Part IV · County Highway D, Grand View, Wisconsin 54839; *corresponding author: [email protected] Western Field Ornithologists A rctic-breeding birds time their arrival on nest-ing areas

295

Part IV

Response to Changes in Climate and the enviRonment

Studies of Western Birds No. 3

296

Abstract: Arctic summers are brief, and there has been strong selection for migratory birds to arrive in Arctic nesting areas as early as possible to time breeding with peak food availability and complete reproduction. The timing of emergence of nesting habitat in spring is, however, extremely variable in the Arctic, and few long-term studies have examined the ability of avian migrants to track spring conditions to assure optimal nesting. Such studies require long-term migration monitoring under variable spring habitat conditions. These conditions were met during our long-term (1977–2008) study of the timing of arrival of shorebirds to their nesting grounds on the central Yukon–Kuskokwim (Y-K) Delta, in western Alaska. Over this period, the timing of arrival on the nesting grounds of 12 species of shorebirds varied significantly, with the Black-bellied Plover (Pluvialis squatarola) generally arriving first (mean arrival of 4 May), the Red Phalarope (Phalaropus fulicarius) usually arriving last (mean arrival of 20 May). The Western Sandpiper (Calidris mauri), Dunlin (Calidris alpina), and Red-necked Phalarope (Phalaropus lobatus), the most common breeding shorebird species we studied, all arrived about the same time each year (7–9 May). From year to year, first arrival of all species varied significantly by more than 2 weeks, but there was no long-term trend in arrival times over the length of our study. Shorebird arrival was highly correlated with the timing of the break-up of ice on the Kashunuk River, which in turn was correlated with decreasing snow cover and increasing ambient temperature. The date of break-up of river ice also varied by year but did not advance significantly during our study. After arriving on the breeding grounds, Arctic-nesting shorebirds rely on local food resources, which on the Y-K Delta they achieve by timing their arrival to coincide with availability of snow-free habitat. Temperatures along the terminal portions of the spring migration route were significantly correlated with both conditions on the breeding ground and the timing of shorebirds’ arrival there, which suggests that shorebirds may use environmental cues during spring migration to regulate its pace. Ours is one of the few multi-decadal studies to evaluate the responses of a suite of migrant species to annual variation in conditions in their Arctic breeding habitat. Shorebirds’ adaptations to variable conditions on the Y-K Delta are probably rooted in frequent changes to the landscape since the last glacial maximum. Such inherent flexibility may serve them well under future scenarios associated with a changing climate.

Keywords: Arctic, arrival timing, breeding, Calidris, Charadriiformes, climate change, phenol-ogy, spring migration

Shorebirds adjust spring arrival schedules with variable environmental conditions: Four decades of assessment on the Yukon–Kuskokwim Delta, AlaskaCraig R. Ely1,*, Brian J. McCaffery2,3, and Robert E. Gill Jr.1

Full citation: Ely, C. R., McCaffery, B. J., and Gill, R. E., Jr. 2018. Shorebirds adjust spring arrival schedules with variable envi-ronmental conditions: Four decades of assessment on the Yukon–Kuskokwim Delta, Alaska, in Trends and traditions: Avifaunal change in western North America (W. D. Shuford, R. E. Gill Jr., and C. M. Handel, eds.), pp. 296–311. Studies of Western Birds 3. Western Field Ornithologists, Camarillo, CA; doi 10.21199/SWB3.16.1U.S. Geological Survey, Alaska Science Center, 4210 University Drive, Anchorage, Alaska 99508, 2U.S. Fish and Wildlife Service, Yukon Delta National Wildlife Refuge, P.O. Box 346, Bethel, Alaska 99559, 3Current address: U.S. Fish and Wildlife Service, 53980 County Highway D, Grand View, Wisconsin 54839; *corresponding author: [email protected]

Western Field Ornithologists

Arctic-breeding birds time their arrival on nest-ing areas to take advantage of the seasonal

availability of a nutrient-rich environment in which to nest and raise young (Barry 1962, Møller 2001, Martin and Wiebe 2004, Yohannes et al. 2010). If birds arrive too early they may find nest sites and food resources still frozen or covered with snow and ice (Meltofte et al. 2007). If birds arrive

too late they may (1) miss the peak of food avail-ability for themselves or their young (Lack 1968, McKinnon et al. 2012, Pearce-Higgins and Green 2014), (2) find reduced access to mates (Kokko 1999, Lanctot et al. 2000), or (3) be forced into delayed departure in autumn, which might com-promise their ability to acquire reserves sufficient for migration and expose them to a greater danger

Shorebirds’ Spring Arrival Schedules: Four Decades of Assessment, Yukon–Kuskokwim Delta, Alaska

297

of predation (Uspenskii 1969, Lank and Ydenberg 2003, Bonier et al. 2007, Ydenberg et al. 2007). Because the timing of breeding in northern cli-mates has such a profound effect on reproductive success and fitness (Lack 1968, Price et al. 1988), there should be strong selection for birds to arrive where they breed at the optimum time (Pearce-Higgins and Green 2014).

The Arctic, defined as northern regions of the globe bounded by the summer isotherm of 10 °C (CAFF 2001), and characterized in western Alaska by tundra, is available for breeding by many organisms only seasonally. Although most bird species nesting in the Arctic have likely been adjusting their breeding schedules to changes in the seasonal availability of resources for millen-nia, they must now try to respond to the recent rapid rise in global temperatures (Root et al. 2003) and increased variability in climate (IPCC 2014). The Arctic is warming at a rate greater than anywhere else on Earth (Serreze et al. 2000, Bintanja and van der Linden 2013, Wendler et al. 2014). This warming, the concomitant changes in habitat phenology, and the ability of birds nesting at high latitudes to keep pace with such change under projected climate change-scenarios have received considerable attention (Visser et al. 1998, Both and Visser 2001, Post et al. 2008, Clausen and Clausen 2013). Most studies have shown that northern plants and animals have generally responded to recent spring and summer warming by advancing the timing of growth and breeding (Walther et al. 2002, Root et al. 2003, Knudsen et al. 2011). Several studies have documented such an advance in the timing of breeding in shore-birds (Pearce-Higgins et al. 2005, Gunnarsson and Tómasson 2011, Liebezeit et al. 2014). There is also evidence, however, that breeding of some Arctic animals has not kept up with the advance of spring (Post et al. 2008, Clausen and Clausen 2013).

The ability of migratory birds to adapt to breeding in a variable environment is dependent, at least in part, on mechanisms controlling the timing of spring migration. Such mechanisms have been the subject of study for over a century. Cooke (1904) and Lincoln (1939) recognized that some spring migrants (notably passerines and some waterfowl breeding in the Arctic and North Temperate Zone) keep pace with the isotherm at 2 °C, and that they migrate more rapidly as they approach breeding areas, especially at northern latitudes where spring arrives more quickly. Not all species, however, employ such a flexible migra-tion strategy, without which they cannot advance

the timing of nesting to match habitat availability (Both and Visser 2001). More generally, however, even species with life cycles regulated by fixed cues or a circannual rhythm may have the capacity to adjust those patterns in response to environmental cues (Gwinner 1996). Species that have a flexible migration strategy may keep pace with changes in the availability of suitable habitat in staging and breeding areas, whereas species that are not able to respond adequately to temporal variation in the availability of resources may fare poorly. The variation among avian taxa in the capacity for such adaptation needs further study.

In this paper, we assess the flexibility in the timing of Arctic shorebirds’ migration to chang-ing habitat conditions in spring. We document the timing of spring arrival for 12 species of shorebirds breeding on the Yukon–Kuskokwim (Y-K) Delta, Alaska, over 28 seasons between 1977 and 2008. The Y-K Delta contains one of the greatest expanses of coastal wetlands in North America (Gill and Handel 1990) and is one of the continent’s most important areas for breed-ing shorebirds (Pitelka 1979, McCaffery et al. 2012). While we have a good understanding of the breeding biology of several of these species (e.g., Holmes 1971a, b, Handel and Gill 2000, Johnson and Walters 2008, Johnson et al. 2009), few studies have documented their migration chronology on the Y-K Delta (Holmes 1971a, Gill and Handel 1990). Although other studies have considered migration timing elsewhere in the Arctic (Gerasimov and Gerasimov 2000, Meltofte et al. 2007 and references therein, Smith et al. 2010, Gunnarsson and Tómasson 2011), most such studies have focused on a single species, or have been of relatively short duration. Therefore, the inferences that can be drawn from them are quite limited. Our study, however, spanned several decades during which we were able to track the annual response of a suite of shorebird species to annual variations in the regional climate.

If the timing of spring thaw and resource availability are relatively and predictably constant through the years (e.g., Alerstam and Hogstedt 1980), then a chronological migration strategy (in which annual migrations are cued by day length) might be the most effective means of capitalizing on available resources. But if habitat availability is not highly predictable from year to year, then a flexible migration strategy might lead to better reproductive outcomes. Here we explore differ-ences in the timing of spring migration of nest-ing shorebirds, the potential influence of climate change on the timing of their migrations, what

298


factors might influence migration timing (including migration distance), and their general responses to migration and breeding in a variable environment.

study aRea and methodsOur study took place on the central Y-K Delta in western Alaska, which lies within the Yukon Delta National Wildlife Refuge (YDNWR, Figure 1). From 1977 to 1998 we monitored birds along the Kashunuk River near the abandoned village of Old Chevak (61° 26′ N, 165° 27′ W) and at a field camp 7 km downriver. Later, from 1999 to 2008, we expanded our effort to include the area around the U.S. Fish and Wildlife Service’s Kanaryarmiut Field Station near the Aphrewn River (61° 22′ N, 165° 08′ W), about 15 km southeast of Old Chevak (Figure 1). In 2005 and 2006, observa-tions were made simultaneously at Old Chevak and Kanaryarmiut. The habitat at both sites is characterized by very low relief and discontinu-ous permafrost overlain by moist and wet tundra. In low-lying areas the vegetation is dominated primarily by graminoids, forbs, and dwarf willow (Salix spp.), on upland terraces (thermokarst pla-teaus 1–2 m higher in elevation) by lichens, moss-es, and dwarf sclerophyllous vegetation (Ely and

Raveling 1984, Babcock and Ely 1994, Jorgenson and Ely 2001). The Kanaryarmiut site has slightly more thermokarst upland than do sites along the Kashunuk River, though both sites are about an equal distance (~20 km) from the coast (Figure 1), and both are within the Kashunuk coastal plain (T. Jorgenson unpubl. data).

enviRonmental dataTemperature data were obtained from daily records at our field sites (1977–1998) and from weather stations elsewhere in south-central Alaska (i.e., Anchorage, Cold Bay, Dillingham, Kodiak, Homer, and Cordova; Figure 1), near areas where shorebirds stop during spring migration (U.S. Dept. Commerce, 1977–2008). We recorded the date that snow cover was ≤10% on uplands just north of our Old Chevak field camp, and the first day that lowlands were inundated by meltwater immediately west of camp (1977–1998; Figure 1). We defined break-up of the Kashunuk River as the date when river ice first moved in our Old Chevak study area (Figure 1). Data on the timing of the river’s break-up and snow cover for 1977–1987 and 2001–2008 augmented data from 1988 to 2000, which has been published by Babcock et

Yukon Delta NWR

Tooksook Bay

Kanaryarmiut

Old Chevak

Homer

Kodiak Island

CordovaDillingham

Bethel

KingSalmon

Kanaryarmiut

Old Chevak

Field Camp

Kashunuk River Aphrewn River10 2 3 4 5 km

Tununak

Newtok

Hooper

Scammon Bay

Chevak

Bay

Cold Bay

Anchorage

123

4

FiguRe 1. Locations of study area on the outer Yukon–Kuskokwim Delta, Alaska, and of spring staging areas in south-central Alaska (1, Prince William Sound; 2, Cook Inlet; 3, Kachemak Bay; 4, Bristol Bay). Dashed ellipses indicate core observation areas.


299

al. (2002). We retrieved data on May sea-sur-face temperatures, El Niño–Southern Oscillation (ENSO), and the Pacific Decadal Oscillation (PDO) from www.beringclimate.noaa.gov/data.

shoRebiRd obseRvationsWe monitored the arrival dates of 12 species of shorebirds that nest commonly or regularly on the outer Y-K Delta and at our field sites (Figure 1). Field observations were made each year by two to four biologists whose primary task during April and May was to document the first arrival of avian migrants. Biologists arrived at the study sites each spring on dates based on the timing of snowmelt reported by pilots and local villagers. This meant that in some years researchers did not arrive until after the arrival of the earliest species. Observations began the day researchers arrived and continued daily until mid- to late May, by which time most species had initiated nests and begun incubating. The observers recorded arrivals at camp, when walking to and from transects, and while walking fixed transects established for assessing waterbirds’ migration chronology. For the initial sightings of each shorebird species we noted the species and number of individuals. We censored the data on a species’ arrival for years in which the species was observed on our first day in the field. These data comprised 4.6% (15/327) of first observations: of the Black-bellied Plover (Pluvialis squatarola) in 1989, 1993, 2001, 2002, and 2003, of the Bar-tailed Godwit (Limosa lapponica) in 1993, of the Ruddy Turnstone (Arenaria interpres) in 2002 and 2003, of the Black Turnstone (A. melanocephala) in 1993, of the Dunlin (Calidris alpina) in 2002, of the Rock Sandpiper (C. ptilocnemis) in 2003, of the Western Sandpiper (C. mauri) in 1993 and 2002, and of Wilson’s Snipe (Gallinago delicata) in 2002 and 2003.

We used the date of first arrival as a metric for characterizing the timing of shorebirds’ breeding season, and assumed that early first arrivals cor-responded with earlier nesting in the overall local population. This assumption could be violated in two ways: (1) if a species actually arrived before the first detection and (2) if the date of first arrival does not actually reflect the timing of migration and subsequent breeding at the population level. With respect to (1), detectability might vary with birds’ habitat, behavior, and the local abundance. We are confident that both habitat and behavior contributed to high detectability at our study sites, where both flocks and actively displaying birds were conspicuous in an early spring landscape characterized by sparse vegetation and low relief.

If population size influences detectability, then we might predict either that species with larger populations are more likely to be detected earlier, on average, and/or that such species’ arrival dates are less variable. To evaluate the potential effect of population size on detectability, we examined cor-relations between local abundance and the timing of first arrivals, the standard error of first arrivals, and the span of dates between earliest and latest first arrivals for each species across the duration of the study. For these analyses, we used abun-dance estimates from an 853-km2 study area that included both Kanaryarmiut Field Station and the lower Kashunuk River (McCaffery et al. 2012). Finally, to determine the correlation and time lag between first arrival and the timing of nesting, observers collected data on first and (in some cases) mean nest-initiation dates for several spe-cies of nesting shorebirds. The Bar-tailed Godwit’s nesting chronology was studied at Old Chevak in 2004, 2005, 2006, and 2008 (BJM unpubl. data). At the Kanaryarmiut Field Station, Western Sandpipers were studied intensively from 1998 to 2006 (Ruthrauff 2002; BJM unpubl. data), Rock Sandpipers from 1998 to 2005 (Johnson et al. 2009), and Dunlin in 1999 (BJM unpubl. data) and from 2004 to 2006 (Jamieson 2011). Finally, we compared arrival times with the nesting chro-nology of the Black Turnstone in 1978, 1979, 1980, 1982, and 2006 at the Tutakoke River, a field station that was 23 km southwest of Old Chevak and 30 km southwest of Kanaryarmiut (Handel and Gill 2000, Handel 2002, P. S. Tomkovich unpubl. data).

migRation ChaRaCteRistiCsThe shorebirds we studied spend the nonbreed-ing season in disparate regions of the northern and southern hemispheres. To assess the possible influence of migration distance on the timing of arrival on the Y-K Delta in spring, we esti-mated the distance of each species’ migration from three different points (north, middle and south) within the nonbreeding range. For all but the two species of phalarope we obtained distributional data from eBird (www.ebird.org/ebird/map/); for phalaropes, we used Cornell Lab’s All About Birds (www.allaboutbirds.org). Data from eBird were filtered to include the period from December to April for the years 1986–2016. We defined the northern and southern limits of the main nonbreeding distributions as the most northern and southern 20- × 20-km eBird blocks in which >10% of observations were of a particular species. The midpoint of a species’ nonbreeding range was

300


the latitudinal midpoint between the northern and southern limits of the range so defined. For the Ruddy Turnstone we considered only the continental nonbreeding distribution along the coasts of the Americas; for the Bar-tailed Godwit we estimated distances of potential nonstop flights from its final spring staging area around the Yellow Sea (Battley et al. 2012). We estimated all distances (km) in Google Earth according to both the line function (= great circle) and path function (smoothed route along coast) from the southern, middle, and northern points of a range to the mouth of the Kashunuk River on the Y-K Delta. For the 12 species considered, distances along the coast were only slightly longer than those along a great-circle route (4.8 ± 3.3% for midpoints) and highly correlated with them (Spearman’s r > 0.986 for all three points), so we restricted analyses to the coastal routes for all species except the Bar-tailed Godwit, whose transoceanic route is known from satellite telemetry (Battley et al. 2012).

data analysesGiven our two study sites’ proximity and similar-ity in habitat, we combined data for analyses. For 2005 and 2006, when we surveyed both Kanaryarmiut and Old Chevak, we used the earliest arrival date for each species, regardless of location. We examined box plots of first arrival dates of each species to assess if they were normally distributed, then tested potentially non-normal distributions with a Shapiro–Wilks test (Sokal and Rohlf 1981; Proc Univariate, SAS Institute 1999). We used Pearson correlation coefficients and corresponding coefficients of determination (r2) (Proc Corr, SAS Institute 1999) to explore relationships among weather and climate variables that might affect arrival timing, to compare arrival times with nesting chronology, and to assess the potential effect of local population size on detect-ability of early migrants. We used Spearman’s rank-correlation coefficients to evaluate the strength of the relationships between mean arrival date on the Y-K Delta and distance of migration from the northern, middle, and southern points of the nonbreeding range for the 12 species (see above; Sokal and Rohlf 1981). We used two-way analysis of variance without interaction to assess the effects of year and species on arrival times and used a Bonferroni adjustment or Fisher’s permuta-tion (Proc Corr Fisher, SAS Institute 1999) when making multiple comparisons (with adjusted p values). To examine temporal trends in environ-mental variables and the arrival of shorebirds, we used linear regression (Proc Reg, SAS Institute

1999). Independent environmental variables test-ed (1977–1998) included the date of break-up of the Kashunuk River, the date when the coverage of snow on uplands dropped below 10%, the onset of the frost-free period, and the date of first maximum temperature >10 °C at the Kashunuk River. As an index of spring phenology, we used heating degree days during April and May at Bethel and other weather stations in south-central and southern Alaska (Figure 1), with a heating degree day defined as the sum of the negative dif-ferences between the mean daily temperature and 65 °F (18.3 °C) (NOAA unpubl. data; www.erh.noaa.gov/cle/climate/info/degreedays.html).

Results

spRing thaw and habitat emeRgenCeThe study area was commonly completely cov-ered with snow and ice when observers arrived in mid-April to early May. Upland areas were the first to become snow-free, because of the scouring effect of strong winds and the propensity for the uplands’ dark heath vegetation to absorb sunlight. The resulting snowmelt filled nearby shallow catchments and inundated lowlands. The progres-sion of the thaw along the Kashunuk River was typical of most of the outer Y-K Delta, but was especially characteristic of the region of the central Y-K Delta between Nelson Island and the Askinuk Mountains. Because slightly elevated areas were freed of snow faster than lower areas they appeared as islands of habitat that were especially attrac-tive to shorebirds and waterfowl when they first arrived (Figure 2).

habitat availability in spRingLocal weather and habitat conditions varied con-siderably from year to year (Table 1, Figure 3). Over the 28 years of our study the date of break-up of ice on the Kashunuk River ranged from 10 May to 10 June and averaged 27 May (± 1.5 days SE). The mean date that uplands were 90% snow-free was 10 May, the earliest 24 April in 1997, and the latest 23 May in 1985. The onset of the frost-free period and the date that maximum daily temperature exceeded 10 °C also varied by 3–4 weeks over the years. Once daytime temperatures exceeded 10 °C, there was a rapid transition from winter to spring, with the landscape changing from solid white to a changing patchwork of white and brown as snow melted and the previous year’s standing dead vegetation became exposed. Generally, lowlands were flooded by meltwater soon after uplands were mostly free of snow,


301

although such flooding was unusually delayed by 10 days in 1977 when cold temperatures returned after the snow melted.

From 1977 to 1998, the years for which data were available for almost all measures (Table 2), indices of local habitat conditions were

significantly correlated. The date of river break-up was especially highly correlated (p < 0.001) with local measures of weather. Most local spring conditions were only weakly correlated with larg-er-scale weather indices. The only significant relationship between any local metric of phenol-ogy and the three broad climate indices (ENSO, PDO, and May sea-surface temperature) was a negative relationship between the timing of river break-up and May sea surface temperatures (r2 = 0.35, p = 0.015, adjusted for 15 comparisons).

timing oF shoRebiRd aRRivalThe date of first arrival varied considerably from year to year (F27,318 = 20.02, p < 0.0001; Figure 3) and from species to species (F11,318 = 31.14, p < 0.0001; Table 3). The Black-bellied Plover was nearly always the first to arrive, whereas the Long-billed Dowitcher (Limnodromus scol-opaceus), Pectoral Sandpiper (Calidris melanotos), and Red Phalarope (Phalaropus fulicarius) were always among the last to arrive (Table 3, Figure 4). The spread of first arrival dates over the 28 years

table 1. Dates of measures of the arrival of spring and arrival of investigators on the outer Yukon–Kuskokwim Delta, Alaska, 1977–2008a.

Parameter Mean SE (d) Min Max

Arrival of investigators 30 April 1.2 16 April 11 MayKashunuk River break-upa

27 May 1.5 10 May 10 June

Uplands ≤10% snow cover

10 May 2.0 24 April 23 May

Onset frost-free period 25 May 2.5 10 May 16 JuneFirst temperature ≥10 °C

16 May 2.4 2 May 28 May

Flooding of lowlands 14 May 1.8 4 May 25 MayaObservation period 1977–2008 for Kashunuk River break-up. No data for 1980–1982, break-up and flooding data only for 1983, and break-up only for 1984. No data for 2007.

table 2. Strength of relationships (r2) between environmental variables at the Old Chevak study site, Yukon–Kuskokwim Delta, Alaska, 1977–1998.

VariableKashunuk River

break-upSnow cover

<10%Onset

frost-freeaFirst temp.

≥10 °CFlooding

of lowlands

Kashunuk River break-up — 0.790***b 0.663*** 0.616*** 0.878***Snow cover <10% — 0.532** 0.590** 0.858***Onset frost-free — 0.322* 0.525**First temperature ≥10 °C — 0.756***Flooding of lowlands —aDate after which temperature continuously >0 °C.bDate of Kashunuk River break-up based on data from 1977 to 2008; of all other variables based on data from 1977 to 1998; * = p < 0.05; ** = p < 0.01; *** = p < 0.001. Bonferroni-adjusted threshold is p = 0.0125 (0.05/4).

FiguRe 2. Typical spring conditions and habitats used by shorebirds arriving on the Yukon–Kuskokwim Delta, Alaska, 1977–2008. (A) Bar-tailed Godwits (upper) and Dunlin (lower) flying over emerging lowland habitat; (B) early-arriving Dunlin roosting on a thermokarst plateau.

302


ranged from 15 days in the Bar-tailed Godwit and Western Sandpiper to 34 days in the Rock Sandpiper. The wide variation among species in levels of interannual variability was due primar-ily to very early arrival by some species in some years rather than extremely delayed first arrivals. The range in the Dunlin’s arrival dates would have been only 16 days except for an extremely early sighting on 23 April in 1996, a week ear-lier than in any other year. Species with broader variability (21–22 days) in first arrival, including the Black-bellied Plover, Black Turnstone, and

Ruddy Turnstone, arrived in late April, nearly a week earlier than the Bar-tailed Godwit, Western Sandpiper, and (except for 1996) Dunlin.

Mean arrival date on the Y-K Delta in spring showed only a marginal relationship with distance of migration from the midpoint of the nonbreed-ing range (Spearman’s r = 0.42, p < 0.10, Table 3) and no discernible relationship with either the northern (r = 0.33) or southern (r = 0.39) limit. This lack of a relationship is exemplified by the similar arrival times of the Bar-tailed Godwit and Rock Sandpiper, despite the fact that the godwit

Year1976 1980 1984 1988 1992 1996 2000 2004 2008

20

1

10

Black-bellied PloverBlack TurnstoneWestern SandpiperDunlinRed-necked PhalaropeBar-tailed Godwit

Apr

ilM

ay

A

1976 1980 1984 1988 1992 1996 2000 2004 2008 20

1

10

20

30

10

Kashunuk River break-upSnow cover 10%Onset frost-free periodFirst 10 oC maximumFirst flooding

Apr

ilM

ayJu

ne

B20

FiguRe 3. (A) Indices of the arrival of spring, illustrating the covariation of arrival with local environmental variables on the outer Yukon–Kuskokwim Delta, Alaska, 1977–1998; data on river break-up through 2008; no data for 1980, 1981, 1982, and 2007, break-up and flooding data only for 1983, and break-up data only for 1984. (B) Annual variation in date of first arrival of six representative (most common and abundant) species of shorebirds. Dashed lines indicate years in which species was present when investigators arrived.


303

is a long-distance migrant and the sandpiper a short-distance migrant (Table 3). Indeed, Bar-tailed Godwits, which winter much farther from the Y-K Delta than Rock Sandpipers, appeared to track conditions on the breeding ground better (r2 = 0.49 vs. 0.12 for godwits and sandpipers, respectively, for relationship between arrival time and timing of river break-up; Table 4).

Relationship between enviRonmental vaRiables and aRRival timesThe arrival of seven species of shorebirds to the Y-K Delta was strongly correlated with at least four of the five environmental variables that we monitored (Table 4; Figure 3). Rock Sandpiper arrival was correlated with only one variable, while for three species (Ruddy Turnstone, Pectoral Sandpiper, and Red Phalarope) arrival dates were not correlated with any of the five variables. In most years, shorebirds arrived at about the time upland areas retained 10% snow cover (Figure 3). There was no relationship between timing of arrival of any species and broader climate indices

such as the PDO, ENSO, or May sea-surface tem-peratures (r2 < 0.22, p > 0.05).

tempoRal tRends in aRRival times and spRing phenologyThere were no significant trends over time in arrival date of any of the 12 species, that is, the slope of the regression line was not signifi-cantly different from 0 (p > 0.05) for any species (Figure 3). There was also no trend in the timing of break-up of the Kashunuk River (i.e., slope not different from 0: p > 0.05) over the course of our study (Figure 3). Although there was a short-term trend for progressively earlier river break-up from the mid-1980s to the late 1990s, extreme interan-nual variation before and after that period, as well as a series of consistently later springs around the turn of the century, resulted in no overall trend. However, if we include data on the timing of river break-up from years preceding our study (back to 1964; C. J. Lensink and C. P. Dau pers. comm.), then there has been a significant advancement in the timing of break-up of the Kashunuk River (r2

table 3. Timing of first arrival and characteristics of spring migration of shorebird species nesting on the outer Yukon–Kuskokwim Delta, Alaska, 1977–2008.

Nesting Migration

Species YearsaMean arrival

SE (days) Range

Relative abundanceb Distancec

Staging areas in south-central Alaskad

Black-bellied Plover 23 4.0 May 0.88 25 April–15 May Low 10.5 (2.3–16.0) PWS, KB, BBBlack Turnstone 27 5.7 May 0.92 26 April–17 May High 4.0 (1.1–7.0) PWS, KA, KB, BBDunlin 28 5.9 May 0.93 24 April–16 May High 4.5 (1.1–7.0) PWS, KB, CI, BBWestern Sandpiper 25 6.7 May 0.78 1–15 May High 8.0 (2.9–13.5) PWS, KB, CI, BBBar-tailed Godwit 27 6.9 May 0.61 2–16 May Low 6.5 (16.0) AA, BBRock Sandpiper 25 8.2 May 1.34 24 April–27 May Low 2.5 (1.1–3.5) PWS, KB, CI, BBWilson’s Snipe 26 8.6 May 1.00 30 April–20 May Low 5.5 (1.6–11.0) PWSRed-necked Phalarope 28 9.0 May 0.92 1–20 May High 11.0 (6.5–14.0) PWS, KB, BBRuddy Turnstone 26 9.1 May 0.95 27 April–17 May Low 10.5 (2.7–16.5); 4 AA, BB, PWSLong-billed Dowitcher 28 10.7 May 0.87 2–21 May Low 5.5 (2.8–10.0) PWS, BBPectoral Sandpiper 27 13.9 May 1.26 2–30 May Low 12.0 (4.5–18.5) PWS, BBRed Phalarope 28 19.7 May 1.34 8 May–4 June Low 11.5 (4.8–16.6) PWSaNumber of years observers were present before the species was first detected in spring (see Methods).bBased on estimates in McCaffery et al. (2012).cValue × 1000 km, for most species from the midpoint of the winter distribution to the Y-K Delta, with distances from northern and southern limits of distribution in parentheses (see Methods). For the Bar-tailed Godwit the distance is from from the midpoint of Yellow Sea staging grounds with the distance from the winter range in New Zealand in parentheses. For the Ruddy Turnstone the distance of 4000 km is from the Hawaiian Archipelago (Gill et al. 2002). Values are rounded to the nearest 100 km if total distance is <3000 km, to 500 km if distance >3000 km.

dStaging areas in south-central Alaska and distances to Y-K Delta study areas: AA (Aleutian Archipelago, 800–1200 km; Gibson and Byrd 2007, REG unpubl. data); PWS (Prince William Sound, including Copper River Delta, 1000 km; Isleib and Kessel 1973, Senner 1979, Iverson et al. 1996, Bishop and Warnock 1998, Takekawa and Warnock 2002, Warnock et al. 2004, REG unpubl. data); KA (Kodiak Archipelago, 800 km; R. A. MacIntosh and J. J. Withrow unpubl. data); KB (Kachemak Bay, 750 km; Matz et al. 2011, G. C. West unpubl. data); CI (Cook Inlet, 700 km; Gill and Tibbitts 1999, Ruthrauff et al. 2013); BB (Bristol Bay estuaries, 350–600 km; Gill and Handel 1990, Warnock et al. 2004, Savage and Payne 2013, REG unpubl. data).

304


= 0.279, p < 0.0001; n = 41). This advancement was highly correlated with a similar advancement over the same period in the timing of break-up of the Kuskokwim River on the inner Y-K Delta at Bethel (r2 = 0.520, p < 0.0001; n = 41).

weatheR at sites along the migRation Route in alaskaMay temperatures (based on heating degree days) at sites where migrating shorebirds may stop over in southern Alaska were highly correlated with the date

30

30

20

20

10

Dunli

n

Black

Turn

stone

Bar

-taile

dGo

dwit

Rock

Sand

piper

Wils

on's

Snipe

Rudd

yTu

rnsto

ne

Red-

neck

edPh

alaro

pe L

ong-

billed

Dowi

tcher

Pecto

ral

San

dpipe

r

R

edPh

alaro

pe

Wes

tern

Sand

piper

Blac

k-be

llied

Plov

er

Species

April

May

June

FiguRe 4. Box plots of dates of first arrival of shorebird species on the outer Yukon–Kuskokwim Delta, Alaska, 1977–2008. Boxes indicate the boundaries of the first and third quartiles, horizontal lines the median, whiskers the range, except for asterisks, the outliers.

table 4. Strength of relationships (r2) between timing of first arrival of shorebird species and environmen-tal conditions on the outer Yukon–Kuskokwim Delta, Alaska, 1977–2008a.

Species Break-up

Uplandsnow cover

≤10%Onset of frost

free periodFirst maximum

temperature ≥10 °CFirst flooding of lowlands

Black-bellied Plover 0.377 0.478 0.516 0.514 0.419Bar-tailed Godwit 0.489 0.379 0.287 0.310 0.394Ruddy Turnstone 0.022 0.006 0.020 0.001 0.020Black Turnstone 0.174 0.435 0.334 0.134 0.396Dunlin 0.294 0.548 0.208 0.143 0.435Rock Sandpiper 0.116 0.064 0.088 0.448 0.014Pectoral Sandpiper 0.096 0.035 0.009 0.054 0.076Western Sandpiper 0.525 0.546 0.447 0.296 0.387Long-billed Dowitcher 0.288 0.468 0.238 0.283 0.319Wilson’s Snipe 0.364 0.358 0.316 0.231 0.249Red-necked Phalarope 0.495 0.550 0.389 0.356 0.473Red Phalarope 0.002 0.045 0.005 0.000 0.015aSnow-cover, flooding, and temperature data for 1977–1998 (n = 17 years); break-up = date when river ice first moved at Old Chevak or at our field site 7 km down river (Figure 1); onset of frost-free period = date temperature was first continuously >0 °C. Light, medium, and dark shading indicate p < 0.05, p < 0.01, and p < 0.001, respectively. For multiple comparisons within a species, Bonferroni-corrected p values are α/5 (the number of comparisons per species).


305

of break-up of the Kashunuk River. The strength of this relationship (r2) was highest for sites closest to the study area such as Bethel (0.74), Dillingham (0.64), and King Salmon (0.68; Figure 1), but still highly significant for areas more distant, including Cold Bay (0.60), Kodiak (0.46), Homer (0.42), and Cordova (0.36). May temperatures at these sites were also good predictors of the time of shorebirds’ arrival on the outer Y-K Delta, as indicated by the significant correlations of May temperatures (heat-ing degree days) with these arrival times (Table 5).

timing oF aRRival Relative to abundanCe and the timing oF nestingLocal population size had no obvious effect on detectability. Correlations between local popula-tion size and mean arrival dates, the standard error of mean arrival dates, and the span of dates between earliest and latest first arrivals for each species generated r2 values of just 0.11, 0.11, and 0.07, respectively. We found positive correlations between the date of first arrival and the date of first nest initiation for all five shorebird species considered, including the Bar-tailed Godwit (r2 = 0.99, n = 3 years), Black Turnstone (r2 = 0.88, n = 5), Dunlin (r2 = 0.81, n = 4), Rock Sandpiper (r2 = 0.80, n = 6), and Western Sandpiper (r2 = 0.89, n = 7). We also found positive correlations between first arrival and mean nest initiation for the three species for which we had that measure, including the Bar-tailed Godwit (r2 = 0.99, n = 3), Dunlin (r2 = 0.99, n = 3), and Western Sandpiper (r2 = 0.56, n = 6).

Shorebirds did not nest immediately after arrival but spent considerable time on the study area before initiating nests, even in years of early

snowmelt. The average time lag (in days) from first arrival to first nest initiation was 14 days for the Bar-tailed Godwit (range 13–15, n = 29 nests over 3 years), 18 days for the Black Turnstone (range 12–23, n = 110 nests over 5 years), 13 days for the Dunlin (range 9–15, n = 147 nests over 4 years), 16 days for the Rock Sandpiper (range 11–20, n = 52 nests over 6 years), and 12 days for the Western Sandpiper (range 9–14, n = 381 nests over 7 years). Time lags between first arrival and mean nest ini-tiation were 21 days (range 19–23, n = 3 years) for the Bar-tailed Godwit, 19 days (range 18–20, n = 3 years) for the Dunlin, and 26 days (range 21–30, n = 6 years) for the Western Sandpiper.

disCussion

Reliability and RelevanCe oF FiRst obseRvations as a metRiC oF timing oF bReedingOur use of the date of first arrival as a metric for characterizing the timing of shorebirds’ breed-ing season implied the assumption that early first arrivals corresponded with earlier nesting in the overall local population. How valid was that assumption? To begin with, the first individual detected obviously may not necessarily have been the first individual to arrive. Although an obser-vation of a bird confirms its presence, failure to detect a bird does not confirm its absence, because the reliability of observational data is dependent on species-specific detection rates (Thompson 2002). Hence, because we do not know the prob-abilities of detection for each species, we do not know with complete confidence when each spe-cies of shorebird first arrived at our study area.

table 5. Strength of relationships (r2) between weather in May (heating degree days) at potential sites of spring stopover in Alaska and timing of arrival of shorebird species on the outer Y-K Delta, Alaska, 1977–2008a.

Species Cordova Anchorage Homer Kodiak Dillingham King Salmon Cold Bay

Black-bellied Plover 0.312 0.146 0.406 0.274 0.626 0.470 0.292Bar-tailed Godwit 0.342 0.267 0.283 0.374 0.416 0.407 0.361Ruddy Turnstone 0.088 0.041 0.041 0.048 0.046 0.010 0.027Black Turnstone 0.193 0.125 0.260 0.181 0.310 0.236 0.111Dunlin 0.242 0.236 0.152 0.210 0.251 0.208 0.192Rock Sandpiper 0.301 0.101 0.258 0.068 0.196 0.107 0.018Pectoral Sandpiper 0.012 0.074 0.029 0.055 0.040 0.071 0.000Western Sandpiper 0.323 0.329 0.308 0.250 0.435 0.365 0.276Long-billed Dowitcher 0.180 0.144 0.248 0.117 0.260 0.223 0.152Wilson’s Snipe 0.200 0.144 0.151 0.114 0.255 0.231 0.253Red-necked Phalarope 0.222 0.184 0.252 0.253 0.414 0.334 0.247Red Phalarope 0.061 0.008 0.164 0.136 0.058 0.045 0.018

aLight, medium, and dark shading indicates p < 0.05, p <0.01, and p <0.001, respectively. For multiple comparisons within a species, Bonferroni-corrected p values are α/7 (the number of comparisons per species).

306


Nonetheless, the wide-open habitats and con-spicuous behaviors of shorebirds at that time of year (Figure 2) certainly increase the probability of early detection and reduce the probability of systematic biases among species in reported arrival dates. Other studies have indicated, however, that population size is a major factor contribut-ing to differences in detectability (Tryjanowski et al. 2005, Lehikoinen and Sparks 2010, Pearce-Higgins and Green 2014). In our study area, the species of shorebirds varied in abundance by an order of magnitude (McCaffery et al. 2012). If local abundance was associated with detectability, we might have expected the observed arrival dates of the more abundant species to be earlier and/or less variable. Our data, however, do not support either hypothesis. Neither the timing of arrival nor annual variation in arrival times was obviously associated with local abundance.

A second objection to using data on first arriv-als to characterize the timing of breeding is that early arrivers could be isolated and essentially random outliers providing little or no predictive information about population-level phenomena. If that were the case, one would expect little or no correlation between first arrival data and the tim-ing of subsequent breeding. That is not, however, what we documented in this study. We found a very strong correlation between first arrival and time of nesting for all five species for which we had at least three years of data on nesting chro-nology. Given the strength of these correlations and the overall conspicuousness of shorebirds in the stark two-dimensional habitats prevalent on the Y-K Delta, we are confident that first arrival dates provide a useful index of the chronology of shorebirds’ local breeding.

what aRe the signs oF spRing?In the Arctic, the breakup of rivers in spring repre-sents the most significant hydrological event of the year with respect to freshwater, nutrient, and sedi-ment dynamics (Beck et al. 2013). Our finding of a strong positive relationship between the timing of break-up of the Kashunuk River and other environmental variables, including percent snow cover and ambient temperatures during April and May (Table 2), indicates that the timing of the river’s break-up is also a good metric for evaluat-ing the onset of spring. Other authors have found this date to reflect seasonal variation in regional climate, including Bieniek et al. (2011), who reported that the timing of break-up of Alaska rivers was highly correlated with surface tempera-tures during April and May, and Muhammad et

al. (2016), who found that timing of break-up of the Mackenzie River in northern Canada was highly correlated with timing of local snowmelt. The analysis of Bieniek et al. (2011) included the Kuskokwim River at Bethel, where during the years of our study (1977–2008), the number of heating degree days in April was strongly related to the date of break-up (r2 = 0.71; p < 0.001). Hence our use of the timing of break-up of the Kashunuk River as an index for spring phenology seems warranted, even in the years when we did not record other environmental data.

inFluenCe oF enviRonmental vaRiables on aRRival timesOur finding that shorebirds’ arrival was highly correlated with local weather variables (Table 4) has been reported elsewhere in the Arctic (e.g., Gunnarsson and Tómasson 2011, Ward et al. 2015). In addition, the high degree of correlation among the local weather variables on the Y-K Delta suggests that migrants had consistent signals of spring. However, larger-scale environmental indices did not appear to be correlated with the timing of shorebird migration. This result paral-lels the findings of Ward et al. (2015) in north-ern Alaska, and of Boyd and Petersen (2006) in Iceland, who reported that the arrival times of less than half the shorebird species migrating to Iceland varied with the phases of the North Atlantic Oscillation. The timing of shorebird migration may be under environmental controls different from those of other species, however. At lower latitudes in North America, the schedule of songbird migration has been found to vary with large-scale environmental indices (MacMynowski and Root 2007, MacMynowski et al. 2007). On the Y-K Delta, however, local weather and migra-tion timing are likely not completely independent of larger-scale phenomena, as evidenced by the delayed snowmelt and delayed arrival of shore-birds in 1992, which were probably affected by ash jettisoned into the atmosphere by the erup-tion of Mt. Pinatubo in 1991 (Ganter and Boyd 2000).

FaCtoRs inFluenCing timing oF aRRival and bReedingWe found that, regardless of migration distance, the birds we studied altered their arrival times to compensate for annual variation in the timing of spring thaw. This result contrasts with other findings that migration distance has a signifi-cant effect on migration timing, although there is disagreement over whether the relationship is


307

positive or negative. Some authors have reported that in warmer springs short-distance migrants are more likely to advance the timing of migra-tion than are species migrating greater distances (Murphy-Klassen et al. 2005, Gunnarsson and Tómasson 2011, Gill et al. 2013, Stutzman and Fontaine 2015), whereas others have reported that long-distance migrants are better able than short-distance migrants to advance migration tim-ing (Jonzén et al. 2006). Many of these studies, however, were conducted at breeding areas in the North Temperate Zone, or at mid-latitude staging areas, and were in general not evaluating arrival in the Arctic. Birds at lower latitudes or at earlier stages of their migration may be under controls different from those at high latitudes or nearing the end of their migration (citations in Newton 2008). It appears to us that shorebirds nesting on the Y-K Delta have considerable flexibility in adjusting the timing of their migrations, especially as they near Alaska.

Our finding no effect of distance on timing of migration is likely related to the fact that in spring most of the species we studied stop in south-central Alaska before moving to the Y-K Delta to breed (Table 3). The principal sites of such staging include the Copper River Delta, Prince William Sound, Cook Inlet (Kachemak and Redoubt bays), the Kodiak Archipelago, and Bristol Bay estuaries, which collectively lie between 600 and 1000 km from the Y-K Delta (Figure 1). These sites provide not only abundant food resources for replenishing reserves after long migrations, but because of their proximity within the flyway to the breeding grounds they may provide environmen-tal cues that allow the birds near the end of their migrations to assess conditions on the breeding grounds. Indeed, the strong correlation we found between environmental conditions on the breed-ing grounds and along the northern portion of the migration route points to a mechanism by which migrants can judge whether to continue migrating or wait until conditions are more favorable (e.g. warmer temperatures, more beneficial winds).

Some of our strongest evidence for this flexibil-ity comes from satellite- and VHF-tagged individ-uals. For example, in some years satellite-tagged Bar-tailed Godwits migrating 6800 km nonstop from the Yellow Sea to the Y-K Delta (Battley et al. 2012), Bristle-thighed Curlews (Numenius tahitiensis) migrating nonstop from at least the Hawaiian Archipelago (T. L. Tibbitts and REG unpubl. data), and Whimbrels (Numenius pha-eopus) migrating 4000 km nonstop from Mexico fly directly to their Alaska breeding grounds. In

other years, the same individuals stop briefly in the Aleutian Islands, at Alaska Peninsula estuaries, or in Cook Inlet before continuing to the breed-ing grounds.

Other evidence of a flexible migration strategy comes from VHF telemetry of small sandpipers, of which the two that breed most abundantly on the Y-K Delta, the Dunlin and Western Sandpiper, migrate up the Pacific coast of North America in short hops each spring (e.g., Iverson et al. 1996, Takekawa and Warnock 2000, Warnock et al. 2004). Most birds stop briefly at sites in south-central Alaska, particularly the Copper River Delta, before flying either directly to the Y-K Delta or in some instances stopping in Cook Inlet or Bristol Bay (e.g., Bishop and Warnock 1998, Gill and Tibbitts 1999, Warnock et al. 2004, Savage and Payne 2013, REG unpubl. data). At these Alaska stopover sites, birds are likely able to pace their migration by evaluating cues of spring as they draw nearer to their breeding grounds.

What remains unclear is whether this strategy is restricted to the American Pacific Flyway or if it is widespread in the Northern Hemisphere. It appears that in regions where weather conditions at distant staging areas are not correlated with conditions on the breeding grounds or where birds have to cross vast ecological barriers and do not have stopover sites near the breeding grounds, birds generally do not show a pheno-logical response to variation in conditions on the breeding grounds, but instead time their arrival on the basis of long-term average conditions (Piersma et al. 1990, Boyd and Petersen 2006, Meltofte et al. 2007, Smith et al. 2010, Gunnarsson and Tómasson 2011).

It is apparent from our study that flexibility in migration schedules, especially as birds near the breeding grounds, contributes to breeding at the optimal time, but there is also evidence that post-arrival adaptations may also facilitate timely breeding. On the Y-K Delta, we found that the lag between first arrival on one hand and first and mean nest initiations on the other averaged about 2 and 3 weeks, respectively. Such an extended interval may provide shorebirds an opportunity to calibrate the initiation of breeding even more finely than can be achieved just by their flexible migration strategies. In addition, in the tundra of western and northern Alaska, the breeding season is long enough for shorebirds to lay replacement clutches when first clutches have been depre-dated (McCaffery and Ruthrauff 2004, Johnson et al. 2009, Gates et al. 2013), and in some cases, even to double-brood (Jamieson 2011). Even

308


at these latitudes, however, there is apparently still a premium on being able to produce addi-tional clutches quickly so that hatching coincides with ample prey for the young (see below). For example, at Barrow, Alaska, a Dunlin can lay the first egg of a replacement clutch in as few as two days following the loss of a prior clutch, and the average interval is 5–7 days (Gates et al. 2013); on the Y-K Delta, the average is about 9 days (Jamieson 2011). Similarly, Bar-tailed Godwits and Western Sandpipers on the Y-K Delta lay replacement clutches in 5–6 days (Woodley 2009, BJM unpubl. data).

maintaining the matChMany studies of animals of both the Arctic and North Temperate Zone have documented that some migrants are unable to alter their migration and breeding times to coincide with advances in the timing of food availability in breeding areas induced by global warming (Visser et al. 1998, 2004, Durant et al. 2007, Post et al. 2008, Both 2010). Most of these studies, including investiga-tions of shorebirds breeding in the Arctic (Tulp and Schekkerman 2008, McKinnon et al. 2012), have focused on monitoring peaks of prey avail-ability relative to the timing of hatch of young. However, for income breeders, migratory species that rely on local food resources for reproduction (Drent and Daan 1980), there are two pulses of food availability that must be tracked for optimal breeding performance: food for breeding adults upon arrival, and food for young at the time of hatching. Our data show that shorebirds breeding on the Y-K Delta have been able to adjust their migration schedules to arrive when breeding habi-tats are favorable for foraging during the prenest-ing period, but it is also likely that their arrival is timed so that young hatch when the density of food is greatest (McKinnon et al. 2012). This is probable because the same environmental factors (especially snow cover and temperatures) govern-ing the emergence of nesting habitat (vegetation) and prey (insects; Emmenegger et al. 2014) are also strongly correlated with when birds make their final flights to breeding areas and start nesting. The results of our study thus support the assertion by Gwinner (1996) that circannual rhythms are not fixed but “allow birds to time seasonal activi-ties like migration or reproduction in temporally constant or unpredictable environments.” The flexibility demonstrated in the species we studied may become increasingly important during this era of rapid global climate change. Indeed, a species’ current range of flexibility in scheduling migration

and breeding may help us predict the likelihood of its being able to adapt to dramatically changing conditions in the decades ahead.

aCknowledgmentsWe thank the U.S. Fish and Wildlife Service, the U.S. Geological Survey, and the University of California, Davis, for funding this long-term work. The collection of these data would not have happened without the insight and support of C. P. Dau and C. J. Lensink, inspirational biolo-gists who worked for the YDNWR in the 1970s. We appreciate the logistic support provided by YDNWR during our study, and the help of the people of the village of Chevak, who shared their knowledge with us, and allowed us to make obser-vations in their “backyard.” We are indebted to numerous biologists who helped collect data over the course of nearly three decades, including B. J. Aldrich, T. Aldrich, C. Babcock, D. Budeau, C. P. Dau, M. Dementyev, A. Fowler, C. M. Handel, C. M. Harwood, S. Jamieson, M. Johnson, D. Mather, J. Morse, T. H. Pogson, D. Raveling, D. R. Ruthrauff, J. S. Sedinger, P. S. Tomkovich, and M. Wege. Reviews of earlier drafts by R. B. Lanctot and D. R. Ruthrauff were most helpful.

liteRatuRe CitedAlerstam, T., and Hogstedt, G. 1980. Spring predict-

ability and the evolution of leap-frog migration. Or-nis Scand. 11:196–200; doi 10.2307/3676124.

Babcock, C. A., and Ely, C. R. 1994. Classification of vegetation communities in which geese rear broods on the Yukon–Kuskokwim Delta, Alaska. Can. J. Bot. 72:1294–1301; doi 10.1139/b94-158.

Babcock, C. A., Fowler, A. C., and Ely, C. R. 2002. Nesting ecology of Tundra Swans on the coastal Yukon–Kuskokwim Delta, Alaska. Waterbirds 25 (Spec. Publ. 1): 236–240.

Barry, T. 1962. The effect of late seasons on Atlantic Brant reproduction. J. Wildl. Mgmt. 26:19–26.

Battley, P. F., Warnock, N., Tibbitts, T. L., Gill, R. E., [Jr.], Piersma, T., Hassell, C. J., Douglas, D. C., Mulcahy, D. M., et al. 2012. Contrasting extreme long-distance migration patterns in Bar-tailed God-wits Limosa lapponica. J. Avian Biol. 43:21–32; doi 10.1111/j.1600-048X.2011.05473.

Beck, R. A., Hinkel, K. M., Eisner, W. R., Whiteman, D., Arp, C. D., Machida, R., Cuomo, C., Liu, H., et al. 2013. Contrasting historical and recent break-up styles on the Meade River of Arctic Alaska in the context of a warming climate. Am. J. Clim. Change 2:165–172; doi 10.4236/ajcc.2013.22016.

Bieniek, P., Bhatt, U. S., Rundquist, L. A., Lindsey, S. D., Zhang, X., and Thoman, R. L. 2011. Large-scale cli-mate controls of interior Alaska river ice break-up. J. Clim. 24:286–297; doi 10.1175/2010JCLI3809.1.


309

Bintanja, R., and van der Linden, E. C. 2013. The changing seasonal climate in the Arctic. Sci. Rep. 3: article 1556; doi 10.1038/srep01556.

Bishop, M. A., and Warnock, N. 1998. Migration of Western Sandpipers: Links between their Alaskan stopover areas and breeding grounds. Wilson Bull. 110:457–624.

Bonier, F., Martin, P. R., Jensen, J. P., Butler, L. K., Ramenofsky, M., and Wingfield, J. C. 2007. Pre-migratory life history stages of juvenile Arctic birds: Costs, constraints, and trade-offs. Ecology 88:2729–2735; doi 10.1890/07-0696.1.

Both, C. 2010. Food availability, mistiming, and cli-matic change, in Effects of Climate Change on Birds (A. Møller, W. Fiedler, and P. Berthold, eds.), pp. 129–147. Oxford Univ. Press, New York.

Both, C., and Visser, M. E. 2001. Adjustment to cli-mate change is constrained by arrival date in a long-distance migrant bird. Nature 411:296–298; doi 10.1038/35077063.

Boyd, H., and Petersen, A. 2006. Spring arrivals of mi-grant waders in Iceland in the 20th century. Ringing and Migration 23:107–115; doi 10.1080/03078698. 2006.9674353.

Clausen, K. K., and Clausen, P. 2013. Earlier Arctic springs cause phenological mismatch in long-distance migrants. Oecologia 173:1101−1102; doi 10.1007/s00442-013-2681-0.

Conservation of Arctic Flora and Fauna (CAFF). 2001. Arctic flora and fauna: Status and conservation. Edita, Helsinki; library.arcticportal.org/1295/.

Cooke, W. W. 1904. Some facts about the migration of birds. Condor 6:115−126.

Drent, R. H., and Daan, S. 1980. The prudent par-ent: Energetic adjustments in avian breeding. Ardea 68:225−252; doi 10.5253/arde.v68.p225.

Durant J. M., Hjermann, D. Ø., Ottersen, G., and Stens-eth, N. C. 2007. Climate and the match or mismatch between predator requirements and resource availabil-ity. Clim. Res. 33:271–283; doi 10.3354/cr033271.

Ely, C. R., and Raveling, D. G. 1984. Breeding biol-ogy of Pacific White-fronted Geese. J. Wildl. Mgmt. 48:823–837.

Emmenegger, T., Hahn, S., and Bauer, S. 2014. Indi-vidual migration timing of Common Nightingales is tuned with vegetation and prey phenology at breeding sites. BMC Ecology 14: article 9; doi 10.1186/1472-6785-14-9.

Ganter, B., and Boyd, H. 2000. A tropical volcano, high predation pressure, and the breeding biology of Arctic waterbirds: A circumpolar review of breeding failure in the summer of 1992. Arctic 53:289–305.

Gates, H. R., Lanctot, R. B., and Powell, A. N. 2013. High renesting rates in Arctic-breeding Dunlin (Calidris alpina): A clutch removal experiment. Auk 130:372−380; doi 10.1525/auk.2013.12052.

Gerasimov, Yu. N., and Gerasimov, N. N. 2000. Infor-mation on the northward migration of the Great Knot Calidris tenuirostris in Kamchatka, Russia. Stilt 36:35−38.

Gibson, D. D., and Byrd, G. V. 2007. Birds of the Aleu-tian Islands, Alaska. Nuttall Ornithol. Club and Am. Ornithol. Union Ser. Ornithol. 1.

Gill, J. A., Alves, J. A., Sutherland, W. J., Appleton, G. F., Potts, P. M., and Gunnarsson, T. J. 2013. Why is timing of bird migration advancing when indi-viduals are not? Proc. R. Soc. B 281:20132161; doi 10.1098/rspb.2013.2161.

Gill, R. E., Jr., and Handel, C. M. 1990. The importance of subarctic intertidal habitats to shorebirds: A study of the central Yukon–Kuskokwim Delta, Alaska. Con-dor 92:702–725.

Gill, R. E., Jr., McCaffery, B. J., and Tomkovich, P. S. 2002. Wandering Tattler (Heteroscelus incanus), in The Birds of North America (A. Poole and F. Gill, eds.), no. 642. Birds N. Am., Philadelphia; doi 10.2173/bna.642.

Gill, R. E., Jr., and Tibbitts, T. L. 1999. Seasonal shore-bird use of intertidal habitats in Cook Inlet, Alaska. Unpubl. rep., USDI, USGS, Biological Resources Di-vision and OCS Study, MMS 99-0012; www.alaska.usgs.gov/science/biology/shorebirds/pdfs/GillTib-bitts_Shorebird%20use%20CookInlet_MMS99.pdf.

Gunnarsson, T. G., and Tómasson, G. 2011. Flexibility in spring arrival of migratory birds at northern latitudes under rapid temperature changes. Bird Study 58:1–12; doi 10.1080/00063657.2010.526999.

Gwinner, E. 1996. Circannual clocks in avian re-production and migration. Ibis 138:47–63; doi 10.1111/j.1474-919X.1966.tb04312.x.

Handel, C. M. 2002. Breeding ecology of the Black Turnstone: Hedging bets in a variable subarctic en-vironment. Ph.D. diss., Univ. of California, Davis.

Handel, C. M., and Gill R. E., Jr. 2000. Mate fidelity and breeding site tenacity in a monogamous sandpiper, the Black Turnstone. Anim. Behav. 60:471−481; doi 10.1006/anbe.2000.1505.

Holmes, R. T. 1971a. Latitudinal differences in the breed-ing and molt schedules of Alaskan Red-backed Sand-pipers (Calidris alpina). Condor 73: 93–99.

Holmes, R. T. 1971b. Density, habitat, and the mating system of the Western Sandpiper (Calidris mauri). Oecologia. 7:191−208.

Intergovernmental Panel on Climate Change (IPCC). 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Core Writing Team, R. K. Pachauri and L. A. Meyer, eds.). IPCC, Geneva, Switzerland; www.ipcc.ch/report/ar5/syr/.

Isleib, M. E., and Kessel, B. 1973. Birds of the North Gulf Coast–Prince William Sound region, Alaska. Bio. Pap. Univ. Alaska 14.

Iverson, G. C., Warnock, S. E., Butler, R. W., Bishop, M. A., and Warnock, N. 1996. Spring migration of Western Sandpipers along the Pacific coast of North America: A telemetry study. Condor 98:10–21.

Jamieson, S. E. 2011. Pacific Dunlin Calidris alpina pa-cifica show a high propensity for second clutch pro-duction. J. Ornithol. 152:1013–1021; doi 10.1007/s10336-011-0691-4.

310


Johnson, M., Conklin, J. R., Johnson, B. L., McCaffery, B. J., and Haig, S. M. 2009. Behavior and reproduc-tive success of Rock Sandpipers breeding on the Yu-kon–Kuskokwim River Delta, Alaska. Wilson Bull. 121:328–337; doi 10.1676/08-010-1.

Johnson, M., and Walters, J. R. 2008. Effects of mate and site fidelity on nest survival of Western Sandpip-ers (Calidris mauri). Auk 125:76–86; doi 10.1525/auk.2008.125.1.76.

Jonzén, N., Lindén, A., Ergon, T., Knudsen, E., Vik, J. O., Rubolini, D., Piacentini, D., Brinch, C., et al. 2006. Rapid advance of spring arrival dates in long-distance migratory birds. Science 312:1959–1961; doi 10.1126/science.1126119.

Jorgenson, T., and Ely, C. R. 2001. Topography and flooding of coastal ecosystems on the Yukon–Kus-kokwim Delta, Alaska: Implications for sea-level rise. J. Coastal Res. 17:124–36.

Knudsen, E., Lindén, A., Both, C., Jonzén, N., Pulido, R., Saino, N., Sutherland, W. J., Bach, L. A., et al. 2011. Challenging claims in the study of migratory birds and climate change. Biol. Rev. 86:928–946; doi 10.1111/j.1469-185X.2011.00179.x.

Kokko, H. 1999. Competition for early arrival in mi-gratory birds. J. Anim. Ecol. 68:940–950; doi 10.1046/j.1365-2656.1999.00343.x.

Lack, D. 1968. Ecological Adaptations for Breeding in Birds. Methuen and Co., Ltd., London.

Lanctot, R. B., Sandercock, B. K., and Kempenaers, B. 2000. Do male breeding displays function to attract mates or defend territories? The explanatory role of mate and site fidelity. Waterbirds 23:155–164.

Lank, D. B., and Ydenberg, R. C. 2003. Death and danger at migratory stopovers: Problems with “predation risk.” J. Avian Biol. 34:225–228; doi 10.1034/j.1600-048X.2003.03250.x.

Lehikoinen, E., and Sparks, T. H. 2010. Changes in migration, in Effects of Climate Change on Birds (A. Møller, W. Fiedler, and P. Berthold, eds.), pp. 89–112. Oxford Univ. Press, New York.

Liebezeit, J. R., Gurney, K. E. B., Budde, M., Zack, S., and Ward, D. 2014. Phenological advancement in Arctic bird species: Relative importance of snow melt and ecological factors. Polar Biol. 37:1309–1320; doi 10.1007/s00300-014-1522-x.

Lincoln, F. C. 1939. The Migration of American Birds. Doubleday, Doran & Co., New York.

MacMynowski, D. P., and Root, T. L. 2007. Climate and the complexity of migratory phenology: Sexes, migratory distance, and arrival distributions. Int. J. Biometeorol. 51:361–373; doi 10.1007/s00484-006-0084-1.

MacMynowski, D. P., Root, T. L., Ballard, G., and Geupel, G. R. 2007. Changes in spring arrival of Nearctic-Neotropical migrants attributed to multi-scalar climate. Global Change Biol. 13:2239–2251; doi 10.1111/j.1365-2486.2007.01448.x.

Martin, K., and Wiebe, K. L. 2004. Coping mechanisms of alpine and Arctic-breeding birds: Extreme weather

and limitations to reproductive resilience. Integr. Comp. Biol. 44:177–185; doi 10.1093/icb/44.2.177.

Matz, G., Lanctot, R. B., West, G. C., Michaud, M., and the Kachemak Bay Birders. 2011. Reassess-ment of a Western Hemisphere Shorebird Reserve Network Site: Kachemak Bay, Alaska. Wader Study Group Bull. 119:9–16.

McCaffery, B. J., and Ruthrauff, D. R. 2004. How in-tensive is intensive enough? Limitations of intensive searching for estimating shorebird nest numbers. Wader Study Group Bull. 103:63–66.

McCaffery, B. J., Bart, J., Wightman, C., and Krueper, D. J. 2012. Shorebird surveys in western Alaska. Studies Avian Biol. 44:19–36.

McKinnon, L., Picotin, M., Bolduc, E., Juillet, C., and Bêty, J. 2012. Timing of breeding, peak food availabil-ity and effects of a mismatch on chick growth in birds nesting in the High Arctic. Can. J. Zool. 90:961–971; doi 10.1139/z2012-064.

Meltofte, H., Piersma, T., Boyd, H., McCaffery, B., Ganter, B., Golovnyuk, V. V., Graham, K., Gratto-Trevor, C. L., et al. 2007. Effects of climate variation on the breeding ecology of Arctic shorebirds. Med-delelser om Grønland Bioscience 59. Danish Polar Center, Copenhagen.

Møller, A. P. 2001. Heritability of arrival date in a migra-tory bird. Proc. R. Soc. London B 268:203–206; doi 10.1098/rspb.2000.1351.

Muhammad, P., Duguay, C., and Kang, K.-K. 2016. Monitoring ice break-up on the Mackenzie River using MODIS data. Cryosphere 10:569–584; doi 10.5194/tc-10-569-2016.

Murphy-Klassen, H. M., Underwood, T. J., Sealy, S. G., and Ashleigh, A. C. 2005. Long-term trends in spring arrival dates of migrant birds at Delta Marsh, Mani-toba, in relation to climate change. Auk 122:1130–1148; doi 10.1642/0004-8038(2005)122[1130:LTISAD]2.0.CO;2.

Newton, I. 2008. The Migration Ecology of Birds. Aca-demic Press, London.

Pearce-Higgins, J. W., and Green, R. E. 2014. Birds and climate change: Impacts and conservation responses. Cambridge Univ. Press, Cambridge, England.

Pearce-Higgins, J. W., Yalden, D. W., and Whittingham, M. J. 2005. Warmer springs advance the breeding phenology of Golden Plovers Pluvialis apricaria and their prey (Tipulidae). Oecologia 143:470–476; doi 10.1007/s00442-004-1820-z.

Piersma, T., Klaassen, M., Bruggemann, J. H., Blomert, A.-M., Gueye, A., Ntiamoa-Baidu, Y., and van Bred-erode, N. E. 1990. Seasonal timing of the spring de-parture of waders from the Banc d’Arguin, Maurita-nia. Ardea 78:123−134.

Pitelka, F. A. 1979. Introduction: The Pacific Coast shorebird scene. Studies Avian Biol. 2:1–11.

Post, E., Pedersen, C., Wilmers, C. C., and Forchham-mer, M. C. 2008. Phenological sequences reveal ag-gregate life history response to climatic warming. Ecology 89:363–370; doi 10.1890/06-2138.1.


311

Price, T., Kirkpatrick, M., and Arnold, S. J. 1988. Di-rectional selection and the evolution of breeding date in birds. Science 240:798–799; doi 10.1126/science.3363360.

Root, T. L., Price, J. T., Hall, K. R., Schnedier, S. H., Rosenweig, C., and Pounds, J. A. 2003. Fingerprints of global warming on wild animals and plants. Nature 421:57– 60; doi 10.1038/nature01333.

Ruthrauff, D. R. 2002. Seasonal and age-related trends in the reproductive output of Western Sandpipers (Calidris mauri) at Kanaryaraq, Alaska. M.S. thesis, Humboldt State University, Arcata, CA.

Ruthrauff, D. R., Gill, R. E., Jr., and Piersma, T. 2013. Coping with the cold: An ecological context for the abundance and distribution of Rock Sandpipers during winter in upper Cook Inlet, Alaska. Arctic 66:269–278.

SAS Institute. 1999. SAS/STAT user’s guide. Release 6.04 edition. SAS Institute, Inc., Cary, NC.

Savage, S. E., and Payne, K. J. 2013. Shorebird temporal and spatial use patterns, Bristol Bay, April–October 2012. Unpubl. rep., Alaska Peninsula/Becharof Na-tional Wildlife Refuge, King Salmon, AK; www.arlis.org/docs/vol1/FWS/A/876052417.pdf.

Senner, S. E. 1979. An evaluation of the Copper River Delta as a critical habitat for migrating shorebirds. Studies Avian Biol. 2:131–145.

Serreze, M. C., Walsh, J. E., Chapin, F. S., III, Oster-kamp, T., Dyurgerov, M., Romanovsky, V., Oechel, W. C., Morison, J., et al. 2000. Observational evi-dence of recent change in the northern high-latitude environment. Clim. Change 46:159–207; doi 10.1023/A:1005504031923.

Smith, P. A., Gilchrist, H. G., Forbes, M. R., Martin, J-L., and Allard, K. 2010. Inter-annual variation in the breed-ing chronology of Arctic shorebirds: Effects of weather, snowmelt and predators. J. Avian Biol. 41:292–304; doi 10.1111/j.1600-048X.2009.04815.x.

Sokal, R. R., and Rohlf, F. J. 1981. Biometry. W. H. Free-man and Co., San Francisco.

Stutzman, R. J., and Fontaine, J. F. 2015. Shorebird mi-gration in the face of climate change: Potential shifts in migration phenology and resource availability. Studies Avian Biol. 47:145−162.

Takekawa, J. T., and Warnock, N. 2000. Long-billed Dowitcher (Limnodromus scolopaceus), in The Birds of North America (A. Poole and F. Gill, eds.), no. 493. Birds N. Am., Philadelphia; doi 10.2173/bna.493.

Thompson, W. L. 2002. Towards reliable bird surveys: Accounting for individuals present but not detected. Auk 119:18−25; doi 10.1642/0004-8038(2002)119 [0018:TRBSAF]2.0.CO;2.

Tryjanowski, P., Kuzniak, S., and Sparks, T. H. 2005.

What affects the magnitude of change in first arrival dates of migrant birds. J. Ornithol. 146:200−205; doi 10.1007/s10336-005-0079-4.

Tulp, I., and Schekkerman, H. 2008. Has prey availability for Arctic birds advanced with climate change? Hind-casting the abundance of tundra arthropods using weather and seasonal variation. Arctic 61:48–60.

U.S. Department of Commerce. 1977−2008. National Centers for Environmental Information. Climatologi-cal data for Alaska. Asheville, NC; www.ncdc.noaa.gov/data-access/land-based-station-data.

Uspenskii, S. M. 1969. Life at High Latitudes: A Study of Bird Life. Mysl’ Publishers, Moscow (Translated in 1984, Amerind Publishing Company, New Delhi).

Visser, M. E., Both, C., and Lambrechts, M. M. 2004. Global climate change leads to mistimed avian repro-duction. Adv. Ecol. Res. 35:89–110; doi 10.1016/S0065-2504(04)35005-1.

Visser, M. E., van Noordwijk, A. J., Tinbergen, J. M., and Lessells, C. M. 1998. Warmer springs lead to mistimed reproduction in Great Tits (Parus ma-jor). Proc. R. Soc. London B 265:1867–1870; doi 10.1098/rspb.1998.0514.

Walther, G. R., Post, E., Convey, P., Menzel, A., Parmesan, R., Beebee, T. J. C., Fromentin, J-M., Hoegh-Guld-berg, O., and Bairlein, F. 2002. Ecological responses to recent climate change. Nature 416:389–395; doi 10.1038/416389a.

Ward, D. H., Helmericks, J., Hupp, J. W., McManus, L., Budde, M., Douglas, D. C., and Tape, K. D. 2015. Multi-decadal trends in spring arrival of avian migrants to the central Arctic coast of Alaska: Effects of environmental and ecological factors. J. Avian Biol. 47:197–207; doi 10.1111/jav.00774.

Warnock, N., Takekawa, J. Y., and Bishop, M. A. 2004. Migration and stopover strategies of individual Dunlin along the Pacific coast of North America. Can. J. Zool. 82:1687–1697; doi 10.1139/z04-154.

Wendler, G., Moore, B., and Galloway, K. 2014. Strong temperature increase and shrinking sea ice in Arc-tic Alaska. Open Atmo. Sci. J. 8:7−15; doi 10.217 4/1874282301408010007.

Woodley, K. 2009. Godwits: Long-haul Champions. Penguin Group, Rosedale, New Zealand.

Ydenberg, R. C., Butler, R. W., and Lank, D. B. 2007. Effects of predator landscapes on the evolutionary ecology of routing, timing and molt by long-dis-tance migrants. J. Avian Biol. 38:523–529.

Yohannes, E., Valcu, M., Lee, R. W., and Kempenaers, B. 2010. Resource use for reproduction depends on spring arrival time and wintering area in an Arctic-breeding shorebird. J. Avian Biol. 41:580–590; doi 10.1111/j.1600-048X.2010.04965.x.

Documents

Part IV · County Highway D, Grand View, Wisconsin 54839; *corresponding author: [email protected] Western Field Ornithologists A rctic-breeding birds time their arrival on nest-ing areas