NORTH CREEK SHELTER: ESCALANTE VALLEY, SOUTHERN UTAH

North Creek Shelter is a deep, multi-component site on the northern Colorado Plateau in southern Utah containing evidence of human occupation from 10,000 years ago to the present. Use was intensive during the Paleoarchaic, Early Archaic, and Formative (Fremont and Anasazi) periods. The current basal date of 9960 ± 30 rcybp (see date Table) is the earliest evidence of human presence on the Colorado Plateau and only Danger Cave and Smith Creek Cave are earlier in Utah. In addition, fine-grained basal sediments containing artifacts and faunal remains accumulated rapidly in the lower levels and hold great potential for documenting human – climate relationships during the Late Pleistocene/Early Holocene transition. Research crews directed by Joel Janetski of the BYU Department of Anthropology are working here to increase understanding of the deep human history of the region.

Overview photo of Northcreek Shelter looking northeast.

Site Setting and Description

North Creek Shelter (elevation 6219 ft) lies on the west edge of Escalante Valley, a small basin bordered on the north by the Aquarius Plateau, on the west by the Table Cliff Plateau, and on the south by the Straight Cliffs and the Kaiparowits Plateau. To the east and southeast are the Escalante River canyons. Three streams merge within about a quarter mile of the site—Upper Valley Creek, Birch Creek, and North Creek—to form the Escalante River. Vegetation is Upper Sonoran with scattered juniper and pinyon trees, various grasses, and low bushes on and off site. The site sprawls along the base of a sheer Dakota sandstone cliff that forms the southern edge of a small mesa. Several Fremont-age granaries are tucked into the cliff face overhead and rock art, both painted and pecked, is abundant and includes historic inscriptions (Native American and Euroamerican) as well as Fremont and Archaic elements. Smaller rock art panels and scatters of cultural debris continue several hundred meters a ong the cliff to the east and north. Historic occupation began in the late 1800s with Mormon pioneer arrival in the valley and the construction of a grist mill using water from North Creek. In sum, the site is massive and clearly multi-component, all of which testifies to the attraction of the locale due to the availability of reliable water and arable land.

Field Work, 2004-2008

Between 2004 and 2007 BYU archaeologists expanded the horizontal exposure at the site to 26 m2 and extended excavations to 4.4 m below ground surface. This work revealed discrete Fremont, Early Archaic, and Paleoarchaic occupations, including multiple Early Archaic and Paleoarchiac use surfaces, some containing hearths and storage and roasting pits (see Plans on this web page). As of 2007 crews had expanded the Early Archaic levels to ~12 m2, but the sample for the earlier Paleoarchaic was limited to just 3 m2. These three grids demonstrated that culturally stained levels extend to ~ 3.85 m below ground surface. Below this charcoal and artifacts diminished rapidly and a probe pushed an additional 75 cm below the deepest excavated level found no charcoal or artifacts.

The 2008 field season focused on expanding the Paleoarchaic sample from the site. Work began on July 14th and continued through August 22nd. Crews consisted of graduate and undergraduate students from BYU and graduate students from University of Nevada, Las Vegas. Funding for the 2008 research came from grants provided by the National Science Foundation and Charles Redd Center for Western Studies at BYU. These grants also provide for ancillary studies including a geological assessment of the site sediments by Tom Morris (Geology Department, BYU) and small mammals by Donald Grayson (Department of Anthropology, University of Washington), both of whom visited the site. In addition, Joan Coltrain (Department of Anthropology, University of Utah) will do oxygen isotope studies of artiodactyl teeth to assess climate change.

As noted, the goal for the 2008 season was to expand the Paleoarchaic levels (9000 to 10,000 BP) from the 3 m2 exposed in 2007 to ~12 m2. This required removing about 1.45 meters of fill from Stratum IVa which had been reached in 2007 (see Profiles in this web page). The season was productive with 10 Paleoarchaic use surfaces mapped and abundant stone tools and faunal remains recovered.

Summary of Human Occupation at North Creek Shelter

North Creek Shelter was intensively occupied during the very early Holocene (~10,000 to 8000 rcybp, Strata II, III, and IV) and provides an opportunity to gain insights into the transition from the late Pleistocene into the early Holocene. The sediments Strata II and III are considered Paleoarchaic in age and the work in 2008 dramatically increased the sample of stone tools, faunal remains, and use surfaces from this time period. Although storage pits are absent, both formal and informal hearths and reddened sediments are present on several Paleoarchaic use surfaces. In addition to the use surfaces containing abundant artifacts, and bone, we excavated three separate levels consisting of flat-lying, heavily burned sandstone slabs covering most of the exposed area. Artifacts were largely non-existent on these surfaces.

Students measure and record two Fremont hearths.

Paleoarchaic stone tools include numerous unifaces, especially steep-edged scrapers. Early and late stage bifaces are common as is debitage. Projectile points from these levels are primarily stemmed (see Artifacts/Points this web page) with a unique stemmed style restricted to the lowest levels, and a few reworked stemmed points reminiscent of Plains types recovered from the upper Paleoarchaic strata. Both local and non-local toolstones were used during this period with exotics including a greenish rhyolite and obsidian, primarily from the Mineral Mountains in central Utah. Several small, notched obsidian flakes came from these lower levels as well (see Artifacts/misc). The function of these tools is unclear at the moment. Flakes are extremely abundant on some use surfaces, although little more can be said as analysis of the tools and debitage from this year is just beginning. Ground stone is almost totally absent in the Paleoarchaic levels. Ornaments were rare , but an incised bone tube bead (see Artifacts misc) came from the F259 use surface (see Plans) adjacent to a large hearth deep in the Paleoarchaic occupation. Faunal data are still being analyzed but small artiodactyls (deer mostly, but some mountain sheep) remains very common.

Site use in Stratum IV (~9000 to 8000 rcybp) differs from the underlying levels in two ways: ground stone is now common, and projectile points styles shift from stemmed to shallow side-notched with a basal concavity similar to Pinto points (see Artifacts/Points in this web page). Several features (pits and hearths) are present on two of the four use surfaces and discontinuous ash smears with associated reddened sediments between the more discrete surfaces document intermittent use. Formal tools include both early and late stage bifaces, drills, and unifaces. As noted, ground stone (thin metates, one-handed manos) is abundant in this stratum. Debitage is common and flake size tends to be small. As with the lower levels, faunal remains are dominated by small artiodactyls, especially deer, although mountain sheep, hares, rabbits, marmots, and birds are present.

Stratum V is primarily Archaic in age, although sediments are heavily bioturbated by both humans and small mammals. Separating burrowed from non-burrowed sediments was a near impossible task. As a consequence, findings are mixed with both atlatl and arrow points and an occasional ceramic sherd recovered from this level. A maize date of 940±40 rcybp came from the upper levels of Stratum V and demonstrates the disturbed nature of this level. Sediments closer to the cliff face, however, appear to have retained their integrity, and dates from these sediments demonstrate an Archaic-age presence rather high in the deposits but which have apparently been disrupted by human and animal activity.

Stratum VI marks Fremont occupations. Formal tools are dominated by arrow points and early and late stage bifaces, ground stone, and grayware ceramics. Carbonized maize cob fragments and chunks of stick or beam-impressed adobe occur throughout this level. Diagnostic features include two clay-rimmed hearths lying at different depths, suggesting multiple occupations. Several features (including a slab-lined hearth dating to ~AD 1000 and a clay-rimmed hearth) lay on the same level suggesting a period of heavy Fremont use (see photo). Portions of constructed features in this level suggest remodeling by Fremont occupants. This kind of site manipulation is reinforced by radiocarbon dates in the 900 to 1100 rcybp range at depths just above a date of 6020±60 rcybp from Stratum V. Although a cultural hiatus of several thousand years is common at sheltered sites on the northern Colorado Plateau, the gap in human use at North Creek Shelter seems excessive and may be confounded by hum an disturbance of site sediments in prehistory. The presence of Anasazi ceramics in Stratum VI, including utilitarian wares, adds to the complexity of human site use during the Formative period.

The uppermost levels of Stratum VI have yield occasional evidences of post-Fremont use in the form of Desert Series arrow points (Desert Side-notched and Cottonwood Triangular) as well as brownware pottery. These tend to be mixed with Fremont age items. No features are yet dated to this period.

Paleoenvironmental Insights

UNLV PhD student David Yoder documents a Fremont granary.

Sediment exposed in the North Creek excavations and preliminary analyses of botanical remains are useful in making some tentative statements about paleoclimates. The deep vertical exposures and site dates have revealed that the rate of deposition was very rapid during the early Holocene and dropped off dramatically after about 8000 years ago. For example, it took about 2000 years to lay down the sediments making up Strata II-IV (~ 2 m) and about 8000 years to deposit the sediments in Stratum V to the ground surface (a little over a meter). The sands and silts that make up the site deposits are most likely the result of colluvial action that brought sediments down vertical cracks in the cliff face to the east and west of our excavation area. Those sediments then flowed across the flat area at the base of the cliff sealing in occupations. The amount of colluvial action is related to precipitation; therefore, the rapid sediment build up in the deeper levels at the site suggest increased precipitation prior to ~8000 years ago and decreased precipitation during the middle and later Holocene.

The sediment data are complemented by charcoal identified for dating purposes. Plant taxa represented by charcoal from the pre-8000 BP levels include Douglas Fir (Pseudotsuga menziesii), hackberry (Celtis sp), mountain mahogany (Cercocarpus sp), cliffrose (Cowania mexicana), probable aspen (Populus sp.), juniper (Juniperus sp), and pine (Pinus sp.) (Puseman 2007). Douglas fir, aspen, and hackberry do not grow on the site today but are found at higher elevations to the north and west. This suggests that, as in southern Utah generally, vegetative belts were depressed here in the very early Holocene.

Dating

Seventeen radiocarbon dates provide a chronological framework for site occupation (see Table of dates). The table presents dates arrayed from the lowest to the upper levels. The integrity of site sediments is supported given that, with one exception, these dates ladder up from the bottom to the top. The exception is an aberrant date of 890±40 or about AD 1100 from a single piece of Atriplex (saltbush) from the lower levels. This small charcoal fragment was most likely dislodged from upper levels during the initial site testing. The dramatic jump from 6020 BP to 1050 BP between F18 and F134 is evidence of the mid-Holocene hiatus that is common at stratified sites in southern Utah. Additional dates from all levels will be processed using funds supplied by the grants.

Table of Northcreek Shelter Radiocarbon dates.

Sample No Material, provenience Depth cmbgs Conventional Radiocarbon age 2 sigma calibrated age 2 sigma cal BP age
Beta 197358 Corn, F14/F18 ~65 940±40 BP AD 1010-1190 760-940 BP
Beta 221411 Corn, F134 ~65 1050±40 BP AD 900-1030 920-1050 BP
Beta 221414 Pooled charcoal, F18 stratum fill ~75 6020±60 BP 5050-4760 BC 7000-6710 BP
Beta 221412 Pooled charcoal, F120 stratum fill ~120 7670±80 BP 6650-6400 BC 8600-8350 BP
Beta 239024 AMS charcoal F229 101 7700±50 BP 6640-6450 BC 8590-8400 BP
PRI-07-102-4364 AMS, juniper twig char, F229 101 7990±30 BP 9000-8720 BP
Beta 207167 Pooled Juniper charcoal, F62 ~ 130 7970±80 BP 7080-6640 BC 9030-8590 BP
Beta 210253 Pooled charcoal F59 ~155 8320±120 BP 7580-7060 BC 9530-9010 BP
Beta 197359 Pooled charcoal, F125 ~160 8310±70 BP 7540-7140 BC 9490-9100 BP
Beta 239023 AMS charcoal F192 187 8310±40 BP 7220-7190 BC 9170-9140 BP
PRI-07-102-4029 AMS, juniper charcoal, F192 187 8860±25 BP 10,160-9860 BP
Beta 194030 Pooled charcoal, F70 ~200 9020±70 BP 8300-8170 BC 10,250-10,120 BP
Beta 195226 Atriplex, F74 ~210 890±40 BP AD 1030-1240 710-920 BP
Beta 207168 Pooled Pinus, F83 251-254 9510±80 BP 9190-8610 BC 11,140-10,560 BP
Beta 221415 AMS charcoal, F156 315 9690±60 BP 9250-9110 BC 11,200-11,060 BP
Beta 239022 AMS charcoal, F190 3.49 9800±50 BP 9310-9220 BC 11,260-11,170 BP
PRI-07-102-3716 AMS Salicacaea charcoal, F190 3.49 9960±30 BP 11,420-11,260 BP
Land owner Jeff Rex.

SUMMARY OF SITE SIGNIFICANCE

Sites with evidence of human occupation earlier than 9000 years ago are rare on the Colorado Plateau and in the Great Basin. The research areas of primary significance at North Creek Shelter are 1) Late Paleoarchaic/Early Holocene paleoclimatic reconstruction, 2) timing of human arrival on the northern Colorado Plateau, 3) insights into late Paleoarchaic/Paleoindian, and 4) Early Archaic lifeways. Fremont, Anasazi and probably historic Native Americans used the site as well. Excavations to date have disturbed about 10 percent of the site suggesting much remains for future work.

ACKNOWLEDGEMENTS

The project has benefitted greatly from the support of the Rex family who own the property and who operate the beautiful Slot Canyons Inn bed and breakfast near the site. They have allowed crews to camp on site and use water and electricity free of charge. In addition, the Grand Staircase-Escalante Monument has graciously provided in-kind support throughout our research here and elsewhere in the Escalante Canyons region. We also are grateful to members of the Utah Valley Chapter of the Utah Statewide Archaeological Society and colleagues who have provided numerous volunteer hours on the project. I especially mention Ren Thomas (Utah Valley USAS) who has been a tireless supporter and volunteer each year at the site. As noted earlier, hard funding for the 2008 work came from the Charles Redd Center for Western Studies. In addition, this material is based upon work supported by the National Science Foundation under Grant No. BCS-0818971. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).