untitled

North American Brant: effects of changes in habitat and

climate on population dynamics

DAVID H. WARD

, AUSTIN REEDw , JAMES S. SEDINGERz, JEFFERY M. BLACK§,

DIRK V. DERKSEN

and PAUL M. CASTELLI}

United States Geological Survey, Alaska Science Center, 1011 E. Tudor Rd., Anchorage, AK 99503, USA, wCanadian Wildlife

Service, 1141 Route de l’Eglise, Ste-Foy, PQ, Canada G1V 4H5, zDepartment of Environmental and Resource Sciences, University

of Nevada Reno, 1000 Valley Rd., Reno, NV 89512 USA, §Department of Wildlife, Humboldt State University, Arcata, CA 95521

USA, }New Jersey Division of Fish and Wildlife, Nacote Creek Research Station, P.O. Box 418, Port Republic, NJ 08625 USA

Abstract

We describe the importance of key habitats used by four nesting populations of nearctic

brant (Branta bernicla) and discuss the potential relationship between changes in these

habitats and population dynamics of brant. Nearctic brant, in contrast to most geese, rely

on marine habitats and native intertidal plants during the non-breeding season, parti-

cularly the seagrass, Zostera, and the macroalgae, Ulva. Atlantic and Eastern High Arctic

brant have experienced the greatest degradation of their winter habitats (northeastern

United States and Ireland, respectively) and have also shown the most plasticity in

feeding behavior. Black and Western High Arctic brant of the Paciﬁc Flyway are the most

dependent on Zostera, and are undergoing a shift in winter distribution that is likely

related to climate change and its associated effects on Zostera dynamics. Variation in

breeding propensity of Black Brant associated with winter location and climate strongly

suggests that food abundance on the wintering grounds directly affects reproductive

performance in these geese. In summer, salt marshes, especially those containing Carex

and Puccinellia, are key habitats for raising young, while lake shorelines with ﬁne

freshwater grasses and sedges are important for molting birds. Availability and

abundance of salt marshes has a direct effect on growth and recruitment of goslings

and ultimately, plays an important role in regulating size of local brant populations.

Keywords: brant, Branta bernicla hrota, Branta bernicla nigricans, breeding, climate change, migration,

molting, North America, winter

Received 15 June 2004; received in revised form and accepted 27 January 2005

Introduction

Brant (Branta bernicla; brent geese in Europe) are small-

bodied geese that migrate long distances from holarctic

nesting areas to temperate wintering habitats of the

northern hemisphere. They are among the most marine

of all geese and are associated with coastal wetlands

throughout their range. Brant are herbivorous, but differ

from other geese in that they still almost completely

rely on native plants through the full annual cycle.

Two of the three recognized subspecies of brant breed

in North America (B. b. nigricans and B. b. hrota). These

subspecies are separated into four breeding popula-

tions based on genetics, location of breeding and

wintering areas, and migration routes (Reed et al.,

1998). Two of the populations occur in the Paciﬁc

Flyway (Fig. 1). Black Brant breed from the western

North America low Arctic to the eastern Russian Arctic,

and winter along the Paciﬁc coast of North America

from Alaska to Mexico. Western High Arctic (WHA)

Brant nest on islands in the western North American

high Arctic and winter mainly in Puget Sound,

Washington (Reed et al., 1998). The other two popula-

tions occur in the Eastern Canadian Arctic (Fig. 1).

Atlantic Brant breed in the eastern Canadian low Arctic

and winter on the Atlantic coast from Massachusetts to

North Carolina, and Eastern High Arctic (EHA) Brant

nest on islands in the eastern Canadian high Arctic and

winter primarily on the coast of Ireland (Reed et al., 1998).

All populations, except EHA Brant, are 20–50%

below levels in the early 1950s and 1960s, when winter

inventories were ﬁrst initiated. Two populations are

considered to be stable. Black Brant number about

Correspondence: David H. Ward, tel: 987-786-3525;

e-mail: [email protected]

Global Change Biology (2005) 11, 869–880, doi: 10.1111/j.1365-2486.2005.00942.x

r 2005 Blackwell Publishing Ltd 869

Fig. 1 Breeding and wintering areas, and migration routes of nearctic brant populations.

870 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

120 000 individuals (Paciﬁc Flyway Council, 2002) and

EHA Brant total about 20 000 birds (Merne et al., 1999;

O. Merne, personal communication). Atlantic Brant, the

largest population in North America, number about

160 000 individuals and are experiencing a period of

growth (Atlantic Flyway Council, 2002), while WHA

Brant are in decline with a population size of about

6000 birds (Paciﬁc Flyway Council, 2002).

Environmental change over the last half of the 20th

century, driven both by human perturbations and/or

natural events, has had a great impact on wetland habitats

in North America and these changes have affected brant.

Effects include changes in distribution, survival and

ﬁtness, and breeding propensity (e.g., Hestbeck et al.,

1991; Cooch et al., 1993; Schamber, 2001; Person et al.,

2003). Although environmental change is ongoing,

current global warming predictions indicate that the rate

of change is likely to accelerate (Folland & Karl, 2001).

Anthropogenic changes to marine and freshwater

wetlands in the temperate regions have caused most

species of geese to abandon or alter their food

preferences during migration and winter from native

plants in natural habitats to agricultural plants and

seeds in cultivated ﬁelds (e.g., Abraham et al., 2005; Fox

et al., 2005; Gauthier et al., 2005; van Eerden et al., 2005).

Species have undergone this adjustment with differing

degrees of success but most with favorable results. In

contrast to other geese, including the European

populations of brant (Madsen et al., 1999), brant

wintering in North America have, for the most part,

not switched to agricultural and cultivated plants.

North American brant still depend on native marine

macrophytes, especially their preferred forage species

of seagrass, Zostera marina. No other species of goose

relies so heavily on a single plant species.

This paper offers insight into the current situation

and conservation challenges faced by brant in North

America. We do this by describing the signiﬁcance of

key wetland habitats in the annual cycle of the four

populations of brant. We characterize the relationship

between brant and their wetland habitats by presenting

examples of variation in brant population dynamics

(i.e., distribution, migration patterns, population para-

meters) with respect to habitat parameters (i.e., abun-

dance, availability, and quality) and climatic variability.

Methods

To compare use of key habitats among the four nearctic

populations of brant during nonbreeding and breeding

periods we summarized data from the literature. We

conducted trend analyses of count data from mid-winter

surveys to examine changes in overall population size

and shifts in winter distribution of Atlantic and Black

brant with respect to variation in winter climatic

conditions. These standardized surveys have been

conducted annually across the entire nonbreeding range

of Atlantic and Black brant populations and have proven

to be reasonably accurate estimates for evaluating trends

in overall population size and winter distribution (Kirby

& Obtrecht, 1982; Sedinger et al., 1994). We used multiple

linear regression of the log-transformed annual totals to

assess trends in overall population size and shifts in

winter distribution of Atlantic and Black brant.

To examine the inﬂuence of climatic variations on

shifts in winter distribution, we tested for a correlation

between annual population counts of brant and a

measure of climatic variability, the North America

Oscillation (NAO) for Atlantic Brant and El Nin

Southern Oscillation (ENSO) for Black Brant, using

the Pearson product–moment correlation coefﬁcients.

The NAO and ENSO are large-scale atmospheric

phenomenon that are associated with the intra- and

inter-annual changes in temperature and precipitation

in the north Atlantic (Hurrell, 1995), and central and

eastern Paciﬁc (Philander, 1990), respectively. A

monthly index value has been developed using several

large-scale atmospheric measures, and in the case of

ENSO, also oceanic measures to monitor the state of

each phenomenon. We used values of the NAO index

averaged over the winter months of December to

March 1960–2003 (Hurrell, 1995; http://www.cgd.ucar.

edu/ jhurrell/noa.stat.winter.html#winter) as an in-

dicator of climate variability during winters in the

northeastern United States. Positive values of the NAO

index are associated with above-normal temperatures

during winters in northeastern North America,

whereas negative values are associated with below-

normal temperatures, and above-normal snowfall

during winters in the region (Hurrell, 1995). For an

indicator of climate variability during winters in

Mexico, we used values from the multivariate index

of the ENSO (MEI) for the months of December and

January combined (http://www.cdc.noaa.gov/ENSO/

enso.mei_index.html). Positive MEI values, or El Nin

events, are associated with above-normal sea surface

temperatures and precipitation along the Paciﬁc coast

of southern California and northern Mexico, and

negative MEI values, or La Nin

a events, are associated

with below-normal sea surface temperatures in this

region (Philander, 1990; Minnich et al., 2000).

Results and discussion

Key habitats during nonbreeding

During the nonbreeding season, nearctic brant are

found exclusively in coastal areas, where they typically

IMPORTANCE OF KEY HABITATS TO NORTH AMERICAN BRANT 871

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

feed within the intertidal zone of shallow, protected

embayments that contain beds of seagrasses (Z. marina,

Ruppia maritima and Z. japonica in North America, and

Z. marina and Z. noltii in western Europe) and/or green

macroalgae (Ulva spp. and Enteromorpha spp.). For all

populations, Zostera is the preferred food during fall

and spring migration and winter (Reed et al., 1998).

Patterns of use of Zostera and other intertidal plants are

primarily dictated by tides and by the plant’s distribu-

tional range and seasonal availability and abundance.

All populations use a network of staging areas

during migration to obtain their nutrient reserves for

the annual cycle. These networks consist of one major

staging area and varying number of ancillary sites.

Eastern populations rely on only a few sites and tend to

concentrate in fall and spring at single major staging

areas that are rich in Zostera; James Bay, Quebec

(Atlantic Brant) and western Iceland (EHA Brant)

(Fig. 1). Black and WHA brant stage at more sites but

also concentrate at a single staging area, Izembek

Lagoon, Alaska, in fall, prior to a 42000 km ﬂight to

their primary wintering area (e.g., Dau, 1992). During

spring migration, western populations generally dis-

perse across several staging areas and travel shorter

distances between stops in a stepping stone approach.

This movement pattern presumably conserves nutrient

reserves for breeding. Moore et al. (2004) showed that

portions of the Black Brant population use no fewer

than 67 different bays and estuaries during spring

migration and variation in bird numbers among eight

of the major staging areas was best explained by

Z. marina abundance and distance to the next large

(containing 4500 ha of Z. marina) estuary to the north

(Fig. 2).

Wintering Black and WHA brant are more dependent

on Zostera than the other two populations, probably

because this seagrass has two to three times the spatial

extent on the Paciﬁc coast (41000 km

) than along the

Atlantic coasts of North America (ca. 600 km

) and

western Europe (o350 km

) (Green & Short, 2003).

Populations from eastern North America still feed on

Zostera when available, but also consume Ulva and

Enteromorpha (green algae now dominates the diet of

wintering Atlantic Brant), as well as, some salt marsh

plants (Spartina alterniﬂora for Atlantic Brant and

Festuca rubra and Puccinellia maritima for EHA Brant)

(Penkala, 1976; Smith et al., 1985; Merne et al., 1999).

Since the 1970s, both eastern populations have used

inland sites that contain cultivated grasslands. Cur-

rently, this behavior is displayed by about 5% of the

wintering Atlantic Brant and is generally conﬁned to

sites within 1–2 km of the coast, where birds feed on

school ﬁelds and golf courses (P. Castelli, unpublished

data). In Ireland, where intertidal foods are more

limited, inland feeding is an important activity for

about 25% of the EHA Brant and birds may travel up to

20 km inland to feed on farmlands, managed grass-

lands, and cereal crops (Merne et al., 1999).

Key habitats during breeding

Nesting. Brant nest predominantly in wet graminoid

meadows adjacent to coastal salt marshes (Reed et al.,

1998). At the low Arctic nesting sites of Black and

Atlantic brant, meadows are relatively extensive allowing

these birds to nest in colonies (Reed et al., 1998), and are

dominated by sedges (Carex spp.) and grasses such as

Poa eminens and Calamagrostis deschampsioides (Jorgenson,

2000). At the Arctic sites of WHA and EHA brant,

graminoid meadows are less extensive and consequently,

nesting is more dispersed, often occurring inland near

freshwater lakes or in braided river beds (Boyd & Maltby,

1979; Merne et al., 1999). For all populations, distribution

of nests is likely inﬂuenced by the typical distribution

of predators, especially arctic foxes (Alopex lagopus),

because brant are incapable of defending their nests

from foxes. Nonetheless, all principal nesting areas

occur in close proximity to salt marshes dominated by

C. subspathacea and Puccinellia spp. (Merne et al., 1999;

Jorgenson, 2000; Person & Ruess, 2003), where females

feed predominantly during incubation (Eichholz &

Sedinger, 1999). Availability of foraging habitat in

close proximity to nests is essential for brant because

females begin breeding with nutrient reserves to meet

only about 20% of their needs during breeding

(Ankney, 1984) and females must continue feeding

throughout the incubation period to maintain their

body mass (Eichholz & Sedinger, 1999).

PSG

456789

ln (ha Zostera)

ln (peak numbers)

Fig. 2 Linear regression of log-transformed data: peak brant

numbers vs. eelgrass abundance at eight pooled spring staging

areas throughout the Paciﬁc Flyway (R

5 0.79, F

1, 6

5 21.91,

P 5 0.003. C, Coos Bay; H, Humboldt Bay; I, Izembek Lagoon; M,

Morro Bay; PSG, Puget Sound and Strait of Georgia area; SQ, San

Quintin Bay; T, Tomales Bay; W, Willapa Bay. The regression

indicates a positive relationship between peak brant numbers

and eelgrass. From Moore et al. (2004).

872 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

Following hatch in the low Arctic, broods also feed

primarily in salt marshes, where they maintain grazing

lawns (Person et al., 2003) containing plant leaves with

high concentrations of protein (Sedinger et al., 2001).

Abundance and quality of plants in salt marshes

govern growth of brant goslings (Sedinger et al., 2001;

Herzog, 2002), which, in turn, controls recruitment

(Sedinger et al., 1995). Consequently, availability of salt

marsh plays an important role in regulating brant

populations (Sedinger et al., 1998; Person et al., 2003). In

the high Arctic, where salt marshes are limited, many

broods are raised along rockier shorelines or at inland

freshwater lakes (O’Briain et al., 1998). Maintaining

population size in this harsh region is also likely

inﬂuenced by available salt marsh habitat; however,

weather is likely equally or possibly more important

and can in some years cause complete breeding failure

(O’Briain et al., 1998).

Molting. Many brant that lose their clutches and those

that do not attempt to nest, emigrate from breeding

areas to high Arctic molting sites in June and July. Brant

molting areas are reused year after year when

undisturbed, and these long-standing traditions are

thought to reduce intraspeciﬁc competition for food

between unsuccessful breeders and nonbreeders and

breeding pairs (Salomonsen, 1968). Two key areas of

concentration for molting Paciﬁc Flyway brant, in

Alaska and Russia, have been identiﬁed and their

habitats described. No exclusive molting areas have

been discovered for Atlantic and EHA brant, but

nonbreeders and failed breeders do molt at nesting

areas on Southampton and Bathurst islands, Canada,

respectively (Reed et al., 1998).

In northcentral Alaska, the large, oriented thaw lakes

north and east of Teshekpuk Lake provide molting

habitat for an average of 14% (17 500 birds) of the

Paciﬁc Flyway population of brant (King & Hodges,

1979; Derksen et al., 1979; King & Derksen, 1979;

E. Mallek, unpublished data), although numbers of

ﬂightless birds vary considerably between years

depending, at least in part, on nest success at

breeding colonies in western Alaska (Sedinger et al.,

1994). Flightless brant feed primarily along moss/peat

shorelines immediately adjacent to open water rather

than along the little used sedge zones more distant from

the security of the lake (Derksen et al., 1982; Weller et al.,

1994). These relatively narrow moss ﬂats support ﬁne

grasses (Deschampsia caespitosa, Dupontia ﬁsheri) and

sedges (Carex spp.) preferred by brant and the birds

spend up to 52% of the 24 h cycle in foraging behavior

(Derksen et al., 1982). Importantly, the availability of

moss/peat ﬂats along shorelines of thaw lakes is very

limited – only about 2% (about 8 km

) of all habitats

classiﬁed in the Teshekpuk Lake molting area (over

4000 km

) consisted of this land cover class (Markon &

Derksen, 1994).

Another important, but less well known, molting

area for Paciﬁc Flyway brant occurs on Wrangel Island,

Russia. An estimated 4200 molting brant were counted

during the ﬁrst survey of the island in 1990 (Ward et al.,

1993). The freshwater lakes used by molting brant on

Wrangel Island have vegetation zones similar to those

of the Teshekpuk Lake area. The largest and most

heavily grazed community was a moist, moss-

dominated zone immediately adjacent to lake

shorelines, where brant primarily foraged on Dupontia

ﬁsheri and F. rubra (Ward et al., 1993).

Variation in brant use of wintering grounds and

migration routes with respect to habitat change

Shifts in distribution and behavior caused by food depletion

or habitat degradation. The nonbreeding distribution of

nearctic brant is largely dictated by the distribution and

abundance of Z. marina (Ganter, 2000; Moore et al.,

2004); therefore, changes in abundance and availability

of Zostera have implications for brant distribution,

survival, and reproductive output. An example of

food depletion resulting in a large-scale, long-term

shift in brant distribution and behavior occurred in

1931–1932 (Table 1) when a pathogenic slime mold,

Labyrinthula zosterae, caused an extensive die-off of

Z. marina in the north Atlantic (Rasmussen, 1977). This

die-off was followed by an apparent 80–90% decline in

the Atlantic Brant population (Cottam et al., 1944; Kirby

& Obrecht, 1982). Loss of Z. marina throughout its

nonbreeding range caused many Atlantic Brant to

move north to areas where Z. marina was more

abundant and/or to switch to alternative foods

(Cottam et al., 1944). Although some birds may have

starved (Cottam et al., 1944), the greatest inﬂuence on

population size was likely through reduced breeding

effort by malnourished birds (Kirby & Obrecht, 1982). A

concurrent rapid decrease in numbers of EHA Brant, as

well as, populations of brant wintering in Europe was

also attributed, in part, to this wasting disease event

(Merne et al., 1999).

Atlantic Brant fed primarily on Z. marina (85% of

their diet) prior to the event. After the onset of the

wasting disease, brant diet consisted mostly of Ulva

(75%), and much less so on Z. marina (9%; Cottam et al.,

1944). The Atlantic Brant population gradually

increased concomitant with the Z. marina recovery

and a sport-hunting moratorium between 1933 and

1952. By the mid-1950s, Atlantic Brant reached the level

present before the Z. marina die-off (Atlantic Flyway

Council, 2002). Lack of historical data makes it

IMPORTANCE OF KEY HABITATS TO NORTH AMERICAN BRANT 873

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

Table 1 Changes in brant population dynamics and use of nonbreeding and breeding areas with respect to habitat change

Event Date/location Population affected

Duration of

effect Effect on brant Population effect Reference

Nonbreeding areas

Wasting disease 1931–1932 North

Atlantic

Atlantic brant EHA

brant

Decadal Starvation

Fewer feeding

opportunities

Flyway-wide shift in

distribution

Reduced breeding

Lower population size

Cottam et al. (1944),

Kirby & Obrecht

(1982)

Other eelgrass

declines

1999–2000 James Bay,

Quebec

Atlantic brant Not yet known Diet switch to salt

marsh vegetation?

Unknown Lemieux & Lalumie

(2001)

Habitat degradation 41970 North Atlantic Atlantic brant EHA

brant

Ongoing? Diet switch to algae

and upland grasses

Fewer feeding

opportunities

Flyway-wide shift in

distribution

Change in migration

corridor

Kirby & Obrecht

(1982), Green & Short

(2003)

Severe weather 1976–1977 1977–1978

Northeastern U.S.

Atlantic brant 2 winters Starvation

Fewer feeding

opportunities

Flyway-wide shift in

distribution

Reduced breeding

Lower population size

Kirby & Obrecht

(1982), Kirby &

Ferrigno (1980)

El Nin

o Southern

Oscillation

1997–1998 North

Paciﬁc

Black brant Usually 1

winter

Fewer feeding

opportunities

Flyway-wide shift in

distribution

Reduced breeding

Greater winter

movement

Lower population size

Ward et al. (1999),

Schamber (2001)

Climate change? 41980 North Paciﬁc Black brant WHA

brant

Ongoing Fewer feeding

opportunities

Flyway-wide shift in

distribution

Ward et al. (1999,

2003), Dau & Ward

(1997)

Oyster aquaculture 1987–1993 Willipa Bay,

Black brant Annual Fewer feeding

opportunities

Local shift in

distribution

Wilson & Atkinson

(1995)

Breeding areas

Habitat degradation 1980–1985 YKD, AK Black brant Decadal Reduced gosling

growth rates

Increased food

competition

Lower carrying

capacity

Reduced breeding

Reduced recruitment

Lower population size

Sedinger

et al. (1993,

1998)

YKD, Yukon–Kuskokwim Delta; EHA, Eastern high arctic; WHA, Western high arctic.

874 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

impossible to know whether Z. marina along the

Atlantic coast has ever recovered to its prewasting

disease abundance and spatial extent. However, based

on behavior of brant, it appears that Z. marina has not

recovered to its full extent. Atlantic Brant have

maintained a diet consisting primarily of Ulva and

alternative foods through the 1970’s and 1980’s

(Penkala, 1976; Smith et al., 1985; Hindman & Ferrigno,

1990) and up to the present (Reed et al., 1998; P. Castelli,

unpublished data).

The dependence of Atlantic Brant on Ulva is likely

related to the increase in abundance and availability of

this macroalgae and a corresponding degradation and

loss of Z. marina habitats along the United States

Atlantic coast (Green & Short, 2003). Over the last 20–30

years, water quality of shallow estuaries and lagoons

has declined because of inputs of nutrients and

sediment loads associated with upland development

and shoreline construction (Short & Wyllie-Echeverria,

1996). Zostera losses have been attributed largely to

nutrient over-enrichment that can stimulate the

proliferation of fast growing phytoplankton and

ﬁlamentous algae like Ulva and Enteromorpha

(McGlathery, 2001). If blooms of macroalgae are

extensive and persistent enough, they can eventually

displace seagrasses as the dominant macrophyte in

eutrophic waters (Hauxwell et al., 2000).

Loss of Z. marina in northeastern North America

may also inﬂuence spring migration routes of Atlantic

Brant. Traditionally, large numbers of these birds

migrated along two main fronts: one up the Atlantic

coast from the United States to the gulf and estuary of

the St. Lawrence River, and the other overland, also to

the St. Lawrence, but further west near the conﬂuence

with the Ottawa River and in eastern Lake Ontario (Fig.

1; Reed et al., 1998). From the St. Lawrence staging areas

both groups move to James Bay. However, a radio

telemetry study (2002–2004) indicated that a minority of

Atlantic Brant now uses the coastal New England route

(P. Castelli, unpublished data), even though it was once

predominant (Palmer, 1976). Extensive declines in

Zostera have occurred along the coastal route (Green

& Short, 2003) and have been especially severe at

Monomoy Point, Cape Cod, Massachusetts (P. Castelli

personal observation), which has traditionally been a

primary spring staging area for Atlantic Brant (Fig. 1;

Palmer, 1976). With the continued low availability of

Zostera along the Atlantic coast and in the St. Lawrence

estuary, and with further recent declines in Zostera, the

coastal route has become less advantageous than the

direct inland ﬂight, even if the latter route offers no

coastal feeding areas before James Bay.

Atlantic Brant have experienced other instances of

food depletion causing large-scale changes in their

distribution and abundance. During the winters of

1976–1977 and 1977–1978, severe weather at the main

brant wintering sites in New York and New Jersey

covered intertidal habitats with ice and prevented birds

from feeding on seagrasses and macroalgae (Kirby &

Ferrigno, 1980). In each of these winters, there was

about a 30% decline in brant use of estuaries in the

north and a corresponding increase in more southerly

wintering areas (Fig. 3; Atlantic Flyway Council, 2002)

where intertidal areas were ice-free. This was also the

ﬁrst time that brant were recorded feeding on

cultivated grasses and clover on golf courses

(Hindman & Ferrigno, 1990). Reduced food resources

in the 1970s combined with hunting pressure and

reduced recruitment were likely responsible for a crash

in the Atlantic population (1960–1979: R

5 0.68,

1, 18

5 38.56, Po0.001; Fig. 3) similar to the 1930s

crash (Kirby & Obrecht, 1982). The population has

steadily increased from its lowest count of 42 000 birds

in 1979 to 165 000 birds in 2003 (1980–2003: R

5 0.45,

1, 22

5 17.80, Po0.001; Fig. 3), but is still 20–25% below

levels of the 1960s (Fig. 3). An important decline in Z.

marina has been reported for James Bay (Lemieux &

Lalumie

re, 2001), but the effects on brant have not been

determined.

We found no correlation between the winter NAO

index and Atlantic Brant wintering in areas north

(r 50.03, P 5 0.84) or south (r 5 0.22, P 5 0.15) of their

core habitats in New Jersey since 1960. Moreover, the

direction of the relationship in each case was opposite

what one would expect if a positive winter NAO index

caused an increase in brant numbers at northern

wintering areas and a negative winter NAO index led

to an increase in brant at southern wintering areas. The

lack of correlation is not surprising given the low

power of these types of analyses; however, except for

extreme climatic events, other factors, such as hunting

pressure, and/or other human activities, may have

greater inﬂuence on winter distribution of Atlantic Brant.

40 000

80 000

120 000

160 000

200 000

240 000

280 000

1960

1963

1966

196

1972

1975

1978

1981

984

1987

199

1996

2002

Number of brant

NJ Northern states Southern states

Fig. 3 Change in abundance and distribution of Atlantic Brant

across the winter range 1960–2003. Trends were evaluated on

log-transformed data using multiple linear regression analyses.

Atlantic Brant populations were assessed across three geogra-

phical regions: NJ (New Jersey), northern states (New York to

Maine), and southern states (Delaware to North Carolina).

IMPORTANCE OF KEY HABITATS TO NORTH AMERICAN BRANT 875

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

Massive die-offs of Z. marina have not been

documented on the Paciﬁc coast; nevertheless, shifts

in distribution of brant related to loss of Z. marina have

occurred. Currently, Black Brant are experiencing a

ﬂyway-wide shift in distribution (Ward et al., 2002) that

is likely related to changing environmental conditions

such as those that inﬂuence the abundance and

availability of Z. marina (e.g., irradiance and water

temperature; Hemminga & Duarte, 2000). Over the last

20 years, there has been a rise in sea-surface

temperatures and sea level in the north Paciﬁc

(Zveryaev & Selemenov, 2000; Cabannes et al., 2001),

and recent assessments of seagrass at the major brant

wintering areas along this coast have revealed a decline

in Z. marina abundance and spatial extent between 1980

and 2000 (Ward et al., 2002, 2003; S. E. Ibarra-Obando,

unpublished data). Between 1980 and 2000 and during

a period of population stability for Black Brant,

inventories showed a negative trend in numbers of

Black Brant wintering in Mexico (R

5 0.35,

2, 20

5 10.59, Po0.01; Fig. 4a) and a positive trend in

numbers in the United States and Canada (R

5 0.69,

2, 20

5 44.71, Po0.01; Fig. 4b). Brant reductions in

Mexico have largely occurred at the southern

wintering sites (Fig. 4a) where Z. marina reaches the

southern extent of its range in the northern hemisphere

and air and sea surface temperatures already limit Z.

marina growth to low intertidal and subtidal areas

(Meling-Lo

pez & Ibarra-Obando, 1999; Cabello-Pasini

et al., 2003). Shifts in distribution also coincided with a

period of increased El Nin

o activity (Fig. 4c). Between

1960 and 2003, we found a weak, but nonsigniﬁcant,

negative correlation (r 50.20, P 5 0.22) between the

winter MEI and number of brant wintering in Mexico.

Nevertheless, seven out of the 10 lowest counts of Black

Brant in Mexico were associated with positive winter

MEI values (0.39–2.74; Fig. 4c). Effects of changing

environmental conditions on Z. marina are exacerbated

during strong El Nin

o events, such as in 1997–1998,

when a sharp rise in sea level (up to a 20 cm above

mean sea level) and increased sea-surface temperatures

(1–3 1C) were associated with dramatic declines (up to

50% decrease) in Z. marina abundance at brant southern

wintering sites (Ward et al., 1999).

Increased numbers of brant at wintering sites has

been most evident in Alaska (Fig. 4B). Recent studies of

the Alaskan wintering population (Dau & Ward, 1997;

D. Mather, unpublished data) indicate that a portion of

the increase may be attributed to an inﬂux of WHA

Brant that traditionally wintered exclusively in Puget

Sound (Paciﬁc Flyway Council, 2002). The reasons for

some WHA Brant shifting their winter distribution

northward are unclear. Estuaries along the Alaska

Peninsula contain extensive beds of Z. marina (Ward

et al., 1997) and support virtually the entire populations

of Black and WHA brant in fall. However, prior to the

1980s, these estuaries were believed to winter fewer

than 1000 brant (Dau, 1992), presumably because Z.

marina beds were often covered by ice and inaccessible

to birds. Brant numbers in Alaska have steadily

increased in winter to 18 000 birds in 2001 (Paciﬁc

Flyway Council, 2002) coincident with a warming trend

in the north Paciﬁc (Zveryaev & Selemenov, 2000).

Warmer temperatures appear to have reduced the

period and frequency of ice cover in coastal areas

along the Alaska Peninsula (Dau & Ward, 1997), thus,

increasing food availability and reducing energy costs

for wintering birds.

(c)

(b)

(a)

0.5

1.5

2.5

1960

1964

1967

1970

1973

1976

1979

1983

1986

1990

1993

1996

1999

2002

MEI index

20 000

40 000

60 000

80 000

100 000

120 000

140 000

160 000

180 000

Number of brant

Winter MEI index (only positive values shown) Mexico

5000

10 000

15 000

20 000

25 000

30 000

1980

1983

1985

1988

1990

1992

1994

1996

1998

2000

2002

Number of brant

Alaska California Washington Oregon

50 000

100 000

150 000

200 000

1960

1964

1967

1970

1973

1976

1979

1983

1986

1990

1993

1996

1999

2002

Number of brant

Northern Mexico Southern Mexico US and Canada

Fig. 4 Change in abundance and distribution of Black Brant

across the entire winter range 1960–2003 (a), within the US 1980–

2003 (b), and in Mexico 1960–2003 with respect to the winter MEI

(c). Black Brant populations were assessed across four geogra-

phical regions: southern Mexico (Mexican states of Baja

California Sur, Sonora, and Sinaloa), northern Mexico (Baja

California), and US/Canada (California, Oregon, Washington,

British Columbia, and Alaska). Three counts in Mexico were

excluded from the analysis because they were either conducted

outside the winter period in February (1962 and 1987) when

Black Brant begin northward migration from Mexico or were

considered unrealistic counts (1981) (Sedinger et al., 1994).

876 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

Shifts in distribution can also occur at a more local

scale such as when human activities inﬂuence seasonal

use patterns. For example, in an important spring

staging area for Black Brant in Washington, Wilson &

Atkinson, (1995) showed that oyster farming activities

were correlated with reductions in Z. marina abundance

and in turn, signiﬁcant decreases in brant use-days.

Variation in habitat condition at nesting areas and its

effect on brant population dynamics

Black Brant nesting on the Yukon–Kuskokwim (Y–K)

Delta declined by 460% between about 1980 and 1985,

primarily as a result of intensive arctic fox predation

(Sedinger et al., 1993). Subsequent increases in the brant

population created a natural experiment by which

regulation of this population could be studied, similar

to the case for several European populations of geese

(Larsson & Forslund, 1994; Loonen et al., 1997; Black et

al., 1998). A retrospective analysis combined with

studies of vegetation dynamics during the population

increase indicated that Black Brant maintain grazing

lawns of C. subspathacea through their foraging activity

(Person et al., 2003). Reduced grazing intensity during

the early 1980s allowed most of these grazing lawns to

‘escape’ to a taller growth form that could no longer be

eaten by goslings. Consequently, carrying capacity was

reduced and density dependent effects on growth and

recruitment of Black Brant were observed at population

sizes 430% below historic levels (Sedinger et al., 1993,

1998). Growth rates of goslings increased steadily

throughout the late 1990s, corresponding to increased

areal extent of grazing lawns (Person et al., 2003).

Association between gosling growth and recruitment

(Sedinger et al., 1995) suggests that local recruitment

rates also increased during this period. These results

directly link recruitment, hence local population dy-

namics, to the availability of intensively grazed salt

marsh vegetation in the vicinity of the breeding colony.

Black Brant goslings on the North Slope of Alaska

grew substantially faster than those on the Y–K Delta,

because there was substantially greater biomass of salt

marsh vegetation on the North Slope of Alaska

compared with the Y–K Delta (Sedinger et al., 2001).

Greater biomass of salt marsh vegetation on the North

Slope of Alaska was associated with lower local

densities of Black Brant there, which is probably a

result of high nest predation rates in the adjacent

oilﬁelds (Sedinger & Stickney, 2000). Black Brant

goslings from the North Slope of Alaska survived their

ﬁrst fall at higher rates than those from the Y–K Delta

(Ward et al., 2004). This was consistent with their higher

growth rates and provides additional support for a

linkage between salt marsh vegetation and population

dynamics of Black Brant. Comparison of Black Brant

populations from the Y–K Delta and the North Slope of

Alaska shows that regulation of local populations

represents a balance between habitat availability and

predation rates. On the Y–K Delta, where nest success

has been high (75–83%) in most years since the mid-

1980s, recruitment is heavily inﬂuenced by gosling size,

which is regulated by availability of salt marsh

vegetation. In contrast, on the North Slope of Alaska,

high nest predation rates reduce the number of goslings

foraging in salt marsh habitats, increasing per capita

food abundance, growth rates, and ﬁrst-year survival.

Links between wintering conditions and population

dynamics

Cross-seasonal effects. Variation in reproductive

performance associated with wintering location and

winter climate strongly suggests that quality of winter

habitat plays an important role in population dynamics

of Black Brant. Schamber, (2001), using observations of

individually marked Black Brant, showed that birds

wintering in southern Baja California were less likely to

nest the next summer than those wintering in northern

Baja California and British Columbia. Individuals

wintering in the southern areas that did breed

initiated their nests later than birds using more

northern wintering areas. This effect was especially

pronounced in the ENSO year of 1997–1998 when fewer

than 10% of Black Brant wintering in San Ignacio

Lagoon, the most southern area studied, were observed

nesting the next summer. This contrasts with 39% of

individuals wintering in Boundary Bay, British

Columbia, and 28% of birds wintering in San Quintin

Bay, Baja California breeding the following summer

(Schamber, 2001). Reduced breeding in the ENSO year

was directly related to reduced production of Z. marina,

especially in more southern bays and estuaries (Ward et

al., 1999; Cabello-Pasini et al., 2003), suggesting that

food abundance at wintering areas had a direct effect

on reproduction at the population level. Numbers of

Black Brant nesting on the Y–K Delta have declined in

each ENSO year since aerial surveys of nesting Black

Brant began in the mid-1980s (W. Eldridge & W. Butler,

unpublished data) indicating that the relationship

between winter food and subsequent reproductive

performance is a general phenomenon.

Reduced ability of adults to breed associated with

declines in winter food availability have the potential to

decrease the size of the Black Brant population that can

be sustained along the Paciﬁc Coast of North America.

Brant management plans for both ﬂyways (Atlantic

Flyway Council, 2002; Paciﬁc Flyway Council, 2002)

recognize this potential and point out the need to

IMPORTANCE OF KEY HABITATS TO NORTH AMERICAN BRANT 877

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

maintain the integrity of Z. marina and other intertidal

plants, particularly in bays that are geographically

isolated from other large staging areas (Moore et al.,

2004), and to reduce human disturbance at goose

feeding and roosting areas where grit is obtained. The

dependence of brant on Z. marina and other intertidal

habitats leaves them vulnerable to the human activities

that increasingly impact shallow bays and estuaries

along North America’s coasts (e.g., oil spills, sediment

runoff, channel dredging, pollution, and mariculture).

Acknowledgements

We thank W. Butler, B. Conant, W. Eldridge, D. Mather, and

J. Voelzer for sharing unpublished data on Black Brant, S. Ibarra-

Obando for providing unpublished seagrass data from Mexico,

and K. Abraham for providing useful comments on Atlantic

Brant. We also greatly appreciate Mary Whalen for assistance in

developing the population maps and T. Lee Tibbitts for helpful

comments on the manuscript.

References

Abraham KF, Jefferies RL, Alisauskas RT (2005) The dynamics of

landscape change and snow geese in mid-continent North

America. Global Change Biology, 11, doi: 10.1111/j.1365-2486.

2005.00943.x.

Ankney CD (1984) Nutrient reserve dynamics of breeding and

molting brant. Auk, 101, 361–370.

Atlantic Flyway Council (2002) Atlantic Brant Management Plan.

Atlantic Flyway Study Committee, U.S. Fish and Wildlife

Service, Washington, DC.

Black JM, Cooch EG, Loonen MJJE et al. (1998) Body size

variation in barnacle goose colonies: evidence for local

saturation of habitats. In: Research on Arctic Geese, Skrifter

200 (eds Mehlum F, Black JM, Madsen J), pp. 129–140. Norsk

Polarinstitutt, Oslo.

Boyd H, Maltby LS (1979) The brant of the western Queen

Elizabeth Islands, N.W.T., Canada. In: Management and Biology

of Paciﬁc Flyway Geese (eds Jarvis RL, Bartonek JC), pp. 5–21.

Oregon State University Book Stores, Corvallis, OR.

Cabannes C, Cazenare A, Le Provost C (2001) Sea level rise

during 40 years determined from satellite and in situ

observations. Science, 294, 840–842.

Cabello-Pasini A, Mun

iz-Salazar R, Ward DH (2003) Annual

variations of biomass and photosynthesis in Zostera marina L.

Along the Paciﬁc coast of Baja California, Mexico. Aquatic

Botany, 76, 31–47.

Cooch EG, Jefferies RL, Rockwell RF et al. (1993) Environmental

change and the cost of philopatry: an example in the Lesser

Snow Goose. Oecologia, 93, 128–138.

Cottam C, Lynch JJ, Nelson AJ (1944) Food habits and manage-

ment of American sea brant. Journal of Wildlife Management, 8,

36–56.

Dau CP (1992) Fall migration of Paciﬁc brant in relation to

climatic conditions. Wildfowl, 43, 80–95.

Dau CP, Ward DH (1997) Paciﬁc Brant Wintering in the Izembek

National Wildlife Refuge: Population Size, Age Composition, and

Breeding Location. U.S. Fish & Wildlife Service Report,

Anchorage, Alaska.

Derksen DV, Eldridge WD, Weller MW (1982) Habitat ecology of

Paciﬁc black brant and other geese moulting near Teshekpuk

Lake, Alaska. Wildfowl, 33, 39–57.

Derksen DV, Weller MW, Eldridge WD (1979) Distributional

ecology of geese molting near Teshekpuk Lake, National

Petroleum Reserve – Alaska. In: Management and Biology of

Paciﬁc Flyway Geese (eds Jarvis RL, Bartonek JC), pp. 189–207.

Oregon State University Book Stores, Corvallis, OR.

Eichholz MW, Sedinger JS (1999) Regulation of incubation

behaviour in Black Brant. Canadian Journal of Zoology, 77,

249–257.

Fox AD, Madsen J, Boyd H et al. (2005) Effects of agricultural

change on abundance, ﬁtness components and distribution of

two Arctic-nesting goose populations. Global Change Biology,

11, doi: 10.1111/j.1365-2486.2005.00941.x.

Folland CK, Karl TR (2001) Observed climate variability and

change. In: Climate Change 2001 – The Scientiﬁc Basis,

Contribution of working group II to the Second Assessment Report

of the Intergovernmental Panel on Climate Change (eds Albritton

DL, Meiraﬁlho LG) pp. 101–165. Cambridge University Press,

New York.

Ganter B (2000) Seagrass (Zostera spp.) As food for brent geese

(Branta bernicla): an overview. Helgoland Marine Research, 54,

63–70.

Gauthier G, Giroux J-F, Reed A et al. (2005) Interactions between

land use, habitat use and population increase in greater snow

geese: what are the consequences for natural wetlands. Global

Change Biology, 11, doi: 10.1111/j.1365-2486.2005.00944.x.

Green EP, Short FT (2003) World Atlas of Seagrasses. UNEP World

Conservation Monitoring Centre, University of California

Press, Berkeley.

Hauxwell J, Cebrian J, Furlong C et al. (2000) Macroalgal

canopies contribute to eelgrass (Zostera marina) decline in

temperate estuarine ecosystems. Ecology, 83, 1007–1022.

Hemminga MA, Duarte CM (2000) Seagrass Ecology. Cambridge

University Press, Cambridge.

Herzog MP (2002) Environmental regulation of growth in black

brant. PhD Thesis, University of Alaska Fairbanks, Fairbanks,

Alaska.

Hestbeck JB, Nichols JD, Malecki RA (1991) Estimates of

movement and site ﬁdelity using mark-resight data of

wintering Canada geese. Ecology, 72, 523–533.

Hindman JL, Ferrigno F (1990) Atlantic Flyway goose popula-

tions: status and management. Transactions of the North

American Wildlife and Natural Resources Conference, 55,

294–311.

Hurrell JW (1995) Decadal trends in the North Atlantic

oscillation: regional temperatures and precipitation. Science,

269, 676–679.

Jorgenson MT (2000) Hierarchical organization of ecosystems at

multiple spatial scales on the Yukon-Kuskokwim Delta,

Alaska, U.S.A. Arctic, Antarctic, and Alpine Research, 32, 221–

239.

King JG, Derksen DV (1979) Alaska goose populations: past,

present and future. Transactions of the 51st North American

Wildlife and Natural Resources Conference, 51, 464–479.

878 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

King JG, Hodges JI (1979) A preliminary analysis of goose

banding on Alaska’s North Slope. In: Management and

Biology of Paciﬁc Flyway Geese (eds Jarvis RL, Bartonek JC),

pp. 176–188. Oregon State University Book Stores, Corvallis,

OR.

Kirby RE, Ferrigno F (1980) Winter, waterfowl, and the

saltmarsh. New Jersey Outdoors, 7, 10–13.

Kirby RE, Obrecht HH, III (1982) Recent changes in the North

American distribution and abundance of wintering Atlantic

Brant. Journal of Field Ornithology, 53, 333–341.

Larsson K, Forslund P (1994) Population dynamics of the

barnacle goose Branta leucopsis in the Baltic area: density-

dependent effects on reproduction. Journal of Animal Ecology,

63, 954–962.

Lemieux C, Lalumie

re R (2001) Complexe La Grande. Suivi

environnemental 2000. Les habitats co

tiers de la baie James. Report

presented to the Direction Expertise et Support technique de

production, Unite

Hydraulique et Environnement, Hydro-

Que

bec by the Groupe Conseil Ge

nivar Inc.

Loonen MJJE, Oosterbeek K, Drent RH (1997) Variation in

growth of young and adult size in barnacle geese Branta

leucopsis: evidence for density dependence. Ardea, 85,

177–192.

Madsen J, Cracknell G, Fox AD (1999) Goose Populations of the

Western Palearctic: A Review of Status and Distribution. Wetlands

International Publication 48. Wetlands International/National

Environmental Research Institute, Wageningen The Nether-

lands/Kalo

, Denmark.

Markon CJ, Derksen DV (1994) Identiﬁcation of tundra land

cover near Teshekpuk Lake, Alaska using SPOT satellite data.

Arctic, 47, 222–231.

McGlathery KJ (2001) Macroalgal blooms contribute to the

decline of seagrass in nutrient-enriched coastal waters. Journal

of Phycology, 37, 453–456.

Meling-Lo

pez AE, Ibarra-Obando SE (1999) Annual life cycles of

two Zostera marina seagrass L. Populations in the Gulf of

California: contrasts in seasonality and reproductive effort.

Aquatic Botany, 65, 59–69.

Merne OJ, Boertmann D, Boyd H et al. (1999) Light-bellied Brent

Goose Branta bernicla hrota. In: Goose Populations of the Western

Palearctic: A Review of Status and Distribution Canada, Wetlands

International Publication 48 (eds Madsen J, Cracknell G, Fox

AD), pp. 298–311. Wetlands International/National Environ-

mental Research Institute, Wageningen The Netherlands/

Kalo

, Denmark.

Minnich RA, Vizcaino EF, Dezzani RJ (2000) The El Nino/

Southern Oscillation and precipitation variability in Baja

California, Mexico. Atmosfera, 13, 1–20.

Moore JE, Colwell MA, Mathis RL

et al. (2004) Staging of Paciﬁc

ﬂyway brant in relation to eelgrass abundance and site

isolation, with special consideration of Humboldt Bay,

California. Biological Conservation, 115, 475–486.

O’Briain M, Reed A, Macdonald SD (1998) Breeding, moulting,

and site ﬁdelity of brant on Bathurst and Seymour Islands in

the Canadian High Arctic. Arctic, 51, 350–360.

Paciﬁc Flyway Council (2002) Paciﬁc Flyway Management Plan for

Paciﬁc Brant. Paciﬁc Flyway Study Committee, U.S. Fish and

Wildlife Service, Portland, OR.

Palmer RS (1976) Handbook of North American Birds, Vol. 2. Yale

Univeristy Press, New Haven.

Penkala JM (1976) Winter food habits and body weights of Atlantic

brant. MS thesis, Rutgers University, New Brunswick, New

Jersey.

Person BT, Herzog MP, Ruess RW et al. (2003) Feedback

dynamics of grazing lawns: coupling vegetation change with

animal growth. Oecologia, 135, 583–592.

Person BT, Ruess RW (2003) Stability of a near Arctic saltmarsh:

community resistance to tidal inundation. Ecoscience, 10,

351–360.

Philander SGH (1990) El Nin

o, La Nin

a and the Southern

Oscillation. Academic Press, San Diego, CA.

Rasmussen E (1977) The wasting disease of eelgrass (Zostera

marina) and its effects on environmental factors and fauna. In:

Seagrass Ecosystems – A Scientiﬁc Perspective (eds McRoy CP,

Helfferich C), pp. 1–51. Marcel Dekker, New York.

Reed A, Ward DH, Derksen DV et al. (1998) Brant (Branta

bernicla). In: The Birds of North America, Vol. 337 (eds Poole

A, Gill F), The Birds of North America, Inc., Philadelphia,

PA.

Salomonsen F (1968) The moult migration. Wildfowl, 19, 5–24.

Schamber JL (2001) Cross-seasonal effects on reproductive perfor-

mance of Paciﬁc black brant. MS Thesis, University of Alaska-

Fairbanks, Fairbanks, Alaska.

Sedinger JS, Flint PL, Lindberg MS (1995) Environmental

inﬂuence on life-history traits: growth, survival and fecundity

in Black Brant (Branta bernicla). Ecology, 76, 2404–2414.

Sedinger JS, Herzog MP, Person BT et al. (2001) Large-scale

variation in growth of Black Brant goslings related to food

availability. Auk, 118, 1088–1095.

Sedinger JS, Lensink CJ, Ward DH et al. (1993) Current status

and recent dynamics of the Black Brant breeding population.

Wildfowl, 44, 49–59.

Sedinger JS, Lindberg MS, Person BT et al. (1998) Density

dependent effects on growth, body size and clutch size in

Black Brant. Auk, 115, 613–620.

Sedinger JS, Stickney AA (2000) Black brant. In: The Natural

History of an Arctic Oil Field Development and Biota (eds

Truett JC, Johnson SR), pp. 221–232. Academic Press, San

Diego, CA.

Sedinger JS, Ward DH, Anthony RM et al. (1994) Management of

Paciﬁc brant: population structure and conservation issues.

Transactions of the 59th North American Wildlife and Natural

Resources Conference, 59, 50–62.

Short FT, Wyllie-Echeverria S (1996) A review of natural and

human-induced disturbance of seagrasses worldwide. Envir-

onmental Conservation, 23, 17–27.

Smith LM, Vangilder LD, Kennamer RA (1985) Foods of

wintering brant in eastern North America. Journal of Field

Ornithology, 56, 286–289.

van Eerden MR, Drent RH, Stahl J et al. (2005) Connecting seas:

Western Palaearctic continental ﬂyway for water birds in the

perspective of changing land use and climate. Global Change

Biology, 11, doi: 10.1111/j.1365-2486.2005.00940.x.

Ward DH, Derksen DV, Kharitonov SP et al. (1993) Status of

Paciﬁc black brant Branta bernicla nigricans on Wrangel Island,

Russian Federation. Wildfowl, 44, 39–48.

IMPORTANCE OF KEY HABITATS TO NORTH AMERICAN BRANT 879

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880

Ward DH, Markon CJ, Douglas DC (1997) Distribution and

stability of eelgrass beds at Izembek Lagoon, Alaska. Aquatic

Botany, 58, 229–240.

Ward DH, Morton A, Tibbitts TL et al. (2002) Recent trends in

eelgrass, Zostera marina, distribution and abundance along the

west coast of Baja California, Mexico. Proceedings of the 5th

International Seagrass Biology Workshop. Ensenada, Baja Cali-

fornia, Mexico.

Ward DH, Morton A, Tibbitts TL et al. (2003) Long-term spatial

change in eelgrass distribution at Bahı

a San Quintı

n, Baja

California, Mexico, using satellite imagery. Estuaries, 26,

1529–1539.

WardDH,SchmutzJA,SedingerJSet al. (2004) Temporal and

geographic survival of juvenile black brant. Condor, 106, 263–274.

Ward DH, Tibbitts TL, Carrera-Gonza

les E (1999) Effects of the

1997–1998 El Nin

o event on brant, Branta bernicla nigricans,

wintering in Mexico. Proceedings of the 6th Neotropical

Ornithological Congress, Monterrey, Mexico.

Weller MW, Jensen KC, Taylor EJ et al. (1994) Assessment of

shoreline vegetation in relation to use by molting black brant

Branta bernicla nigricans on the Alaska coastal plain. Biological

Conservation, 70, 219–225.

Wilson UW, Atkinson JR (1995) Black brant and spring-staging

use at two Washington coastal areas in relation to eelgrass

abundance. Condor, 97, 91–98.

Zveryaev I, Selemenov KM (2000) Decadal scale changes in

annual cycle of the north Paciﬁc sea-surface temperatures.

International Journal of Climatology, 20, 1639–1651.

880 D. H. WARD et al.

r 2005 Blackwell Publishing Ltd, Global Change Biology, 11, 869–880