Strontium isotope (87Sr/86Sr) data from archaeological sites in Utah, USA

The information presented here includes the results of strontium isotope analysis on 75 baseline samples from nine Fremont sites in Utah. The baseline samples are of lagomorphs and rodents with limited foraging ranges. The baseline ranges for each site were calculated with two standard deviations. Also included are the raw strontium isotopic data for 30 large game samples from Wolf Village, a Fremont site in Utah. Additional data include a map showing the location of the sites in this study, box plots portraying the local ranges of nine Fremont sites in Utah, and an individual value plot comparing the Wolf Village large game samples to the strontium baseline for the site. These data compliment the discussions and interpretations found in “Identifying Strontium Baselines and Large Game Animal Trade at Fremont Sites through Strontium Isotope (87Sr/86Sr) Analysis” [1].


Data
The data includes strontium isotope ratios ( 87 Sr/ 86 Sr) from 75 small mammal teeth from the archaeological record of nine Fremont sites in Utah (Table 1; see Fig. 1 for regional map and local baseline ranges for each of the sites). All small mammal teeth were analyzed to identify strontium baselines in their respective areas. Other data includes strontium isotope ratios ( 87 Sr/ 86 Sr) from 30 large game samples from the Wolf Village site ( Table 2). The baseline and large game sample data are included in a supplemental CSV file. Box plots of local ranges for the nine Fremont sites indicates that local strontium ranges overlap at some areas far apart from one another and vary between some sites in close proximity (Fig. 2). The Wolf Village baseline was previously identified using muskrat incisors [1,2]. Recent strontium data using squirrel incisors from the same site indicate that small land mammals provide more precise baselines than semi-aquatic mammals such as muskrats (Figs. 3e5).

Sampling strategy
The purpose of the baseline samples was to identify strontium local ranges for three main areas in Utah, including: (1) the regions around Utah Lake, (2) central Utah, and (3) the Parowan Valley in southeastern Utah. All areas had large Fremont settlements. Sites around Utah Lake include Hinckley Mounds (42UT111), Woodard Mound (42UT102), and Wolf Village (42UT273). In addition, Nephi Mounds (42JB02) is somewhat close to the Utah Lake sites and could be accessed by the Fremont through Goshen Canyon. Sites in central Utah include Nawthis Village (42SV633), Five Finger Ridge (42SV1686), and Icicle Bench (42SV1372). Sites in the Parowan Valley in southeastern Utah include Parowan (42IN100) and Paragonah (42IN43). The multiple sites in each of the three main areas can be compared to identify variability in local strontium levels.
To avoid contamination from modern fertilizers and air pollutions, rodents and lagomorphs from the archaeological record are useful to identify local 87  Value of the Data Data is useful to other scholars who are interested in strontium isotope studies in Utah This data will allow data from future research to be compared to the already identified strontium baselines in Utah The data shows potential complications with testing teeth from semi-aquatic mammals to identify regional baselines samples are from local small game. Lagomorph specimens were more common at dryland sites in central and southwestern Utah (i.e., Nawthis Village, Parowan, and Paragonah) and muskrat samples more common at wetland sites in northern Utah (i.e., Hinckley Mounds, Woodard Mound, Wolf Village, and Nephi Mounds) ( Table 1). The initial goal for this project was to test ten samples from seven sites, but limitations on suitable specimens from Five Finger Ridge and the Parowan site required that additional specimens from nearby sites also be tested (i.e., Icicle Bench and Paragonah respectively). This was done with the assumption that sites near one another (<5 km apart) would share similar strontium baseline ranges. Thus, the local baselines for the Five Finger Ridge, Icicle Bench, Parowan, and Paragonah sites were all identified with less than ten samples.

Sample description
Seventy-five small game samples were analyzed to identify strontium baselines at nine sites. All baseline samples were incisors taken from intact mandibles from the archaeological record of each respective site in this study. To ensure all samples were from different individuals, all teeth from the same species were removed from mandibles of the same side. In addition, 30 large game samples from the Wolf Village site were also analyzed. These include 13 deer (Odocoileus hemionus) specimens, four pronghorn (Antilocapra americana) specimens, eight bighorn sheep (Ovis canadensis) specimens, and five worked bone gaming piece specimens constructed from long bones of unidentified large game (Artiodactyla) ( Table 2). All tested large game specimens are from Fremont contexts. Like with the baseline samples, all deer samples were extracted from intact mandibles of the same side. Due to limited pronghorn and bighorn sheep specimens, a tooth sample from all suitable mandibles were analyzed so it is possible some are from the same individual.

Pre-treatment and strontium isotope analysis
All small mammal and large game samples were analyzed at the Strontium Isotope Geochemistry Laboratory at the University of Utah. Pretreatment was done at the Biogeochemistry Laboratory at the University of Utah. Small mammal samples are too small to manually remove dentine from enamel, so whole tooth specimens were analyzed. For large game samples, teeth were extracted from the mandible using a Dremel Lithium-Ion cordless drill (10.8 V, Model 800). Samples were examined under a Bausch & Lomb StereoZoom 5 (zoom range 0.8Âe4.0Â) microscope to ensure the dentine and discoloration were removed from the samples, leaving as much tooth enamel as possible (at least 0.05 g). The Dremel Lithium-Ion drill was also used to remove at least 0.05 g from each of the worked bone gaming pieces.
Samples were pretreated with 5% acetic acid (CH 3 COOH) and then rinsed three times in quadrupole de-ionized water (4 Â H 2 O). These methods are effective at removing contaminants from samples

Data analysis
Local ranges for nine Fremont sites were calculated with two standard deviations (Table 3). A box plot visual displays the homogeneity of strontium baselines at sites in northern Utah (Fig. 2a). While Nephi Mounds and Hinckley Mounds both vary, strontium values overlap between Wolf Village and the other three sites in northern Utah. Likewise, strontium values from sites in central Utah overlap with one another, specifically Nawthis Village and Fiver Finger Ridge. In contrast, sites close to one another such as Five Finger Ridge and Icicle Bench are highly variable (Fig. 2b). Similarly, both sites in the Parowan Valley have highly variable strontium baseline values despite their close proximity to one another (Fig. 2c).
Muskrat (Ondatra zibethicus) specimens may not reflect accurate strontium baselines since muskrats are semiaquatic animals. Muskrats may be influenced by nonlocal strontium coming from various geological formations through rivers and streams. Therefore, their bones and teeth may no longer represent local strontium baselines. Unfortunately, a previous baseline for Wolf Village was based on strontium levels in semiaquatic muskrat [2]. To test whether muskrat provide inaccurate or less precise strontium baselines than terrestrial mammals, five squirrel (Spermophilus sp.) incisors from Wolf Village were analyzed. The Wolf Village baseline using squirrels is more precise than the muskrat baseline and contains no outlier specimens outside the local strontium range (Fig. 3). Comparing a random sample of five muskrat specimens (Sample ID No. 1494-60, -62, -63, -64, -65) to the five squirrel specimens further indicates that the squirrel baseline is more precise (Fig. 4).    Comparisons between the muskrat and squirrel baselines to the Wolf Village large game specimens provide differing results (Fig. 5). Almost all large game samples fall within the muskrat baseline, indicating that most large game were local to the Wolf Village site. In contrast, the squirrel baseline suggests that approximately half the large game samples at Wolf Village were local to the area. In addition, most muskrat specimens fall outside the squirrel baseline suggesting that the muskrat samples are either of non-local individuals or no longer reflect local strontium signatures.

Funding sources
Funding for this project was generously provided by the Charles Redd Center for Western Studies, the Grace Elizabeth Shallit Memorial Grant, the Warren Van Pelt Student Grant, and the Department of Anthropology at Brigham Young University. The Charles Redd Center for Western Studies funded the publication of this research.