MIgRATION AND WINTERINg AREAS Of gLAUCOUS

The Condor 113(2):340–351  The Cooper Ornithological Society 2011

Migration and Wintering Areas of Glaucous-Winged Gulls From South-Central Alaska Scott A. H atch1, Verena A. Gill, 2

and

Daniel M. Mulcahy

U.S. Geological Survey, Alaska Science Center, 4210 University Drive, Anchorage, AK 99508 Abstract. We used satellite telemetry to investigate the migration patterns and wintering areas of Glaucouswinged Gulls (Larus glaucescens) from Middleton Island, Alaska, where this species’ population increased tenfold from the 1970s to the 1990s. Fall migration spanned 11 weeks, including numerous stopovers en route, apparently for feeding. Spring migration from wintering sites to Middleton Island was shorter (4 weeks) and more direct. One juvenile spent several months in southern Prince William Sound. An adult spent several months near Craig, southeast Alaska, while three others overwintered in southern British Columbia. For all four wintering adults use of refuse-disposal sites was evident or strongly suggested. Commensalism with humans may have contributed to the increase on Middleton, but a strong case can also be made for a competing explanation—regional recruitment of gulls to high-quality nesting habitat in Alaska created after the earthquake of 1964. An analysis of band returns reveals broad overlap in the wintering grounds of gulls from different Alaska colonies and of gulls banded on the west coast from British Columbia to California. The seasonal movement of many gulls from Alaska is decidedly migratory, whereas gulls from British Columbia, Washington, and Oregon disperse locally in winter. Key words: Alaska, Glaucous-winged Gull, Larus glaucescens, Middleton Island, migration, migratory connectivity, satellite telemetry, wintering areas.

Migración y Áreas de Invernada de Larus glaucescens del Centro y Sur de Alaska Resumen. Usamos telemetría satelital para investigar los patrones de migración y las áreas de invernada de Larus glaucescens en la isla Middleton, Alaska, donde la población de esta especie aumentó 10 veces desde la década de 1970 hasta la de 1990. La migración de otoño duró 11 semanas, incluyendo numerosas paradas en ruta, apa rentemente para alimentarse. La migración de primavera desde las áreas de invernada hacia la isla Middleton fue más corta (cuatro semanas) y más directa. Un individuo joven pasó varios meses en el sur de Prince William Sound. Un adulto pasó varios meses cerca de Craig, al sudeste de Alaska, mientras que otros tres extendieron el período de invernada en el sur de British Columbia. Para los cuatro adultos invernantes el uso de los sitios de disposición de residuos fue evidente o fuertemente sugerido. El comensalismo con los humanos puede haber contribuido al aumento en Middleton, pero también se puede argumentar una explicación alternativa contundente—el reclutamiento regional de las gaviotas como respuesta al hábitat de anidación de alta calidad en Alaska creado luego del terremoto de 1964. Un análisis de recuperación de anillos revela una gran superposición en los sitios de invernada de las gaviotas de las distintas colonias de Alaska y de las gaviotas anilladas en la costa este desde British Columbia hasta California. El movimiento estacional de muchas gaviotas desde Alaska es indudablemente migratorio, mientras que las gaviotas de la Columbia Británica, Washington y Oregón se dispersan localmente en invierno.

INTRODUCTION Seabirds are usually monitored at breeding colonies in summer, but population trends are strongly influenced by factors operating in winter. Mortality is often concentrated in the winter months, at locations far removed from the breeding grounds (e.g., Hatch 1987, Aebischer and Coulson 1990), and small changes in annual survival produce large changes in the net replacement rates of seabirds and other long-lived species (Sæther and Bakke 2000). This is important from a management perspective, because when population change is

detected at a colony, a logical follow-up question is, or should be, “where does this population go in winter, and what there could be accounting for the trend?” Clearly, it is not enough to know, in general terms, the range over which a species winters, but rather to know where the individuals from a particular colony go in the nonbreeding season. The linkage of summer and winter locations and events in the annual cycles of individuals and populations has recently come under the rubric “migratory connectivity” (Webster et al. 2002, Webster and Marra 2005, Norris et al. 2006, Marra et al. 2006). While it has long been recognized that conditions

Manuscript received 19 November 2009; accepted 13 August 2010. 1 E-mail: [email protected] 2 Current address: U.S. Fish and Wildlife Service, Marine Mammal Management, 1011 East Tudor Road, Anchorage, AK 99503. The Condor, Vol. 113, Number 2, pages 340–351. ISSN 0010-5422, electronic ISSN 1938-5422.  2011 by The Cooper Ornithological Society. All rights reserved. Please direct all requests for permission to photocopy or reproduce article content through the University of California Press’s Rights and Permissions website, http://www.ucpressjournals.com/ reprintInfo.asp. DOI: 10.1525/cond.2011.090224

340

10_MS090224.indd 340

5/13/11 3:01:33 PM

Gull Migration and Wintering 341

Figure 1. Map of western North America showing locations mentioned in the text.

during one phase of the annual cycle (breeding, wintering, migration) can have carryover effects during subsequent phases (Salomonsen 1955, Fretwell 1972, Boulet and Norris 2006a), only now is it becoming possible to gather efficiently the data needed to explore such connections in many species of birds. New technologies, including intrinsic (genetic and isotopic) markers and telemetry (bird-borne satellite transmitters, GPS, solar geolocating devices), are rapidly replacing leg banding as preferred methods for gathering crucial information on seasonal movements (Hobson 2005, Croxall et al. 2005, Boulet and Norris 2006b, Hatch et al. 2010). Migratory connectivity has been formally defined as “the extent to which individuals from the same breeding area migrate to the same nonbreeding area and vice versa” (Webster et al. 2002). Strong connectivity describes a situation in which most individuals from a discrete breeding population move seasonally to a shared nonbreeding location, whereas weak connectivity arises when sympatric breeders disperse during migration to multiple wintering sites of approximately equal importance to the population (Webster et al. 2002, Boulet and Norris 2006a). In our view, the power of this formalism will be realized mainly via the comparative method—by using populations with known, contrasting histories (i.e., popula-

10_MS090224.indd 341

tion trends) and information on seasonal space and habitat use (shared versus allopatric breeding and wintering areas) to infer the stage or stages of the annual cycle most responsible for trends at the population level (e.g., Hatch et al. 2000). In Europe and North America, many populations of gulls (Larus spp.) have increased markedly in recent decades, a pattern often attributed to commensal relations with humans, because gulls readily adapt to artificial food sources such as garbage dumps and discharges from vessels and fish-processing plants (Harris 1970, Drury 1973, Camphuysen et al. 1995, Belant 1997). Gulls are hazardous to aircraft near airports (Sodhi 2002), are potential vectors of human pathogens (Spaans and Bloekpol 1991), and may be viewed as a general nuisance in urban settings (Belant 1997). As predators on the eggs and young of many species, gulls at artificially high population levels affect other seabirds negatively (Thomas 1972). The Glaucous-winged Gull (Larus glaucescens) is the most numerous species of Larus in the northeastern Pacific, with an estimated 200 000 pairs breeding from the Kamchatka coast, Aleutian Islands, and southern Bering Sea south to coastal British Columbia (BC) and Washinton state (Vermeer and Irons 1991, Vermeer et al. 1993). The winter distribution overlaps the breeding range broadly but extends also to central

5/13/11 3:01:34 PM

342 Scott A. Hatch

et al .

and southern California; small numbers reach Baja California and the Sea of Cortez in Mexico (Hayward and Verbeek 2008). Numbers of the Glaucous-winged Gull in the Strait of Georgia, BC, doubled from 1960 to 1986 (Vermeer and Devito 1989, Vermeer 1992), but later censuses found declines in both the Strait of Georgia and on the west coast of Vancouver Island (Vermeer et al. 1992, Sullivan et al. 2002). Recent declines are also documented for several sites in the Aleutian Islands (Byrd et al. 2005). A large colony of Glaucous-winged Gulls (>10 000 individuals counted in 2009) nests on Middleton Island, Alaska (Fig. 1), but that is a recent development. None bred on Middleton in 1956 (Rausch 1958), and surveys in the mid-1970s found small colonies totaling fewer than 1000 birds (Hatch et al. 1979). The population expanded rapidly in the 1980s, peaking at more than 12 000 individuals in 1993. Subsequently, numbers have fluctuated widely between about 6000 and 10 000 (S.A. Hatch, unpubl. data). Middleton now supports one of the largest breeding aggregations of Glaucous-winged Gulls anywhere (Drent and Guiguet 1961, Sowls et al. 1978, Speich and Wahl 1989), and the rate of increase sustained between 1984 and 1993 (29% per year) is exceptional in gulls (cf. Kadlec and Drury 1968, Harris 1970, Greenlaw and Sheehan 2003). In view of the dramatic increase of gulls on Middleton, we wanted to know the birds’ specific wintering area(s), a possible key to understanding population growth. Traditionally, linkages between the summer and winter ranges of seabirds are established by a laborious process of banding at colonies and the slow accumulation of recoveries away from the banding sites (e.g., Woodbury and Knight 1951, Butler et al. 1980). However, with the advent of satellite transmitters small enough to be carried by large and medium-sized seabirds, the same information can now be obtained faster and more efficiently (Petersen et al. 1995, Hatch et al. 2000, Mosbech et al. 2006, Hatch et al. 2010). For species capable of carrying a device weighing 20–60 g, the main obstacle at present is finding a suitable attachment—one that will last through all or most of an annual cycle, without affecting the bird’s normal activity or jeopardizing its survival. In 2000 and 2001, we tagged adult and juvenile gulls on Middleton and tracked their post-breeding movements away from the island. We find only one previous application of satellite telemetry to Larus gulls in the open literature, to the Lesser Black-backed Gull (Larus fuscus) in Europe (Pütz et al. 2007, 2008). To complement our telemetry study and provide a context of seasonal movements by Glaucous-winged Gulls from other colonies, we examined band-recovery data for this species obtained from the North American Bird Banding Laboratory. METHODS Middleton Island (59.4° N, 146.3° W) is located near the shelf edge in the north-central Gulf of Alaska, approximately 115 km

10_MS090224.indd 342

from the Alaska mainland. The nearest stationary source of anthropogenic food is Cordova, Alaska, about 125 km away. We believe artificial food sources are beyond the routine foraging range of incubating and chick-rearing gulls from Middleton, although that assumption is as yet unconfirmed by telemetry or other means. We initiated this study by tagging two adult gulls on Middleton in July 2000. The following year, in August, we tagged two more adults and two late-stage, pre-flight chicks. We captured adults not within nesting habitat but on beaches with a wildlife net gun. Thus, the breeding status (actively breeding, nonbreeding, or failed) of the adults at the time of their capture is unknown. Juveniles were captured by hand within colonies. The telemetry devices we used were implantable platform transmitter terminals (PTTs) supplied by Microwave Telemetry, Inc. (Columbia, MD). Battery power (and thus overall size and weight of the package) was of two types—a singlebattery unit (40 g) or a double-battery model (60 g) designed for increased transmitter life. PTTs were surgically implanted in the coelom according to a modified version (Mulcahy and Esler 1999) of a protocol developed by Korschgen et al. (1984, 1996). Briefly, the transmitter was inserted into the abdominal cavity through a mid-ventral incision, with the antenna exiting through the skin at the synsacrum (i.e., upward, to one side of the tail). Surgeries (performed by DMM) lasted about 20 min from induction of general anesthesia (2–5% vaporized isoflurane) to reclosure of the skin. Birds were monitored for normal post-operative recovery for 1–2 hr, then released in the vicinity of their colonies. Single-battery transmitters used in 2000 had a duty cycle of 6 hr on/12 hr off for the first 20 cycles (15 days), then switched to 6 hr on/120 hr (5 days) off for the duration of the deployment. In 2001, both PTT types had a duty cycle of 6 hr on/12 hr off for 6 cycles (4.5 days) before switching to 6 hr on/120 hr off (double battery) or 6 hr on/192 hr (8 days) off (single battery) for the rest of the study. Thus, the majority of tracking data collected are clusters of positions obtained at roughly weekly intervals (5–8 days) or less frequently in cases where no usable location was obtained during a weekly “on” period. In presenting data for migrating birds, we highlight “stopover” sites, defined as places from which a PTT transmitted in more than a single duty cycle, indicating local residence lasting 5–8 d or longer. PTTs included sensors for temperature and remaining battery voltage. If the indicated body temperature of a bird turned abruptly cold, we concluded that mortality had occurred. We denote the outcome as “indeterminate” if the last signals received from a PTT indicated both a warm body and residual battery life. In the remaining cases, temperature data remained normal as reported voltages approached the lower limit for PTT function, indicating a depleted battery. We obtained raw location data from the Argos system, which assigns each reported position to one of seven classes of predicted accuracy (Argos 1989). We culled implausible

5/13/11 3:01:34 PM


Table 1. Deployments of six PTTs on Glaucous-winged Gulls captured on Middleton Island, Alaska in 2000 and 2001. Date Last location

Raw

Filtered

Final status

17 Jul 2000

11 Jan 2001

178

357

335

3.1

17 Jul 2000

15 May 2001

302

451

421

40

4.2

8 Aug 2001

20 Jun 2002

316

253

247

1160

60

5.2

8 Aug 2001

22 Sep 2002

410

368

328

960 680

40 60

4.2 8.8

8 Aug 2001 9 Aug 2001

13 Aug 2001b 14 Nov 2001

5 97

93 169

89c 166

Battery depleted Battery depleted Battery depleted Battery depleted Mortality Indeterminate

Age

Device (g)

Load (%) a

Deployed

Ad1

Adult

1310

40

3.1

Ad2

Adult

1285

40

Ad3

Adult

960

Ad4

Adult

Juv1 Juv2

Fledgling Fledgling

Bird ID

Locations Days contact

Body mass (g)

a

Implanted PTT as a percentage of body mass. Additional locations received intermittently from stationary PTT still active through November 2002. c Includes 45 locations obtained before mortality and 44 locations after mortality. b

data by the Douglas Argos-Filter Algorithm, a program for PC-SAS that flags improbable locations on the basis of userdefined distance and velocity thresholds (Douglas 2006). We applied the distance-angle-rate filter with r (maximum sustainable rate of movement) = 80 km hr−1 and d (maximum redundant distance) = 15 km (Douglas 2006). The filter also retained locations in quality classes 1, 2, or 3 (most accurate) irrespective of other criteria. We plotted location data in ArcGIS 9.2, extended with Hawth’s Analysis Tools, version 3.27. For comparison with gull movements during winter months (November–February), we added a feature class consisting of domestic waste-disposal sites in BC. The latter we extracted from the freely available GIS data product CanVec (edition 1.0.2, 2008), distributed by the Centre for Topographic Information, Natural Resources Canada (http://geogratis.ca/). We obtained historical banding and band-encounter data for Glaucous-winged Gulls by request to the Bird Banding Laboratory, Patuxent, Maryland. These data comprised bands applied to chicks or adults, mostly at colonies, that were subsequently encountered elsewhere through resighting or recovery of dead birds—360 records for gulls banded in Alaska since 1964 and 19 828 records for gulls banded outside of Alaska since 1922. To better portray wintering areas, as opposed to resightings of gulls at colonies or stops during migration, we selected for analysis only band encounters in the months November–February. RESULTS PTT performance

Implanted PTTs were 3.1 to 8.8% of a gull’s body mass (Table 1), the high figure representing a 60-gram unit placed in an immature gull that probably gained considerable body mass after

10_MS090224.indd 343

surgery, prior to fledging. We obtained few data from one juvenile that died about 5 days after it was released. Otherwise, we tracked the gulls from 97 to 410 days. Argos locations obtained during the study totaled 1691. The filtering program rejected 105 locations (6.2%), leaving 1586 for analysis of gull distribution and movements. Adult gulls

Local movements in summer. Both adults tagged in July 2000 (Ad1 and Ad2) made repeated trips between Middleton Island and the Wooded Islands, about 80 km to the northwest, before vacating Middleton for the season (Fig. 2a). Ad1 left the island on or about 17 August on its southward migration, whereas Ad2 remained an additional month in the Prince William Sound region, visiting sites near Hinchinbrook Island, Egg Island off the Copper River mouth, and Kayak Island, before finally heading south on or about 14 September. Adults tagged in 2001 either remained in the immediate vicinity of Middleton before departure on fall migration (Ad3) or moved north to spend time at sites in the Prince William Sound region (Egg Island, and especially Valdez) before heading south around 18 September (Ad4; Fig. 2a). Three of four adults tagged were tracked back to the breeding area in the following spring (Fig. 2b). Between arrival in early-mid April and final signal reception (mid May and mid June, respectively), Ad2 and Ad3 roamed widely from Middleton, visiting Kayak Island, the Copper River mouth and associated barrier islands, or the outer islands of Prince William Sound. In neither case, nor in the case of Ad4—which was tracked through the whole of its second summer upon returning to Middleton—was it possible to ascertain the bird’s breeding status in the second summer from the telemetry data. Ad4 spent considerable time in the immediate vicinity of Middleton in

5/13/11 3:01:34 PM

344 Scott A. Hatch

et al .

Figure 2. Local movements of Glaucous-winged Gulls tagged on Middleton Island in 2000 (Ad1, Ad2) and 2001 (Ad3, Ad4): (a) before departure on southward migration in fall, and (b) upon completion of northward migration in 2001 (Ad2) and 2002 (Ad3, Ad4).

10_MS090224.indd 344

5/13/11 3:01:35 PM


Figure 3. (a) Fall migration and a protracted stopover near Craig, Alaska by gull Ad1, last heard from on 11 January 2001. (b) Encircled locations correspond to an open landfill that existed at Craig in 2001.

2002, suggesting possible breeding, but also made excursions to the Copper River delta and outer Prince William Sound in June and July. Ad4 vacated Middleton about 6 August and proceeded to Valdez, where it spent several weeks before beginning southward migration about 17 September—an itinerary nearly identical to the one it followed in late summer 2000. Fall migration. Although Ad1 was tracked only through mid January, it clearly had a winter strategy different from that of the three adults described below. Ad1 spent most of its time between departure from Midddleton and mid January in the vicinity of Craig, Alaska, a residency interrupted only by a single known excursion to the shelf edge on 14 December (Fig. 3). The bird was last heard from in Hecate Strait, having possibly departed its wintering site at Craig for winter quarters farther south. A rapid drop in battery voltage indicated this deployment was likely curtailed by early battery failure (Table 1). During its fall migration, Ad2 stopped in southeast Alaska (24–29 September), on the outer coast of the Queen Charlotte Islands, BC (4–29 October), and at two sites on the outer coast of Vancouver Island, BC (8–23 November), before completing its southward movement near Port Angeles, Washington, around 28 November (Fig. 4a). Ad3 also made extended stops in southeast Alaska during September, reaching the southern terminus of its fall migration in southern Vancouver Island around 3 November (Fig. 4b). Ad4 had one extended stopover (4–14 October) on northern Chicagof Island, southeast Alaska, and another on the east coast of Vancouver Island (30 October–20 November) before completing its fall migration near Delta, BC, around 1 December (Fig. 4c).

10_MS090224.indd 345

Fall migration of the three preceding individuals (Ad2, Ad3, Ad4) lasted some two and a half months (77 days), beginning around 5 September, on average, and ending around 20 November (Table 2). Winter locations and habitat use. Two of three gulls (Ad3, Ad4) tracked to southern BC spent the winter months (November–mid March) in the Strait of Georgia, including the urbanized areas of Vancouver, Richmond, and Delta and coastal sites on the southern east side of Vancouver Island (Fig. 4d). It appeared that favored habitats included mudflats at the Fraser River mouth, Boundary Bay on the border between BC and Washington, and mixed urban and agricultural lands lying between those natural features. The principal landfills serving the Vancouver metropolitan area lie approximately at the center of that region. Excursions throughout the winter may have included visits to other landfills at Nanaimo and Parksville on Vancouver Island, Hornby Island, and the Vancouver suburbs of Burnaby and Surrey on the mainland. Ad4, having ended its southerly movement at the base of Dungeness Spit in the Strait of Juan de Fuca, spent most of the ensuing winter on the northwest coast of Vancouver Island, especially the sparsely settled region of Esperanza and Nuchalitz inlets. Ad4 also made mid-winter trips to the edge of the continental shelf west of Vancouver Island (Fig. 4d). Spring migration. On its northward migration, Ad2 made one extended stopover (2–18 April) between the Alsek River and Yakutat Bay, Alaska, whereas Ad3 was not detectably delayed in transiting from southern BC to Middleton Island between 22 March and 7 April (Fig. 4a, b). Ad4 moved slowly

5/13/11 3:01:36 PM

346 Scott A. Hatch

et al .

Figure 4. Migration routes, timing, and stopping places of adult gulls Ad2 (a), Ad3 (b), and Ad4 (c) tagged on Middleton Island. Plain dots and solid connecting lines depict southward migration in fall; circled dots and dashed lines depict northward migration in the following spring. Dates identify places where gulls stopped for 5–8 days or longer, as indicated by PTT positions spanning more than a single duty cycle. In (d), the wintering areas and local movements in Washington and southern British Columbia are shown for Ad2 (solid triangles), Ad3 (circled dots), and Ad4 (plain dots). Unfilled squares are sites of solid-waste disposal.

10_MS090224.indd 346

5/13/11 3:01:37 PM


Table 2. Dates of migration of Glaucous-winged Gulls tagged on Middleton Island, Alaska.a Fall migration Bird ID Ad2 Ad3 Ad4 Mean a

Year tagged 2000 2001 2001

Spring migration

Departure

Arrival

Days en route

Departure

Arrival

Days en route

14 Sep 13 Aug 18 Sep 5 Sep

28 Nov 2 Nov 1 Dec 20 Nov

75 81 74 77

18 Mar 22 Mar 22 Mar 21 Mar

23 Apr 7 Apr 22 Apr 17 Apr

36 16 31 28

Approximate departure and arrival dates based on satellite signals received at 5- or 8-day intervals.

from Icy Bay to the Copper River delta, Alaska, on its trip north in 2002 (12–17 April), but otherwise it moved without notable delays between the Strait of Georgia and Middleton Island (Fig. 4c). Departure dates of all three gulls on their spring migration in 2001 and 2002 were similar, averaging 21 March (range 18–22 March). The duration of spring migration averaged 28 days, range 16–36 days (Table 2). Juvenile gulls

One of two juvenile gulls tagged before fledging in August 2001 died just 5 days after release, so we obtained no meaningful data on its movements. However, locations obtained both before

and after the chick’s death reflect the accuracy of Argos locations expected in the rest of the study (Fig. 5). Those removed by the data-filtering program excluded, plotted locations of the stationary PTT had a mean distance (±SD) from their centroid of 2.3 ± 2.9 km (range 0.07–12.0 km, n = 89). A second fledgling (Juv2) departed Middleton around 19 August and spent the following 3 months in southern Prince William Sound. It ranged over approximately 100 km from east to west, from wilderness sites on the western mainland to Hartney Bay near Cordova (Fig. 5). Stockdale Harbor at the north end of Montague Island was its principal place of residence during the period. When the last signals were received from Juv2 on 14 November, the battery’s voltage remained

Figure 5. Positions of two juvenile gulls tagged on Middleton Island in August 2001. Circled dots depict location error for a relatively immobile (8–13 August 2001) or stationary PTT (Juv1) on the north end of Middleton. Plain dots and solid connecting lines show the movements of Juv2, 9 August–14 November 2001.

10_MS090224.indd 347

5/13/11 3:01:37 PM

348 Scott A. Hatch

et al .

Figure 6. (a) Locations of recoveries from November to February (solid dots) of 103 Glaucous-winged Gulls banded in Alaska (filled squares) between 1964 and 2002. Open circles are winter recoveries of four individuals banded on Middleton Island. (b) Encounters and recoveries in winter (November–February) of Glaucous-winged Gulls (n = 5272) banded outside of Alaska, 1922–2007. Unfilled circles are sites of banding in British Columbia, Washington, Oregon, and California.

high and the bird’s body temperature was normal. It is unclear, therefore, whether signal loss resulted from death of the bird or failure of the PTT. Banded gulls

From 1970 to 2006, 103 Glaucous-winged Gulls banded at 13 locations in the Gulf of Alaska were encountered later, dead or alive, at presumed wintering sites in the months November–February. Wintering gulls ranged from southeastern Alaska to central California (Monterey Bay), with broad latitudinal overlap among gulls from different colonies of origin (Fig. 6a). Four individuals in the set were banded on Middleton Island. Three of those were encountered in Washington (Puget Sound, Strait of Juan de Fuca, and Columbia River near Kennewick); the fourth was seen at Half Moon Bay, California (Fig. 6a). For comparison with gulls of Alaska origin, we plotted band-encounter locations for birds marked in the southern part of the breeding range or in the winter range. From 1922 to 2007, 5272 Glaucous-winged Gulls banded at 52 locations in BC, Washington, Oregon, and California were subsequently encountered from November to February. Apart from

10_MS090224.indd 348

a few stragglers reported from inland sites, these birds wintered mainly in the lower reaches of the Columbia River and coastally from southern BC to southern California (Fig. 6b). Clearly, the winter range of gulls banded in the southern part of the breeding range overlaps the winter range of birds from Alaska broadly but excludes the region from southeast Alaska to central BC and extends to southernmost California. DISCUSSION Satellite telemetry was successful in revealing broad patterns of migration and winter habitat use of gulls from Middleton Island. Trials with juvenile gulls met with limited success, however, and additional work with that group will be needed to determine whether first-year birds behave differently from adults. Verbeek (1986) estimated that >50% of young Glaucouswinged Gulls die within 4 months after fledging. The behavior of two adults soon after release—repeated flights between Middleton and the Wooded Islands—was somewhat puzzling and leads us to question whether those individuals belonged to the breeding population on Middleton or were transients. A colony of a few hundred gulls, at most,

5/13/11 3:01:38 PM


exists at the Wooded Islands (Sowls et al. 1978; P. M. Meyers, pers. comm.), so it is possible the birds we tagged were breeding on the Wooded Islands but happened to be loitering on a beach at Middleton at the time of capture. The disparity of numbers (>10 000 gulls at Middleton versus ~200 at the Wooded Islands) argues for a Middleton origin, but unfortunately the bird tracked back to Alaska in the spring of 2001 made no definitive visit to either Middleton or the Wooded Islands before its signal expired in April. Whether or not the two adults tagged in 2000 belonged to the Middleton population, we suggest their late-summer movements reflect a more generalized pattern of gulls visiting distant colonies of their species at the end of one breeding season and beginning of the next. This pattern involved multiple colonies including Port Etches (Hinchinbrook Island), Egg Island, west Kayak Island, Yakutat, and the Alsek River mouth (Sowls et al. 1978). We surmise that the reason for such visits was social attraction—a tendency for gulls to investigate neighboring colonies in their region opportunistically, not only to range in search of food. Though our sample is small, our extended telemetry of three adults permits a broad characterization of migration patterns in fall and spring. Southward movement was clearly more protracted, spanning about 11 weeks on average, than the 4 weeks of northward migration. Spring departure from the wintering grounds was relatively synchronous, and a more or less direct return to Middleton resulted in a relatively synchronous arrival at the breeding site. The attraction at various stopping points during fall migration can only be guessed, but the typical situation—the head of a bay near a freshwater outflow—suggests the possibility of the birds feeding on spawning salmonids. The analysis of encounters of banded Glaucous-winged Gulls reveals broad overlap in winter habitat use between birds from Alaska and those banded elsewhere. Whereas birds from south-central Alaska clearly migrated in spring and fall, the post-breeding movements of gulls banded in BC, Washington, Oregon, and California are better characterized as dispersive. Winter recoveries of four individuals banded on Middleton Island agree with our telemetry results, apart from one bird seen on the central California coast. Clearly, larger numbers of band returns and/or more telemetry data may enlarge the perceived core winter range of birds breeding on Middleton. An individual from Middleton (Ad1) that spent several months (September–December 2000) in the vicinity of Craig, Alaska, occurred repeatedly at three discrete locations, one of which coincided with an open dump that existed in the community at that time (H. Fleury, Craig public works manager, pers. comm.). There can be little doubt the gull was foraging regularly at the dump. Similarly, the home ranges of two other adults in southern BC included several municipal dumps as likely food sources, although the precision of our location data cannot show a direct association unequivocally.

10_MS090224.indd 349

The evidence from satellite telemetry and banding is that migratory connectivity in the Glaucous-winged Gull is weak— birds from different breeding grounds (southeast Alaska to Oregon and Gulf of Alaska) are mixed seasonally on shared wintering grounds. Contrast the situation recently reported for the Northern Fulmar (Fulmarus glacialis), in which birds from major North Pacific colonies winter in colony-specific areas (Hatch et al. 2010). Therefore, changes on the wintering grounds (commensal relations with humans, future climate change, for example) could affect the species’ population over a wide portion of its breeding range. On the other hand, the wide variability in gulls’ migratory behavior (distinct, long-distance migration versus shorter-range dispersal in winter) implies substantial genetic variation open to shaping by natural selection as changing conditions might demand (Webster et al. 2002). It is reasonable to conclude that a commensal relationship of gulls with human activity in winter has benefited the Middleton Island population and may be a factor in the dramatic increase in numbers there. The overlap of wintering sites among gulls from various colonies in south-central Alaska raises the question whether neighboring colonies have undergone similar changes. A colony on Egg Island, 100 km to the north, rivals or exceeds the Middleton colony in size and is thought to have increased since the 1990s, possibly because of the birds feeding in summer at a fish-processing outfall in Cordova (P. M. Meyers, pers. comm.). Unfortunately, the monitoring of Glaucous-winged Gull colonies in Alaska is inadequate to reveal what has happened at most of them over the last 20–30 years. Middleton Island is a rare exception. An alternative explanation for the increase on Middleton has more to do with changes in the breeding habitat than in winter ecology. Middleton is unique in having been drastically altered by the 1964 Alaska earthquake. It was uplifted approximately 3.5 m (Plafker 1970), and, following gradual successional changes on previously submerged land, there is now an abundance of high-quality nesting habitat for groundnesting birds that did not exist previously (Gill et al. 2004). Thus the increase on Middleton may have occurred largely through recruitment of gulls to newly available habitat. Indeed, the maximum rate of increase on Middleton (29% per year between 1984 and 1993) far exceeds the 6% annual increase of Glaucous-winged Gulls documented on Protection Island, Washington, under conditions of high productivity and survival (Reid 1988) and suggests the increase on Middleton involved immigration from other colonies. Gulls’ demonstrated habit of visiting neighboring colonies in the off season implies that a notable improvement in conditions at any given breeding colony would not go unnoticed. To fully resolve such questions about gull population changes, more work—geographically broad-based and longterm—will be needed to identify contributing elements of summer and winter ecology. This study demonstrates the utility of satellite telemetry, implantable transmitters in particular, for revealing spatial connections across seasons.

5/13/11 3:01:38 PM

350 Scott A. Hatch

et al .

ACKNOWLEDGMENTS We gratefully acknowledge the work of more than 50 banders of Glaucous-winged Gulls in and outside of Alaska. Rick Sinnott, Alaska Department of Fish and Game, kindly loaned the net gun we used to capture adult gulls. David Irons and Rob Butler read and commented on an earlier draft of this paper. Any mention of trade names is for descriptive purposes only and does not imply endorsement by the U.S. government.

LITERATURE CITED Aebischer, N. J., and J. C. Coulson. 1990. Survival of the Kittiwake in relation to sex, year, breeding experience and position in the colony. Journal of Animal Ecology 59:1063–1071. A rgos. 1989. Guide to the Argos system. Argos CLS, Toulouse, France. Belant, J. L. 1997. Gulls in urban environments: landscape-level management to reduce conflict. Landscape and Urban Planning 38:245–258. Boulet, M., and D. R. Norris. 2006a. The past and present of migratory connectivity. Ornithological Monographs 61:1–13. Boulet, M., and D. R. Norris [eds.]. 2006b. Patterns of migratory connectivity in two nearctic−neotropical songbirds: new insights from intrinsic markers. Ornithological Monographs 61. Butler, R. W., N. A. M. Verbeek, and R. G. Foottit. 1980. Mortality and dispersal of the Glaucous-winged Gulls of southern British Columbia. Canadian Field-Naturalist 94:315–320. Byrd, G. V., H. M. R enner, and M. R enner. 2005. Distribution patterns and population trends of breeding seabirds in the Aleutian Islands. Fisheries Oceanography 14:139–159. Camphuysen, C. J., B. Calvo, J. Durinck, K. Ensor, A. Follestad, R. W. Furness, S. Garthe, G. Leaper, H. Skov, M. L. Tasker, and C. J. N. Winter. 1995. Consumption of discards by seabirds in the North Sea. Netherlands Institute for Sea Research, Texel, the Netherlands. Croxall, J. P., Silk, J. R. D., Phillips, R. A., A fanasyev, V., and Briggs, D. R. 2005. Global circumnavigations: tracking yearround ranges of nonbreeding albatrosses. Science 307:249– 250. Douglas, D. C. [online]. 2006. The Douglas Argos-filter algorithm. U.S. Geological Survey, Alaska Science Center, Anchorage, Alaska. (12 August 2010). Drent, R. H., and C. J. Guiguet. 1961. A catalogue of British Columbia sea-bird colonies. British Columbia Provincial Museum Occasional Paper 12. Drury, W. H. 1973. Population changes in New England seabirds. Bird-Banding 44:267–313. Fretwell, S. D. 1972. Populations in a seasonal environment. Prince ton University Press, Princeton, NJ. Gill, V. A., S. A. H atch, and R. B. Lanctot. 2004. Colonization, population growth and nesting success of Black Oystercatchers Haematopus bachmani following a seismic uplift. Condor 106: 791–800. Greenlaw, J. S., and W. J. Sheehan. 2003. The current status and history of Ring-billed Gulls (Larus delawarensis) in Maine. Northeastern Naturalist 10:55–66. H arris, M. P. 1970. Rates and causes of increase of some British gull populations. Bird Study 17:325–335. H atch, S. A. 1987. Adult survival and productivity of Northern Fulmars in Alaska. Condor 89:685–696. Hatch, S. A., T. W. Pearson, and P. J. Gould. 1979. Reproductive ecology of seabirds at Middleton Island, Alaska. U.S. Department of Commerce, NOAA, OCSEAP Annual Reports 2:233–308.

10_MS090224.indd 350

H atch, S. A., P. M. Meyers, D. M. Mulcahy, and D. C. Douglas. 2000. Seasonal movements and pelagic habitat use of murres and puffins determined by satellite telemetry. Condor 102:145–154. H atch, S. A., V. A. Gill, and D. M. Mulcahy. 2010. Individual and colony-specific wintering areas of Pacific Northern Fulmars (Fulmarus glacialis). Canadian Journal of Fisheries and Aquatic Sciences 67:386–400. H ayward, J. L., and N. A. Verbeek [online]. 2008. Glaucouswinged Gull (Larus glaucescens), no 59. In A. Poole [ed.], The birds of North America online. Cornell Lab of Ornithology, Ithaca, NY. . Hobson, K. A. 2005. Stable isotopes and the determination of avian connectivity and seasonal interactions. Auk 122:1037–1048. K adlec, J. A., and W. H. Drury. 1968. Structure of the New England Herring Gull population. Ecology 49:644–676. Korschgen, C. E., S. J. Maxson, and V. B. Kuechle. 1984. Evaluation of implanted radio transmitters in ducks. Journal of Wildlife Management 48:982–987. Korschgen, C. E., K. P. K enow, A. Gendron-Fitzpatrick, W. L. Green, and F. J. Dein. 1996. Implanting intra-abdominal radiotransmitters with external whip antennas in ducks. Journal of Wildlife Management 60:132–137. M arra, P. P., D. R. Norris, S. M. H aig, M. Webster, and J. A. Royle. 2006. Migratory connectivity, p. 157–183. In K. Crooks and M Sanjayan [eds.], Connectivity conservation. Cambridge University Press, New York. Mosbech, A., G. Gilchrist, F. M erkel, C. Sonne, A. Flagstad, and H. Nyegaard. 2006. Year-round movements of Northern Common Eiders Somateria mollissima borealis breeding in arctic Canada and west Greenland followed by satellite telemetry. Ardea 94:651–665. Mulcahy, D. M., and D. Esler. 1999. Surgical and immediate postrelease mortality of Harlequin Ducks (Histrionicus histrionicus) implanted with abdominal radio transmitters with percutaneous antennae. Journal of Zoo and Wildlife Medicine 30:397–401. Norris, D. R., M. B. Wunder, and M. Boulet. 2006. Perspectives on migratory connectivity. Ornithological Monographs 61:79–88. Petersen, M. R., D. C. Douglas, and D. M. Mulcahy. 1995. Use of implanted satellite transmitters to locate Spectacled Eiders at-sea. Condor 97:276–278. Plafker, G. 1970. Tectonics of the March 27, 1964 Alaska earthquake. Geological Survey Professional Paper 543-I. Pütz, K., C. R ahbek, P. Saurola, K. T. Pedersen, R. Juvaste, and A. J. Helbig. 2007. Satellite tracking of the migratory pathways of first-year Lesser Black-backed Gulls Larus fuscus departing from the breeding grounds of different subspecies. Vogelwelt 128:141–148. Pütz, K., A. J. Helbig, K. T. Pedersen, C. R ahbek, P. Saurola, and R. Juvaste. 2008. From fledging to breeding: long-term satellite tracking of the migratory behaviour of a Lesser Black-backed Gull Larus fuscus intermedius. Ringing & Migration 24:7–10. R ausch, R. 1958. The occurrence and distribution of birds on Middleton Island, Alaska. Condor 60:227–242. R eid, W. V. 1988. Population-dynamics of the Glaucous-winged Gull. Journal of Wildlife Management 52:763–770. Sæther, B-E., and Ø. Bakke. 2000. Avian life history variation and contribution of demographic traits to the population growth rate. Ecology 81:642–653. Salomonsen, F. 1955. The evolutionary significance of bird migration. Biologiske Meddelelser 22:1–62. Sodhi, N. S. 2002. Competition in the air: birds versus aircraft. Auk 119:587–595. Sowls, A. L., S. A. Hatch, and C. J. Lensink. 1978. Catalog of Alaskan seabird colonies. U.S. Fish and Wildlife Service, Biological Services Program, FWS/OBS−78/78, Anchorage, AK.

5/13/11 3:01:38 PM


Spaans, A. L., and H. Bloekpol. 1991. Concluding remarks: superabundance in gulls: causes, problems and solutions. Acta Congressus Internationalis Ornithologici 20:2396–2398. Speich, S. M., and T. R. Wahl. 1989. Catalog of Washington seabird colonies. U.S. Fish and Wildlife Service, Biological Report 88(6). Sullivan, T. M., S. L. Hazlitt, and M. J. F. Lemon. 2002. Population trends of nesting Glaucous-winged Gulls, Larus glaucescens, in the southern Strait of Georgia, British Columbia. Canadian FieldNaturalist 116:603–606. Thomas, G. J. 1972. A review of gull damage and management methods at nature reserves. Biological Conservation 4:117–127. Verbeek, N. A. M. 1986. Aspects of the breeding biology of an expanded population of Glaucous-winged Gulls in British Columbia. Journal of Field Ornithology 57:22–33. Vermeer, K. 1992. Population growth of the Glaucous-winged Gull (Larus glaucescens) in the Strait of Georgia, British Columbia. Ardea 80:181–185. Vermeer, K., and K. Devito. 1989. Population trend of nesting Glaucous-winged Gulls in the Strait of Georgia, p. 88–93. In K. Vermeer, and R. W. Butler [eds.], The ecology and status of marine and shoreline birds in the Strait of Georgia, British Columbia. Canadian Wildlife Service, Ottawa.

10_MS090224.indd 351

Vermeer, K., and D. B. Irons. 1991. The Glaucous-winged Gull on the Pacific coast of North America. Acta Congressus Internationalis Ornithologici 20:2378–2383. Vermeer, K., K. H. Morgan, and P. J. Ewins. 1992. Population trends of Pelagic Cormorants and Glaucous-winged Gulls nesting on the west coast of Vancouver Island. Canadian Wildlife Service Occasional Paper 75:60–64. Vermeer, K., D. B. Irons, E. Velarde, and Y. Watanuki. 1993. Status, conservation, and management of nesting Larus gulls in the North Pacific, p. 131–139. In K. Vermeer, K. T. Briggs, K. H. Morgan, and D. Siegel-Causey [eds.], The status, ecology, and conservation of marine birds of the North Pacific. Canadian Wildlife Service, Ottawa. Webster, M. S., P. P. M arra, S. M. H aig, S. Bensch, and R. T. Holmes. 2002. Links between worlds: unraveling migratory connectivity. Trends in Ecology and Evolution 17:76–83. Webster, M. S., and P. P. M arra. 2005. The importance of understanding migratory connectivity and seasonal interactions, p. 199–209. In R. Greenberg and P. P. Marra [eds.], Two worlds: the ecology and evolution of migration. Johns Hopkins University Press, Baltimore. Woodbury, A. M., and H. K night. 1951. Results of the Pacific gull color-banding project. Condor 53:57–77.

5/13/11 3:01:38 PM