sign in sign up
<<
Home , Foraging, Galiteuthis, Seabird

2636 SEABIRD FORAGING

Ballance L.T., D.G. Ainley, and G.L. Hunt, Jr. 2001. Seabird Foraging Ecology. Pages 2636-2644 in: J.H. Steele, S.A. Thorpe and K.K. Turekian (eds.) Encyclopedia of Sciences, vol. 5. Academic Press, London.

SEABIRD FORAGING ECOLOGY

L. T. Balance, NOAA-NMFS, La Jolla, CA, USA Introduction D. G. Ainley, H.T. Harvey & Associates, San Jose, CA, USA Though bound to the land for , most G. L. Hunt, Jr., University of California, Irvine, seabirds spend 90% of their life at where they CA, USA forage over hundreds to thousands of kilometers in a matter of days, or dive to depths from the surface ^ Copyright 2001 Academic Press to several hundred meters. Although many details of doi:10.1006/rwos.2001.0233 seabird reproductive have been successfully SEABIRD FORAGING ECOLOGY 2637 elucidated, much of their life at sea remains a mesoscales (100}1000km, e.g. associations with mystery owing to logistical constraints placed on warm- or cold-core rings within current systems). research at sea. Even so, we now know a consider- The question of why associate with different able amount about seabird foraging ecology in water types has not been adequately resolved. At terms of foraging , behavior, and strategy, as issue are questions of whether a seabird responds well as the ways in which seabirds associate with or directly to habitat features that differ with water partition prey resources. mass (and may affect, for instance, thermoregula- tion), or directly to prey, assumed to change with Foraging Habitat water mass or current system. Environmental Gradients Seabirds predictably associate with a wide spectrum of physical marine features. Most studies Physical gradients, including boundaries between implicitly assume that these features serve to in- currents, eddies, and water masses, in both the hori- crease prey or availability. In some zontal and vertical plane, are often sites of elevated cases, a physical feature is found to correlate dir- seabird abundance. Seabirds respond to the strength ectly with an increase in prey; in others, the of gradients more than the presence of them. In causal mechanisms are postulated. To date, the shelf , e.g. the eastern Bering Sea shelf general conclusion with respect to seabird distribu- and off the California , cross-shelf gradients tion as related to oceanographic features is that are stronger than along-shelf gradients, and sea- seabirds associate with large-scale currents and re- distribution and abundance shows a corresponding gimes that affect physiological temperature limits strong gradient across, as opposed to along the and/or the general level of prey abundance (through shelf. At larger scales the same pattern is evident, ), and with small-scale oceano- e.g. crossing as opposed to moving parallel with graphic features that affect prey dispersion and boundary currents. availability. Physical gradients can affect prey abundance and availability to seabirds in several ways. First, they Water Masses can affect nutrient levels and therefore primary pro- In practically every ocean, a strong relation between duction, as in eastern boundary currents. Second, sea bird distribution and water masses has been they can passively concentrate prey by carrying reported, mostly identiRed through temperature planktonic organisms through , downwell- and/or salinity proRles (Figure 1). These correlations ing, and convergence. Finally, they can maintain occur at macroscales (1000}3000km, e.g. associ- property gradients (fronts, see below) to which prey ations with currents or ocean regimes), as well as actively respond. In the open ocean, where currents

30 30 Pterodroma macroptera V IS Pterodroma incerta H HS 28 Pterodroma mollis 28 S ISS 26 ISS 26 24 24 c 22 22 LTSE 20 20 18 18

Temperature (°C) Temperature 16 16 bulwerii 1_10 _ 14 Pterodroma baraui 11 30 14 LTSW Bulweria fallax 31_100 _ 12 101 300 12

0 0 36.5 36.0 35.5 35.0 34.5 34.0 36.0 35.5 35.0 34.5 Salinity (‰) Salinity (‰)

Figure 1 Distribution of gadfly in the Indian Ocean corresponding to various regimes of surface temperature and salinity: (A) warm-water species, and (B) cool-water species. The relative size of symbols is proportional to the number of sightings. Symbols for water masses as follows: VHS, very high salinity; HS, high salinity; IS, intermediate salinity; ISS, intermediate salinity south; C, common water; LTSE, low temperature southeast; LTSW, low temperature southwest. (Redrawn from Pocklington R (1979) 51: 9}21.) 2638 SEABIRD FORAGING ECOLOGY and dynamic processes are less active, prey behavior 3000 should be the principal mechanism responsible for seabird aggregation. In these cases, locations of ag- 2500

gregations are unpredictable, and this has important ) 1 _ Sitting consequences for the necessary for 2000 seabirds to locate and exploit them. In contrast, in Flying continental shelf systems, currents impinge upon 1500 topographically Rxed features, such as reefs, creating physical gradients predictable in space and 1000 time to which seabirds can go directly. Thus, the Abundance (ind km Abundance Rrst and second mechanisms are the more important in 500 shelf systems, and aggregations are so predictable that seabirds learn where and when to be in order to eat. 0 0 4 8 12 16 20 Fronts _ Much effort has been devoted successfully to identi- 5 fying correlations between seabird abundance and _15 fronts, or those gradients exhibiting dramatic 8.8 _ 8.0 change in temperature, density, or current velocity 25 7.2 (Figure 2). Results indicate a considerable range of _35 variation in the strength of seabird responses to 6.4 Depth (m) fronts. This may be due to the fact that fronts _45 inSuence sea-bird distribution only on a small scale. _ The factors behind the range of response is of 55 interest in itself. _65 Nevertheless, fronts are important determinants 0 5 10 15 20 of prey capture. Two hypotheses have been pro- Distance (km) posed to account for this: (1) that frontal zones enhance primary production, which in turn in- Figure 2 The aggregation of foraging at fronts, in creases prey supply, e.g. boundaries of cold- or this case the area where an elevation in bottom topography warm-core rings in the Gulf Stream; and (2) that forces a transition between stratified and well-mixed water. The bottom panel shows isotherms. (Redrawn from Hunt GL et al., frontal zones serve to concentrate prey directly into Marine Ecology Progress Series 141: 1}11.) exploitable patches, e.g. current rips among the Aleutian Island passages. general has long been related to seabird abundance Topographic Features and species composition. Depth-related differences Topographic features serve to deSect currents, in primary explain large-scale patterns and can be sites of strong horizontal and vertical between shelf and oceanic waters. Within shelf sys- changes in current velocity, thus, concentrating prey tems themselves, several hypotheses explain changes through a variety of mechanisms (Figure 2). For in species composition and abundance with depth. example, are often sites of seabird ag- First, primary production may be diverted into one gregation, likely related to the fact that they are also of two food webs, benthic or pelagic, and this may sites of increased density and heightened migratory explain differences in organisms of upper trophic activity for organisms comprising the deep scatter- levels in inner versus outer shelf systems, e.g. the ing layer. A second example are topographic eastern Bering Sea. Alternatively, the fact that inter- features in relatively shallow water, including de- actions between Sow patterns in the upper water pressions in the tops of reefs and ridges across the column and bottom topography will be strong in slope of marine escarpments, which may physically inner shelf areas but decoupled in outer shelf areas, trap euphausiids as they attempt to migrate down- may result in differences in predictability of prey ward in the morning. A third example is the down- and consequently, differences in the species that stream, eddy effect of islands that occur in strong exploit them, e.g. most coastal shelf systems. Fi- current systems. nally, depending on diving ability and depth, certain Depth gradient itself is sometimes correlated with seabirds may be able to exploit bottom sub- increased abundance of seabirds, and water depth in strate, whereas others may not. SEABIRD FORAGING ECOLOGY 2639

1.0 1.0 Spring 1983 Spring 1983 0.8 Pack ice 0.8 Open water 0.6 0.6 0.4 0.4 0.2 0.2 * 0 * 0 PENA PETS PENC PRAN PTKG PENA PETS PENC PRAN PTKG FUAN PEAN PETC PEBL FUAN PEAN PETC PEBL

1.0 1.0 0.8 Winter 1988 0.8 Winter 1988 Pack ice Open water 0.6 0.6 * 0.4 0.4 0.2 0.2 0 0 PENA PETS PENC PRAN PTKG PENA PETS PENC PRAN PTKG FUAN PEAN PETC PEBL FUAN PEAN PETC PEBL

Figure 3 The correspondence of various seabird species either to pack ice or open water immediately offshore of the ice; Scotia- Weddell Confluence, Southern Ocean. Abbreviations: PENA, AdeH lie ; FUAN, Antarctic ; PETS, snow ; PEAN, Antarctic petrel; PENC, chinstrap penguin; PETC, pintado petrel; PRAN, Antarctic ; PEBL, ; PTKG, Kerguelan petrel. (Redrawn from Ainley et al. (1994) Journal of Ecology 63: 347}364.)

Sea Ice Foraging Behavior

A strong association of individual species and char- Most seabird species take prey within a half meter acteristic assemblages exists with sea ice features. of the sea surface. This they accomplish by captur- On one hand, certain species are obligate associates ing prey that either are airborne themselves (as a re- with sea ice; on the other, sea ice can limit access to sult of escaping subsurface predators, see below), or the , and in some cases, aggregation of are shallow enough that, to grasp prey, the bird dips seabirds near the ice margin is a simple response to its below the surface (dipping) or crashes into this barrier (Figure 3). the surface and extends its head, neck and upper Sea ice often enhances foraging opportunities. body downward (surface plunging, contact dipping). First, there is an increased abundance of meso- Other species feed on dead prey Soating at the pelagic organisms beneath the ice, believed to surface. Another group takes prey within about be a phototactic response of these organisms to 20 m of the surface either by Sying into the water to reduced light levels. Second, primary production is continue Sight-like movements below the sur- enhanced beneath the ice or at its edge, and in face (pursuit plunging) or by using the momentum turn leads to increased abundance of primary of an aerial dive (plunging). and secondary consumers. The abundance and Finally, a number of seabirds can dive using either degree of concentration of this sympagic fauna their or feet for propulsion. Dive depth is varies with ice age, ice structure, and depth to related to body size, which in turn relates to physio- the bottom. As a result, Arctic ice, often multi- logical capabilities of diving. The deepest dives re- year in nature, has a speciose sympagic fauna as corded by a bird (penguin) reach 535 m, but most compared to Antarctic ice, which is often annual. species are conRned to the upper 100 m. A highly The ice zone can be divided into at least three specialized group of species capture prey by stealing , a region of leads within the ice itself, the them from other . ice edge, and the zone seaward of the ice. Each zone is exploited by different seabird species, and the Foraging Strategies relative importance of zones appears to differ be- tween Arctic and Antarctic , with the sea- The issue of how seabirds locate their prey is ward zone being particularly important in Antarctic far from completely understood. Below is a sum- systems. mary of known strategies, most of which depend on 2640 SEABIRD FORAGING ECOLOGY sea-birds locating some feature which itself serves to Third, vulnerability of individual Rsh in a school reliably concentrate prey. may increase with the number of birds feeding in the Sock. Finally, Socks are highly visible signals of Physical Features the location of a prey patch, e.g. species keying on Physical features are important to foraging sea- frigate birds circling high over schools. birds, because they serve to concentrate prey in Certain species are disproportionately responsible space and time, and because they often occur under for these signals, simply through their highly visual S predictable circumstances (Figure 2). This issue was ight characteristics. Such species are termed ‘cata- addressed above. lysts’. There is strong evidence that seabirds follow these visual signals, in some cases distinguishing Subsurface Predators between searching and feeding catalyst species, and between those feeding on a single prey item and Subsurface predators commonly drive prey to the those feeding on a clumped patch. surface because the air}water interface acts as a boundary beyond which most prey cannot escape. Nocturnal Feeding Under these circumstances, seabirds can access these same prey from the air. A wide array of subsurface Many Rshes and invertebrates remain at depth dur- predators are important to seabirds: large predatory ing the day and migrate to the surface after dark. Rshes (e.g. tuna), particularly in relatively barren During crepuscular or dark periods, surface densit- waters of the tropics; marine mammals, both ies of prey can be 1000 times greater than during cetaceans and pinnipeds; and marine birds themsel- the day. This migration is more signiRcant in low ves. than high latitudes, and in oceanic than neritic Subsurface predators increase the prey available waters. Many studies indirectly infer nocturnal feed- to birds in at least three ways. First, they drive prey ing from the presence of vertically migrating species to the surface. Second, they injure or disorient prey, in seabird diets or from circadian activity patterns. which then drift to the surface and are accessible to Little direct evidence exists. surface foragers. Third, they leave scraps on which Among the indirect data on how seabirds might seabirds forage, particularly when the prey them- locate prey at night is a considerable body of in- selves are large. formation on olfaction. In particular, members of Feeding subsurface predator schools can be highly the avian order possess olfactory visible and the degree of association of birds with lobes disproportionately large compared to the total these prey patches is often great. One investigation brain size and compared to most vertebrates. Ex- found 79% of the variability in seabird density periments show a marked ability of these birds to could be explained by the number of gray whale differentiate among odors and, especially, to Rnd mud plumes; this visibility may have been respon- sources of odors that are trophically meaningful. sible for the higher correlation than typically re- ported for other studies attempting to relate seabird Maximization of Search Area and prey abundance (see below). Feeding opportunities in the open ocean are often patchily distributed requiring seabirds to travel over Feeding Flocks large areas in search for prey. This ability to search Seabirds in most of the world’s oceans exploit clum- large areas is tightly linked to adaptations for Sight ped prey by feeding in multispecies Socks. Studies in proRciency. Seabirds capable of wide-area search all latitudes report that seabirds in Socks, often in exhibit morphological adaptations of the wing and a very few disproportionately large aggregations, tail that enhance energy-efRcient Sight. They also account for the majority of all individuals seen feed- modify their Sight behavior to take advantage of ing. Although Socks can result from passive ag- wind as an energy source. , , , gregation at a shared , evidence indicates and alcids, the pre-eminent seabird that seabirds beneRt in some way from the presence divers (see above), sacriRce wide-area search capa- of other individuals. First, as noted above, some bilities for what is needed for diving: high density seabird species act as subsurface predators, making and small wings. Therefore, divers are limited to prey available to surface-feeding seabirds, e.g. alcids areas of high prey availability. driving prey to the surface for larids in coastal An investigation into the relationship between Alaska. Second, a small number of seabird species wind and foraging behavior has revealed taxon- are kleptoparasitic, obtaining their prey from other speciRc preferences in Sight direction dependent on seabirds, e.g. jaegers, and a few other species. wind direction, and in turn, to wing morphology SEABIRD FORAGING ECOLOGY 2641 and presumed prey distribution. Procellariiformes, nonbreeders, which comprise at least half the typical primarily oceanic foragers, preferentially Sy across sea-bird population). the wind, whereas and Charadrii- formes, the majority of which forage over shelf and slope waters, preferentially Sy into and across head- Resource Partitioning winds. Because prey on a global scale are more Food is considered to be an important resource patchy in oceanic than shelf and slope waters, regulating seabird populations. Accordingly, much S across-wind ight may allow Procellariiformes to research has focused on identifying differences in cover more area at lower energetic cost, whereas the way coexisting species exploit prey. At sea, the S headwind ight allows slower ground speeds, pos- fact that different oceanic regimes or currents sup- sibly increasing the probability of detecting prey and port different prey communities as well as different decreasing response time once a prey item is detec- sea-bird communities has been used to support the ted. Flying up- or across-wind among procellariids idea that the geographic range of seabird species is R also maximizes probabilities of nding prey using a response to the presence of speciRc prey. Contrast- olfaction. ing these patterns, however, several colony-based studies report high diet overlap between species, Associations With Prey leading to speculation that dietary differences may reSect differences in foraging habitat as opposed to R Positive or signi cant correlations have been identi- prey selection. Evidence from at-sea research indi- R ed between seabird and prey abundance, in the cates broad overlap in the species and/or sizes of Bering Sea, eastern North Atlantic, Barents Sea, and prey taken by coexisting seabird species, often des- in locations throughout the Antarctic, although pite species-speciRc feeding methods, body size or rarely at scales smaller than about 2}3 km. In some habitat segregation. studies, however, the correlation is weak to nonexistent, or a negative correlation is reported. Prey Selection From these results, several general principles have arisen. First, the strength of the correlation increases Under certain circumstances seabirds do make with the spatial scale at which measurements are choices as to what prey they will attempt to capture made. Second, correlations between planktonic- (Figure 4). Among breeding species of the western feeding seabirds and their prey are lower than cor- North Atlantic and , in cases where a prey relations between piscivorous seabirds and their stock, such as or sandlance, are being ex- prey. The reasons for this pattern may relate to ploited by a large array of species, birds key in on Rsh that provide the highest energy package, in this differences in patch characteristics dependent on R prey species, or to differences in search mode of case sh in reproductive condition. In the Bering various predators. Third, correlations are not as Sea, breeding auklets have been observed to ignore strong as expected and in many cases, a correlation smaller to take the most energy-dense is found only with repeated surveys. Many hypothe- copepod species available. Finally, in the Antarctic, ses have been proposed to explain the latter, includ- during winter with almost constant darkness or ing: (1) seabirds are unable consistently to locate near-darkness when the mesopelagic large prey patches; (2) prey are sufRciently abundant is near the surface most of the time, seabirds that seabirds do not need to locate largest prey have been documented to avoid smaller prey patches; (3) prey are actively avoiding seabirds; (4) (euphausiids) to take larger and more energy-rich prey patches are continuously moving so that a time prey (myctophids; Figure 5). However, although this lag exists between patch formation, discovery by the is evidence for active prey selection, in these cases seabird, and measurement by the researcher; (5) there is broad dietary overlap among seabird extremely large prey patches are disproportionately predators. important to seabirds so that they spend much of Prey/Predator Size their time searching for or in transit to and from such patches; and (6) our means to measure prey Body-size differences among coexisting sea bird spe- patches (usually hydroacoustically) is a mismatch to cies have been used to imply diet segregation by the biology and attributes of the predators. Finally, prey size (Figure 5). In general, discounting pen- different seabird species may respond on the basis of guins but realizing there are a number of exceptions, threshold levels of prey abundance, and these thre- the larger seabirds (i.e. those '1500 g) tend to take sholds vary seasonally as well as with reproductive Rsh and ; the smaller species tend to take status of the bird (breeders require more food than juvenile or larval Rsh and squid, along with zoo- 2642 SEABIRD FORAGING ECOLOGY

100 one assemblage associated with pack-ice covered 26 July 1985 waters and the other with ice-free waters (Figure 3). 2 _ 50 The species composition of these assemblages cha-

km nges little over time; assemblage distribution tracks

Least auklets 0 the distribution of ice features in the absence of 250 differences in prey communities between the two 3 304.0 _ 243.2243.2 304.0 200 habitat types, and there is little spatial overlap 150 between the two assemblages. Spatial segregation, 100 with respect to species or assemblages, can also 50 occur along environmental gradients with respect to Copepods m 0 KD6 KD5 KD4 KD3 KD2 KD1 KE2 KE3 physical, chemical, and biological features of a sea- 0 bird’s habitat, in both the horizontal and vertical 30.8

31.6 31.6 30.0 dimension. This has been well documented espe- 31.2 31.2 31.8 32.0 10 31.4 31.4 cially in shelf waters, where differences in foraging 32.2 32.4 31.6 32.2 33.0 habitat, particularly as determined by depth, lead to 32.4 32.0 20 32.8 differences in diet; it is also evident in the pelagic Depth (m) tropical waters along productivity gradients. 32.4 30 A recurrent theme is that seabird species sort out along prey density gradients, regardless of prey identity. Such segregation has been recorded even Figure 4 Aggregation of least auklets over concentrations of the copepods Eucalanus bungii and Neocalanus spp. (light within the same prey patch, with certain species bars), and Calanus marshallae (dark bars). All three are exploiting the center and others the periphery. The confirmed prey of least auklets. Birds virtually ignored huge idea that oligotrophic waters, having reduced prey concentrations of C. marshallae, which were much closer to availability, can only be exploited by highly aerial the breeding colony on King Island to preferentially feed on species with efRcient locomotion, whereas produc- larger and presumably more energy-rich Neocalanus spp. (Redrawn from Hunt GL Jr and Harrison NM (1990) Marine tive waters are necessary for diving species is one Ecology Progress Series 65: 141}150.) that occurs in a wide variety of studies conducted in tropical, temperate, and polar systems. planktonic invertebrates. It comes down to energetic Finally, segregation can occur with respect to cost-efRciency of foraging and the morphology of time, speciRcally with respect to those species that the seabird foraging apparatus: the bill, which picks feed at night versus during the day. A few seabird one prey item at a time (the lone exception, perhaps, species are adapted to feed only at night. being one or two species groups, e.g. prions, that may Rlter-feed). Some degree of dietary size segrega- and tion is apparent among the few studies that have The sea bird Socking community in the North Paci- investigated all species breeding at single sites, e.g. Rc comprises species having complementary forag- a tropical oceanic island. Other studies at sea report ing behaviors, thus, indicating a degree of little, if any, dietary separation, even though integration within the community. In particular, the a 1000-fold difference in seabird size can exist. The feeding behavior of catalyst and diving species could implication is that seabirds often forage opportunis- be interpreted as mutualistic, catalysts signaling the tically depending on the availability of prey in their location of a prey patch, and divers increasing or preferred habitat, and that differences in habitat are maintaining prey concentration. Certain authors more important than differences in prey selection in have speculated that this relationship could have facilitating predator coexistence. resulted from coevolution of behavior designed to Habitat Type/Time increase the mutualistic beneRt of the association. Kleptoparasitism was also proposed to stabilize these Species or assemblages often segregate according to feeding Socks by forcing alcids to forage at the edges habitat with little evidence of interactions among ofapreypatchwheretheyarelessvulnerableto seabirds that signiRcantly inSuence their pelagic dis- piracy, thus, maintaining patch density and ultimate- tributions. Instead, the implication is that species ly, increasing prey availability to all Sock members. respond to physical and biological characteristics of environments according to their individual needs Morphological or Physiological Factors and Sight or diving capabilities. Spatial segregation can occur with respect to simple habitat features. Differential resource use is sometimes ascribed to For example, the Antarctic avifauna is divided into species-speciRc morphological or physiological SEABIRD FORAGING ECOLOGY 2643

0 0.2 0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0

86 STWI 4 EUSU 83 PETS 2 ELAN(Cont'd) 86 TEAR 3 83 STWI 2 83 STWI 1 NOCO 86 PETS 2 86 PENE 3 GAGL/KOLO 88 PETS 2 88 PENE 3 86 PEBL 4 88 FUAN 2 PSGL 88 PEBL 5 88 PENA 3 KOLO 88 PETS 3 83 PETH 4 86 TEAN 3 86 PEAN 4 GOAN/PSGL 86 PRAN 4 88 PEAN 2 GOAN 83 TEAR 2 83 FUAN 4 86 STWI 2 83 FUAN 3 ELAN/ORRO/KOLO 86 PEAN 2 86 FETC 4 GAGL 83 PRAN 2 86 FUAN 4 83 PETC 2 86 FUAN 2 83 PETC 5 83 PETH 1 ELAN/GAGL 88 PETS 1 83 PEAN 3 ELAN/PASC 88 FUAN 1 83 PRAN 1 88 PEBL 2 83 STBB 4 ELAN/PRBO/SATH 88 PEBL 1 86 PTKG 4 ELAN/ANUR/NOCO 86 PRAN 2 83 PETS 3 ELAN/PASC 83 PRAN 4 83 PEAN 2 83 PEBL 4 83 STWI 3 ELAN 83 PRAN 3 88 PEAN 5 83 FUAN 2 86 PETS 3 83 PETC 3 88 PETS 5 83 PETC 4 88 PETS 4 83 PETC 1 88 PEAN 1 83 PEBL 1 83 STWI 4 83 FUAN 1 ELAN/ 88 FUAN 5

Figure 5 Diet overlap among Antarctic seabirds. Even though a 1000-fold difference in predator size existed, there was no appreciable separation of diet for many sea-bird species. The vertical stippled line indicates the level at which diet is considered to be similar on the basis of prey species overlap. Bird species to the left of bars, prey species to the right. Each bird species name is preceded by the year in which collection was made, and followed by a number that denotes habitat (1, open water; 2, sparse ice; 3, heavy ice). Bird species: PENE, emperor penguin; PENA, AdeH lie penguin; FUAN, Antarctic fulmar; PETS, ; PEAN, Antarctic petrel; PENC, chinstrap penguin; PETC, pintado petrel; PRAN ; PEBL, blue petrel; PTKG, ; STWI, Wilson's -petrel; STBB, black-bellied storm-petrel; TEAR, Arctic ; TEAN, Antarctic tern. Prey names: ANUR, Anuropis spp.; EUSU Euphausia superba; ELAN, Electrona ; GAGL, ; GOAN, Gonatus antarcticus; KOLO, Kondakovia longimana; NOCO, Notolepis coatsi; ORRO, Orchomene rossi; PASC, Pasiphaea scotia; PRBO, Proto- myctophum bolini; PSGL, Psychroteuthis glacialis; SATH, Salpa thompsoni. (Redrawn from Ainley DG et al. (1992) Marine Ecology Progress Series 90: 207}221.) factors affecting Sight or diving capabilities. Several examples exist. First, differ in their ability to feed successfully in dense Socks over predatory Interference competition apparently does occur be- Rshes as a function of a given species’ ability to tween seabirds at sea. It is referred to most often in hover for prolonged periods of time. Second, differ- the context of feeding Socks, taking the form of ential metabolic demands may be responsible for aggressive encounters, and collisions between feed- species-speciRc differences in the threshold prey ing birds. The proximate limiting resource identiRed density to which alcids respond. Finally, differential in many of these cases is access to prey, i.e. space Sight costs correlate with species-speciRc patterns in over the prey patch. In another situation, shear- resource use, e.g. along productivity gradients in the waters in the North PaciRc feed by pursuit plunging tropics, or the amount of foraging habitat that can in large groups, by which they disperse, decimate, or be exploited (near-shore vs. offshore). drive prey deeper into the water column thereby Ultimately, many of these morphological and reducing the availability of prey to surface-feeding physiological adaptations are driven by body species. This same mechanism has been proposed size. Body size inSuences depth of dive capabilities, for tropical , which by plunge diving may cost of transport, and basal metabolic rate. also drive prey beyond the reach of surface feeders. Additionally, body size can frequently be used Despite widespread discussion of trophic competi- to predict the outcome of interference competition tion, supporting data are sparse and some evidence (below). indicates it to be not important in structuring some 2644 SEABIRD MIGRATION sea-bird communities. For example, one study in the Further Reading Antarctic found no habitat expansion of the pack- Ainley DG, O’Connor EF and Boekelheide RJ (1984) The ice assemblage into adjacent open waters seasonally Marine Ecology of Birds in the Ross Sea, Antarctica. vacated by another community (Figure 3), a shift that American Ornithologists’ Union, Monograph 32. might be expected if competition affected community Washington, DC. structure and habitat selection. In that study, sufRcient Ainley DG and Boekelheide RJ (eds) (1990) Seabirds of epipelagic prey were available in the ice-free waters the : Ecology, Dynamics and Structure to be exploited successfully by sea birds (Figure 5). of an Upwelling-system Community. Stanford, CA: Stanford University Press. Competition with Ashmole NP (1971) Seabird ecology and the marine envi- ronment, In Farner DS, King JR and Parkes KC (eds) Many of the forage species sought by sea birds are R Avian Biology, vol. 1, pp. 223}286. New York: the same sought by industrial sheries. The result is Academic Press. S con ict, particularly in eastern boundary currents, Briggs KT, Tyler WB, Lewis DB and Carlson DR (1987) where clupeid Rshes are dominant and are of ideal Bird Communities at Sea off California: 1975}1983. size and shape to be consumed by sea birds. The Studies in Avian Biology, No. 11. Berkeley, CA: tracking of bird populations with Rsh stocks has Cooper Ornithological Society. been especially well documented in the Benguela Burger J, Olla BL and Winn WE (eds) (1980) Behavior of and Peru currents, where not only have Rsh stocks Marine ,vol.4:Birds. New York: Plenum Press. been heavily exploited but so have deposits Cooper J (ed.) (1981) Proceedings of the Symposium on accumulated by the sea birds. The bird populations Birds of Sea and Shore. African Seabird Group. Cape have responded closely to geographic, temporal Town, Republic of South Africa. Croxall JP (ed.) (1987) Seabirds: Feeding Ecology and and numerical variation in the Rsh stocks. Well R Role in Marine Ecosystems. London: Cambridge documented, also, have been sh stocks and avian University Press. predator populations in the North Sea. There, com- Furness RW and Greenwood JJD (1983) Birds as moni- mercial depletion of predatory Rsh beneRted sea- tors of environmental change. London and New York: bird populations by reducing competition for forage Chapman & Hall. Rsh; when Rsheries turned to the forage Rsh them- Furness RW and Monaghan P (1987) Seabird Ecology. selves, seabird populations declined. In some areas, London: Blackie. it has been proposed to use statistical models of Montevecchi WA and Gaston AJ (eds) (1991) Studies of predator populations as an indicator of Rsh-stock high latitude seabirds 1: Behavioural, Energetic and status independent of Rshery data, for instance, the Oceanographic Aspects of Seabird Feeding Ecology. Convention for the Conservation of Antarctic Living Canadian Wildlife Service, Occasional Papers 68. Ottawa, . Marine Resources. Much information is needed to Nettleship DN, Sanger GA and Springer PF (1982) Mar- calibrate seabird responses to prey populations before ine Birds: Their Feeding Ecology and Commercial seabirds can be used reliably to estimate prey stocks. Fisheries Relationships. Canadian Wildlife Service, Special Publication. Ottawa, Canada. See also Vermeer K, Briggs KT and Siegel Causey D (eds) (1992) Ecology and conservation of marine birds of the Benguela Current. Canary and Currents. temperate North PaciRc. Canadian Wildlife Service, Seabird Migration. Seabird Responses to Climate Special Publication. Ottawa, Canada. Change. Seabirds and Fisheries Interactions. Sea Whittow GC and Rahn H (eds) (1984) Seabird Energetics. Ice: Variations in Extent and Thickness. New York: Plenum Press.

SEABIRD MIGRATION

L. B. Spear, H. T. Harvey & Associates, has been studied since ancient times, particularly San Jose, CA, USA among nonmarine species. Studies of migration and

Copyright ^ 2001 Academic Press navigation of nonmarine species have become quite sophisticated, examining in detail subjects including doi:10.1006/rwos.2001.0238 orientation and navigation, and physiological and morphological adaptations. In contrast, studies of Introduction migration among marine birds have been fewer and is one of the most fascinating phe- more simplistic. Indeed, until recently, much of the nomena in our living environment, and accordingly information on migration of seabirds had come