Series: Alaska Park Science - Volume 14 Issue 1: Resource Management in a Changing World

Differential Effects of Coastal Erosion on Colonial-Nesting Sea Birds on the St. Matthew Islands, Alaska

By David R. Klein and Richard Kleinleder

The St. Matthew Islands, which include St. Matthew, Hall, and Pinnacle islands and are a part of the Alaska Maritime National Wildlife Refuge, are the most remote lands in the entire 50 states (Figure 1). They support perhaps a million
or more colonial-nesting and ground-nesting bird species. Coastal erosion through wave action as sea levels rose during the early Holocene, flooding much of the Beringian coastal plain, played a major role in creating the coastal landscapes that have been so propitious for colonial sea birds by providing secure nesting sites on the coasts of the Pribilof and the St. Matthew archipelagoes. Refuge expeditions to the St. Matthew Islands have returned about every five to seven years to monitor the rich bird populations and other life forms and their habitats. The occurrence of increased coastal erosion in recent decades on Alaska’s north and northwest coasts associated with climate change influences (ACIA 2005) raised our interest in the possible consequences of coastal erosion on colonial-nesting sea birds on the St. Matthew Islands
(Figure 2). Pronounced erosion on specific portions of the coastal cliffs of these islands where sea bird colonies were located has been observed in the last century. The biologist G. Dallas Hanna described one of the regions of significant erosion on a visit to the St. Matthew Islands in 1916: “The earth and cliffs are torn and tumbled in the greatest confusion. New slides are seen and the beach line boulders are not much rounded. In some places rocks are constantly falling, making it dangerous to go beneath the cliffs” (Hanna 1920).

The St. Matthew Islands are largely of volcanic origin, their exposed coastal rocks dating from ~60 to ~77 million years ago (Dawson 1893; Patton et al. 1975). The rocky coastal cliffs of the St. Matthew Islands have provided the locations where over a million sea birds have been able to establish their colonial-nesting sites (A. Sowls pers. comm) (Figures 3a and 3b). Following an assignment by the Biological Survey of the U.S. Department of Agriculture to do a reconnaissance survey of the mammals of St. Matthew Islands in the summer of 1916, G. Dallas Hanna (1920) provided a most perceptive description of the coastal landscape of St. Matthew:

The mountains are cut into by the sea on every side of the island, making long stretches of towering cliffs, between which the sea has built up beaches of such an extent as to give the impression that the island is much older than the Pribilofs. The cliffs display wonderful geological formations. There are blues, yellows, greens and bright reds in layers or dikes. The large number of cliffs with their grand scenic display are notable as the nesting places of countless sea birds. Of all the places I have visited St. Matthew Island is rivaled in this respect only by that incomparable bird cliff on St. George Island; the ledges on St. Matthew are more nearly perpendicular and thus afford less favorable nesting sites.
driftwood on a rocky slope
Accelerated erosion of fractured basalt on the north-eastern coast of St. Matthew Island, August 1, 2012.

D. Klein

Hanna’s asigned focus while on St. Matthew and Hall islands was primarily the mammals; thus, his comments on birds were limited to the obvious dense aggregations of birds nesting on the steep ciffs, primarily murres, kittiwakes, and fulmars. He made no mention of the large colonies of crevice-nesting least (Aethia pusilla) and crested auklets (A. cristatella) that are largely obscure to the casual visitor to these islands when birds are hidden from view and on their nests. Hanna disagreed with John Muir’s suggestion, made when visiting St. Matthew in 1899 with the Harriman expedition, that glacial scouring accounted for the rounded nature of the mountain landscape (Merriam 1901-10). In this regard Hanna noted that (The action of ice (meaning glacial ice) on these islands seems inconsequential.) While there is no evidence of the St. Matthew Islands being overridden by a glacial advance during the last glacial maximum of the Pleistocene as suggested by Muir, Potter et al. (1975) did identify and map several small cirque basins in the mountains of St. Matthew extending from the northeastern coast inland, providing evidence that a few small glaciers were present in these mountains during full glaciation when sea levels were lowered 328 feet (100 meters) and the St. Matthew Islands were part of Beringia and the Alaska mainland.

Accelerated erosion of portions of the rocky coasts of the St. Matthew Islands was documented during the 2005 expedition to these islands (Renner and Jones 2005). In view of the accelerated coastal erosion rates on Alaska’s northern coasts in recent years (ACIA 2005) and the potential for even higher coastal erosion rates in the future in a dynamic climate (Anisimov et al. 2007; Hinzman et al. 2005), we investigated effects of coastal erosion on the colonial-nesting sites of sea birds during our 2012 expedition to these islands.

aerial images of a green island with few trees
Fig 3. Coastal land forms of the St. Matthew Islands; the southeast coast of St. Matthew Island (left) and the northeast coast of St. Hall Island (right).

D. Klein (left) and M. Romano (right)

Methodology

During July 29 to August 7, 2012, we collected representative samples of rock types at sea bird colonies where we were able to gain access on St. Matthew and Pinnacle islands. The rock samples that we collected on Pinnacle were from a small rock fan composed of rocks that had eroded and fallen from the steep rocky slope above; thus, they included rocks representative of the basic rock types of the island. All rock samples were examined for specific rock type and density at the Geochronology Laboratory at the University of Alaska Fairbanks. We took reference photos at the rock collection sites, which enabled correlation of rock type, density, and possible vulnerability to coastal erosion with the geomorphically detailed Reconnaissance Geologic Map of St. Matthew Island produced by Patton et al. (1975) in conjunction with their 1971 geological work on the island. Turbidity plumes in the adjacent sea, visible on 1948 U.S. Navy aerial photos, and satellite imagery from southwestern Hall and northwestern St. Matthew were used in defining locations of active erosion of silicic pyroclastic deposits. Locations and relative size of sea bird colonies were available from Renner and Jones (2005) and Alaska Seabird Information Series (USFWS 2006).

four diagrams of rocks with text too small to read
Figure 4. Diagrammatic sketches of volcanic sequences exposed at selected coastal locations on St. Matthew Island

From Patton et al. 1975

Results and Discussion

Assessing Effects of Coastal Erosion on Colonial-Nesting Sea Birds

In recent decades, as a consequence of global climate warming, sea levels have risen more than 11.8 inches (30 centimeters). Duration of the annual presence of sea ice at the St. Matthew Islands has declined markedly, at least through 2011 (NASA 2013). The southern extent of winter sea ice however, has shown little change from the long-term mean. As a consequence of the delay in formation of sea ice and associated shore-fast ice in early winter, wave action enhanced by the extreme storms of early winter have created conditions that can accelerate coastal erosion on the St. Matthew Islands where rock types are sensitive to erosion. Luchin et al. (2002) reported on a pronounced decrease beginning in the late 1970s in mean seasonal duration of sea ice in the northern Bering Sea, presumably a consequence of accelerated climate warming throughout the Arctic (IASC 2005). This decline in duration of Bering Sea ice and associated warming of ambient temperatures was most pronounced in the eastern Bering Sea where warmer North Pacific waters moving with northeastward flowing currents also accounted for warming of ambient temperatures there

(Grebmeier et al. 2006; Klein and Shulski 2009). In the western Bering Sea however, cooling was associated with southward flowing ice and waters (Stabeno et al. 2005). The mean annual extent of sea ice and annual duration of sea ice in the region of the St. Matthew Islands, and the associated climate inclusive of both cooler and warmer years, appears to have been more greatly influenced by the North Pacific Ocean via the Pacific Decadal Oscillation and El Niño events than by Arctic Ocean influences (Kitaysky and Golubova 2000; Luchin et al. 2002).

simple map diagraming how crevice-dwelling birds live in colonies around edge of islands
Figure 5. Location of sea bird colonies in relation to coastal geology, rock type, and its relative vulnerability to erosion on the St. Matthew Islands.

The surficial geology of the St. Matthew Islands testifies to its volcanic origin (Potter et al. 1975; Wittbrodt et al. 1989). The nature of the coastal rock however, varies in its morphology, hardness, stratification, and metamorphic history, accounting for wide local variation in its vulnerability to coastal erosion. Following periods of active volcanism in the late Cretaceous and early Tertiary ages, ~ 77 to ~ 60 million years ago (Patton et al. 1975), both St. Matthew and Hall islands have remained horizontal and relatively free of tilting and folding by tectonic movements (Wittbrodt et al. 1989). This is more evident on Hall and northern St. Matthew than in the central and southern portions of St. Matthew. The volcanic rock forms, transformed through erosion processes, provide the sites now occupied by myriads of colonial-nesting sea birds.

These coastal rocks are a product of the volcanic past when basalt and andesite flows were interlayered with pyroclastic fine- and course-grain, ash-rich tuffs, sometimes including dacite blocks of up to 3.2 feet (1 meter) or more in diameter (Potter et al. 1975; Coombs and Bacon 2012).  Diagrammatic profiles of the rock strata that reflect these volcanic sequences at specific coastal localities on St. Matthew Island are shown in Figure 4.

These coastal profiles illustrate how the less dense, and more erodible rock derived from pyroclastic flows, overridden by the harder rock from lava flows of basalt and andesite, have over time and with differential responses to erosion by the sea, created nesting habitats for both the crevice nesting auklets and the typical cliff ledge-nesting murres (Uria aalg and U. lomvia), kittiwakes (Rissa tridactyla), and fulmars, Fulmarus glacialis.

three images, one aerial of an island, another closeup of rocks and third of two large birds
Fig 6. Erosion has generated habitat for crevice-nesting auklets

Ian Jones (middle), R. Kleinleder (right)

The rocky nesting habitat of the crevice-nesting small auklets, primarily the least auklet and crested auklet on St. Matthew and Hall islands is generally the product of past coastal erosion.  These nesting colonies are often associated with massive slumps into the adjacent sea where highly erodible thick layers of intermediate and silicic pyroclastic deposits had been overridden by thin basalt and andesite flows (Coletti 2012; Figure 4). The relation between sea bird nesting colony location and volcanic rock types on the
St. Matthew Islands is illustrated in the map in Figure 5. This has presumably been an active process as rising sea levels throughout the Holocene engulfed the mountainous landforms that ultimately became the St. Matthew Islands. Within the pyroclastic deposits a fine bentonite-like material, white and multicolored and with an affinity for water, often forms layers in the coastal strata, which act as a lubricant that facilitates movement and the slumping of denser basalt and andesite in the above strata.  These massive slumps and their active nature have been described by early visitors to the St. Matthew Islands (Elliott 1882; Dawson 1883). Pronounced turbidity resulting from the continuing leaching and erosion of the fine light-colored tuffs into the adjacent sea is apparent in 2009 satellite imagery (Aleutian and Bering Sea Islands Land Conservation Cooperative 2009; Figure 6a). Auklets choose nesting sites among and under large boulders that remain
on the surface of slopes or benches where the slumps have occurred (Figure 6b). Earlier visitors to the St. Matthew Islands have noted the continuing slow movement of these slumping and fractured rock strata toward and into the sea (Elliott 1882; Hanna 1920).

Rock falls from the head or edge of the slumps often pose risks to the nesting auklets, their eggs, and young as well as the biologists attempting to monitor the extent of auklet nesting within the slump areas (Renner and Jones 2005).

This is where thin layers of the basalt and andesitic mafic flows are intermixed with or abut softer and more erodible pyroclastic strata (Table 1). These are the locations where coastal erosion is active in the softer, more water-permeable pyroclastic materials and therefore have generated massive earth slumps, which are often topped by fields of fractured basalt, andesitic, and dacite boulders.

a rocky slope partially slumping into the ocean
Figure 7. This relatively small slump into the sea on the east coast of Hall Island was shown through satellite imagery to be actively expanding into the sea

D. Klein

Rock Sample Type-MorphologyDensity
SM 1: Altered rhyolite, alteration mineralogy is epidote and secondary biotite 2.87
SM 2: Altered porphyritic rhyolite, alteration mineralization is epidote, chlorite, and secondary biotite. Rock was altered after intrusion of hot fluid circulating through overlying rock. Original rock was rhyolite with visible feldspar crystals. 3.00
SM 3: Basalt from old crater rim 2.50
SM 4: Vesicular basalt, greenstone from flow in wet environment 2.83
SM 5: Rhyolite and dacite, hypabyssal 2.35
SM 6: Altered porphyritic from tuff 2.67
Mean density of collected St. Matthew Island rocks 2.72
P 1: Altered epidote, pyrite, with some chalcopyrite 2.79
P 2: Fine tuff, altered porphyritic and marine hardened 3.01
P 3: Flow-banded rhyolite, alteration and mineralogy is pyrite and epidote 2.64
P 4: Flow-banded rhyolite, alteration and mineralogy is pyrite and epidote, marine hardened 3.20
P 5: Marine hardened intrusion in basalt 3.91
Mean density of collected Pinnacle Island rocks 3.11

The resulting boulder fields sloping to the sea provide relatively secure crevice-nesting habitat for the small auklets. These auklet nesting colonies are therefore the product of past as well as continuing coastal erosion.

Although accelerated coastal erosion was observed to be ongoing at the existing auklet colonies on Hall and northern St. Matthew islands, contributing to dynamic changes in the surface landscape at these coastal locations, we observed no detrimental consequences for the nesting auklets except at the Glory of Russia colony. There, where the adjacent mountain slopes are composed of highly disrupted pyroclastic material, a massive mud slide of fine ash-fall tuff has in recent decades bisected the colony, covering extensive areas that had been used by breeding auklets in the 1970s (Art Sowls, pers. comm. in Renner and Jones 2005).

Auklet nesting habitat is, however, being generated on Hall and was being used by some nesting auklets in 2005 (Figure 7). Rock types where murres, kittiwakes, and fulmars—the cliff-ledge nesters—concentrate are consistently among the much harder rock types derived from the massive andesitic and basalt flows (Figure 8; Table 1). These high-density, hard rocks are resistant to erosion by the sea, in contrast to the dacite, rhyolite, and andesitic tuff breccias whose erosion creates habitat for nesting auklets. Thus, it is the erosion resistance of the thick and massive basalt and andesitic flows where they interface with the sea that generates the sheer cliffs where the most extensive nesting habitat for murres, kittiwakes, and fulmars are found. These rock faces also contain crevices storm surges (Figure 10).

two images of birds on rocky slopes
Fig 8 (left). Sea cliffs resulting from ancient lava flows. Fig 9 (right). MacKay’s buntings are endemic to the St. Matthew Islands

D. Klein (left) and R. Kleinleder (right)

The resulting eroded slopes of relatively thin plate-like rocks do not provide secure nesting habitat for either cliff-ledge or crevice-nesting sea birds though single pairs of pigeon guillemots (Cepphus columba) find isolated nesting sites (Figures 11a and 11b). Large nesting colonies are absent at such locations. Coastal erosion on the St. Matthew Islands is highly variable in relation to the rock types present, their volcanic origin, morphological history, and nature of the adjacent coastal landscape (Figures 12a and 12b).

Pinnacle Island is composed of strata of rock types similar to those of St. Matthew and Hall islands; however, the strata of basalt, andesitic, and pyroclastic tuffs of Pinnacle are tilted to a nearly vertical position in contrast to the predominant horizontal rock strata of St. Matthew and Hall (Figure 13). The volcanic rocks of Pinnacle are denser and harder than similar rock types on St. Matthew, a presumed consequence of their composition and tectonic deformation history following their volcanic origin.

man walking through a rocky landscape
Fig 10. Near Glory of Russia Cape, light-colored pyroclastic rock and dark basalt-andesite at right. Both rock types show considerable ongoing active erosion

R. Kleinleder

Potter et al. (1975) were limited in their assessment of Pinnacle Island’s geology to observations made from shipboard. The island’s geology nevertheless, as seen from the ship, appeared to them to be composed of mafic dikes of low silica andesite, basalt, dacite, and rhyolite, which had been morphologically altered prior to the tectonic action that resulted in the emergence of Pinnacle Island from the sea.

We were able to get ashore on Pinnacle briefly in 2012
to collect rock samples, including rocks fallen from the
near vertical rock slope above; thus, they included rocks representative of at least three of the primary rock types of the island. Table 1 shows the comparative density analysis
of similar coastal rock types collected from Pinnacle and
St. Matthew. The higher mean densities of the rock samples collected from Pinnacle of 3.11 versus 2.72 for those from St. Matthew as also determined by Barnes and Eastlund (1968) are consistent with the geology Potter et al. (1975) had suggested for Pinnacle, though they had not been able to land on the island. The rocks we collected at Pinnacle, derived originally from the dark, massive basalt and andesitic flows as well as the generally lighter colored rocks of pyroclastic origin, had become hardened through metamorphic alteration prior to their uplift. The lighter colored rocks derived from the volcanic tuffs were considerably much denser and harder than those from comparable rock types on St. Matthew (Table 1). The harder rocks of Pinnacle Island and its vertical rock strata rising abruptly from the sea have been much more resistant to coastal erosion in the past and currently than has been
the case with the coastal rocks in horizontal strata on St. Matthew and Hall islands. Thus, while the cliff-ledge nesting murres, kittiwakes, and fulmars are abundant nesters on Pinnacle, there is an absence of least and crested auklet nesting colonies. Without vegetated lowlands and coastal beaches on Pinnacle Island, a consequence of the highly erosion-resistant geomorphology of Pinnacle both in the past and at present, the island lacks suitable habitat for the endemic singing vole
(Microtus abbreviatus), as well as the arctic fox, (Alopex lagopus), and the more recently arrived red fox (Vulpes vulpes). In the absence of the threat of predation by foxes on Pinnacle Island, birds, especially murres, fulmars, and kittiwakes, are able to nest on the steep slopes of the island from just above the wave splash zone to the ridge tops of the island. Hence the rock type and subsequent geomorphology of Pinnacle Island allows both prime nesting habitat for murres, fulmars, and kittiwakes and for these birds species to take full advantage of all nesting sites due to the lack of nest predators.

two images of birds on rocky slopes
Fig 11. Fractured and eroded basalt and andesite (left). Pigeon guillemots find isolated sites in this type of eroding coast where single pairs nest (right)

D. Klein (left) and R. Kleinleder (right)

Conclusions

It was the Holocene rise of sea levels that brought about the present insularity of the St. Matthew Islands by the mid-Holocene. The accelerated rise in sea levels in recent decades, a consequence of climate warming, has resulted in shortening of the duration of seasonal sea ice, a corresponding sea level rise, and increased frequency and severity of extreme storm events, especially in early winter before sea ice has reformed, all of which collectively have accounted for greater erosive force of wave action on the St. Matthew Islands.

The nature of the coastal rocks and variable tectonic history on these volcanic islands places a first order control on the geomorphology of the individual islands, their pronounced differences in vulnerability to erosion, and thus their relative security as sites for colonial-nesting birds. This variability in coastal rock types is primarily associated with specific eruptive events during the volcanically active period of the late Tertiary and early Cretaceous ages, namely mafic basalt and andesitic lava flows versus volcaniclastic deposits of ash of varying texture and subsequent rock type and density. Both the initially harder flow rocks and softer ash-derived rocks also underwent differential further hardening or fracturing whether above or below sea level during the cooling process and through metamorphosis in relation to their position to regional structures.

Whereas coastal erosion has been an ongoing process on the St. Matthew Islands in the past, the shape of the present shoreline is largely a product of variation in rock lithology and their topography at the interface of land and sea. The marine birds nesting colonially on the St. Matthew Islands, although dependent on proximity to the sea for the food source it provides during nesting and rearing of young, also require secure nesting locations on the rocky coasts of these islands.

The harder rocks, more resistant to erosion by the sea, now form the points and headlands of these eroded island coastlines. The increased erosive force of the open sea and its longer mean annual ice-free period in recent decades appears, nevertheless, to have had minor consequences for the colonial sea birds that nest on these islands. The hard rocks of the cliffs where colonies of the cliff-ledge nesting murres, kittiwakes, and fulmars are located are highly resistant to erosion by the sea and have shown little change in their extent despite the increased potential of the erosive force of the sea in recent decades. The large coastal slumps where thin basalt and andesite flows were underlain by fine pyroclastic materials have, through their long-term erosion of the soft underlying pyroclastic strata, created nesting habitat for the crevice-nesting small least and crested auklets; the slumps are most common on the coasts of western Hall and northern St. Matthew. Erosion of the fine-grained pyroclastics at these auklet colony nesting sites is continuing, as is evidenced by turbidity in the adjacent sea visible in recent satellite imagery; existing auklet nesting habitat seems to be minimally affected. In fact, small numbers of auklets appear to be pioneering establishment of a new nesting colony in a recently expanding slump area on northwestern Hall Island.

Thus, although these slump areas remain active and their erosion appears to be accelerated by rising sea levels and increased seasonal duration of wave action, new auklet nesting habitat is correspondingly being generated.

two images of rocky, rough terrain
Fig 12. Nesting habitat for auklets (left) and fulmars (right)

R. Kleinleder (left) and R. Klein (right)

Acknowledgements

We appreciate the competent and safe logistic support provided by the entire crew of the M/V Tiglax during the 2012 expedition to the St. Mathew Islands. Field camp and logistic support was provided by the Alaska Maritime National Wildlife Refuge. Carla Tomsich, a PhD student in the geology department at University of Alaska Fairbanks (UAF) provided assistance in identification of rock types. Jeff Benowitz, a research geologist with UAF, assisted in measurement of rock density at the Geochronology Laboratory at UAF and provided a review of a draft manuscript from a geological perspective. We appreciate the use of Dan Mann’s notes on landform geology following his review of the 1948 U.S. Navy aerial photo coverage of the St. Matthew Islands. Satellite imagery was made available by Douglas Burn through the Aleutian and Bering Sea Islands Land Conservation Cooperative 2008-2014. Marc Romano provided excellent photo coverage of bird colony locations on the coast of Hall Island. Laura Weaver prepared the map with bird icons (Figure 5) showing relation of colonial-nesting colonies to coastal geomorphology. We appreciate reviews of the manuscript by Heather Renner, Art Sowls, Tony DeGange, and Marc Romano.

Glossary of Geological Terms

  • Andesite: volcanic rock containing feldspar microliths and crystals of plagioclase
  • Basalt: dark volcanic rock of mafic composition, fine grained with varying portions of calcium and sodium, primary component of lava and magma
  • Dacite: felsic to intermediate volcanic rock containing hornblende with plagioclase
  • Diorite: course-grained intermediate plutonic rock composed of plagioclase, pyroxene, and/or amphilode
  • Epidote: gray-green semitransparent crystalline rock of monoclinic volcanism
  • Granite: coarse-grained plutonic rock composed of orthoclase, plagioclase, and quartz
  • Hornblendite: mafic or ultra-mafic cumulate rock with greater than 90 percent hornblende
  • Hypabyssal: consolidated or partly crystalline from fusion at moderate depths underground
  • Mafic: volcanic rocks high in silicate and heavier minerals such as magnesium and ferric elements
  • Obsidian: volcanic glass
  • Porphyritic: rock type, usually granitic, with crystals embedded in a fine-grained mass
  • Pyrite: rock composed of iron sulfide, often called “fool’s gold”
  • Pyroclastic: rock types composed of ash and breccia resulting from explosive volcanic eruptions
  • Rhyolite: felsic volcanic rock
  • Tuff: porous rock composed of scoria and ash formed in proximity to volcanic craters

References



Arctic Climate Impact Assessment (ACIA). 2005.
Arctic Climate Impact Assessment: Scientific Report. Cambridge: Cambridge University Press. 1042.

Aleutian Island and Bering Sea Islands Land Conservation Cooperative (ABSILCC). 2009.
GeoEye-1 Satelllite Imagery. August 9, 2009. Worldview-2 Satellite Imagery. Anchorage: U.S. Fish and Wildlife Service.

Anisimov, O., D. Vaughan, T. Callaghan, C. Furgal, H. Marchant, T. Prowse, H. Vilhjálmsson, and J. Walsh. 2007.
Polar regions (Arctic and Antarctic). Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.

Barnes, D., and M. Eastlund. 1968.
Gravity Map of St. Matthew and Hall Islands. Miscellaneous Field Studies Map MF 642, U.S. Geological Survey open-file Report, page 8, 31 tables.

Coletti, H. 2012.
Possible effects of a volcanic eruption on the nearshore marine environment. Alaska Park Science 11(1):79-81. Anchorage: National Park Service.

Coombs, M., and C. Bacon. 2012.
Using rocks to reveal inner workings of magma chambers below volcanoes in Alaska’s National Parks. Alaska Park Science 11(1):27-33. Anchorage: National Park Service.

Dawson, G. 1883.
Geological notes on some of the coasts and islands of Bering Sea and vicinity. Bulletin of the Goelogical Society of America 5:138.

Elliott, H. 1875.
Polar bears. Harpers Weekly 19(957):1-2.

Elliott, H. 1882.
A monograph of the seal islands of Alaska. Washington, DC: U.S. Commission on Fish and Fisheries. 162.

Grebmeier, L., et al. 2006.
A major ecosystem shift in the northern Bering Sea. Science 311:1461-1464.

Hanna, G.D. 1920.
Mammals of the St. Matthew Islands, Bering Sea. Journal of Mammology 1(3):118-122.

Hinzman, L., et al. 2005.
Evidence and implications of recent climate change in northern Alaska and other arctic regions. Climatic Change 72(3):251-298.

Klein, D., and M. Shulski. 2009.
Lichen recovery following heavy grazing by reindeer delayed by climate change. Ambio 38:11-16.

Klein, D., and A. Sowls. 2011.
History of polar bears as summer residents on the St. Matthew Islands, Bering Sea. Arctic 64:429-436.

Luchin, V., I. Semiletov, and G. Weller. 2002.
Changes in the Bering Sea region: atmosphere—ice—water system in the second half of the twentiethcentury. Progress in Oceanography 55:23-44.

Merriam, C., ed. 1901-1910.
The Harriman Alaska Expedition (the Harriman Alaska Series). Vol. 1-5. New York: Doubleday. Page and Company.

Patton, W. Jr., et al. 1975.
Reconnaissance geologic map of St. Matthew Island, Bering Sea. Miscellaneous Field Studies Map MF-641. Washington, DC: U.S. Geological Survey. U.S. Government Printing Of-fice.

Renner, H., and I. Jones. 2005.
Mapping distribution and relative density of auklets at selected colonies on Hall and St. Matthew Islands. U.S. Fish and Wildlife Service Report. Alaska Maritime National Wildlife Refuge 05/20.

Smith, J., G. Gray, and N. Kinsman. 2012.
Annotated bibliography series in support of coastal community hazard planning—Northwest Alaska Division of Geological and Geophysical Surveys Miscellaneous Publication 147:5.

Stabeno, P., G. Hunt Jr., J. Napp, and J. Schumacher. 2005.
Physical forcing of ecosystem dynamics on the Bering Sea Shelf. The Sea. Vol. 14. Chapter 30. Eds. A.R. Robinson, and K. Brink.

U.S. Fish and Wildlife Service (USFWS). 2006.
Alaska Seabird Information Series. Anchorage: U.S. Fish and Wildlife Service. 43 information sheets.

Wittbrodt, P., D. Stone, and D. Turner. 1989.
Paleomagnetism and geochronology of St. Matthew Island, Bering Sea. Canadian Journal of Earth Science 26:2116-2129