hall 1999). In New York, often feeds in tidal inlets, over shoals, or in tide-rips (Duffy 1986, Safina and Burger 1988); most breeding colonies near inlets and numbers at each site correlated with volume of water drained by the inlet (Buckley and Buckley 1980). In Great Lakes, feeds singly within 200 m of shore; feeding flocks not reported (Moore 1993, Burness et al. 1994, D. J. Moore). Birds breeding on smaller lakes appear to prefer eutrophic waters for foraging (Pinkowski 1980, LaBarr 1996). In s. Caribbean, feeds over coral flats, in bays, inlets, salt lagoons and freshwater ponds (Voous 1957, van Halewyn 1985); also over open sea (LeCroy 1976).

Little information for staging birds, migrants, or wintering birds. Around Cape Cod, MA, many feed in tidal inlets or between islands (Trull et al. 1999), also to 20 km offshore, sometimes with marine mammals (P. Trull). Off se. U.S., migrants recorded mainly over inner and middle shelf waters <40 m deep; prefer more turbid waters (Haney and Stone 1988). In Trinidad, Guyana, and Suriname, usually in close association with human fishing offshore, along shore or in turbid coastal waters (Blokpoel et al. 1982, 1984); band recoveries from shrimp and fishing boats up to 180 km offshore. In e. and s. Brazil, apparently mainly off-shore (Harrington et al. 1986; Hays et al. 1997, 1999). In Peru, some feed on high-energy beaches (Blokpoel et al. 1989).

Flock Size And Foraging Associations

Often feeds singly or in small groups, but in marine environments, flocks ≥1,000 form over schools of predatory fish that drive prey fish to the surface (Duffy 1986, Safina and Burger 1988); up to 1 birds diving will attract others from up to 1 km away. Near a breeding colony in New York, flock size increased with density of prey in upper 3 m of water column and with activity of predatory fish near surface (Safina and Burger 1988). More successful in denser flocks (Duffy 1986), presumably because flock size increases with prey availability (Safina and Burger 1988). In suitable areas, pairs maintain linear feeding territories of 150–250 m along shoreline and exclude all other conspecifics (Nisbet 1983d, Kirkham 1986); hence, birds along shore widely spaced and overdispersed. In Great Lakes and St. Lawrence River, feeds individually or in pairs (Waltz 1988, Moore 1993, D. J. Moore).

On Atlantic Coast, forms mixed feeding aggregations with Roseate or Arctic Terns; sometimes with Laughing Gulls (Larus atricilla) or other gulls. Common Terns usually out-compete Roseates in mixed groups (Duffy 1986, Safina 1990); no studies of feeding interactions with Arctic Terns in North America. In addition to predatory fish, recorded feeding over foraging Red-throated Loons (Gavia stellata; ICTN), Double-crested Cormorants (Phalacrocorax auritus; Heinemann 1992), and marine mammals (Voous 1957, P. Trull). No in-formation on foraging associations in winter quarters.

Food Capture And Consumption

Takes food almost exclusively while on wing. Feeds by plunge-diving, diving-to-surface, or contact-dipping. Birds holding feeding territories commonly dive from perches on boats or docks (Kirkham 1986); non-territorial birds rarely do so. Flying insects (e.g., Hymenoptera) taken by hawking in air, sometimes at heights up to 60 m; other insects (e.g., moths, beetles) taken from water surface. In breeding colonies, adults frequently steal fish from conspecifics, either forming groups to pursue adult carrying fish in air, or swooping down to seize fish as parent passes it to chick. Large chicks frequently steal fish from neighboring small chicks; gangs of large chicks sometimes prevent small chicks from feeding altogether. Intraspecific kleptoparasitism more frequent in dense colonies and at times when food is limiting (Ludwigs 1998). Often attempts to steal fish from Arctic Terns or Least Terns (Sterna antillarum). Often picks fish from ground when dropped in breeding colonies.

Height of plunge-dives usually 1–6 m; off New York, mean 2.6 m ± 1.3 SD (n = 18; Duffy 1986). Immersion time usually 0.6–1.4 s; off New York, mean 0.8 s ± 0.3 SD (n = 213; Duffy 1986). Dive height and immersion time positively correlated; both significantly lower than those of Sandwich Terns (Sterna sandvicensis) or Roseates, but not significantly different from Arctics (Dunn 1972). Dive depths not measured precisely, but maximum depth estimated as 50 cm (Dunn 1972); fish taken from 65 cm depth in artificial feeding situation (J. J. Hatch).

Dive frequency and capture success widely variable depending on abundance, depth, and behavior of prey, and interference by other terns. In flock-feeding situations off New York, mean frequencies 0.4 successful dives, 0.8 unsuccessful dives, 0.65 aborted dives/min (n = 1,540 min; Safina 1990). Non-territorial birds in Nova Scotia made 0.27 successful dives, 0.40 unsuccessful dives, 0.09 aborted dives/min (n = 933 min; Kirkham 1986). Mean capture success in these studies about 0.33 fish/dive. Adults foraging off United Kingdom had lower capture success (0.22 vs. 0.39 fish/dive) and lower capture rates (0.23 vs. 0.50 fish/min) when sea surface was calm than when moderately choppy (Dunn 1973a). However, foraging techniques changed with increasing wind speed; both diving and capture rates declined when wind speed >7 m/s (Taylor 1983).

Fledglings in Nova Scotia made 0.1 successful dives, 1.2 unsuccessful dives, and 0.7 aborted dives/min (n = 40 min; Kirkham 1986); capture success only 0.08 fish/dive. Migrants (including Arctic Terns, but probably mainly Commons) of unspecified age in New Brunswick in Aug–Sep made 0.7 dives, 0.8 dips, and 1.0 aborted dives/min (n = 75 min) when feeding on fish; 0.9 dives, 2.7 dips, and 0.4 aborted dives/min (n = 82 min) when feeding on euphausiids (Braune and Gaskin 1982b). Migrants of unspecified age in Virginia in Aug had mean capture success of 0.32 fish/dive (n = 81 dives; Reed et al. 1982).

Successful birds catch single prey items in bill, fly up and shake water from plumage, turn prey in bill with tongue, and swallow head-first within a few seconds of capture. Occasionally carry multiple fish (2–4) to young (Hays et al. 1973); not known whether or how fish are caught sequentially in successive dives.

Foraging Range

No quantitative data based on systematic area surveys. Based on trip times (see below), boat surveys, and/or radiotelemetry, most breeding birds feed within 20 km of colony-sites, often much less if numbers small and/or prey locally abundant (Nisbet 1983d, Duffy 1986, Safina 1990, Burness et al. 1994, D. J. Moore, ICTN). At Hamilton Harbour, most birds flew either 0.9 km to a small pond (30% of trips) or 1–8 km to foraging sites on Lake Ontario; mean trip distance 2.4–4.2 km, max. 20 km (n = 99 males, >1,000 trips; Moore 1993, 2001). Longer trips known; a few birds breeding at Bird I., MA, made triangular feeding flights of at least 60 km, including 15 km return flight overland with fish (Heinemann 1992). Some birds from Bird I. defended feeding territories up to 19 km away (Nisbet 1983d, ICTN).

Trip times (intervals between departure from nest and return with fish) vary from <1 min to ≥2 h. Mean trip times reported by Kirkham (1986) for birds feeding chicks: Ball I., Nova Scotia: males, 21 min ± 16 SD (n = 556); females, 13 min ± 14 SD (n = 523); Bird I., MA: both sexes, 16 min ± 12 SD (n = 520). Mean trip times for birds breeding in Hamilton Harbour, Lake Ontario, varied from 19 to 41 min, depending on year and chick age (D. J. Moore).

Generalist and opportunist; >55 species of fish and >35 invertebrate taxa recorded as prey in North America; many others in South America. Adults feed on a wide variety of fish and invertebrates in all areas, but chicks are fed mainly fish at coastal colonies, and almost exclusively fish at freshwater sites. Diets vary spatially, annually, weekly, daily, and even hourly in relation to diurnal and tidal cycles, and activity of predatory fish.

Most quantitative data are for food items fed to chicks, identified visually without collection. In several studies, prey lengths were estimated relative to terns’ bill-length; in a few studies, prey masses and/or energy contents were estimated from lengths. These estimates of biomass and energy content are subject to substantial error (Galbraith et al. 1999), but published estimates are cited here without further qualification.

In s. Québec, major food items fed to chicks were American sand lance ((Ammodytes americanus, 37% by frequency), capelin (Mallotus villosus, 19%), and crustaceans (36%, n = 787; Chapdelaine et al. 1985). In s. New Brunswick, stomachs of adults collected in breeding season (n = 13) contained insects (mostly Coleoptera and Diptera, 77%), threespined sticklebacks (Gasterosteus aculeatus, 69%), crustaceans (23%), and annelids (8%; Mills 1957). In s. Nova Scotia, major food items fed to chicks were Atlantic herring (Clupea harengus, 23%), sticklebacks (Gasterosteus spp. and/or Pungitius pungitius [Newfoundland stickleback], 18%), shrimps (15%), pollock (Pollachius virens, 12%), striped mummichog (Fundulus heteroclitus, 12%), shad (Alosa spp., 7%), capelin (6%), and American sand lance (5%; n = 6,582; Kirkham 1986). In Gulf of Maine east of 66°W, major food items fed to chicks were Atlantic herring (42%), hake (white hake [Urophycis tenuis], fourbeard rockling [Enchelyopus cimbrius], etc., 40%), pollock (6%), butterfish (Poronotus triacanthus, 3%), and invertebrates (mostly insects and amphipods: 4%, n = 21,925; Hall 1999). In same area, stomachs of adults (n = 155) collected in breeding season contained fish (46%), shrimps (40%), other crustaceans (10%), and insects (11%; Mendall 1936). In s. Gulf of Maine, major food items fed to chicks were American sand lance (43%), hake (34%), Atlantic herring (12%), and butterfish (5%, n = 9,317; Hall 1999). At Bird I., MA, food items fed to mates included shrimps (73%), Atlantic silversides (Menidia menidia, 16%), American sand lance (4%), and Atlantic herring (3%; n = 1,392), but fish made up about 60% of biomass (Nisbet 1973a, Kirkham 1986). At same site, food items fed to chicks included American sand lance (40%), shrimps (27%), Atlantic silversides (16%), and Atlantic herring (13%; n = 2,189; Kirkham 1986); later in season wide variety including Atlantic silversides, cunner (Tautogolabrus adspersus), Atlantic herring, American sand lance, and northern pipefish (Syngnathus fuscus; Nisbet 1983d). At Cedar Beach, NY, major food items fed to chicks were American sand lance (36%), bluefish (Pomatomus saltatrix, 12%), clupeids (12%), northern pipefish (10%), bay anchovy (Anchoa mitchelli, 8%), butterfish (6%), and Atlantic mackerel (Scomber scombrus, 5%; n = 2,165; Safina et al. 1990); males brought more sand lance and bluefish than females, but fewer of other fish species (Wagner and Safina 1989). At 2 sites in Rhode Island, major food items to chicks included grass shrimps (Hippolyte spp., 30%), killifish (Fundulus spp., 26%), and Atlantic silversides (17%, n = 552; Custer et al. 1986).

At all coastal sites, diet includes variable numbers of invertebrates: Gammarus and Thysanoessa spp. in Gulf of St. Lawrence (Chapdelaine et al. 1985), isopods and euphausiids (Meganyctiphanes norvegica) in Bay of Fundy and Gulf of Maine, shrimps (Crangon spp.), prawns (Penaeus spp.), and crabs (e.g., green crab [Carcinus maenas] or mole crab [Eremita talpoida]) south of Cape Cod. Insects taken sporadically, especially Hymenoptera (ants and stingless bees), Diptera, and June beetles (Phyllophaga spp.). Long-finned squid (Loligo pealei) occasionally taken in numbers in Massachusetts. Mollusks, berries, and other vegetable matter sometimes found in stomachs (McAtee and Beal 1912, Mendall 1936, Moser and Lee 1992) probably from secondary consumption or accidental ingestion.

In Saginaw Bay, Lake Huron, stomachs of adults (n = 611) collected in breeding season contained carpenter ants (Camponotus pennsylvanicus, 39%), lake shiners (Notropis spp., 27%), mayflies (Hexagenia spp., 15%), and yellow perch (Perca flavescens, 12%; n = 2,589 items); by volume, most important prey species were lake shiner (53%), yellow perch (27%), trout perch (Percopsis omiscomaycus; 8%), and carpenter ants (6%; n = 1,054 ml; Manuel 1931). At Hamilton Harbour, Lake Ontario, major food items fed to chicks were emerald shiner (Notropis antherinoides, 34%), fathead minnow (Pimephales promelas, 22%), alewife (Alosa pseudoharengus, 16%), rainbow smelt (Osmerus mordax, 16%), threespined stickleback (5%), and salmonids (Salmo spp. and Oncorhynchus spp., 6%; n = 5,644; Moore 1993, D. J. Moore). At Port Colborne, Lake Erie, major food items fed to chicks were emerald shiner (48%), rainbow smelt (42%), and unidentified larval fish (10%; n = 1,156; Burness et al. 1994).

At sea off N. Carolina, stomachs of adults (n = 51) collected at unspecified dates contained fish (82%), insects (23%), crustaceans (6%), and squid (4%; Moser and Lee 1992).

In n. Argentina in Dec–Mar, food items (n = 1,846) identified in pellets of adults included 55% fish (Argentine anchovy [Engraulis anchoita], 28%; Marini’s anchovy [Anchoa marinii], 16%; drum [Paralonchurus brasiliensis], 4%) and 45% insects (predatory diving beetles [Rhantus signatus and Lancetes marginalis], 26%; dragonflies [Aeshna spp.], 11%; Mauco et al. 2001). In Netherlands Lesser Antilles, diet includes Engraulis spp., Cyprinodontidae, and Blenniidae (Voous 1957). In Peru, reported feeding on sand crabs (Eremita analoga; Blokpoel et al. 1992).

Lengths of prey fed to chicks typically 30–90 mm; range from <5 mm for some invertebrates (insects and crustaceans) to 130 mm for clupeids, 150 mm for capelin and American sand lance, and >180 mm for northern pipefish. See Kirkham 1986 and Galbraith et al. 1999 for details. At Hamilton Harbour, Lake Ontario, mean length of fish fed to chicks 43 mm; mean mass of fish brought to chicks 1.6–4.8 g, varying among years and chick ages; mean energy content 13 kJ (n = 4,280 fish; Moore 2001).

For detailed information on diets in Europe and Asia, see Glutz von Blotzheim and Bauer 1982, Cramp 1985, Il’icev and Zubakin 1990, Higgins and Davies 1996 .

At most colonies, individual birds switch frequently (sometimes several times each day) among different foraging locations, hunting techniques, and types and sizes of prey (Nisbet 1983d, Safina et al. 1990, Becker et al. 1993a, Moore 1993, Burness et al. 1994). Moore (2001) showed these choices depend on levels and variability of food availability in different locations, and can be changed experimentally by manipulating brood size (i.e., demand for food). No evidence that birds actively choose prey of specific sizes (i.e., reject prey of other sizes) when at foraging grounds. When feeding distant mates or young, however, adults usually swallow small items and carry larger items back to colony (Taylor 1979, ICTN); in latter case they orient towards destination within seconds of emerging from the water. Average trip times increase with increasing prey size (Kirkham 1986), but less rapidly than in proportion to biomass; hence, largest food items are most profitable in terms of biomass/unit time. However, large items disproportionately attract kleptoparasites (Hopkins and Wiley 1972, Dunn 1973b, Hatch 1975) and are often lost before they can be fed to chicks. Parents feed smaller food items to small than to large chicks (Kirkham 1986; Wiggins and Morris 1987; Safina et al. 1990; Moore 1993, 2001).

No food storage. In breeding area, takes primarily live prey; fish dropped in colony usually ignored if on ground for more than a few minutes.

In 6 prey species from Lake Ontario, 78–91% of total mass was water; 6–13% of dry mass was lipid (n = 5–36/species; Moore et al. 2000). Two fish samples brought to chicks in coastal New York contained 64% and 74% water; dry matter included 11% and 23% lipid, 13% and 15% ash (Ricklefs and White 1981). In 6 prey species in Germany, 22–43% of dry mass was protein, 9–28% inorganic matter (n = 29–144/species; Massias and Becker 1990). Captive chicks grew fastest when fed herrings or other marine fish; did not grow adequately when fed shrimps (Crangon crangon) or threespined sticklebacks, despite similar or greater intakes of protein, lipid, and total energy (Massias and Becker 1990). Egg-laying females at Bird I., MA, frequently eat fragments of mollusk shells, suggesting temporary calcium deficits (Nisbet 1997). No other information on nutrition.

Estimates of daily energy expenditure (DEE) of adults, obtained with doubly-labeled water technique, were 355 kJ/d during incubation (range 281–418 kJ/d, n = 10; Klaassen et al. 1992), 454 kJ/d during early chick-feeding (range 269–533 kJ/d, n = 14; Galbraith et al. 1999). Energy requirement for egg production estimated as 175 kJ/egg; males delivered 161–183 kJ/d (61–76% of total energy requirement of female) during egg-laying (Moore et al. 2000). Total energy requirement of chicks during first 25 d of life estimated as 4,850 kJ; DEE increased from about 20 kJ/d at hatching to 200–250 kJ/d at ages 15–25 d (Klaassen 1994). Maximum energy required for growth about 66 kJ/d at ages 15–20 d (Ricklefs and White 1981). Additional details on energetics of growth presented by Ricklefs and White (1981), Drent et al. (1992), Klaassen (1994), and Klaassen et al. (1994). For energetics of flight, see Wakeling and Hodgson 1992, Alerstam 1985 .

Basal metabolic rate (BMR) of chicks in range 1.5–2.5 ml O₂(g·h)^-1(equivalent to 30–50 J (g·h)^-1), peaking in mid-growth (Klaassen 1994). BMR of adults reported as 2.08 ml O₂(g·h)^-1(equivalent to 130 kJ/d) by Klaassen (1994), but lower values suggested by Galbraith et al. (1999). Resting heart rate about 230 beats/min (Würm and Hüppop 1998).

Coefficient of temperature regulation (percentage of difference between adult body temperature and air temperature that chick maintains after 30 min at 20°C) increased from 50% in newly hatched chicks to 70%, 85%, and 90% at ages 1–3 d (Ricklefs 1979, calculated from data of LeCroy and Collins 1972). Body temperature of chicks >7 d old varied between 37° and 43°C, peaking at midday (Ricklefs and White 1981). Parents brood chicks continuously until age 3 d; older chicks in hot or cold weather or rain. Unattended chicks seek shade in hot weather, or make scrapes through hot surface layer to cool sand below. Parental brooding estimated to reduce chicks’ thermoregulatory costs by 34–38% (Klaassen 1994). Brooding adults dissipate heat by gular fluttering; when air temperature exceeds 30–35°C, they fly to water, wet feet and belly, and return to nest (Mes et al. 1978, Grant 1981).

Drinks mainly on wing, gliding with wings slightly raised and dipping bill several times into water surface. Coastal birds drink salt water; do not seek fresh water even when available nearby. Most chicks do not drink until after fledging and may be water-stressed when evaporative losses are high (Ricklefs and White 1981). Small chicks apparently reabsorb water and sodium (Na⁺) from the cloaca. Most Na⁺excreted through nasal gland, which makes up about 0.1% of body mass throughout growth, secreting fluid with Na⁺concentration about 500 mEq/l (Hughes 1968). Rate of turnover of body water in adults feeding small chicks 1.30/d ± 0.15 SD (n = 14; Galbraith et al. 1999). No in-formation on water or Na⁺economy of freshwater breeders.

No formal studies of pellet-casting, but frequently or habitually ejects loose boluses of fish bones, scales, otoliths or spines, or exoskeletons of crustaceans or insects (Hughes 1968, Vermeer 1973, Massias and Becker 1990). Plastic particles ejected in pellets (Hays and Cormons 1973).

Incubating adults fly off nest to defecate 5–50 m away; feces deposited indiscriminately in water or territories of other birds (Nisbet 1983b). Chicks <3 d old walk off nest, defecate 10–30 cm away, and return to nest (Cullen 1956). Adults foraging in their own feeding territories habitually fly over land to defecate (Nisbet 1983b). Adults attacking predator or human intruder habitually defecate at lowest point of dive, frequently hitting intruder with feces (Nisbet 1983b). Defecation rate of hand-raised chicks about 17 mg dry mass (g body mass·d)^-1(Ricklefs and White 1981). Energy density of feces 10–11 kJ/g dry mass; feces accounted for about 8% of energy intake when fed on high-quality food, ranging up to 21% when fed on shrimps (Massias and Becker 1990).