an introduction to the biocomplexity of sanak island...

673

An Introduction to the Biocomplexity of Sanak Island,Western Gulf of Alaska1

Herbert D. G. Maschner,2,12 Matthew W. Betts,3 Joseph Cornell,4 Jennifer A. Dunne,5

Bruce Finney,4 Nancy Huntly,4 James W. Jordan,6 Aaron A. King,7 Nicole Misarti,2

Katherine L. Reedy-Maschner,8 Roland Russell,9 Amber Tews,10 Spencer A. Wood,11

and Buck Benson2

Abstract: The Sanak Biocomplexity Project is a transdisciplinary research effortfocused on a small island archipelago 50 km south of the Alaska Peninsula in thewestern Gulf of Alaska. This team of archaeologists, terrestrial ecologists, socialanthropologists, intertidal ecologists, geologists, oceanographers, paleoecolo-gists, and modelers is seeking to understanding the role of the ancient, historic,and modern Aleut in the structure and functioning of local and regional ecosys-tems. Using techniques ranging from systematic surveys to stable isotope chem-istry, long-term shifts in social dynamics and ecosystem structure are presentin the context of changing climatic regimes and human impacts. This paperpresents a summary of a range of our preliminary findings.

The greater North Pacific region hostsone of the world’s most important fisheries.Recent dramatic decreases in the numbers ofmany species in this region have perplexed bi-ologists, caused sweeping and underinformedmanagement decisions, and are provokingcultural disintegration of many Alaska Nativevillages. But twentieth-century data showclearly that there have been other periods oflow productivity for species, often followedby periods of abundance (Trites et al. 2007;H.D.G.M., A.W.T., K.L.R.-M., M.W.B.,A.T., and M.L., unpubl. data), as do histori-cal, paleoecological, and archaeological datafrom the last 5,000 years (e.g., Moser 1902,Finney 1998, Maschner 1998, Finney et al.2000, 2002, Savinetsky et al. 2004, Maschneret al. 2008). The global problem is thus: un-der varying population cycles of humans,salmon, groundfish, sea mammals, birds, andmany other species in the North Pacific re-gion, and with varying harvests by the highertrophic levels, how do we sustain populations,ecosystems, and the peoples and cultures whodepend on them for their social, political,economic, and cultural survival?

Most modern investigations of this ques-tion have had limited temporal data. For ex-ample, studies of possible long-term cyclesin populations of Steller sea lions in the

Pacific Science (2009), vol. 63, no. 4:673–709: 2009 by University of Hawai‘i PressAll rights reserved

1 This research was funded with generous supportfrom NSF BE/CNH 0508101, NSF ARC 0326584,NSF ARC 9996415, and NSF BCS 0119743. Manuscriptaccepted 3 February 2009.

2 Department of Anthropology and the Center forArchaeology, Materials, and Applied Spectroscopy (CA-MAS), Idaho State University, 921 South 8th Avenue,Stop 8005, Pocatello, Idaho 83209-8005.

3 Archaeology and History Division, Canadian Mu-seum of Civilization, 100 Laurier Street, Box 3100,Station B. Gatineau, Quebec, J8X 4H2 Canada.

4 Department of Biological Sciences, Idaho StateUniversity, 921 South 8th Avenue, Stop 8007, Pocatello,Idaho 83209-8007.

5 Santa Fe Institute, 1399 Hyde Park Road, Santa Fe,New Mexico 87501.

6 Antioch University New England, 40 Avon Street,Keene, New Hampshire 03431-3552.

7 Ecology and Evolutionary Biology, University ofMichigan, Ann Arbor, Michigan.

8 Department of Anthropology, Idaho State Univer-sity, 921 South 8th Avenue, Stop 8005, Pocatello, Idaho83209-8005.

9 Columbia University, Center for InternationalEarth Science Information Network (CIESIN), 61 Route9W, Palisades, New York 10964.

10 1013 South 800 East, Salt Lake City, Utah 84105.11 Department of Zoology, University of British Co-

lumbia, 2370 University Boulevard, Vancouver, BritishColumbia V6T 1Z4, Canada.

12 Corresponding author (e-mail: [email protected]).

North Pacific were handicapped by limitedand short-term data, although cycles may bemuch longer (Ocean Studies Board and PolarResearch Board 2003). Further, attempts tomodel the effects of human fishing on thepopulation dynamics of cod, pollock, or sealions, for example, typically begin with an as-sumed prefishing or prehunting state (e.g.,Castellini 1993). Yet indigenous peoples haveharvested large quantities of these species forthousands of years, requiring that we assumethat there is no ecologically meaningful timeperiod without humans, at least not sincedeglaciation 16,000 years ago. Thus, archaeo-logical data provide the essential temporalperspective for such an analysis.

From both anthropological and ecologicalviewpoints, and building on the work ofCrumley (1994 and papers therein), the rolesof humans must be considered as one ofmany linkages within ecosystems that includeother biotic components, as well as abioticconstraints and drivers. A sweeping article inScience ( Jackson et al. 2001) suggested large-scale roles of humans in ecosystems, arguingthat changes in marine ecosystems are com-plex, systemic, and have a long historical con-text (e.g., see papers in Rick and Erlandson2008). However, most of the suggested rela-tionships between trophic structure, speciesextinctions, marine productivity, and humanexploitation remain untested. Our researchwas developed to directly test the sorts of re-lationships suggested by Jackson et al. (2001),as well as provide fundamental data on therole of humans in ecosystem dynamics in theNorth Pacific over millennia. This projectfocuses on whole-ecosystem complexities ofthe region, which are founded on predator-prey and other food-web interactions, cul-tural harvesting strategies, long-term changesin the North Pacific ecosystem and climate,and direct human impacts on coastal environ-ments.

We have constructed a transdisciplinaryapproach to studying humans as part of thenorthern ecosystem, incorporating the mod-ern and prehistoric, the terrestrial and ma-rine, the local and regional, and used bothempirical and theoretical exploration. Wesuggest that this multidimensional approach

is not only possible but necessary to our un-derstanding of the region. The implicationsof this approach are profound and requireintegration of anthropology, archaeology,geology, ecology, mathematics, climatology,and history; the perspective of many spatialand temporal scales; and the seamless merg-ing of theoretical approaches from manyfields. Because indigenous peoples have beenharvesting resources in the North Pacific forthousands of years, we must include them inmodels and reconstructions of this ecosystem,in which they may function as ecosystemengineers, as more ordinary components offood webs and landscapes, or as passive re-sponders to a world that is primarily drivenby largely external forces such as climate andgeomorphic evolution, or some combinationof all three. We expect this human role tovary temporally and spatially, with culturalcontext, and our work provides a frameworkand a test case for understanding the historyof people as components of ecosystems.

This project is being conducted in theSanak Island archipelago, 50 km south of thetip of the Alaska Peninsula (Figure 1), anddraws on past and ongoing work on thenearby western Alaska Peninsula and Aleutianregions. Sanak Island is about 120 km2 inarea, with 92 km of shoreline, and is the larg-est island of its archipelago. The islands havea 6,000-year archaeological record. Theywere populated by a number of villages atRussian contact, were the center of sea-otterharvesting in the nineteenth century andcod harvesting in the early twentieth century,and supported active cattle ranches until the1960s. The islands are surrounded by a mas-sive reef system that supported some of thelargest populations of groundfish and seamammals in the region. Millions of tons ofcod were harvested here in the early twenti-eth century, and, today, commercial fisher-men from False Pass and King Cove fishhere for cod in spring or halibut in summer.These islands supported the last major popu-lations of sea otters to be commercially har-vested. At least three river drainages haveruns of sockeye, humpback, and chumsalmon. In the last 100 years, the landscapehas been altered by large herds of now feral

674 PACIFIC SCIENCE . October 2009

Figure 1. Map of western Gulf of Alaska. Sand Point, False Pass, Nelson Lagoon, and King Cove are modern Aleutcommunities. Sanak and Pauloff Harbor are abandoned.

cattle that are an important part of localsubsistence in the region. Finally, local peo-ples lived on these islands for many yearsbefore their abandonment in the 1960s. Fam-ilies in Sand Point, King Cove, and FalsePass, Alaska, have provided a wealth of localknowledge about the landscape. Given the ar-chaeological, historic, and ethnographic data,the local knowledge of the region, and therelative isolation of this archipelago from themainland peninsula, the Sanak Island regionis a perfect location to investigate the com-plex dynamics of human-landscape relationsand the role that humans have played in thestructure of the North Pacific ecosystemover the last several thousand years.

Humans function within ecosystems inmany ways; they are one of many consumers,whose roles within food webs may range fromstrong top predators, to omnivorous and gen-eralized consumers, to builders and husbandsof domesticated or industrialized resourcesupply systems. They can be ecological engi-neers ( Jones and Lawton 1995, Jones et al.1997, Power et al. 1998), changing the physi-cal structure of ecosystems, and can link ter-restrial and marine ecosystems, as do otherorganisms (e.g., seabirds or salmon) andabiotic processes (Polis and Hurd 1996, Poliset al. 1997). For instance, they may alter ormediate movements of resources from ma-rine to terrestrial regions and may modifygeological or hydrological linkages deliber-ately or as unanticipated side effects of theiractivities. Humans draw from the EcosystemServices of nature (e.g., supply of freshwaterand food, medicinals, esthetics), and thesehave played strong historical roles in patternsof human settlement, health, and culture(Dale et al. 2000). Nevertheless, humanshave only rarely been evaluated explicitly asintegral components of ecosystems (e.g., Ellisand Swift 1988, Ellis and Galvin 1994) withproperties that can be compared with thoseof other biotic and abotic components withinan evolving system, the trajectory of whichmay be disproportionately influenced by peo-ple and their behaviors. The interdependenceof people and their ecosystems makes explicitstudy and evaluation of humans as integralparts of ecosystems critical to study. Much

ecological theory addresses the dynamicsand interactions of species within ecosystems(e.g., Chesson and Huntly 1994, 1997, Jonesand Lawton 1995, Polis and Winemiller1996, King and Schaffer 2001, Dunne et al.2002, Turchin 2003, Worm and Myers 2003,Williams and Martinez 2004a), and we areapplying such approaches to the study of theAleut (sometimes referred to as Unangan)and their ecosystems of the western Gulf ofAlaska.

Evidence suggests that humans have beenkey components of the regional ecosystemfor millennia. Standing crop, diversity, spe-cies composition, and the abundances of edi-ble and medicinal plants all differ betweenareas inhabited by Aleuts and those thathave not been (Bank 1953, Shacklette 1969,Amundsen 1972, Johnson 2003). The distri-butions of plants typically influence the dis-tributions of animals (Morrison et al. 1998),including mammals and birds that, in turn,influence human populations and cultures.Additional ecosystem-level interactions arecaused by cultural systems of exploitation,such as fisheries and the keeping of domesticlivestock. Moreover, humans have repeatedlyremoved and exterminated species, whichmay have altered food webs and ecosystemsprofoundly. For instance, in the 1780s, Rus-sian documents reported three species offoxes on Sanak Island, but in the 1920s,no foxes were reported. In recent decades,arctic foxes were reintroduced only to be ex-terminated in the last few years. In 1823, theRussian Orthodox Church removed the Aleutpeople from the islands to protect one of thelast commercially viable sea otter populationsin the North Pacific. We suspect that this re-moval again reorganized ecosystems, whichagain were altered when Scandinavian fisher-men and their Aleut wives recolonized Sanakin the 1870s.

Our research is different than the moretraditional (and highly successful) approachespreviously attempted in the region. Yesner’s(1980, 1981, 1987, 1988, 1998) studies, forexample, were some of the first to identifythe role of the Aleutian ecosystem in thestructure of ancient Aleut society. He wasalso one of the first to attempt to identify


human impacts on that ecosystem. Simenstadet al. (1978) made an early foray into thecomplex dynamics between human harvestingand the restructuring of the marine ecosys-tem in the Aleutians. Building on these earlystudies, one focus of our work has been thelinkages involving humans within the Sanakecosystem. We are investigating major bioticcomponents, both terrestrial and marine, ofthe food webs and ecosystems that includethe Aleut, and include primary abiotic driverssuch as climate (e.g., Pacific Decadal Oscilla-tion, Aleutian Low) and geomorphic factors.We are testing hypotheses as to the link-ages of humans within ecosystems, beginningwith a series of descriptors of the effects ofhumans on terrestrial and marine populationsand the landscape, and following through toanalyze potential ramifying effects of theseto other ecosystem components and pro-cesses. Hypotheses are being addressed fromcontemporary to ancient times, using sites onSanak and the western Alaska Peninsula thatdate back to 6,000 years ago to address pat-tern and process in deep time.

A second focus is the dynamic importanceof humans to other components of the foodweb and landscape over the last several mil-lennia, as well as the dynamics of the humanpopulation itself. Here, the emphasis is onpopulation and community dynamics, as pos-sible to reconstruct from long-term indicatorsand contemporary, more precise, measures,and from models. This focus more directlytests Jackson et al.’s (2001) assertion that hu-mans may have cascading effects on marine(or other) food webs that are realized overdecades, centuries, and millennia and overlarge spatial scales. This work additionallyexamines population and social dynamics ofhumans in correlation with resource andlandscape dynamics over millennia. Thus, weare directly addressing the conditions underwhich humans have sustainable patterns ofresource supply and exploitation. We are in-vestigating pattern and process at multiplepoints in time and in space, as well as thelinkages and transitions through time andacross space.

The modern history of Sanak Island offersa final important research focus, a unique op-

portunity to view both of the above researchthemes in finer detail and with comparisonsthat include modern domestic agriculture(e.g., cattle) and technology (e.g., canneries,regulated commercial fisheries). Here, we areinvestigating contrasts between traditionaland modern domestic-technological exploita-tion ecosystems. Given this context, we pro-vide some of our preliminary results here.Although still in the early phases of analyses,some distinct patterns are beginning toemerge from these investigations.

The first section of our preliminary find-ings covers the ethnohistory and history ofthe last 200 years. This is followed by a geo-logical and paleoenvironmental review. Thearchaeological data are then used to assessthe temporal distribution of sites and the dis-tribution of sites with shell middens to recon-struct the population history of the island, toinvestigate cod and otariid histories, to com-pare the modern intertidal ecosystem withthe prehistoric exploitation of shellfish, totrack regional paleoenvironmental data usingmodern and prehistoric samples for isotopicanalysis. We conclude with our first explora-tion into the reconstruction and analysis offood webs.

Finally, restating our research question inmore specific and testable form: What havebeen the roles of prehistoric, historic, andmodern peoples in the structure and func-tioning of the North Pacific ecosystem, andis it possible for that role to continue and toviably sustain the communities that live inthis ecosystem today? This paper provides afirst assessment of a number of specific find-ings that contribute directly to this funda-mental question of the project, especially inregard to how humans shaped the NorthPacific region, and, in turn, how the complex-ities of the North Pacific conditioned the be-haviors, adaptations, and histories of thepeople who lived there.

preliminary results

An Economic and Social History

An analysis of the ethnohistoric data, as wellas many hours of interviews with former

Biocomplexity of Sanak Island . Maschner et al. 677

residents of the island, demonstrates the dy-namic and volatile economic history of theSanak archipelago. This historic volatility ispartially a product of changing politicaleconomies and partly a product of the real-ities of human impacts on these islandscreating changing economic and social op-portunities (Reedy-Maschner 2001, 2004,2007).

The inhabitants of this region today, atthe time of historic contact, and as far as wecan surmise for at least the last 6,000 years,are the Aleut. In the eighteenth century, theAleut lived in large, sedentary villages. Theywere organized into lineage-based householdsthat included from six to 10 families in a sin-gle dwelling, perhaps larger in the easternAleutian Islands (Lantis 1984, McCartney1984, Liapunova 1996, McCartney andVeltre 1999). The Aleut were ranked, withhereditary nobility, a middle class made upof kinsmen to the nobility, people withoutproperty or kinsmen, and slaves, who were ina separate class altogether (Townsend 1980,1983, Veniaminov 1984). Because most trans-portation and communication were by kayak,the Aleut participated in trade and warfareover extremely long distances (Maschner andReedy-Maschner 1998). Ethnographic andarchaeological investigations have demon-strated that the Aleut lived corporately, shar-ing supplies, work areas, and other aspectsof household maintenance (Hoffman 2002).Economically, they were oriented specificallytoward the sea and utilized most edible re-sources. Only on the Alaska Peninsula andadjacent nearby islands are terrestrial re-sources found and only in a few time periodsdid the Aleut regularly exploit caribou, bear,and smaller terrestrial mammals found lo-cally.

The Aleut of the Sanak Islands were aranked polity of perhaps a few hundred livingin several villages across the islands beforeRussian contact. They were in an antagonistictrade relationship with the Alaska PeninsulaAleut and frequently at war with KodiakIslanders. The first historical reference to theislands is in 1762, describing their sea otterwealth, which initially provoked Russianinterest and hostilities (Black 1999:61–64,

O’Leary 2002), and later, peaceful contactsthat included baptisms and trade (Black1999:66). In 1778, Captain James Cook an-chored his ship Discovery off Sanak Islandand took so many large halibut that he namedthe group the ‘‘Halibut Islands’’ (Beaglehole1967). But the Russians had a foothold onAlaska, and after 1808 Sanak was reorganizedinto an artel, the last of six primary artels, orworkers cooperatives under Imperial Russia,for the purposes of mass harvest and exportof sea-otter skins. Artels were led by Russianbaidarshchiks (captains), with Russian andAleut employees supplying the company withskins.

Sea-otter hunting was conducted usingAleut labor, methods, and gear, in parties ofup to 15 baidarkas (kayaks) in the calmersummer months (Veniaminov 1984:330). Rec-ords from that era are sparse, but Khlebni-kov’s report in 1818 ranks Sanak as a topproducer compared with the other artels(1994:170). Aleut hunters were also held indebt to the Russian America Company. In1823, after a decline in otters on the islands,hunters of the Sannakh Artel were trans-ferred to Bel’kovskaia Bay on the mainlandand placed under the Unga baidarshchik,even though Sanak continued to be huntedfrom a distance.

Sanak was not resettled until after theU.S. purchase of Alaska in 1867. Unga andBelkofski Aleuts gradually returned to Sanakalongside Euro-American hunters who werehunting the few remaining sea otters. Formervillages were resettled, and new villages wereestablished. The Alaska Commercial Com-pany built a trading post on the north endof the island at Sanak village. Scandinavianimmigrants took a particular interest in theregion, first for sea otters (until they werehunted out in the 1890s), and then more in-tensively for the codfish industry. These menmarried locally and have left an indelible leg-acy within the eastern Aleut community.

The cod fisheries began in the Aleutians inthe 1870s, with Scandinavian dory fishermenarriving in the Shumagin Islands, Sanak Is-lands, and up to Port Moller in the BeringSea. Schooners from San Francisco to Van-couver sailed to the region stacked with do-


ries and fishermen who fished for cod withhandlines (Shields 2001). This developedinto a shore-based fishery, and codfish sta-tions for salting and storing in barrels wereestablished at Company Harbor (townsite ofSanak), Pavlof Harbor (now Pauloff ) andseveral other locations on Sanak Island, aswell as several on the Alaska Peninsula andin the Shumagin Islands (Figure 1). By 1915,there were 17 shore stations operating in theregion taking over 1 million fish each year(Shields 2001:20). Scandinavian fishermenand Aleuts shared communities to some ex-tent and worked alongside one another.

After 1915, fish began to disappear aroundthe Shumagins and Sanak, and by 1930 shorestations began to close (Shields 2001). But by1899, salmon canneries had been built by thePacific American Fisheries Company on theAlaska Peninsula, offering a viable economicalternative, but with local environmental con-sequences (Figure 2). Villages were estab-lished around these canneries, and graduallyresidents of Sanak and several other formersea otter and codfish based communitiesmoved to these new villages, which remainthe viable fishing villages of King Cove,Sand Point, False Pass, and Nelson Lagoontoday.

Fox farming and cattle industries were alsoattempted on the Sanak Islands with mixedsuccess but allowed some residents to stayuntil 1980. Today, the island supports wildcattle and horses from those economic ex-periments; the Aleut harvest cattle annuallyfor subsistence. Sanak Island is now ownedand managed by the Sanak Corporation andPauloff Harbor Tribe in Sand Point, andmany Aleuts in the region continue to visitthe island and fish its waters.

The historic Aleut were adept at takingadvantage of changing social, political, andenvironmental dynamics. Regardless of theglobal political and economic dominance ofthe time, the Aleut translated their long his-tory of harvesting the North Pacific andsouthern Bering Sea into adaptations thatreflected the larger forces around them butmaintained their identify as coastal foragers.As we show in the following sections, thisflexibility was present in the prehistory ofthe Sanak region as well.

Environmental Background

geological context. The Sanak Islandarchipelago is located at the edge of thecontinental shelf above the Aleutian Trench

Figure 2. Long-term effects of the sockeye salmon harvest on Sanak Island, 1911 to 1927. Today there are fewer than3,500 fish returning to the combined streams of Sanak Island (data from Rich and Ball 1930).


subduction zone and is separated from theAlaska Peninsula by 50 km of open water.The island group is geographically part ofthe greater eastern Aleutian Island archipel-ago, but its location south of the westernAlaska Peninsula effectively separates it fromthe Bering Sea, a physiographic characteristicthat distinguishes it from the rest of the Aleu-tians. The Sanak archipelago is geologicallyrelated to the Shumagin Island group to thenortheast (Moore 1974) and is composedprincipally of rocks of the Cretaceous-agedShumagin Formation—continental shelfslope and turbidite deposits consisting ofmedium- to fine-grained sandstone inter-bedded with mudstone. Fault-bound Jurassic-to Cretaceous-aged chert, volcanic and ma-rine sedimentary rocks occur between Sanakand Long Island (Figure 3). Sanak Islandproper is topographically dominated by SanakPeak, a Tertiary intrusive granitic massif atthe northwestern end of the island. Althoughits bedrock geology is distinct from that of

the Alaska Peninsula, its proximity is criticalin the paleoenvironmental history and dy-namics of the Sanak archipelago, primarilydue to the influence of the peninsula onglaciation of the regional continental shelfand the proximity and eruption frequency ofits closely spaced Holocene volcanic centers,and seismic province. New paleoenvironmen-tal evidence from Sanak Island (sea level andvegetation and climate proxies) is remarkablebecause it provides spatial context and tempo-ral depth to the record of postglacial environ-mental change in the region.

Before this project, our view of the glacialhistory of the western Gulf of Alaska wasdominated by a handful of mapping or mod-eling studies (Detterman 1986, Mann and Pe-teet 1994, Dochat 1997, Wilson and Weber1999), none of which included field observa-tions in the Sanak Islands. The Sanak archi-pelago was overrun by ice from a source areato the north during the last glacial maximum,but field evidence indicates that ice thickness

Figure 3. Sanak Island paleoenvironmental sites: lake and peat core sites (circles), glacial striations or grooves cut ongranodiorite (arrows), and ice override (drumlinoid) features (dashed lines). Stippled areas indicate location of beachridge complexes.


did not exceed 70 m above modern sea level(asl) ( Jordan et al. 2007), an important down-ward revision of ice thickness modeled byMann and Peteet (1994). Glacial striationsand ice flow indicators oriented north-southacross Sanak, the spatial pattern and limitedthickness of till on the island, and till lithol-ogy distinct from that observed in Cold andMorzhovoi bays on the peninsula all indicatethat the Sanak Group was outboard of a re-gional ice sheet axis ( J.W.J., unpubl. data),the thickness and position of which was re-constructed by Mann and Peteet (1994) andsuggested by Dochat (1997) to have beencentered near Sanak. Basal ages from lakescored on Sanak during our project indicatethat ice had disintegrated from the archipel-ago by about 16,000 cal years B.P., severalmillennia earlier than on the peninsula basedon the current chronology there. (All dates inthis paper are calibrated years B.P., unless thedate concerns near-modern events.)

New insights from field evidence of glaci-ation on Sanak are critical for understandingthe record of postglacial sea-level change inthe archipelago and across the adjacent shelf.A thin last glacial maximum ice margin atSanak constrains our estimates of glacial-isostatic depression and uplift to values lowerthan those of the peninsula and is consistentwith the elevation of marine terraces docu-mented on the island (4 to 5 m asl) relativeto the isostatically uplifted shorelines ob-served in areas around Pavlof, Cold, andMorzhovoi bays (25 m þ) (Figure 1) (Wilsonet al. 1998, Jordan 2000, Jordan et al. 2007).Shoreline and ice thickness data from theAlaska Peninsula (Mann and Peteet 1994,Jordan 2000) included in the regional post-glacial rebound model of James (2001) sup-port his findings of relatively low mantleviscosity and limited glacial-isostatic reboundvalues relative to those of previous global-scale models (cf. Peltier 1994). An importantimplication of James’ work for the westernGulf of Alaska and the Sanak archipelago isthat current-day vertical motions due to icemass changes in the late Pleistocene are ex-pected to be of the magnitude of @1 mm orless per year ( James 2001), suggesting thatsea-level fluctuations recorded by late Holo-

cene terraces and beach ridges on Sanak aredriven by nonisostatic processes (e.g., chang-ing storm regimes, seismic events, tsunamis,eustatic sea level change). A well-developedshore platform cut in sedimentary bedrockdeposits around the islands suggests verticalstability of the crust at time scales on the or-der of one to two millennia (Trenhaile 2002).Sand and gravel storm beach ridge sets havebeen deposited in bay heads on Sanak duringthe past 1,000 years, particularly in Sandy Bay(Figure 3), which was rapidly infilled with anextensive sandy beach ridge plain during thattime. Historic records and field observations(Lander 1996, Fryer et al. 2004, Jordan et al.2007) document the effect of tsunamis onSanak; run-up heights exceeded 10 m duringthe 1788 and 1946 events, and although thoseevents were locally catastrophic, the impactalongshore was constrained by the directionfrom which the tsunamis originated.

paleoenvironment. Pollen stratigra-phies were compiled from lake cores (Figure3) and represent the vegetational history ofSanak Island as it evolved in response toclimatic and environmental change sincedeglaciation. Artemisia, adapted to cool/dryconditions, was abundant during initial degla-ciation. Artemisia was also abundant duringthe subsequent Younger Dryas. Cyperaceae,generally adapted to wetter climates, in-creased as Artemisia decreased. Ericaceaepeaked between @7,000 and 4,000 cal B.P.,before the onset of neoglaciation. Sedgesand grasses then dominated from 4,000 to1,000 B.P., and Ericaceae decreased. Sedgesdeclined after 1,000 B.P., and Ericaceae in-creased. The Sanak pollen data are in generalagreement with sites on the peninsula (Table1 and associated references) but extend theregional paleovegetation record by severalmillennia. Changes in pollen stratigraphy areregionally conspicuous during the Holocene,and suggest a shift at@4500 BP from gener-ally warmer (and drier) to cooler conditionsin the eastern Aleutian Islands, the ShumaginIslands, and on the central Alaska Peninsula(Heusser et al. 1985). Localized dunes wereactive at that time based on the lack of stablesurfaces in regional eolian deposits. After3,200 B.P. vegetation changes indicate the


onset of cooler and wetter conditions thatinitiated neoglacial advances on the peninsulaand higher peaks of the Aleutians, and the re-gional stabilization of dunes. Modern coastaltundra developed after 1,000 B.P., broadlyreflecting the influence of the region’s mari-time climate. Multiple tephras preserved inlake cores imply that vegetation on Sanakresponded periodically to this disturbancemechanism (Heusser 1990), and organic car-bon analysis of a peat core from UnimakCove (Figures 3, 4) indicates cyclical inter-ruption of marsh productivity during thepast 5,500 years, which may relate to periodsof storminess and the delivery of inorganicshoreline sediment onto the marsh.

Archaeological Research

The archaeology of Sanak Island was com-pletely unknown before our first preliminarysurvey in 2002. Since that time, over 120 ar-chaeological sites spanning nearly 6,000 yearshave been surveyed, mapped, and most havebeen tested (Figure 5). The goals of our ar-chaeological research have been to create alocal prehistory of the Sanak Islands thatcould be correlated with the better-knownprehistory of the western Alaska Peninsulato investigate macroregional changes in so-cial, economic, and political interactions;

identify regional changes in land use and sitedistributions; reconstruct the population his-tory of the island based on measures of sitesize, density, and numbers of house depres-sions; sample middens for a fine-grained fau-nal record to be analyzed in the context ofmodern species distributions and in light ofthe paleoclimatic record; to use faunal data,and the isotopic analysis of those specimens,to track changes in the structure and func-tioning of the marine ecosystem; and to usethe faunal data to reconstruct prehistoricfood webs and to assess changes in the mod-ern marine ecosystem. The first two of thesegoals are reviewed in this section, and the re-maining goals are discussed in the followingsections. This is considered a very prelimi-nary assessment of the archaeological data inlight of these project goals.

Critical to all of these analyses is that set-tlement archaeology is feasible in the region.House depressions are clearly visible on thesurface, and most sites are single component,or if multicomponent, the components arespatially or stratigraphically distinct. This al-lows population reconstructions based on sitesize or surface depressions to be possible. Italso contributes to our ability to distinguishspatial and temporal trends in the settlementsystem.

The archaeology of the western Alaska

TABLE 1

Climatic Regime Shifts in the Western Gulf of Alaska and Sanak Archipelago

Period Age Range Approximate Climatic Conditions

Early Holocene 9,000–6,200 B.P. Cool and dry6,200–4,500 B.P. Warm and wet

Neoglacial 4,500–3,200 B.P. Cool and wet3,200–2,100 B.P. Cool and wet (increased storminess)

Pre-Medieval climatic anomaly 2,100–1,700 B.P. Warmer1,700–1,100 B.P. Cold

Medieval climatic anomaly 1,100–700 B.P. Warmer and dryLittle Ice Age 700–100 B.P. Cold and wetRecent era 100 B.P.–Present day Very warm

Note: Shading highlights periods recording important changes in air and sea-surface temperature, storminess, and enhanced oceanproductivity reconstructed from various climatic proxies for the North Pacific and Gulf of Alaska (Huesser et al. 1985, Mason andJordan 1993, Mann and Hamilton 1995, Calkin et al. 2001, Hu et al. 2001, Mann 2001, Finney et al. 2002, Jordan and Krumhardt2003, D’Arrigo et al. 2005, Loso et al. 2006, Misarti 2007). There is a cool period at approximately 5,400 B.P. and a warm period be-tween 3,000 and 2,600 B.P. that have been documented in multiple Arctic records. Neither of these is currently visible in our regionalproxy data, but both are possibly present in our settlement and midden data.


Peninsula has been the focus of intensiveresearch for nearly 15 years (Maschner 1998,1999a,b, 2000, 2004a,b, 2008, Maschner andReedy-Maschner 1998, 2005, Jordan 2000,2001, Jordan and Maschner 2000, Maschnerand Jordan 2001, 2008, Hoffman 2002,Maschner and Bentley 2003, Maschner andHoffman 2003, Maschner et al. 2008, 2009).When these data are combined with previousstudies from the region (Weyer 1930, Work-man 1966, McCartney 1969, 1974, 1984,1988, Okada 1980, Yesner 1985, Johnson1988, 1992, Johnson and Winslow 1991, Hol-land 1992, Johnson and Wilmerding 2001), arather complete regional overview is possible.

Extending from Unimak Pass in the westto the central Alaska Peninsula in the east isa region that shares a number of importanthistorical trajectories in technology, houseform, art, and other factors that permit itsclassification as an archaeological region—

one that shared some sort of social, political,and probably ethnic identity. Everything thatwe have discovered so far for Sanak Islandfalls squarely within our understanding ofthe prehistory of the western Gulf of Alaskaregion. Over 150 Accelerator Mass Spec-trometry (AMS) dates on charcoal fromhouse floors, shell middens, and other fea-tures demonstrate a continuous 6,000-yearoccupation of the island, matching our re-gional prehistory for the mainland to thenorth (Figure 6).

The earliest sites are small groups of housedepressions with square stone slab box-hearths along one wall. These houses mayhave been covered with skins as opposed tomore permanent structures. Artifacts includelarge laurel-leaf lance blades, cruciform-based bone and ivory harpoons, and a smallnumber of microblades (Maschner 2008).There are few sites with faunal preservation

Figure 4. Total organic carbon (TOC) measurements for the Unimak Cove peat core. Low organic carbon is a proxymeasure of increased storminess; higher TOC is a measure of calmer oceanic conditions.


between 6,000 and 4,000 years ago becausethere appears to have been little earlyemphasis on shellfish. The earliest preservedmidden deposits are dated to approximately4,500 B.P.

By 3,800 years ago, shell middens arefound across the island, much like the main-land. Although there does not appear to be agreater number of sites, the economic focushas clearly shifted to a greater diversity ofspecies. Shell middens again disappear after3,300 years ago, only to become common af-ter 2,600 years ago. In fact, there are regular

disruptions in the use of shellfish, and a plotof all dates on preserved middens demon-strates this clearly (Figure 7). Thus, althoughthere appears to be a continuous occupationof the island based on Figure 6, the distribu-tion of shell-based middens is discontinuousat best.

At approximately 2,400 B.P., on both themainland and the islands, there is a period ofvillage formation where much larger groupsand larger households arise for a 400-yearspan. Although villages are smaller on theislands than on the mainland, the trend is

Figure 5. Map of the Sanak Islands showing archaeological sites in the project area.


much the same. On both Sanak Island andthe Alaska Peninsula, the period between1,900 and 1,000 B.P. is poorly known, andalthough there are a large number of sites,

large shell middens are rare. The years be-tween 1,000 B.P. and 600 B.P. are tumultu-ous, with the rapid rise and fall of villages,a brief abandonment of the region, and thedevelopment of the historic house form(Maschner 1999a,b). About 400 years ago,the development of large, corporate multi-roomed houses is continuous across the entireregion as is a massive population increase(Hoffman 2002, Maschner and Hoffman2003). Currently, the analysis of the archaeo-logical data is only in its infancy.

A brief summary of the regional climatehistory based on our data and a review of theliterature shows a strong association betweenclimate shifts and the distribution of shellmiddens in time (e.g., Table 1), especially inregard to storminess described earlier (Figure4). It will become obvious in the next sectionthat the demographic trends on the islandpartly mimic the distribution of shell middensas well and the preliminary climatic historypresented here. One major disruption in thecultural sequence on the western Alaska Pen-insula mainland is a disappearance of mostvillages and their populations during the Me-dieval climatic anomaly approximately 750B.P. (Maschner et al. 2009). All indicationsare that Sanak Island went through a similarpopulation decline. A full evaluation andanalysis of these trends is in preparation.

Regionally, the artifacts and house styles,with a trend toward increasing technologicalcomplexity, increasing house size, and in-creasing village size, show the same patternsas those on the Alaska Peninsula (Maschner1999a,b, Maschner and Bentley 2003, Masch-ner and Hoffman 2003), although the trendsare not necessarily linear and our first im-pression is that the trends appear to be con-ditioned by marine productivity. Specificartifact types at key chronological transitionsover the last 6,000 years point to regular con-tact between the Sanak archipelago and thewestern Alaska Peninsula, a connection thatwas observed among historic peoples by earlyexplorers in the region as well. But these dataalso lead us to recognize, especially in har-poon and end blade types, that there were re-gional interactions spanning over 1,000 km ofthe North Pacific Ocean from Kodiak Islandin the east to Umnak Island in the west.

Figure 6. A temporal density plot of 2s ranges of all ra-diocarbon dates in the Sanak archipelago. All radiocarbondates in the paper are presented as calibrated years beforepresent.

Figure 7. Histogram of mean calibrated dates of indi-vidual shell midden components in the Sanak archipel-ago.


A Population Prehistory of Sanak Island

Critical to our long-term understanding ofthe role of humans in the structure and func-tioning of the Sanak ecosystem, and, in turn,the role of that ecosystem in human adaptivecapabilities is modeling the history of humanpopulations on the island. One commonmethod for estimating changes in humanpopulations from archaeological data is toconvert analyses of temporal frequency distri-butions of house or settlement size into esti-mates of human population (Hassan 1978,Chamberlain 2006). This method relies uponethnographic parallels that allow past popula-tion densities to be estimated from modern orhistorical information concerning the rela-tionship between house size and populationdensity (Hassan 1981, Kolb 1985) or settle-ment size and population density (Wiessner1974). An important assumption is thathigher frequencies of dated sites representhigher population densities (Brantingham etal. 2004, Holdaway et al. 2005). Temporalfrequency distributions, however, may beaffected by taphonomic bias where the lossof data increases with time, which Surovelland Brantingham (2007) suggested may belargely responsible for the positive curvilinearfrequency distributions often observed inarchaeological, paleontological, and geologi-cal data. One potential improvement to pop-ulation models based on temporal frequencydistributions could be to focus on short-termvariations, which presumably occur at differ-ent rates than either the time-dependent ef-fects of taphonomic bias or the long-termtrends in other factors that affect human pop-ulations, such as climate change. Surovell andBrantingham (2007) ended their paper withan intriguing aside concerning the possibilityof extracting population trends from biaseddata by converting temporal frequency datainto ‘‘dimensionless’’ ratios or proportions,such as the ratio of juvenile to adult humans,which should preserve information on demo-graphic change despite the potential effects oftaphonomic bias.

Another, largely ignored, problem withpopulation estimates based on temporal fre-quency distribution models is the uncer-

tainty introduced by carbon dating. Moderntechniques of carbon dating can producevery precise date estimates, but these esti-mates always come with an estimated errorrange. Usually, only the midpoint carbondate is used to place an archaeological sampleinto one time period or another. Given thefact that the estimated error associated withany given carbon date may well be plus orminus 100 years or so, it is possible to con-struct multiple estimates of population den-sity, based on the estimated error of all thecarbon dates. One method for dealing withthis problem is to use the estimated error as-sociated with each carbon date to define apossible window of occupation for each sitebased upon probabilistic resampling.

To estimate changes in human occupationon Sanak Island we developed a sophisticatedmodel called ‘‘wiggle.sites’’ in R (version2.4.1, R Core Development Team, 2006)to calculate a ‘‘human occupation index’’(HOI) using temporal frequency data. Thewiggle.sites model combines the idea that itshould be possible to extract persistent short-term demographic trends from temporalfrequency distributions despite potential taph-onomic bias with the idea of a dimensionlessindex that can describe trends in human pop-ulation over time without having to estimateprecisely the magnitude of change.

The model uses probabilistic resamplingto assign an ‘‘occupation window’’ fromwithin the estimated carbon date range foreach village. The size of the window deter-mines the amount of overlap in time betweensites, with a longer window resulting in moreoverlap and a greater smoothing of the index.In this case we used a window length of 100years. Using a smaller window typically re-sults in less smoothing but does not substan-tially change the shape of the resulting index.

After assigning an occupation window toeach site the model calculates the total occu-pied village area active in each 10-year period.The model uses a conversion factor of oneperson per 100 m2 of site area (from regionalethnographic data) to estimate populationdensity from the occupied area. The modelrepeats the entire process 1,000 times and re-ports the results as 25th, 50th, and 75th quan-


tiles. Because the goal of the model is to showhow human occupation changes over time,the shape of the resulting index is more im-portant than the absolute numbers derivedfrom the index.

The HOI for Sanak Island shows threerelatively well-defined periods of human pop-ulation growth and decline (Figure 8). Thefirst period is between approximately 3,000and 1,800 B.P. with a peak around 2,000B.P. The second period peaks at about 1,000B.P., and the third period peaks at about400 B.P. The 25th quantile shows that, giventhe possible range of carbon dates for all thesites, the population of Sanak could havebeen zero at some points during the timeline,such as between approximately 1,800 and1,200 B.P. and even 1,100 to 700 B.P., but italso shows that using even the most conserva-tive estimate of population still reveals thethree distinct periods of population growthand decline.

One potential issue with the HOI is that itdoes not appear to accurately track the pau-city of human occupation on both Sanak and

the lower Alaska Peninsula, ca. 1,000 B.P. and700 B.P. (see later in this section). This is be-cause there are several small sites on Sanakthat all date to the period surrounding ca.1,000 B.P., and the probabilistic model isunable to track the sudden and precipitousdecline in sites directly after that time.

Like many temporal frequency distribu-tions, the HOI does exhibit what appears tobe an exponential increase over time, whichmay be due in part to taphonomic bias. How-ever, although taphonomic bias might bepartly responsible for the differences in peakpopulation between the three periods notedearlier, bias cannot account for the generalshape of the HOI. Instead, the results showthat regardless of possible taphonomic bias,the Aleut population on Sanak Island proba-bly experienced large fluctuations over time.

Given the HOI we can now ask questionsthat link changes in human population onSanak with changes in ecology, climate, andculture. As discussed earlier, the distributionsof sites, and the distributions of sites withmiddens, is at least partially correlated with

Figure 8. The Human Occupation Index (HOI). The HOI is a model of possible ranges of human population onSanak Island based on site area and numbers of houses.


the occupation index. But this is not tauto-logical. There are many sites, some quitelarge, that have no middens leading to largepopulation estimates between the middenpeaks described in the previous section. Theimplication is that although there is some re-lationship between population size and coolerclimates, and a relationship between coolerclimates and midden use, there is a largeamount of variation that is not measured bya simple climate model.

Faunal Studies

The archaeofaunal remains described in thefollowing analyses are derived from seven dis-crete archaeological contexts (many more arecurrently being analyzed) that form a time se-ries spanning the period ca. 4,500 to 500 B.P.All were recovered from single stratigraphicunits dated by multiple radiocarbon determi-nations on wood charcoal. These depositswere screened through 8 mm (1/4 inch)mesh, a sieve gauge we consider appropriateto the ‘‘specimen sizes’’ for the taxa under in-vestigation (e.g., Cannon 1999).

To assess the changes in relative taxo-nomic abundance through time, we used theclimatic model for the western Gulf of Alaskapresented in Table 1. From a marine ecosys-tem perspective, these perturbations may bereferred to as macrolevel regime shifts (Ben-son and Trites 2002, Polovina 2005, Finneyet al. in press) and are used here to assess ma-jor shifts in the distributions of key resourcesand their harvests.

cod population histories. The mid-dens on Sanak Island contain important in-formation on the population histories of fishtaxa. For example, fish bones are allometri-cally linked to animal size and weight, andtherefore measurements from preserved fishbones can be utilized to reconstruct lengthfrequency distributions for prehistoric fishpopulations (e.g., Wheeler and Jones 1989).The ability to do such analysis is critical be-cause fish grow throughout their life span,and thus most life history traits in fish stocksare correlated with size (Reiss 1989). Size istherefore an important metric in modernfisheries management research (e.g., Shin

et al. 2005). As a result, allometric analysiscan provide a crucial means to compare thepreindustrial fisheries record with modernfisheries management data.

As outlined in Maschner et al. (2008) (seealso Betts et al. in press a), we reconstructeda deep time series of Pacific cod (Gadusmacrocephalus) length frequencies over our se-quence and compared this with modern fish-eries lengths taken from the Gulf of Alaska in2005 (Thompson et al. 2006). Here we pro-vide an overview of this analysis and providean updated and in-depth discussion of rela-tionships between changes in cod size andclimatic shifts.

Measurements on the cod remains werederived from the length of the ascending pro-cess of premaxillas and the anterior-posteriorlength of the centra of abdominal (trunk)vertebrae. We estimated the size of eachmeasured element using regression equationsdeveloped by Orchard (2001, 2003). Samplesizes for the Sanak assemblages ranged be-tween a low of 10 to a high of 507 elements.The modern average length of Pacific codwas obtained from data obtained from long-line surveys done by the Alaska FisheriesScience Center in the Gulf of Alaska in 2005(n ¼ 3,308) (Thompson et al. 2006). Al-though data from baited jigs (the fishing tech-nique used by the prehistoric Aleut) were notavailable, modern longline-obtained cod arelikely to be most comparable with fish takenby prehistoric jigging practices. The meanlength recorded for the 2005 longline survey,which occurs as frequencies in size bins, wascalculated using procedures in Gedamke andHoenig (2006).

Figure 9 displays a time series of averagePacific cod lengths spanning the preindustrial(prehistoric) to recent period. To us, thegraph indicates that: (1) preindustrial meanlength varies in a cyclical fashion of increas-ing and then decreasing size, in apparentmulticentennial cycles; and (2) the modernmean length of Pacific cod appears to fallwithin the range of precontact (prehistoric)variability. In fact, t-tests reveal that the mod-ern mean is not significantly different fromthat of five of the precontact assemblages(P < :01).


This latter point is not an unexpected cor-relation; we know from modern managementrecords that Gulf of Alaska cod populationsare not overexploited (e.g., Thompson et al.2006, 2007) and therefore should not exhibita significant harvesting-mediated size reduc-tion (for a discussion of the relationship be-tween average size and exploitation pressuresee Shin et al. [2005]). We have previouslysuggested (Maschner et al. 2008) that modernfisheries practices are largely maintaining thepreindustrial population structure of Pacificcod, and, therefore, using sized-based indica-tors as a metric, modern cod harvest practicesappear to be ‘‘sustainable.’’

What is perhaps equally intriguing is theapparent oscillation in cod size over the last4,500 years. When we compare this profilewith our climatic model (light gray bars arecold regimes, dark gray bars are warmer re-gimes), we see a correlation (Figure 9). Ingeneral, the trends appear to suggest an in-crease in mean length with warm regimesand a decrease in mean length with cold re-gimes. We draw specific attention to the cor-

respondence between size and the Medievalclimatic anomaly (1,000–700 B.P.) and LittleIce Age (700–100 B.P.), respectively the twoclimatic periods immediately preceding themodern era. Following these trends, it istempting to speculate that the slight increasein cod length at the end of the sequence mayin fact reflect a stock response to globalwarming, similar to the increase in mean sizethat occurred during the Medieval climaticanomaly. The implications of this correlationare unclear, but we propose that global warm-ing may be a complicating factor in the recentpopulation histories of Pacific cod.

otariids, climate, and human inter-

actions. We assess the relationship be-tween Steller sea lions (Eumetopias jubatus)and humans by tracking the relative frequen-cies of otariid bones in middens and compar-ing shifts in the resulting profiles. Wemeasure shifts in the frequency of Stellersea lions over the sequence by means of sev-eral abundance indices, a procedure that wenote is necessary to control for potentialperturbations in the frequency of the com-

Figure 9. Gulf of Alaska reconstructed mean cod lengths with the modeled climatic reconstruction demonstratingthat there is a relationship between climatic regimes and cod length, especially after 2,000 B.P. (adapted from Bettset al. in press a and Maschner et al. 2008). Light gray bars are cooler regimes; dark gray bars are warmer regimes.


parison taxa and thus eliminate the issue of‘‘closed arrays’’ (e.g., Betts and Friesen 2006;see Lyman [2008] for discussion about closedarrays).

An abundance index (AI) represents anormed ratio of a highly ranked taxon (A), interms of body size, to a lower-ranked taxon(B), measured as AI ¼ A/Aþ B. Decreasingindex values imply a decline in the abundanceof a taxon, and higher values indicate an in-crease in abundance. Because the measuresincorporate caloric relationships based onbody size (i.e., larger-bodied animals have ahigher caloric return), shifts in the index canreflect changes in return/encounter rate andthus foraging efficiency (e.g., Hildebrandtand Jones 1992, Broughton 1994, 1997,1999, Janetski 1997, Butler 2000, Cannon2000, Grayson 2001; for discussion see Ly-man [2003], Betts and Friesen [2006]). As aresult, when judiciously calculated, AIs canbe used to track the occurrence of resourcedepressions, defined as a reduction in return/

encounter rate caused by availability of prey(e.g., Charnov et al. 1976, Stephens andKrebs 1986).

In Figure 10, we assess changes in twootariid abundance indices against changesin climate. Here the dark boxes representwarmer and less-productive climatic regimes,and the lighter boxes indicate cooler andstormier periods (from Table 1). A relativelyclear correlation occurs between the climaticmodel and the post-Neoglacial sequence afterca. 2,100 B.P. Specifically, we note an in-crease in the otariid abundance index duringa cold period from ca. 1,700 to 1,100 B.P., adecline in otariid abundance during the Me-dieval climatic anomaly ca. 1,100–700 B.P.,and an increase in otariid abundance duringthe Little Ice Age ca. 700 to 100 B.P.

We interpret this correspondence as ota-riid populations responding to climate andmarine productivity shifts. In fact, the graphsuggests that during times of warmer andless-productive climatic regimes, otariids

Figure 10. Otariid abundance indices (AI, Index Value on chart) as compared with climatic model. Cooler regimes(light gray bars) appear to positively influence otariid abundance (from Betts et al. in press b, Lech et al. in press;H.D.G.M., A. W. Trites, K.L.R.-M., M.W.B., A.T., and M. Livingston, unpubl. data). Dark gray bars are warmerregimes.


experienced a form of resource depression,likely a classic ecological depression (e.g.,Charnov et al. 1976). In general the two AIscorrespond well, although the otariid–sea ot-ter AI appears to correspond more favorablywith the climatic model (Table 1), especiallyafter the post-Neoglacial period ca. 2,100B.P. Specifically, the otariid–sea otter AI in-dicates an increase in the otariid abundanceduring a cold period from ca. 1,700 to 1,100B.P., a decline in otariid abundance duringthe Medieval climatic anomaly ca. 1,100–700B.P., and an increase in otariid abundanceduring the Little Ice Age ca. 700 to 100 B.P.The discordance between the two AIs, ca.2,300 B.P. and 1,500 B.P., is explored in a de-tailed manner in Betts et al. (in press b); how-ever it is believed that the discordance is theresult of disproportionate decreases and in-creases in phocid seals during those periodsrelative to other taxa. Given the close rela-tionship of phocid seals to ice platform, itmay be noteworthy that the decrease in pho-cid seals ca. 2,300 B.P. (resulting in a slightincrease in the otariid-phocid AI) was associ-ated with stormier environments, and the po-tential increase in phocid seals (resulting in adecrease in the otariid-phocid AI) was associ-ated with very cold environments (see Table1). It may be that the otariid–sea otter indexmore closely tracks the natural abundance ofotariids, especially as this relates to climaticshifts (see Betts et al. in press b for discus-sion).

Using the human occupation index de-scribed in Figure 8, the archaeological recordindicates that human populations were rela-tively stable during the Neoglacial but beganto rise steadily, and at a relatively predictablerate for hunter-gatherers, during the pre-Medieval climatic anomaly ca. 2,100–1,100B.P. During the Medieval climatic anomaly,populations declined sharply. This was fol-lowed by a notable increase in populationduring the Little Ice Age.

Comparing our schematic populationcurve (derived from Figure 8) against otariidabundance indicates interesting correlationsbetween otariid and human populations (Fig-ure 11). The trend during the Neoglacial isagain murky, but post-Neoglacial there is a

positive correspondence between the numberof humans and the number of otariids. To us,this suggests two things. First, it appears thathumans were not overexploiting sea lionsin the past, at least on Sanak Island. If theywere, declines in otariids should be associatedwith an increase in the number of predators(humans), the hallmark of a classic exploita-tion depression (Charnov et al. 1976). Theexact opposite is evident here. The secondimplication of this relationship is that humansand sea lion populations appear to respond toclimate changes in exactly the same fashion,at least post-Neoglacial. This strongly impliesthat changes in climate and oceanic produc-tivity were an important force in the popu-lation histories of large-bodied, top-tieredpredators, which includes both humans andsea lions.

Stable Isotopes of C and N and Changing PacificRegimes

Stable isotopes have been used in ecologicaland archaeological studies to trace diet andgeographic location for the past few decades.In this study, we integrated both paleoeco-logical studies and archaeological studies todetermine changes through time in marineecosystems utilized by the Aleut for the last4,500 years. Stable isotope signatures are pre-served in many different types of tissue, in-cluding bone collagen, and bone is often wellpreserved in shell middens in coastal Alaska.Bone is also particularly well suited to studiesthat span hundreds or thousands of yearsbecause it has a slower turnover rate thanorgans, blood, and muscle tissue and there-fore reflects a longer period in an organism’slife (Ambrose and Norr 1993, Lambert andGrupe 1993). This allowed us to compare anorganism’s overall trophic levels without con-cerns about seasonal or short-term foragingarea changes. d13C and d15N are used in tro-phic level and dietary reconstructions, andin theory, if changes are discerned in eitherisotope, they can be linked to changes in for-aging habits (Post 2002). Samples of Pacificcod, sockeye salmon, harbor seal, northernfur seal, sea otter, and Steller sea lion werecollected from archaeological middens span-


ning the last 4,500 years on Sanak Island.Bone collagen was extracted from 250 indi-viduals from all six species and analyzed ford13C and d15N.

In general, mean d13C and d15N of all sixspecies from 4,500 to 200 cal years B.P. main-tain expected trophic positions when com-pared with modern prey species (Figure 12).Three species, salmon, cod, and sea otter,were compared within six archaeologicaltime periods. There are significant changesin isotope concentrations within that timeframe for salmon and sea otter but not forPacific cod. Sea otters had a significantdifference in d13C (single-factor ANOVA,F ¼ 3:644, df ¼ 120, P ¼ :07) and in d15N(single-factor ANOVA, F ¼ 3:767, df ¼ 120,P ¼ :003), and changes in d13C of salmonwere significant (single-factor ANOVA,F ¼ 3:012, df ¼ 32, P ¼ :028), but nochanges were found in d15N of salmon (Mis-

arti 2007; Misarti et al. in press). d15N is uti-lized as a marker for trophic position, andd13C is more often linked to the type of plantat the base of a food web, geographic locationof an organism, and productivity levels (Fryand Sherr 1984, Schoeninger and DeNiro1984, Michener and Schell 1994, Doucettet al. 1996, Hobson et al. 1996, Hirons et al.2001, Schell 2001, Post 2002, McCroy et al.2004). These data suggest that otters, withchanges in both carbon and nitrogen isotopes,may have been feeding at slightly differenttrophic levels through time. For example, seaurchins, a favored prey item, are secondaryproducers in a kelp forest ecosystem whilepelagic fish, also otter prey species, are notonly in higher trophic positions but also havephytoplankton as the base of their food web(Estes and Duggins 1995, Watt et al. 2000,Steneck et al. 2002, Estes et al. 2003, Bodkinet al. 2004, Reisewitz et al. 2006). The

Figure 11. Otariid abundance indices (AI, Index Value on chart) compared with the relative human occupation index(HOI) from Figure 8 (without scale). Although there are many more sites included in the HOI than in the faunal data,patterns do emerge. The increase in the otariid AI during the Little Ice Age, the decrease during the Medieval climaticanomaly, and the higher AI value during the general population increase between 1,700 and 1,100 B.P. (another coolperiod) track closely, indicating some relationship between otariid abundance, human populations, and climate after2,000 B.P. (see also Betts et al. in press b). Before 2,000 B.P., there is no obvious relationship.


changes in sea otter carbon and nitrogen iso-topes may be explained by a change in per-centage of diet represented by each of theseprey items. Only d13C changed in salmonover time, suggesting that there was a changein productivity throughout the Gulf of Alaskaor a change in geographic area inhabited bythe fish but not a change in trophic positionof salmon. The lack of change in either d13Cor d15N of cod in the archaeological recordis interesting. Studies over the past 50 yearshave noted a change in diet for Pacificcod during regime shifts in the northeasternPacific, but no change in foraging location

has been recorded (Albers and Anderson1985, Yang 2004, Ciannelli and Bailey 2005).If all prey items, even though changingthroughout time, held the same trophic posi-tions and were in the same foraging areas,then there would be no discernable changein either d13C or d15N.

The Intertidal Landscape

The intertidal habitats of the Sanak archipel-ago provide a wealth of human resources thatwere a key source of food for residents dur-ing times of low oceanic productivity or

Figure 12. Trophic position of archaeological mammals and fish with some selected modern prey species. All samplescorrected for fractionation from muscle tissue to bone collagen. Prey data compiled from Schoeninger and DeNiro(1984), Duggins et al. (1989), Boutton (1991), Hobson and Welch (1992), Hobson et al. (1997), Satterfield (2000),Hirons (2001), Kline and Willette (2002). Archaeological data from Misarti (2007).


decreased offshore access such as storms. Pri-marily these intertidal regions are rockybenches and cobble beaches, with occasionalsand or pebble shores. Food resources of theintertidal areas are relatively plentiful, quicklyrenewed, and predictable; all ethnographicand archaeological sources agree that there isthus little doubt that the intertidal has beencritical to the history of this region. The pre-served middens, described earlier, provide in-formation on the structure of the prehistoricintertidal.

Results from experimental collections onSanak indicate that an hour of collection ofcontemporary intertidal resources (i.e., shell-fish) provides up to 23 kg of meat mass (anaverage of 4 kg and a median of 316 g acrossall shellfish taxa collected). Given such con-temporary resource abundance it would befeasible to meet a reasonable daily proteinrequirement for humans (e.g., 40 g/day [Er-landson 1988]), especially in concert witha readily accessible source of carbohydrate(e.g., seeds or bulbs from terrestrial plants;or even marine algae, which contain roughly3 kcal per gram of dry mass [Paine 1971]).Some of these resources are quickly gatheredin large volume (e.g., littorine snails, marinealgae); others (e.g., sea urchins) are slower toharvest but may provide a substantial foodvalue in a reasonable amount of time (i.e.,1 hr of harvesting time could provide wellin excess of one individual’s requirements).Thus, similar to elsewhere in Alaska (e.g., Er-landson 1988), contemporary intertidal com-munities could provide substantial resourcesfor human populations on Sanak Island. So,do we see evidence that Sanak Islanders reliedupon the intertidal for such resources, and isthere evidence that these intertidal communi-ties have changed through time—whether anartifact of human behavior or otherwise?

Using archaeological and contemporarybiological surveys, we now explore whetherthere have been shifts in the compositionof intertidal communities around the islandover the last 5,000 years. Our contemporarysurveys involve exhaustive vertical transects(sampled using 0.25 m2 quadrats in which allmacroscopic organisms were recorded) pro-viding a robust sample of the intertidal

biological communities around Sanak Island.For comparison between ancient and moderndata we use whole-organism biomass, calcu-lated from samples of modern abundanceand ancient shell mass (using experimentallyderived regressions to back-calculate animalbiomass from shell mass), as our response.There are clear differences in the biomasscomposition of modern intertidal communityassemblages and ancient shellfish middens onthe island (Figure 13). This pattern is drivenprimarily by differences in abundances of chi-tons, snails, and mussels. These three taxaaccount for 66% of the dissimilarity betweenmodern and ancient assemblages in a similar-ity percentage (SIMPER) analysis (Clarke1993), which quantifies species’ contributionsto sample dissimilarity (further results notpresented). Snails and mussels are moreabundant in the ancient midden records, butchitons are more characteristic of the modernintertidal assemblages. What is driving thesedifferences? We ought to be able to explainthe differences between modern and ancientassemblages by shifts in either (1) intertidalhabitat types (e.g., from rocky to sandy); (2)preferences of harvesters; or (3) productivity,climate, or weather that alter ecosystems orthe accessibility of certain habitats. Here weexplore the first two drivers of change anddraw attention to earlier sections of thispaper for discussion of the climatic changeson the island.

First, there is no evidence of shifts inthe proportions of rocky versus sedimentedshoreline habitat. The midden remains aredominated by rocky shore species, with noevidence of sandy shore taxa dominating theassemblage, throughout the course of avail-able data. A substantial shift in shorelinehabitat types would be expected to be evidentin the relatively continuous midden recordavailable here; thus we conclude that it is un-likely that there have been large-scale shiftsof the fundamental habitat types around theisland over the last 5,000 years (see similaranalyses with opposite results in Graham etal. [2003] or Masters and Gallegos [1997]).

To assess the harvesting preferences forparticular taxa, we quantified return on col-lection effort using empirical data from Sanak


Figure 13. Comparison of ancient midden (a) and modern intertidal (b) abundances of major taxa around SanakIsland. The length of each wedge represents the relative abundance of the taxon in the sample unit (a vertical transectsurvey of the modern assemblage and one midden sample of the ancient assemblage).

Island. Total meat weight collected over time(taxon-specific timed collections by knowl-edgeable individuals repeated at differentlocations at appropriately low tide levels) isnormalized by modern abundance (from sur-veys mentioned earlier) for each group of taxato estimate a standardized return on collect-ing. The results indicate that three taxa areexpected to be preferred: snails, mussels, andchitons. This helps explain the abundance ofsnails in the middens relative to the modernshores; snails provide more meat for a giventime spent collecting. Further, littorinid snailsare often found in massive aggregations andoccur on wave-sheltered high intertidalshorelines that are accessible much of theyear. It is interesting that although chitonsare a rewarding and thus likely a preferredshellfish, they are less abundant throughoutthe midden record. If the midden record re-flects the natural abundance of chitons, thisresult indicates that rocky intertidal commu-nities on the island may have varied over thecourse of settlement history.

The intertidal record for Sanak Island in-dicates that: (1) humans relied heavily uponthe intertidal for food resource needs (basedon the rich and extensive midden record onthe island); (2) the intertidal communitiescould nutritionally support substantial hu-man populations on the island (based onreconstructions of prehistoric biological com-munities); and (3) although the moderncommunities appear different from ancientmidden assemblages, there is little evidencefor clear temporal trends in the harvesting ofintertidal communities over the course of the5,000-year midden record on Sanak Island(data forthcoming). This implies that thestructure and dynamics of the intertidal eco-system has likely had a strong influence onhuman history on Sanak Island, yet humanhistory does not appear to have had a power-ful influence on the intertidal system.

Role of Humans in the Intertidal Food Web

Through a combination of field observationsof the current intertidal communities ofSanak Island, additional trophic (feeding) in-formation culled from the literature, detailed

zooarchaeology of Sanak Island middens,ethnographic data, and interviews with mod-ern Aleuts in the region, a detailed picture ofthe Sanak Island intertidal food web, and theposition of humans in it, begins to emerge(Figure 14). Initial results reveal a species-rich and complex food web with 725 feedinglinks among 164 taxa represented, makingthis the most diverse intertidal food web evercompiled and one of the few food webs toexplicitly include humans as a node (see alsoLink 2002).

The Sanak Island intertidal food web is acumulative community food web—it repre-sents the feeding relationships among co-occurring taxa that live or feed in the inter-tidal, as integrated over space (multiple Sanakintertidal sites) and time (the last 5,000 years).For initial analyses of the network structureof trophic interactions among species thatlive or feed in the Sanak intertidal system,we assume that the composition of theSanak Island intertidal communities has notchanged dramatically over the last few thou-sand years in terms of what species arepresent. For example, all intertidal speciescollected in the middens are also present inthe current intertidal community. Althoughabundances of those and the other speciesclearly change through time, abundance dataare not necessary to analyze basic propertiesof food-web structure. This initial version ofthe web is incomplete because it still lackssome of the terrestrial taxa, particularly birds,that feed in the intertidal, and inasmuch as nofood web can ever be entirely complete be-cause there are always species and links thatremain unidentified. However, it does give auseful first look at the structure of the foodweb and how the Aleut fit into it.

On average, each species in this initialversion of the Sanak intertidal food web has4.4 trophic links to other species (i.e., linksper species, or L=S ¼ 4:4) through links toresource (prey) species plus those from con-sumer (predator) species. The connectanceðCÞ of the food web, which describes the pro-portion of possible links among all speciesthat actually occur, usually calculated as linksper species2 ðL=S2Þ, is 0.027, or 2.7%. Thisfalls at the low end of the range observed for


other food webs of @2% to 30% (Dunneet al. 2004). The mean trophic level of thespecies in the food web is 2.2. Trophic levelrepresents how many steps energy must taketo get from an energy source to a species. Byconvention, primary producers are assigneda trophic level of 1, and obligate herbivoreshave a trophic level of 2. Trophic level canbe calculated in a variety of ways; here weuse the short weighted trophic level algo-rithm (Williams and Martinez 2004b). Onaverage, the shortest chain of feeding linksconnecting each pair of species is 2.6 linkslong. This ‘‘path length’’ gives an indicationof how quickly trophic effects can spreadthrough a food web, with 2.6 indicating alonger mean path, and thus longer potentialpropagation time, than seen in most otherwebs (Williams et al. 2002). Twenty-threepercent of the taxa in the web are basal (i.e.,autotrophs or detrital categories), including adiverse set of algae taxa, 65% are inverte-brates, 9% are fishes, and 3% are mammals

(including humans). There is excellent reso-lution of basal and invertebrate taxa, unlikemost published marine and estuary food webdata sets (Dunne et al. 2004). In terms of tro-phic roles, 14% of the taxa are herbivores ordetritivores, 6% are cannibals, and 56% areomnivores (i.e., feed on taxa at more thanone trophic level). The harbor seal (Phocavitulina) has the highest trophic level (3.7),and the most common resource is detritus,which is consumed by 60 taxa.

Until they left the island in the late 1970s,humans played important roles in the Sanakintertidal food web. They fed directly on 40taxa, making them the most general con-sumer of intertidal taxa, with the next mostgeneral species, a sea star (Pycnopodia helian-thoides), consuming 35 taxa. They were strongomnivores, feeding on everything from algaeto a variety of invertebrates to the highesttrophic level taxon, the harbor seal, makingtheir own trophic level intermediate at 2.9.In addition to the 40 taxa humans fed on

Figure 14. Sanak Island intertidal food web. Spheres represent 164 taxa and links represent 725 feeding interactions.Basal taxa (autotrophs such as algae as well as detrital categories) are shown at the bottom, with the highest trophiclevel taxa shown at top. Aleut are indicated by the labeled black sphere and are located in the background due to theirhigh number of links (40) to resource taxa. Image produced with FoodWeb 3D, www.Foodwebs.org.


directly, 92 additional taxa occurred withinhuman-topped food chains, with only 31 taxa(only one basal species) having no direct orindirect trophic linkages to humans. Thus,humans played unique roles in the SanakIsland intertidal food web through their ex-treme generality and omnivory, and therewas a large potential for humans to affect theintertidal ecosystem directly and indirectlythrough hunting and gathering, as well asthrough mechanisms such as food storage,technology, planning, and prey switching.

Future work will include analyses of thetopological roles of humans in the other ma-jor habitat types of Sanak Island (i.e., terres-trial, freshwater, marine) and comparisons ofthe structure of those habitat-specific webswith each other, with the larger metaweb en-compassing all four habitats, and with otherfood webs (Dunne et al. 2008). Early-stageresearch is making use of the initial intertidaltopological analyses to create scenarios formodeling the dynamics of complex ecologi-cal trophic networks (Brose et al. 2006) toexplore how humans’ roles and capabilitiestended to promote or inhibit species andcommunity persistence and other measuresof ecosystem stability in this type of pre-industrial coupled human-natural system.

discussion and conclusions

The late Holocene paleoecology of the west-ern Gulf of Alaska, and the Sanak Islandregion in particular, has shown considerablevariation in general climate regimes, particu-larly in regard to the Pacific Decadal Oscilla-tion (PDO). Trites et al. (2007) have shown,as have Finney et al. (in press), Misarti(2007), and others, that these changing cli-matic regimes alter regional marine produc-tivity in a number of complex ways and thatbasic marine productivity is key to under-standing the population histories of mosthigher mammals (including humans) andother species (Maschner et al. 2009).

We suspect that the human occupation in-dex, which is identical to population trends inthe greater Alaska Peninsula region (Masch-ner et al. 2009), highly similar to those in theentire North Pacific (Maschner 1997, Ames

and Maschner 1999, Fitzhugh 2003), and par-allel with new data from the South Pacific(Nunn 2007), is directly a product of rapidand rather radical regime shifts affectingprimary productivity. This appears to be es-pecially true of the higher trophic levels, af-fecting human and sea mammal populationsdirectly, and is even perceptible in the shiftsin the size of Pacific cod (e.g., Maschneret al. 2008).

In historic times this does not appear tohave had the same impact. The ethnohistoricdata imply that Sanak Island populations werenot so constrained or influenced by marineproductivity but rather by the mitigatingeffects of global economic and political re-gimes. But marine productivity does appearto influence which economic forces have pre-cedence at any one time. For example, thecollapse of the nineteenth-century sea otterstocks, as well as the decline in the Atlanticcod fishery (both human caused), certainlyled to the success of the early twentieth-century cod fishery in the North Pacific. Butthe collapse of the North Pacific cod fisheryin the late 1930s, leading to cattle and othereconomic diversification on Sanak, and cod’sresurgence in the mid-1970s as a dominantfishery appears to have been driven by oce-anic regime shifts (Maschner et al. 2008;Reedy-Maschner et al. unpubl. field notes).

It should be recognizable that there issome correlation between the human occupa-tion index (Figure 8), the distribution of siteswith preserved shell middens (Figure 7), andthe total organic carbon data (Figure 4). For-aging models would predict that higher pop-ulations would result in an expansion of thesubsistence base, requiring an increase inthe use of lower-ranked, small-package foods,such as shellfish. The correlation between theHOI and the distribution of middens impliesthis to be true. But the periods with the great-est numbers of middens occur during cooleroceanic regimes, the periods with the highestmarine productivity. The total organic car-bon (TOC) data provide a key to this discrep-ancy. Decreasing TOC in nearshore coresis likely a product of increasing storminess.Storms are the only condition that keeps tra-ditional hunters and fishermen from going


offshore. So although marine productivity isup during cooler periods, increasing stormi-ness results in fewer days out hunting andfishing, requiring the use of lower-rankedspecies during storm periods. Conversely,this implies that during warmer, less-stormyperiods, the decreased marine productivityprovided sufficient resources, except in theworst of climatic regime shifts.

Elsewhere, we have argued (Maschneret al. 2009; H.D.G.M., A.W.T., K.L.R.-M.,M.W.B., A.T., and M.L., unpubl. data) thatone of the key factors in Aleut interactionswith sea mammals was their constant needfor Steller sea lion skins, the only viable cov-ering for ocean-going kayaks. We wonderif the close relationship over the last 1,500years between the human population indexand the otariid AI is not simply a product ofthe fact that Aleut males had to harvest up tosix Steller sea lions every year to cover theirkayaks, meaning that the actual sea lion har-vest is not dictated by diet but by the numberof Aleut males owning kayaks.

Likewise, it is possible that the changesin the archaeological record of otter nitrogenand carbon stable isotopes are driven by Aleutharvesting practices (Misarti 2007). Sea ottersare a more-localized, nearshore species, andhunting may have changed population struc-ture over time. If population size affects theavailability of otter prey then the Aleut mayhave inadvertently affected what food itemswere available to otters depending on howheavily otters were hunted at any one time.Salmon on the other hand, although heavilyharvested along the Alaska Peninsula, seemto have been more affected by marineproductivity and climate regimes. Perhapssalmon were less impacted prehistorically bynumbers harvested by the Aleut and more bychanges to the marine ecosystems out in thedeep ocean.

The archaeological data concerning Pacificcod allow us to draw two critical conclusions.The first is that minor variations in the popu-lation structure and size structure of Pacificcod appear to be driven by overall marineproductivity, as measured in changing cli-matic regimes. The lack of change over4,500 years in the stable isotope data for cod

is another line of evidence that leads us tothe same conclusion. But more important, thesize structure of modern populations doesnot appear to have been affected by moderncommercial fishing, a strong case for the sus-tainability for that species and fishery (e.g.,Maschner et al. 2008), a situation unlike thehistory of the North Atlantic (Betts et al. inpress a).

In nearshore areas, it appears that humansharvested large quantities of shellfish dur-ing particular time periods, especially 3,600,2,200, and 400 years ago, and that these shell-fish were important contributions to the diet.Our first attempts to place the local and pre-historic peoples in the structure of a dynamicfood web show that humans have complexinteractions at multiple levels. Harvestingover 40 taxa of intertidal species, the ancientAleut are the most connected species in theintertidal food web. This means that re-moving the Aleut from the food web, as oc-curred both in the 1820s and then again inthe 1970s, probably had measurable organiz-ing effects on the structure of the intertidalsystem.

It is also clear from the species harvested(M.W.B. and H.D.G.M., unpubl. data) thathumans did have local impacts on the struc-ture of the intertidal zone, harvesting lower-ranked species as the intertidal zone wasincreasingly exploited. On the other hand,there were periods when, despite large humanpopulations, shellfish were rarely harvested ornot harvested at all. But it is important tonote that although the modern intertidalzone is different from the intertidal regimesharvested by the prehistoric Aleut between5,000 and 400 years ago, the modern regimedoes not appear to be a product of human ac-tivities but rather may be more closely causedby changing oceanic conditions.

These observations are at best preliminary.We are just now to the stage in the investiga-tions that more complex modeling efforts willbe possible and we may with confidence be-gin addressing in detail the hypotheses putforward at the beginning of this paper. Butone observation is clear—modern 25- to 50-year fisheries and climate data sets for theNorth Pacific region represent little of the


dramatic shifts in climate, productivity, andhuman feedbacks that have occurred over thelast 5,000 years. Although there is some evi-dence forthcoming that humans have hadlong-term impacts on the structure of the in-tertidal zone, or even on offshore fisheries,what is clear is that much of the behavioralecologies of many species such as sea lions,or of the interactions among species aroundvillage sites, are products of both passive andactive human engineering of those behaviors.As such, at least for the last few thousandsof years, there is no ‘‘natural’’ North Pacificecosystem without humans as one of the keyspecies in that ecosystem.

acknowledgments

We gratefully acknowledge the support andaccess provided by the Sanak Corporationand the Pauloff Harbor Tribe. We would es-pecially like to thank the many undergraduateand graduate students who helped with field-work and laboratory analyses used in this pa-per; among these the contributions of DietaLund, Garrett Knudsen, Krystal Williams,Krista Williams, and Julie Kramer are ac-knowledged. We also thank the Idaho StateUniversity Office of Research and the Re-search Vice-President for financial and infra-structure support. All opinions, findings, andconclusions or recommendations expressedin this material are those of the authors anddo not necessarily reflect the views of the Na-tional Science Foundation.

Literature Cited

Albers, W. D., and P. J. Anderson. 1985. Dietof Pacific cod, Gadus macrocephalus, andpredation on the northern pink shrimp,Pandalus borealis, in Pavlof Bay, Alaska.Fish. Bull. 83:601–610.

Ambrose, S. H., and L. Norr. 1993. Experi-mental evidence for the relationship ofthe carbon isotope ratios of whole dietand dietary protein to those of bone colla-gen and carbonate. Pages 1–38 in J. B.Lambert and G. Grupe, eds. Prehistorichuman bone archaeology at the molecularlevel. Springer-Verlag, New York.

Ames, K., and H. D. G. Maschner. 1999.Peoples of the Northwest coast: Theirprehistory and archaeology. Thames andHudson, London.

Amundsen, C. C. 1972. Amchitka bioenvir-onmental program: Plant ecology ofAmchitka—final report. Battelle MemorialInstitute, Columbus, Ohio.

Bank, T. P. 1953. Ecology of prehistoricAleutian village sites. Ecology 34:246–264.

Beaglehole, J. C. 1967. The journals of Cap-tain James Cook on his voyages of discov-ery: The voyage of the Resolution andDiscovery, 1776–1780. 2 vols. CambridgeUniversity Press, Cambridge.

Benson, A. J., and A. W. Trites. 2002. Eco-logical effects of regime shifts in the Be-ring Sea and eastern North Pacific. FishFish. 3:95–113.

Betts, M., and T. M. Friesen. 2006. Decliningforaging returns from an inexhaustibleresource?: Abundance indices and belugawhaling in the western Canadian Arctic. J.Anthropol. Archaeol. 25:59–81.

Betts, M., H. D. G. Maschner, and D. Clark.In press a. Zooarchaeology of the ‘Fishthat Stops’: Using archaeofaunas to con-struct long-term time series of Atlanticand Pacific cod populations. In M. Mossand A. Cannon, eds. The archaeologyof North Pacific fisheries. University ofAlaska Press, Anchorage.

Betts, M., H. D. G. Maschner, and V. Lech.In press b. A 4500 year time-series of ota-riid abundance on Sanak Island, westernGulf of Alaska. In T. Braje and T. Rick,eds. Sea dogs of the North Pacific: The ar-chaeology and historical ecology of seals,sea lions, sea cows, and sea otters. Univer-sity of California Press, Berkeley.

Black, L. 1999. Sanak. Pages 59–77 in L.Black, S. McGowan, J. Jacka, N. Taksami,and M. Wright, eds. The history and eth-nohistory of the Aleutians East Borough.The Limestone Press, Kingston, Ontario,Canada.

Bodkin, J. L., G. G. Esslinger, and D. H.Monson. 2004. Foraging depths of seaotters and implications to coastal marinecommunities. Mar. Mamm. Sci. 20:305–321.


Boutton, T. W. 1991. Stable carbon isotoperatios of natural materials: II. Atmo-spheric, terrestrial, marine, and freshwaterenvironments. Pages 173–185 in D. C.Coleman and B. Fry, eds. Carbon isotopetechniques. Academic Press, San Diego.

Brantingham, P. J., K. W. Kerry, A. I.Krivoshapkin, and Y. V. Kuzmin. 2004.Time-space dynamics in the early-UpperPaleolithic of northern Asia. Pages 255–283 in D. B. Madsen, ed. Entering Amer-ica: Northeast Asia and Beringia beforethe Last Glacial Maximum. University ofUtah Press, Salt Lake City.

Brose, U., R. J. Williams, and N. D. Marti-nez. 2006. Allometric scaling enhances sta-bility in complex food webs. Ecol. Lett.9:1228–1236.

Broughton, J. 1994. Declines in mammalianforaging efficiency during the late Holo-cene, San Francisco Bay, California. J.Anthropol. Archaeol. 13:371–401.

———. 1997. Widening diet breadth, de-clining foraging efficiency, and prehistoricharvest pressure: Ichthyofaunal evidencefrom the Emeryville Shellmound, Califor-nia. Antiquity 71:845–862.

———. 1999. Resource depression and in-tensification during the late Holocene,San Francisco Bay: Evidence from theEmeryville Shellmound vertebrate fauna.Univ. Calif. Anthropol. Rec. 32.

Butler, V. 2000. Resource depression on theNorthwest coast of North America. Antiq-uity 74:649–661.

Calkin, P. E., G. Wiles, and D. J. Barclay.2001. Holocene coastal glaciation ofAlaska. Quat. Sci. Rev. 20:449–461.

Cannon, M. 1999. A mathematical model ofthe effects of screen size on zooarchaeo-logical relative abundance measures. J. Ar-chaeol. Sci. 26:205–214.

———. 2000. Large mammal relative abun-dance in Pithouse and Pueblo PeriodArchaeofaunas from southwestern NewMexico: Resource depression among theMimbres-Mogollon? J. Anthropol. Ar-chaeol. 19:317–347.

Castellini, M. 1993. Report on marine mam-mal working group. Pages 4–13 in Is itfood?: Addressing marine mammal and

seabird declines. Alaska Sea Grant Rep.93-01.

Chamberlain, A. 2006. Demography in ar-chaeology. Cambridge Manuals in Archae-ology. Cambridge University Press,Cambridge.

Charnov, E., G. Orians, and K. Hyatt. 1976.Ecological implications of resource de-pression. Am. Nat. 110:247–259.

Chesson, P., and N. Huntly. 1994. Temporalhierarchies of variation and the mainte-nance of diversity. Plant Species Biol.8:195–206.

———. 1997. The roles of harsh and fluctu-ating conditions in the dynamics of eco-logical communities. Am. Nat. 150:519–553.

Ciannelli, L., and K. M. Bailey. 2005. Land-scape dynamics and resulting species in-teractions: The cod-capelin system in thesoutheastern Bering Sea. Mar. Ecol. Prog.Ser. 291:227–236.

Clarke, K. R. 1993. Non-parametric multi-variate analyses of changes in communitystructure. Aust. J. Ecol. 18:117–143.

Crumley, C. L., ed. 1994. Historical ecology:Cultural knowledge and changing land-scapes. School of American ResearchAdvanced Seminar Series. School ofAmerican Research Press, Santa Fe, NewMexico.

Dale, V. H., S. Brown, R. Haeuber, N. T.Hobbs, N. Huntly, R. Naiman, W. Rieb-same, M. Turner, and T. J. Valone. 2000.Ecological principles and guidelines formanaging the use of land. Ecol. Appl.10:639–670.

D’Arrigo, R., E. Mashig, D. Frank, R. Wil-son, and G. Jacoby. 2005. Temperaturevariability over the past millennium in-ferred from northwestern Alaska treerings. Clim. Dyn. 24:227–236.

Detterman, R. L. 1986. Glaciation of theAlaska Peninsula. Pages 151–170 in T. D.Hamilton, K. M. Reed, and R. M. Thor-son, eds. Glaciation in Alaska: The geo-logic record. Alaska Geological Society,Anchorage.

Dochat, T. 1997. The glacial history of thelower Alaska Peninsula. Ph.D. diss., Uni-versity of Wisconsin, Madison.


Doucett, R. R., G. Power, D. R. Barton, R. J.Drimmie, and R. A. Cunjak. 1996. Stableisotope analysis of nutrient pathways lead-ing to Atlantic salmon. Can. J. Fish. Aquat.Sci. 53:2058–2066.

Duggins, D. O., C. A. Simenstad, and J. A.Estes. 1989. Magnification of secondaryproduction by kelp detritus in coastalmarine ecosystems. Science (Washington,D.C.) 245:170–173.

Dunne, J. A., R. J. Williams, and N. D.Martinez. 2002. Network structure andbiodiversity loss in food webs: Robustnessincreases with connectance. Ecol. Lett.5:558–567.

———. 2004. Network structure and robust-ness of marine food webs. Mar. Ecol.Prog. Ser. 273:291–302.

Dunne, J. A., R. J. Williams, N. D. Martinez,R. A. Wood, and D. E. Erwin. 2008.Compilation and network analyses ofCambrian food webs. PLoS. Biol. 5:e102.doi:10.1371/journal.pbio.0060102.

Ellis, J. E., and K. Galvin. 1994. Climatepatterns and land use practices in thedry zones of Africa. BioScience 44:340–349.

Ellis, J. E., and D. M. Swift. 1988. Stabilityof African pastoral ecosystems: Alternateparadigms and implications for develop-ment. J. Range Manage. 41:450–459.

Erlandson, J. M. 1988. The role of shellfishin prehistoric economies: A protein per-spective. Am. Antiq. 53:102–109.

Estes, J. A., and D. O. Duggins. 1995. Seaotters and kelp forests in Alaska: General-ity and variation in a community ecologicalparadigm. Ecol. Monogr. 65:75–100.

Estes, J. A., M. L. Riedman, M. M. Staedler,M. T. Tinker, and B. E. Lyon. 2003. Indi-vidual variation in prey selection by seaotters: Patterns, causes and implications.J. Anim. Ecol. 72:144–155.

Finney, B. P. 1998. Long-term variabilityin Alaskan sockeye salmon abundance de-termined by analysis of sediment cores.North Pac. Anadromous Fish. Comm.Bull. 1:388–395.

Finney, B. P., J. Alheit, K. C. Emeis, D. B.Field, D. Guitierrez, and U. Struck. Inpress. Paleoecological studies on variabil-

ity in marine fish populations: A long-term perspective on the impacts of climatechange on marine fisheries. J. Mar. Syst.GLOBEC Special Issue.

Finney, B. P., I. Gregory-Eaves, M. S. V.Douglas, and J. P. Smol. 2002. Fisheriesproductivity in the northeastern PacificOcean over the past 2,200 years. Nature(Lond.) 416:729–733.

Finney, B. P., I. Gregory-Eaves, J. Sweetman,M. S. V. Douglas, and J. Smol. 2000. Im-pacts of climatic change and fishing onPacific salmon abundance over the past300 years. Science (Washington, D.C.)290:795–799.

Fitzhugh, J. B. 2003. The evolution of com-plex hunter-gatherers: Archaeological evi-dence from the North Pacific. Kluwer-Plenum, New York.

Fry, B., and E. B. Sherr. 1984. d13C Mea-surements as indicators of carbon flow inmarine and freshwater ecosystems. Con-trib. Mar. Sci. 27:13–47.

Fryer, G. J., P. Watts, and L. F. Pratson.2004. Source of the great tsunami of 1April 1946: A landslide in the upper Aleu-tian forearc. Mar. Geol. 203:201–218.

Gedamke, T., and J. M. Hoenig. 2006. Esti-mating mortality from mean length datain nonequilibrium situations, with applica-tion to the assessment of goosefish. Trans.Am. Fish. Soc. 135:476–487.

Graham, M. H., P. K. Dayton, and J. M.Erlandson. 2003. Ice ages and ecologicaltransitions on temperate coasts. TrendsEcol. Evol. 18:33–40.

Grayson, D. 2001. Explaining the develop-ment of dietary dominance by a single un-gulate taxon at Grotte XVI, Dordogne,France. J. Archaeol. Sci. 28:115–125.

Hassan, F. A. 1978. Demographic archaeol-ogy. Pages 49–103 in M. B. Shiffer, ed.Advances in archaeological method andtheory. Vol. I. Academic Press, New York.

———. 1981. Demographic archaeology.Academic Press, New York.

Heusser, C. J. 1990. Late Quaternary vegeta-tion of the Aleutian Islands, southwesternAlaska. Can. J. Bot. 68:1320–1326.

Heusser, C. J., L. E. Heusser, and D. M.Peteet. 1985. Late-Quaternary climate


change on the American North Pacificcoast. Nature (Lond.) 315:485–487.

Hildebrandt, W., and T. Jones. 1992. Evolu-tion of marine mammal hunting: A viewfrom the California and Oregon coasts. J.Anthropol. Archaeol. 11:360–401.

Hirons, A. C. 2001. Trophic dynamics ofpinniped populations in Alaska using stablecarbon and nitrogen isotope ratios. Ph.D.diss., University of Alaska, Fairbanks.

Hirons, A. C., D. M. Schell, and B. P. Finney.2001. Temporal records of d13C and d15Nin North Pacific pinnipeds: Inferencesregarding environmental change and diet.Oecologia (Berl.) 129:591–601.

Hobson, K. A., D. M. Schell, D. Renouf, andE. Noseworthy. 1996. Stable carbon andnitrogen isotopic fractionation betweendiet and tissues of captive seals: Implica-tions for dietary reconstructions involvingmarine mammals. Can. J. Fish. Aquat. Sci.53:528–533.

Hobson, K. A., J. L. Sease, R. L. Merrick, andJ. F. Piatt. 1997. Investigating trophicrelationships of pinnipeds in Alaska andWashington using stable isotope ratios ofnitrogen and carbon. Mar. Mamm. Sci.13:114–132.

Hobson, K. A., and H. E. Welch. 1992. De-termination of trophic relationships withina high arctic marine food web using d13Cand d15N analysis. Mar. Ecol. Prog. Ser.84:9–18.

Hoffman, B. W. 2002. The organization ofcomplexity: A study of late prehistoricvillage organization in the eastern Aleutianregion, Alaska. Ph.D. diss., University ofWisconsin, Madison.

Holdaway, S., P. Fanning, and J. Shiner.2005. Absence of evidence or evidence ofabsence? Understanding the chronologyof indigenous occupation of western NewSouth Wales, Australia. Archaeol. Oceania40:33–49.

Holland, K. M. 1992. Rethinking Aleutianprehistory as viewed from a lithic data-base. Ph.D. diss., Arizona State University,Tempe.

Hu, F. S., E. Ito, T. A. Brown, B. B. Curry,and D. R. Engstrom. 2001. Pronouncedclimatic variations in Alaska during the

last two millennia. Proc. Natl. Acad. Sci.U.S.A. 98:10552–10556.

Jackson, J. B. C., M. X. Kirby, W. H. Berger,K. A. Bjorndal, L. W. Botsford, B. J. Bour-que, R. H. Bradbury, R. Cooke, J. Erland-son, J. A. Estes, T. P. Hughes, S. Kidwell,C. B. Lange, H. S. Lenihan, J. M. Pan-dolfi, C. H. Peterson, R. S. Steneck, M. J.Tegner, and R. R. Warner. 2001. Histori-cal overfishing and the recent collapse ofcoastal ecosystems. Science (Washington,D.C.) 293:629–638.

James, T. S. 2001. Postglacial rebound analy-sis for better determination of earthquake-related deformation and rheology at sub-duction zones. USGS/NEHRP Award01HQGR0005. Program Element II: Re-search on Earthquake Occurrence and Ef-fects. http://earthquake.usgs.gov/research/external/reports/01HQGR0005.pdf

Janetski, J. 1997. Fremont hunting and re-source intensification in the eastern GreatBasin. J. Archaeol. Sci. 24:1075–1088.

Johnson, D. 2003. Species composition, di-versity, and productivity of plant commu-nities on Prehistoric Aleut village sites onthe western Alaska Peninsula. M.S. thesis,Idaho State University, Pocatello.

Johnson, L. L. 1988. Archaeological surveysof the outer Shumagin Islands, Alaska,1984, 1986. Arct. Anthropol. 25:139–170.

———. 1992. Prehistoric settlement patternsand population in the Shumagin Islands.Anthropol. Pap. Univ. Alsk. 24:73–88.

Johnson, L. L., and E. G. Wilmerding. 2001.Bringing down the house: Modeling con-struction and deconstruction of Aleutsemisubterranean houses. Pages 127–149in D. E. Dumond, ed. Archaeology in theAleut zone of Alaska: Some recent re-search. Univ. Oreg. Anthropol. Pap. 58.

Johnson, L. L., and M. A. Winslow. 1991.Paleoshorelines and prehistoric settle-ments on Simeonof and Chernabura Is-lands, outer Shumagin Islands, Alaska. InL. L. Johnson, ed. Paleoshorelines andprehistory: An investigation of method.CRC Press, Boca Raton, Florida.

Jones, C. G., and J. H. Lawton. 1995. Link-ing species and ecosystems. Chapman andHall, New York.


Jones, C. G., J. H. Lawton, and M. Shachak.1997. Positive and negative effects of or-ganisms as physical ecosystem engineers.Ecology 78:1946–1957.

Jordan, J. W. 2000. Coastal paleogeographyof the western Alaska Peninsula. Ph.D.diss., University of Wisconsin, Madison.

———. 2001. Late Quaternary sea levelchange in southern Beringia: Post-glacialemergence of the western Alaska Penin-sula. Quat. Sci. Rev. 20:509–523.

Jordan, J. W., B. P. Finney, and H. D. G.Maschner. 2007. Geomorphic history ofthe Sanak Islands: Implications for ecosys-tem change and human settlement. Societyfor American Archaeology, 72nd AnnualMeeting, Austin.

Jordan, J. W., and A. Krumhardt. 2003. Post-glacial climate and vegetation history ofthe western Alaska Peninsula. Alsk. J.Anthropol. 1:16–33.

Jordan, J. W., and H. D. G. Maschner. 2000.Coastal paleogeography and human occu-pation of the lower Alaska Peninsula. Geo-archaeology 15:385–414.

Khlebnikov, K. T. 1994. Notes on RussianAmerica, Parts II–V: Kad’iak, Unalashka,Atkha, the Pribylovs. Compiled by R. G.Liapunova and S. G. Fedorova. M. Ram-say, transl. R. A. Pierce, ed. LimestonePress, Kingston, Ontario, Canada. [Origi-nally published in 1979 in Russian.]

King, A. A., and W. M. Schaffer. 2001. Thegeometry of a population cycle: A mecha-nistic model of snowshoe hare demogra-phy. Ecology 82:814–830.

Kline, T. C., and T. M. Willette. 2002.Pacific salmon (Oncorhynchus spp.) earlymarine feeding patterns based on N-15/N-14 and C-13/C-12 in Prince WilliamSound, Alaska. Can. J. Fish. Aquat. Sci.59:1626–1638.

Kolb, C. C. 1985. Demographic estimatesin archaeology: Contribution from ethno-archaeology on Mesoamerican peasants.Curr. Anthropol. 26:581–599.

Lambert, J. B., and G. Grupe. 1993. Prehis-toric human bone: Archaeology at the mo-lecular level. Springer-Verlag, New York.

Lander, J. F. 1996. Tsunamis affectingAlaska: 1737–1996. NGDC Key to Geo-

physical Research, Documentation No.31. National Geophysical Data Center,Boulder, Colorado.

Lantis, M. 1984. Aleut. Pages 161–184 in D.Damas, ed. Handbook of North AmericanIndians. Vol. 5. Arctic. Smithsonian Insti-tution Press, Washington, D.C.

Lech, V., M. Betts, and H. Maschner. Inpress. An analysis of seal, sea lion, and seaotter consumption patterns on Sanak Is-land, Alaska: A 4500 year record on Aleutconsumer behavior. In T. Braje and T.Rick, eds. Sea dogs of the North Pacific:The archaeology and historical ecology ofseals, sea lions, sea cows, and sea otters.University of California Press, Berkeley.

Liapunova, R. G. 1996. Essays on the ethnog-raphy of the Aleuts (at the end of the eigh-teenth and the first half of the nineteenthcentury). J. Shelest, transl. University ofAlaska Press, Fairbanks.

Link, J. 2002. Does food web theory work formarine ecosystems? Mar. Ecol. Prog. Ser.230:1–9.

Loso, M. G., R. S. Anderson, S. P. Anderson,and P. J. Reimer. 2006. A 1500-year rec-ord of temperature and glacial responseinferred from Varved Iceberg Lake,Southcentral Alaska. Quat. Res. 66:12–24.

Lyman, R. L. 2003. The influence of timeaveraging and space averaging on the ap-plication of foraging theory in zooarch-aeology. J. Archaeol. Sci. 30:595–610.

———. 2008. Quantitative paleozoology.Cambridge University Press, New York.

Mann, D. H. 2001. Climate during the pastmillennium. Weather 56:91–102.

Mann, D. H., and T. D. Hamilton. 1995.Late Pleistocene and Holocene paleo-environments of the North Pacific coast.Quat. Sci. Rev. 14:449–471.

Mann, D. H., and D. M. Peteet. 1994. Extentand timing of the last glacial maximum insouthwestern Alaska. Quat. Res. 42:136–148.

Maschner, H. D. G. 1997. Settlement andsubsistence in the later prehistory of Te-benkof Bay, Kuiu Island. Arct. Anthropol.34:4–99.

———. 1998. Salmon run volatility, subsis-tence, and the development of North Pa-


cific societies. Pages 11–28 in HokkaidoMuseum of Northern Peoples, ed. Pro-ceedings of the 12th International AbashiriSymposium: Salmon fishery in the northand its change through time. Associationfor the Promotion of Northern Cultures,Abashiri, Hokkaido, Japan.

———. 1999a. Prologue to the prehistory ofthe lower Alaska Peninsula. Arct. Anthro-pol. 36:84–102.

———. 1999b. Sedentism, settlement andvillage organization on the lower AlaskaPeninsula: A preliminary assessment.Pages 56–76 in B. Billman and G. Fein-man, eds. Settlement pattern studies inthe Americas: Fifty years since Viru.Smithsonian Institution Press, Washing-ton, D.C.

———. 2000. Catastrophic change and re-gional interaction: The southern BeringSea in a dynamic world system. Pages252–265 in M. Appelt, J. Berglund, andH. C. Gulløv, eds. Identities and culturalcontacts in the Arctic. Proceedings from aConference at the Danish National Mu-seum, Copenhagen, 30 November–2 De-cember 1999. Danish National Museumand Danish Polar Center, Copenhagen.

———. 2004a. Redating the Hot SpringsVillage site. Alsk. J. Anthropol. 2:100–116.

———. 2004b. Traditions past and present:Allen McCartney and the Izembek Phaseof the western Alaska Peninsula. Arct.Anthropol. 41:98–111.

———. 2008. Fishtails, ancestors, and old is-landers: Chirikof Island, the Alaska Penin-sula, and the dynamics of western Alaskaprehistory. Alsk. J. Anthropol. 6:171–184.

Maschner, H. D. G., and R. A. Bentley. 2003.The power law of rank and household onthe North Pacific. Pages 47–60 in R. A.Bentley and H. D. G. Maschner, eds.Complex systems and archaeology: Empir-ical and theoretical applications. Univer-sity of Utah Press, Salt Lake City.

Maschner, H. D. G., M. W. Betts, K. L.Reedy-Maschner, and A. W. Trites. 2008.A 4500 year time series of Pacific cod(Gadus macrocephalus): Archaeology, re-gime shifts, and sustainable fisheries. Fish.Bull. 106:386–394.

Maschner, H. D. G., B. Finney, J. Jordan, A.Tews, N. Misarti, and G. Knudsen. 2009.Did the North Pacific ecosystem collapsein AD 1000? In H. D. G. Maschner,O. W. Mason, and R. McGhee, eds. Thenorthern world AD 900–1400. Universityof Utah Press, Salt Lake City. Pp. 33–57.

Maschner, H. D. G., and B. W. Hoffman.2003. The development of large corporatehouseholds along the North Pacific Rim.Alsk. J. Anthropol. 1:41–63.

Maschner, H. D. G., and J. W. Jordan. 2001.The Russell Creek manifestation of theArctic Small Tool Tradition on the west-ern Alaska Peninsula. Pages 151–172 inD. E. Dumond, ed. Archaeology in theAleut zone of Alaska: Some recent re-search. Univ. Oreg. Anthropol. Pap. 58.

———. 2008. Catastrophic events and punc-tuated culture change: The southern Be-ring Sea and North Pacific in a dynamicglobal system. In D. Papagianni, R. H.Layton, and H. D. G. Maschner, eds.Time and change: Archaeological and an-thropological perspectives on the longterm. Oxbow, Oxford, United Kingdom.Pp. 95–113.

Maschner, H. D. G., and K. L. Reedy-Maschner. 1998. Raid, retreat, defend (re-peat): The archaeology and ethnohistoryof warfare on the North Pacific. J. Anthro-pol. Archaeol. 17:19–51.

———. 2005. Aleuts and the sea. Archaeol-ogy. March/April, 63–70.

Mason, O. K., and J. W. Jordan. 1993.Heightened North Pacific storminess dur-ing synchronous late Holocene erosion ofNorthwest Alaska beach ridges. Quat. Res.40:55–69.

Masters, P. M., and D. R. Gallegos. 1997.Environmental change and coastal adapta-tions in San Diego County during theMiddle Holocene. Pages 11–22 in J. M.Erlandson and M. A. Glassow, eds. Ar-chaeology of the California coast duringthe Middle Holocene. Institute of Ar-chaeology, University of California, LosAngeles.

McCartney, A. P. 1969. Prehistoric Aleut in-fluences at Port Moller, Alaska. Anthropol.Pap. Univ. Alsk. 14:1–16.


———. 1974. Prehistoric cultural integrationalong the Alaska Peninsula. Anthropol.Pap. Univ. Alsk. 16:59–84.

———. 1984. Prehistory of the Aleutianregion. Pages 119–135 in D. Damas, ed.Handbook of North American Indians.Vol. 5. Arctic. Smithsonian InstitutionPress, Washington, D.C.

———. 1988. Maritime adaptation in south-ern Alaska. Pages 19–51 in H. Okada, ed.Proceedings of the International Sympo-sium on Maritime Adaptations in theNorth Pacific. City of Abashiri, Hokkaido,Japan.

McCartney, A. P., and D. W. Veltre. 1999.Aleutian Island prehistory: Living in insu-lar extremes. World Archaeol. 30:503–515.

McCroy, C. P., R. Gradinger, A. M.Springer, B. Iverson, S. Iverson, and S.Budge. 2004. Stable isotopes reveal foodweb shifts due to arctic sea ice decline.Paper presented at AGU Joint Assembly,Montreal, Canada.

Michener, R. H., and D. M. Schell. 1994.Stable isotope ratios as tracers in marineaquatic food webs. Pages 136–156 in K.Lajtha and R. H. Michener, eds. Stableisotopes in ecology and environmentalscience. Blackwell Scientific Publication,London.

Misarti, N. 2007. Six thousand years ofchange in the Northeast Pacific: An inter-disciplinary view of maritime ecosystems.Ph.D. diss., University of Alaska, Fair-banks.

Misarti, N., B. Finney, H. Maschner, and M.Wooller. In press. Changes in a NortheastPacific marine ecosystem over 4500 years:Evidence from stable isotope analysis ofbone collagen from archaeological mid-dens. The Holocene.

Moore, J. C. 1974. Geologic and structuralmap of the Sanak Islands, southwesternAlaska. U.S. Geol. Surv. Misc. Invest. Ser.Map I-817.

Morrison, M. L., B. G. Marcot, and R. W.Mannan. 1998. Wildlife-habitat relation-ships, concepts and applications. 2nd ed.University of Wisconsin Press, Madison.

Moser, J. F. 1902. Alaska salmon investiga-

tions in 1901. Pages 173–398 in Bull. U.S.Fish Comm. Vol. XXI for 1901.

Nunn, P. D. 2007. Climate, environment,and society in the Pacific during the lastmillennium. Vol. 6. Developments inearth and environmental sciences. ElsevierScience, Amsterdam.

Ocean Studies Board and Polar ResearchBoard. 2003. The decline of the Stellersea lion in Alaskan waters: Untanglingfood webs and fishing nets. National Aca-demies Press, Washington, D.C.

Okada, H. 1980. Prehistory of the AlaskaPeninsula as seen from the Hot SpringsSite, Port Moller. Pages 103–112 in Y.Kotani and W. B. Workman, eds. AlaskaNative culture and history. Senri Ethno-logical Studies No. 4. National Museumof Ethnology, Osaka.

O’Leary, M. 2002. Estimating casualties forthe Aleut resistance, 1763–1775. Paperpresented at the Annual Alaska Anthropo-logical Association meeting, April 2002,Anchorage, Alaska.

Orchard, T. 2001. The role of selected fishspecies in Aleut palaeodiet. M.S. thesis,University of Victoria, Victoria, BritishColumbia.

———. 2003. An application of the linearregression technique for determininglength and weight of six fish taxa: Therole of selected fish species in Aleut pa-laeodiet. Archaeopress, Oxford, UnitedKingdom.

Paine, R. T. 1971. The measurement and ap-plication of the calorie to ecological prob-lems. Annu. Rev. Ecol. Syst. 2:145–164.

Peltier, W. R. 1994. Ice-age paleotopog-raphy. Science (Washington, D.C.) 265:195–201.

Polis, G. A., W. B. Anderson, and R. D. Holt.1997. Toward an integration of landscapeand food web ecology: The dynamics ofspatially subsidized food webs. Annu. Rev.Ecol. Syst. 28:289–316.

Polis, G. A., and S. D. Hurd. 1996. Linkingmarine and terrestrial food webs: Allochth-onous input from the ocean supports highsecondary productivity on small islandsand coastal land communities. Am. Nat.147:396–423.


Polis, G. A., and K. O. Winemiller. 1996.Food webs: Integration of patterns anddynamics. Chapman and Hall, New York.

Polovina, J. J. 2005. Climate variation, regimeshifts, and implications for sustainable fish-eries. Bull. Mar. Sci. 76:233–244.

Post, D. M. 2002. Using stable isotopes to es-timate trophic position: Models, methods,and assumptions. Ecology 83:703–718.

Power, M. E., D. Tilman, J. A. Estes, B. A.Menge, W. J. Bond, L. S. Mills, G. Daily,J. C. Castilla, J. Lubchenco, and R. T.Paine. 1998. Challenges in the quest forkeystones. BioScience 46:609–620.

Reedy-Maschner, K. L. 2001. Aleut identityand indigenous commercial economies:Local responses under global pressures inthe eastern Aleutians. Alsk. J. Anthropol.1:62–82.

———. 2004. Aleut identity and indigenouscommercial fisheries. Ph.D. diss., Cam-bridge University, Cambridge, UnitedKingdom.

———. 2007. The best-laid plans: Limitedentry permits and limited entry systemsin eastern Aleut culture. Hum. Organ.66:210–225.

Reisewitz, S. E., J. A. Estes, and C. A. Simen-stad. 2006. Indirect food web interactions:Sea otters and kelp forest fishes in theAleutian archipelago. Oecologia (Berl.)146:623–631.

Reiss, M. J. 1989. The allometry of growthand reproduction. Cambridge UniversityPress, Cambridge, United Kingdom.

Rich, W. H., and E. M. Ball. 1930. Statisticalreview of the Alaska salmon fisheries. PartI: Bristol Bay and the Alaska Peninsula.Bull. U.S. Bur. Fish. Vol. XLIV, 1928.

Rick, T., and J. Erlandson. 2008. Humanimpacts on ancient marine ecosystems: Aglobal perspective. University of Califor-nia Press, Berkeley.

Satterfield, F. R. 2000. A comparison of sock-eye salmon (Oncorhyncis nerka) in two cli-mate regimes in the North Pacific Oceanusing stable carbon and nitrogen isotoperatios. Institute of Marine Science, Fair-banks, Alaska.

Savinetsky, A. B., N. K. Kiseleva, and B. F.Khassanov. 2004. Dynamics of sea mam-

mal and bird populations of the BeringSea region over the last several millennia.Palaeogeogr. Palaeoclimatol. Palaeoecol.209:335–352.

Schell, D. M. 2001. Carbon isotope ratiovariations in Bering Sea biota: The roleof anthropogenic carbon dioxide. Limnol.Oceanogr. 46:999–1000.

Schoeninger, M. J., and M. DeNiro. 1984.Nitrogen and carbon isotopic compositionof bone collagen from marine and terres-trial animals. Geochim. Cosmochim. Acta48:625–639.

Shacklette, H. T., ed. 1969. Vegetation ofAmchitka Island, Aleutian Islands, Alaska.U.S. Geol. Surv. Prof. Pap. 648.

Shields, Captain E. 2001. Salt of the sea: ThePacific Coast cod fishery and the last daysof sail. Pacific Heritage Press, Lopez Is-land, Washington.

Shin, Y.-J., M.-J. Rochet, S. Jennings, J. G.Field, and H. Gislason. 2005. Using size-based indicators to evaluate the ecosys-tem effects of fishing. ICES J. Mar. Sci.62:384–396.

Simenstad, C. A., J. A. Estes, and K. W. Ken-yon. 1978. Aleuts, sea otters, and alternatestable-state communities. Science (Wash-ington, D.C.) 200:403–411.

Steneck, R. S., M. H. Graham, B. J. Bourque,D. Corbett, J. M. Erlandson, J. A. Estes,and M. J. Tegner. 2002. Kelp forest eco-systems: Biodiversity, stability, resilienceand future. Environ. Conserv. 29:436–459.

Stephens, D., and J. Krebs. 1986. Forag-ing theory. Princeton University Press,Princeton, New Jersey.

Surovell, T. A., and P. J. Brantingham. 2007.A note on the use of temporal frequencydistributions in studies of prehistoric de-mography. J. Archaeol. Sci. 34:1868–1877.

Thompson, G. G., M. W. Dorn, and D. G.Nichol. 2006. Assessment of the Pacificcod stock in the Gulf of Alaska. Pages147–220 in Stock assessment and fisheryevaluation report for the groundfish re-sources of the Gulf of Alaska. NorthPacific Fisheries Management Council,Anchorage, Alaska.

Thompson, G. G., J. N. Ianelli, M. W. Dorn,and M. Wilkins. 2007. Assessment of the


Pacific cod stock in the Gulf of Alaska.Pages 169–194 in Stock assessment andfishery evaluation report for the ground-fish resources of the Gulf of Alaska. NorthPacific Fisheries Management Council,Anchorage, Alaska.

Townsend, J. B. 1980. Ranked societies ofthe Alaskan Pacific Rim. Senri Ethnologi-cal Studies 4:123–156.

———. 1983. Pre-contact political organiza-tion and slavery in Aleut societies. Pages120–132 in E. Tooker, ed. The develop-ment of political organization in NativeNorth America. Proc. Am. Ethnol. Soc.1979.

Trenhaile, A. S. 2002. Rock coasts, with par-ticular emphasis on shore platforms. Geo-morphology 48:7–22.

Trites, A. W., A. J. Miller, H. D. G. Masch-ner, M. A. Alexander, S. J. Bograd, J. A.Calder, A. Capotondi, K. O. Coyle, E. D.Lorenzo, B. P. Finney, E. J. Gregr, C. E.Grosch, S. R. Hare, G. L. Hunt, J.Jahncke, N. B. Kachel, H.-J. Kim, C.Ladd, N. J. Mantua, C. Marzban, W. Mas-lowski, R. Mendelssohn, D. J. Neilson,S. R. Okkonen, J. E. Overland, K. L.Reedy-Maschner, T. C. Royer, F. B.Schwing, J. X. L. Wang, and A. J. Win-ship. 2007. Bottom-up forcing and thedecline of Steller sea lions (Eumetopiasjubatus) in Alaska: Assessing the ocean cli-mate hypothesis. Fish. Oceanogr. 16:46–67.

Turchin, P. 2003. Historical dynamics: Whystates rise and fall. Princeton Studies inComplexity. Princeton University Press,Princeton, New Jersey.

Veniaminov, I. 1984. Notes on the islands ofthe Unalaska District. [1840]. L. T. Blackand R. H. Goeghegan, transls. R. A.Pierce, ed. Alaska history, No. 27. TheLimestone Press, Kingston, Ontario, Can-ada.

Watt, J., D. B. Siniff, and J. A. Estes. 2000.Inter-decadal patterns of population anddietary change in sea otters at AmchitkaIsland, Alaska. Oecologia (Berl.) 124:289–298.

Weyer, E. M., Jr. 1930. Archaeological mate-rial from the village site at Hot Springs,

Port Moller, Alaska. Am. Mus. Nat. Hist.Anthropol. Pap. 31:239–279.

Wheeler, A., and A. Jones. 1989. Fishes.Cambridge University Press, Cambridge,United Kingdom.

Wiessner, P. 1974. A functional estimator ofpopulation from floor area. Am. Antiq.39:343–349.

Williams, R. J., E. L. Berlow, J. A. Dunne,A. L. Barabasi, and N. D. Martinez. 2002.Two degrees of separation in complexfood webs. Proc. Natl. Acad. Sci. U.S.A.99:12913–12916.

Williams, R. J., and N. D. Martinez. 2004a.Stabilization of chaotic and non-permanent food-web dynamics. Eur. Phys.J. B 38:297–303.

———. 2004b. Trophic levels in complexfood webs: Theory and data. Am. Nat.163:458–468.

Wilson, F. H., and F. R. Weber. 1999. Qua-ternary geology of the Cold Bay and FalsePass quadrangles, Alaska Peninsula. Geo-logic Studies in Alaska by the U.S. Geo-logical Survey. U.S. Geol. Surv. Prof.Pap. 1633:51–71.

Wilson, F. H., F. R. Weber, T. M. Dochat,T. P. Miller, and R. L. Detterman. 1998.Revised geologic map of the Cold Bayand False Pass quadrangles, Alaska Penin-sula. Open File Report 97-866. U.S. Geo-logical Survey, Washington, D.C.

Workman, W. B. 1966. Prehistory at PortMoller, Alaska Peninsula, in the lightof fieldwork in 1960. Arct. Anthropol.3:132–153.

Worm, B., and R. A. Myers. 2003. Meta-analysis of cod-shrimp interactions revealstop-down control in oceanic food webs.Ecology 84:162–173.

Yang, M. S. 2004. Diet changes of Pacific cod(Gadus macrocephalus) in Pavlof Bay associ-ated with climate changes in the Gulf ofAlaska between 1980 and 1995. Fish. Bull.102:400–405.

Yesner, D. R. 1980. Maritime hunter-gatherers: Ecology and prehistory. Curr.Anthropol. 21:727–750.

———. 1981. Archaeological applications ofoptimal foraging theory: Harvest strategiesof Aleut hunter-gatherers. Pages 148–170


in B. Winterhalder and E. A. Smith, eds.Hunter-gatherer foraging strategies. Uni-versity of Chicago Press, Chicago.

———. 1985. Cultural boundaries and eco-logical frontiers in coastal regions: An ex-ample from the Alaska Peninsula. Pages51–91 in S. W. Green and S. M. Perlman,eds. The archeology of frontiers andboundaries. Academic Press, New York.

———. 1987. Life in the Garden of Eden:Causes and consequences of the adoptionof marine diets by human societies. Pages

285–310 in M. Harris and E. Ross, eds.Food and evolution. Temple UniversityPress, Philadelphia, Pennsylvania.

———. 1988. Effects of prehistoric humanexploitation on Aleutian sea mammal pop-ulations. Arct. Anthropol. 25:28–43.

———. 1998. Origins and development ofmaritime adaptations in the Northwest Pa-cific region of North America: A zooarch-aeological perspective. Arct. Anthropol.35:204–222.


an introduction to the biocomplexity of sanak island...

Documents