(]:{]£00~&"' ~®&®©@ ' Susitna Joint Venture 0 Document Number c jq_j_~ Please Return To [ DOCUMENT CONTROL n EARLY MAN IN AMERICA [ From a Circum- Pacific Perspective [ [ 0 ~ t~ [ Edited by Alan Lyle Bryan G D c c [j Qccasional Papers No 1 of the Department of Anthropology, E University of Alberta Publisher: ·~. c Archaeological Researches International u Edmonton, Alberta, Canada C c DRY CREEK: A LATE PLEISTOCENE Wm. Roger Powers Thomas D. Hamilton HUMAN OCCUPATION IN CENTRAL ALASKA Anthropology Program Geology Department University of Alaska University of Alaska [ Fairbanks, Alaska Fairbanks, Alaska [ c INTRODUCTION (Fig. 2). The remaining units appear to be culturally sterile. [ The Dry Creek site. discovered by C. E. Holmes in Nine radiocarbon dates on charcoal have been obtained 1973, is located in central Alaska about 180 km south­ from the Loess 1-5 sequence (Fig: 2). These show severe west of Fairbanks (Fig. 1 ). It occupies a late Pleistocene inconsistencies which are puzzling because careful strati­ [ outwash terrace at the forest-tundra ecotone within the graphic mapping shows smearing within each paleosol Nenana Valley, close to the north flank of the Alaska complex near the bluff face, but no mixing of any com­ Range. The following discussion is based on preliminary plex with any other. The two oldest samples (SI-1544 [ field work carried out in 1973 and 1974. and Sl-1938) which would appear to date around 19,000 The archaeological components of,the Dry Creek site and 24,000 years BP, probably should be rejected are stratified within a two m section of eolian sediments because of their very small size and resulting high and paleosols which overlie glacial outwash deposits counting errors. Both samples had to be heavily diluted [ (Fig. 2). Test excavations here have isolated three during radiocarbon analysis, and are considered suspect cultural components of late Pleistocene age and one of for this reason (R. Stuckenrath, written communication). Sample Sl-1936, yielding a date of 12,080:!:. 1025 BP, ~ Holocene age. Vertebrate fossils have been discovered in L the two lowest components, and datable charcoal is also has a very high counting error, and may also be present in most of the paleosols. The absence of severe questioned. The remaining dates, although still somewhat postdepositional disturbance of the sediments has inconsistent, suggest that the highly deformed Paleosol 3 [ helped make Dry Creek an optimal locality for the study is broadly of very late Pleistocene to early Holocene age. of human adaptation to late Quaternary environments in, This interpretation is consistent with an inferred episode the North. Thus, among known Alaskan sites of this of intense periglacial activity associated with the final antiquity, Dry Creek is unique. readvance of the Riley Creek glacier about 10,000 years [ ago (Thorson and Hamilton 1977). The remaining date SITE GEOLOGY on Paleosol 2 (9340 ±. 195 - Sl-2329) is consistent with the date of 10,690:!:. 250 (SI-1561) on Palesol 1 and The outwash at the base of the section was deposited the date of 11,120 ±. 85 (SI-2880) for the first archaeo­ c logical component. The two oldest paleosols evidently close to the front of the Healy glacier when it stood at its maximum position (Fig. 1 ). This event probably formed during a period of relatively mild conditions occurred during early Wisconsin (Zyrianka) time which date between approximately 13,000 and 10,500 c (more than 50,000 BP). A hiatus of unknown duration years BP elsewhere in Alaska (Hamilton 1974a). separated outwash deposition from subsequent loess Although 10,690 ±. 250 BP had previously been taken accretion. as the "official" age of the main occupation of Dry Creek c The lower loess units (loesses 1-5, Fig. 2) and inter· (Hamilton 1974b). it now is clear that this date merely vening paleosols (Paleosols 1-3) appear to represent late suggests an approximate minimum age for the two glacial tundra'or tundra-steppe conditions. Glacial oldest occupations. Early human occupation at the site meltwater streams probably still extended down the evidently began about 11,000 years ago and continued c i'ntermittently for more than a millennium. Nenafla Valley, maintaining broad barren floodplains which served as source areas for loess. Forests probably The upper sequence of loess beds, windblown sand c were absent from the entire region, and the character of sheets, and paleosols was deposited under environmental L the fossil tundra soils, rodent burrows and frost structures conditions markedly different from those of the older at the Dry Creek site indicate that this area at least was soils and sediments. Buried subarctic brown soils treeless. Loesses 2, 3 and 4 contain stone chips, artifacts containing large charcoal fragments indicate forest c and other evidence for prehistoric human occupation growth on the sandy sediments which were localized 6 72 [ c [ [ 0 5 c ea E' F3 [ [ [ . -~ ~ -~ -~~==-=~~ . D n 63°45 I . - I ·- - ~ 6 [ c D Northern limits of glacial advances n u: :. :: 'lr' ..,.. Carlo Reodvance c A W W -w- Riley Creek :rr "" • • • 1 ·-- Riley Creek I c , , 1 1 ~~ Healy 0 Altitudes above 1000 m asl 149° c (3280 ft) F b c' D 73 [. c [ Generalized Archeolooic 14 [ em Unit c dotes from paleosols sect1on component-; 0 -.:.; ..-t l SAND 4 ;oc. [ [ so -!"'"""I L«~,,~ .3430.!75, 3655!60 IV [ • 4670 ±95 eH6270:ttto 8600±460 D 10,600!580 l6900195 r19,050~·500 Lr tOO~ ~ Ill [ 9340±195; 12,080:tl025 23.930±9300 b II [ 150 SAND I ...... .11,120±85 c LOESS 2 - E LOESS I c 200 Figure 2. Ge.n.eralized stratigraphic section of the Dry Creek site, showing radiocarbon dates and r~ I:J location of archaeological components. [;j L D 74 [ [ along the bluff edge. Four concordant radiocarbon rear of the trench and centered around an accumulation dates (Fig. 21 indicate that forest became established at of cobbles and small boulders with charcoal scattered the site during middle Holocene time and that inter­ diffusely throughout. This feature is clearly a hearth. c mittent human occupation occurred between about 4700 Scattered around this hearth are excellent examples of and 3600 radiocarbon years BP. side scrapers, a small obsidian biface with extremefy worn edges plus other artifacts and an incredibly high frequency [ ARCHAEOLOGICAL COMPONENTS of flaking debris. It is in this part of the site that Component II is represented by a thin occupation clearly Four stratigraphically distinct cultural horizons or. delineated as a single surface. · n compo~nents have been distinguished at the site. The Component II also has added significance in that there total number of remains for these components is 2,827 is a great deal of vertebrate material scattered around the of which 99 percent are flakes, blades, cores and tools of hellrth areas and away from it. One specimen from the [ diverse lithologies, and 1 percent water-rounded cobbles very rear of the trench is a mandibular portion of Bison and small boulders recovered from positions within the sp. 1 Other teeth were recovered from the edge of the loess and presumed to have cultural significance. We bluff in direct association with a large biface and a chopper. will concern ourselves only with the three oldest compo­ This specimen has been identified as belonging to a [ nents in the following discussion. horse (Equus sp.). Component 1: N = 310 Component Ill: N = 578 Depth: 170-190 em below datum. Stratigraphic Depth: 100-120 em below datum. Stratigraphic [ position: loess 2 (Fig. 2). Content: 282 flakes, Position: loess 4 (Fig. 2). Content: 573 flakes, one large blade-like flake, one flake core, three one blade-like flake, three blades, one biface. retouched flakes, one small triangular biface in This component is concentrated in two distinct areas: three pieces, two possible bur ins, one chopper and one, just back from the bluff edge in the front area of nL.; six scrapers, associated with ten stones and three the trench where scattered flakes and a biface fragment bone fragments. were located; and a second, at the edge of the bluff where r The remains of this component were found primarily there was an extremely local concentration of flakes and b away from the bluff edge, although there were two large three blades. flake clusters and scattered flakes near the cliff edge. The remains of cultural activity at the Dry Creek site Unidentifiable faunal remains were discovered here in cluster in two separate areas in all three components: the [ direct association with a retouched flake and several bluff edge and several meters back from the front of the other unworked flakes. bluff. The distribution of the cultural remains hints Component II: N = 1,791 strongly at the possibility of at least two separate activity [ Depth: 130-160 em below datum. 95 percent of areas: a workshop at the front of the bluff and a habita­ the remains between 145-155 em. Stratigraphic tion area further back from the bluff. This situation is Position: soil stringers in loess 3 (Paleosol 1, most striking in Component II. Fig. 2). Content: 1,716 flakes, six retouched c flakes, three blade-like flakes, five blades, 11 micro­ DISCUSSION blades, one end scraper, six side scrapers (three classic Siberian skreblol, two choppers, four Knowledge of the late Pleistocene prehistory of c elongate bifaces , two biface fragments, one asym­ Alaska is practically nonexistent. Some·authors have metric lozenge-shaped edge ground point, two postulated the presence of hunting Qroups dating back to wedge-shaped microblade cores, 19 large stones and about 13,000 to 15,000 BP. The presence of fractured pebbles, two an vi I stones, one hammer stone, and bison calcanei and a horse scapula at the Trail Creek caves L ten bone specimens.