For funding and logistical support, many thanks to the Social Sciences and Humanities Research Council of Canada, Polar Continental Shelf Project, Aurora Research Institute, and Northern Scientific Training Program. The map was produced by Michael O’Rourke, and the abstract was translated by Marie-Annick Prévost. I am also grateful to my collaborators Claire Alix, Cathy Cockney, Peter Dawson, and Brian Moorman. My greatest thanks go to all members of the field crews and visiting experts: Chuck Arnold, Danielle Desmarais, Jesse Fraser, Becky Goodwin, Colleen Hughes, Adam Jahraus, Rémi Méreuze, Mike O’Rourke, Letitia Pokiak, Lawrence Rogers and Rose Scott.
1The long-term history of the circumpolar North is currently facing increasing threats from changing climates, which are causing widespread destruction of archaeological sites. Archaeologists working in many northern regions are aware of this issue, and are addressing it in diverse ways. This paper will present a brief overview of one relevant project: Arctic Cultural Heritage At Risk (Arctic Char). Arctic Char is a multi-pronged program of research centred in the culturally unique Mackenzie Delta region, northwestern Canada, home to modern Inuvialuit communities. It is intended to investigate spatial patterns in the threats to the archaeological record in this region, and to begin to mitigate them through mapping and excavation of critically important sites.
2Over the past two decades, processes associated with climate change have emerged as a paramount concern for cultural heritage at a global scale. Nowhere is this more true than in the circumpolar North, where climate change is well-documented and is already causing massive and ongoing destruction of archaeological sites through four interrelated processes. First, permafrost and permanent snow packs are melting at unprecedented rates (Jorgenson et al. 2010), exposing long-frozen sites to increased chemical impacts, microbial activity and rapid drying, and in turn leading to accelerated destruction of delicate organic remains (Elberling et al. 2011). Second, sea levels are rising (Vafeidis et al. 2008), in a process which is projected to continue and accelerate through the present century, thus exposing coasts to increased erosion. Third, weather patterns are changing, and in some cases are expected to lead to increased storminess (Atkinson 2005). Importantly, these three factors often amplify each other, leading to increased cumulative impacts. In addition to these three direct natural processes, sea ice loss is leading to increasing human activities, particularly relating to Arctic shipping (Ho 2010). This in turn contributes to increased mining and port construction, damaging wake impacts on shorelines, and increasing ecotourism which often includes visits to fragile archaeological sites.
3The potential impacts of these processes are regionally variable across the circumpolar world, and some regions stand out as being particularly vulnerable. In Canada, at the top of the list is the outer Mackenzie Delta region, which is vulnerable for two main reasons. First, much of the western Arctic is currently sinking due to sediment load and isostatic factors (Lajeunesse and Hanson 2008). When combined with -climate-change induced eustatic rise in sea levels, the relative sea level is rising at a greater pace than most other regions in the Arctic (Lantuit and Pollard 2008). Second, current climate models indicate that the western Arctic contains many areas where permafrost is expected to thaw at a relatively rapid rate (Lantz and Kokelj 2008), in part due to the presence of extensive ground ice which can melt rapidly, leading to destabilization of surrounding soil (Dallimore et al. 1996). Importantly, modern climate change did not initiate the massive coastal erosion in the region; rather, processes that were already underway are currently being accelerated.
4As is true for most areas in the Arctic, impacts of these factors are further exacerbated by the fact that most past human settlement occurred on the coasts due to an economic reliance on marine species, leading to sites being situated where erosion is most severe. For example, several of the largest and most important Inuvialuit settlements in the region, including the very early Thule Inuit site of Washout on Herschel Island (Friesen and Hunston 1994), and the important precontact bowhead whale hunting site of Nuvugaq on the Tuktoyaktuk Peninsula (Gordon 1971), have already been completely destroyed by erosion. Other key coastal sites such as the large Inuvialuit villages of Kitigaaryuit (a National Historic Site; McGhee 1974) and Kuukpak (Arnold 1988, 1994; Friesen and Arnold 1995) are actively eroding; and the World Monuments Fund placed the Inuvialuit and Eurocanadian structures on Herschel Island, just west of the study area, on its 2008 “Watch List” of the 100 most endangered sites in the world, due to climate change impacts (WMF 2008).
5The archaeological record of the Mackenzie Delta region is not worth researching only because it is under severe threat. Rather, it is worth saving, and studying, because it is critically important to Inuvialuit culture, to the history of northern Canada, and to broader anthropological issues relating to hunter-gatherer economy, relations with the environment, and social organization.
6Our current knowledge of regional culture history comes from three sources (see overviews in Alunik et al. 2003; Betts 2008; Friesen 2013; McGhee 1974). First, Inuvialuit elders retain extensive detailed knowledge of many aspects of life in the region, both from their own experiences and as passed down from earlier generations (Hart 2011; Lyons 2010; Nagy 1994). Second, beginning with Franklin’s (1828) second voyage to the Arctic in 1826, we have a sporadic documentary record, supplemented in later periods by the Inuvialuit historian Nuligak’s (1966) autobiography, and professional anthropological fieldwork, particularly that of Stefansson (1919) and Petitot (1876, 1887). Third, archaeologists have been working in the region since the 1950s, establishing a solid regional chronology as well as engaging in excavation of a number of important sites. Despite these sources, the culture history of the region is poorly understood relative to adjacent Arctic regions.
7Leaving aside the generally ephemeral record of Paleo-Inuit and earlier peoples, the culture history of the region explodes in complexity with the arrival of Thule Inuit around AD 1200 (Friesen and Arnold 2008). Thule Inuit are the earliest direct ancestors of Inuvialuit, and developed from earlier Birnirk and Punuk tradition peoples in the Bering Strait region of Siberia and Alaska. The earliest Thule sites in the Mackenzie Delta are located in coastal locations situated to maximize access to bowhead and beluga whales, though a variety of other resources were also exploited. These earliest Inuit lived in villages composed of small, single-family semi-subterranean driftwood houses (Friesen 2009). During the 14th and 15th centuries, a major shift occurred, with higher populations, occupation of a greater diversity of locations including some inland settlements, a move to larger houses, and shifts in material culture (Betts 2008; Friesen 2013; Morrison 1997, 2000). These changes probably result from a combination of internal social dynamics linked to large, permanent population aggregations (Friesen 1999), a process of economic intensification linked in part to acquisition of new technologies, including fish nets (Betts and Friesen 2004), and climate change which saw cooling and increasing sea ice (D’Arrigo and Jacoby 1992).
8By the 15th century, the archaeological record indicates that Inuvialuit lifeways were similar to those recorded by early Europeans in the 19th century. Inuvialuit lived in a region extending from the Alaska–Yukon border in the west to Cape Bathurst in the east. They were divided into around eight named regional groups occupying tightly defined territories, each of which was centred on a particularly large settlement, and each of which engaged in a somewhat different annual cycle based on seasonally available resources (Alunik et al. 2003). The Mackenzie Delta region itself, which is the focus of Arctic Char, was home to three regional groups: the Kuukpangmiut whose central village of Kuukpak was on Richards Island; the Kitigaaryungmiut whose central village of Kitigaaryuit was located on the southwestern portion of the Tuktoyaktuk Peninsula; and the Nuvugarmiut, whose central village was located at Nuvugaq near the northeastern tip of the Tuktoyaktuk Peninsula. The first two of these groups had economies which revolved around the late summer beluga whale hunt; the third was positioned to hunt the much larger bowhead whales. All were large, complex settlements which would have held hundreds of individuals during their summer peak occupations, with population dispersing in other seasons to obtain fish, caribou, seals, waterfowl, and other resources. Inuvialuit were the most populous and socially complex Inuit society in Arctic Canada, with formal leadership roles, complex architecture, well-developed trade in prestige goods, and elaborate decoration on material culture. By the late 19th century, epidemic losses (Usher 1971), migrations from Alaska (Bockstoce 1986), and introduced Euroamerican ideas and material culture (Friesen 2013) contributed to a rapid reorganization of Inuvialuit life into its modern form.
9The culture history of the Mackenzie Delta is still poorly understood, due to the sparse ethnographic record and the fact that relatively few archaeologists have specialized in its study. This is unfortunate, as it is critically important both because of the unique nature of Inuvialuit culture, and because it is one of the few areas in Arctic Canada which has an uninterrupted archaeological record lasting from early Thule to the historic period. We also do not understand fully the nature of variability within the region, or the nature of interactions between the major Inuvialuit regional groups.
10As a result of the importance of the region’s archaeological record, combined with the severe threats it faces, the Arctic Char project was developed as a collaboration between university-based scientists and the Inuvialuit Cultural Resource Centre (Icrc) (fig. 1). Icrc, under the direction of Cathy Cockney, is the Inuvialuit agency formally charged with stewardship of language, culture, and heritage, and is engaged in an array of activities; however it does not have the resources to pursue large-scale archaeological research.
11Over the planned six year duration of the project, the research team will collaboratively engage in three main categories of research. First, is the collection of baseline data on permafrost, surficial geology, coastal erosion, recent Inuvialuit land use patterns, wildlife distributions, known archaeological site distributions, and other factors. These are being drawn together to create a Geographic Information Systems (Gis) model for understanding threats to archaeology across the study area. Second, is extensive helicopter survey of selected coastal zones. These surveys are designed in part to locate unrecorded sites. However, since the 1950s (MacNeish 1956) the region has been surveyed many times by a series of archaeologists; thus, the primary goal of the current surveys is to assess the condition of known sites, in order to help guide decisions about which ones should be excavated, with particular attention paid to evidence for erosion, instability, and thawing of permafrost soils. Third, is excavation of selected site(s). Using newly acquired information on the degree of threat to all revisited sites in the region, combined with assessments of the importance of each site based on multiple criteria, portions of sites will be excavated with a goal of preserving critical information for the long term. Excavation will incorporate standard archaeological practice for recovery of artifacts, ecofacts, architectural information, and spatial data; but will also involve several more specialized elements. Fine-grained studies of permafrost will be used to understand the impacts of permafrost thaw on anthropogenic sediments, through comparison between on- and off-site locations. In addition, 3D laser scanning will be used to capture high resolution digital images of selected archaeological features as a way to produce a permanent, archived record for research and presentation to the public.
12Arctic Char is in its early stages, with a second field season just completed. Detailed results are not yet available, but a brief outline of progress follows.
Map of the study area, indicating overall survey area as well as four more intensive survey zones. Kuukpak is in Zone 1; McKinley Bay is in Zone 4
(Data Sources : Geodata.gov ; Geobase.ca- © Department of Natural Resources Canada. All rights reserced. Projection : NAD83 Northwest Territories Lambert. Map drafted by Michael O’Rourke)
13In July of 2013, a three-person survey team spent six days visiting the most important archaeological sites in the region by helicopter. In 2014, since fieldwork was concentrated primarily on excavation, only one day of survey was performed, with revisits to a few specific threatened sites. Given the project’s mandate, the 2013 survey was mainly intended to assess the condition of the major sites in the region, subject to constraints of time and/or amount of helicopter fuel available. For all sites, a first step involved making sure that the site matched the existing information in the official site record forms, since in some cases location information is vague or contradictory. Next, sites were assessed for evidence of erosion. This involved observation of any actively eroding banks or other features in the immediate vicinity of the site. Where possible, sites’ current statuses were compared to maps, photos, or written descriptions produced by archaeologists in past decades, potentially allowing quantification of rates of erosion. At two sites with actively eroding coastal bluffs, wooden stakes were inserted into the ground at measured distances perpendicular to the bluff face, so that accurate erosion rates could be measured in future years.
14Following assessment of site condition, a brief survey of each site was conducted, to produce an initial description of, for example, numbers of house pits or status of surface artifacts and architectural timbers. In several cases, radiocarbon samples were collected, and in one case artifacts that had eroded from the site were surface-collected from the beach. Between revisits to known sites, the entire coastline was flown slowly, at low altitude. A total of 19 sites were assessed during the 2013 survey. Some were judged to be stable, and in at least one case adjacent shorelines appear to be prograding (building up) rather than being destroyed. However, most are actively eroding, and in several cases this is occurring at an alarming rate.
15One example of the latter category is the site of McKinley Bay, located at the northeastern edge of the survey area near the tip of the Tuktoyaktuk Peninsula (fig. 2). It was previously researched by Charles Arnold in 1991 and Matthew Betts (2007) in 2004. The site contains a minimum of 11 semi-subterranean houses, plus several other features. It is critically important to an understanding of precontact lifeways, because it is the last remaining substantial site in the traditional Inuvialuit bowhead whale-hunting area of the Tuktoyaktuk Peninsula. The only other similar site in the region, Nuvuqaq (Atkinson Point), has been completely destroyed by erosion. During the 2013 site visit, we re-located all of the houses on Betts’ 2004 site map. Two houses that were intact in 2004 are now threatened with erosion. One of these, House 6, was now largely destroyed, with floor and wall logs seen in the eroded bluff face. A second, House 5, was observed to be directly at the bluff edge. Based on Comparison to Betts’ map, the rate of erosion has been approximately 8 metres in 9 years. This is equivalent to slightly less than a meter a year, though the process may be accelerating, given that the bluff edge appears to be breaking into larger slump blocks than was the case in 2004.
The McKinley Bay site. House mounds can be seen at centre and right; eroding bluff is at lower left, with locations of Houses 5 and 6 indicated
(© Max Friesen)
16In 2013, two lines of stakes were placed into the McKinley Bay site surface perpendicular to the eroding face, at measured distances, to allow accurate calculation of erosion rates in future years. During a re-visit in 2014, measurements confirmed that at both lines of stakes, the erosional face had retreated by an average of approximately 1 metre in one year. As a result, House 5, which in 2013 was intact but near the edge, was now eroding; architectural wood from the walls and floor could be seen eroding down the sandy bluff face. House 6, which was partly gone in 2013, appeared to have experienced another 2 metres of further horizontal damage (fig. 3).
Lawrence Rogers examines house timbers eroding from McKinley Bay House 6
(© Max Friesen)
172014 was the first year of large-scale excavations, with efforts concentrated at the site of Kuukpak. Selection of this site was straightforward, since Kuukpak ranks very high on both dimensions used to prioritize excavation: significance to Inuvialuit heritage and degree of threat. Until the late 19th century, Kuukpak was the central, principal village of the Kuukpangmiut, one of the largest precontact Inuvialuit regional groups. Its location was ideal for beluga hunting, with extensive shallows directly adjacent to the village. The main occupation area with semi-subterranean houses stretches for almost a kilometer along the shore of the Mackenzie River East Channel. Although it is difficult to estimate the number of semi-subterranean houses that once existed here, it was almost certainly the largest Inuvialuit site ever occupied. Previous excavations by Arnold (1988, 1994), and current observations of eroding banks, indicate the site is extremely rich in terms of architecture, artifacts, and animal bones. Thus, in terms of significance, the site is unmatched. In the second dimension of assessment, degree of threat, large parts of the site are visibly eroding, with architectural wood originating as house floors, walls, and entrance tunnels seen in collapsing banks and sliding down slopes. Below these areas, long sections of beach are covered in a dense carpet of bones (mainly beluga) and artifacts, further attesting to the scale of site destruction (fig. 4).
18Kuukpak is a large, complex site with a variety of archaeological contexts including houses, middens, and storage pits, however our top priority was to excavate a complete three-alcove igluryuaq (sod house). This house form (sometimes referred to as a “cruciform house”) is unique to the Mackenzie Delta region, and consists of a rectangular central floor area with raised benches / alcoves on three sides, and a sunken entrance tunnel extending from the fourth wall. It appears prominently in Inuvialuit histories and the ethnohistoric record, and one has even been reconstructed in Tuktoyaktuk, indicating its central importance to Inuvialuit heritage. Complete or partial archaeological examples have been excavated previously at several sites, including Kuukpak (Arnold 1988), Kitigaaryuit (McGhee 1974), Tuktoyaktuk (MacNeish 1956), and McKinley Bay (Betts 2007), confirming their existence across much of the region and yielding many important details of their architecture. However, none has been fully excavated in recent times in a manner that allows their complete architecture to be mapped and recorded in fine detail.
The Kuukpak beach at low tide, showing large numbers of beluga whale bones
(© Max Friesen)
19In 2014, therefore, we selected a large, deeply buried house near the middle of the site (fig. 5). It was not under immediate threat, but is the closest complete house to the actively eroding area, and it is unclear how soon the spur of land it is on will begin to erode. When the bottom level of the house was finally exposed, a remarkably complete and well-preserved igluryuaq was revealed. The main floor area is roughly 3 m wide and 2,5 m long. Posts at each corner represent the main roof supports universally used in Mackenzie Delta region house construction. The rear half of the main floor is surfaced with small diameter driftwood logs running almost its full width. The front half of the main floor is divided into two functionally different areas. On the west side is a well-defined hearth. This hearth was rebuilt at least once; an upper hearth was framed with driftwood logs burnt on their inner faces, and a lower hearth was indicated by an oval area of white clay. The east side of the floor is the location of the katak (entry to the tunnel). The katak was not fully defined in 2014, however the floor dips low here and is clearly aligned with the entrance tunnel depression south of the house. Just to the north of the katak is a very large flat rock, which may have been situated to add a stable platform for people entering and exiting the house. Surrounding the floor are the three alcoves, each of which is trapezoidal in shape, with greater width at the front where the alcove meets the main floor. The alcoves were raised approximately 20-30 cm above the floor level and were for the most part surfaced with whole logs; when in use, they would have been covered with insulating vegetation and skins. The entrance tunnel was not fully excavated, however it extended a minimum of 5 m south of the front wall of the house. When fully exposed, the igluryuaq was mapped, photographed, recorded with photogrammetry, and laser scanned to obtain a high-resolution 3D image.
The Kuukpak igluryuaq fully excavated. The rear alcove is at lower left
(© Max Friesen)
20Arctic Char has begun to produce significant results in terms of both a regional understanding of threats to the archaeological record, and site-specific data relating to house architecture, artifact typology and function, and economic organization as indicated in the very large archaeofaunal assemblage from Kuukpak. It is continuing to function as a collaborative enterprise, with meaningful interactions between Inuvialuit organizations and individuals and “southern” university-based researchers and students. At the same time, the research to date has made very clear the daunting scale of the problem as a whole. Surveys have indicated that many sites are already largely or completely destroyed, and it is clear that whatever excavations occur will only scratch the surface of the region’s rich history.
21Future plans include continued survey in 2015, with the addition of a new region to the east of the Tuktoyaktuk Peninsula. This new region was added as a result of discussions with the Tuktoyaktuk Hunters and Trappers Committee, whose members are concerned about the status of several important sites there. In 2016, further excavation work is planned at the Kuukpak site. Despite 2014’s intensive field season, the Kuukpak igluryuaq was not fully excavated – the tunnel has not yet been defined, and the nature of the architecture under the floor is also unclear. Other site areas under more direct threat will also be excavated. Fieldwork is expected to continue until at least 2018, with each additional field season salvaging another small portion of the region’s irreplaceable history.