Evidence for a winter-snowpack derived water source for the Fremont maize farmers of Range Creek Canyon,Utah,USA

Abstract

We present results of multiproxy analysis of a sediment core collected from Billy Slope Meadow, a spring-fed wet meadow in Range Creek Canyon, Utah. Range Creek Canyon was the home to Fremont maize farmers between roughly 1200 and 800 cal BP (AD 750–1150). Stable carbon isotope analysis of core sediments from Billy Slope Meadow indicate the Billy Slope Meadow site was used as a field for maize agriculture during that time. Some scholars have suggested the florescence of the Fremont culture may have been driven by increased summer precipitation, which improved the economic profitability of dry farming maize. But analysis of pollen, macroscopic charcoal and sediment geochemistry from Billy Slope Meadow, and a comparison with a local tree-ring chronology indicate the Fremont period in Range Creek Canyon was probably marked by reduced summer precipitation, and not an invigorated monsoon. The Fremont maize farmers of Range Creek Canyon therefore likely used winter snowpack-derived water from Range Creek for maize agriculture. This observation has significant implications, as using creek water rather than direct precipitation and runoff necessitates the construction of dams irrigation infrastructure, limited evidence for which has been reported by archaeologists working in the Fremont region.

Keywords

Fremont Tavaputs prehistoric agriculture Colorado Plateau Medieval Climate Anomaly Range Creek Canyon

Introduction

The Fremont archaeological complex was a part time foraging and part time maize-farming group which lived and farmed in an area approximately congruent with the modern border of the northern and central regions of Utah during the period from approximately AD 0–1300 (1950–650 calibrated radiocarbon years before present, cal BP), with a peak in radiocarbon date frequency around AD 1000 (950 cal BP; Massimino and Metcalfe, 1999; Madsen and Simms, 1998). The impetus for the rise of farming in the Southwest has been discussed for decades, and has usually been framed in one of two ways. The first explanation for the rise of farming in the Southwest involves climatic changes favoring maize agriculture. Many have hypothesized that increases in summer precipitation made maize farming a more productive economic strategy than foraging in certain areas and at different times in the Southwest, especially during the Medieval Climate Anomaly (MCA) from approximately AD 900–1300 (1050–650 cal BP; e.g. Benson, 2011; Coltrain and Leavitt, 2002; Mann et al., 2009; Matson et al., 1988; Petersen, 1994). A second explanation for the rise of maize farming involves using models from evolutionary ecology and foraging theory (often referred to as Human Behavioral Ecology, HBE; Bird and O’Connell, 2006; Broughton and Cannon, 2010; Kennett and Winterhalder, 2006). This approach has focused on the high costs of maize farming relative to wild resource foraging and suggests that maize farming would only occur in the context of wild resource depression (Barlow, 2002; Broughton et al., 2010; Cannon, 2000, 2001).

Those who adopt the first approach generally accept that a predictable, perhaps amplified monsoon season contributed to the rise of maize agriculture among Ancestral Puebloans in the Four Corners region during the MCA. Subsequent long term failure of the monsoon during prolonged droughts may have then led to widespread abandonment of previously farmed areas and regional re-organization (Benson and Berry, 2009; Benson et al., 2006, 2007; Matson et al., 1988; Petersen, 1994). Some scholars have suggested increases in summer precipitation may also have driven agricultural productivity over the area inhabited by the Fremont to the north during parts of the MCA. This hypothesis is supported by some tree ring records from the western sites across the Great Basin showing wet intervals during the MCA (e.g. Graybill, 1990; Leavitt, 1994; Stine, 1990; Stuiver, 1984). Leavitt (1994) for example analyzed stable carbon isotope and ring-width values from the White Mountains bristlecone pine chronology and found evidence for an extreme wet period between AD 1080 and AD 1129 (870–821 cal BP). Coltrain and Leavitt (2002) hypothesize that this wet interval may have been region-wide, increasing available summer moisture for maize farming activities across the Fremont area.

Faunal evidence from Homestead Cave in the northern Bonneville Basin of western Utah is also consistent with the suggestion of an overall increase in moisture during this interval (Broughton and Smith, 2016). Dramatic increases in mammal and fish bones and artiodactyl fecal pellets occur in stratum XVII, which dates to approximately 950 cal BP (AD 1000). The increase in fish remains suggests lake level for the Great Salt Lake increased sufficiently to freshen the lake to the level required for Utah chub (Gila atraria) to re-colonize the lake. The increase in small mammal bones and artiodactyl fecal pellet abundance also indicate a generally mesic period at this time (Byers and Broughton, 2004; Grayson, 2000) with expanding artiodactyl populations on the landscape. However, these data don’t necessarily indicate an increase in summer precipitation specifically. In the historic period, the largest transgressions of the Great Salt Lake (GSL) as witnessed during the AD 1983–1986 highstand resulted from increased winter precipitation. But, cold, snowy winters generally have detrimental effects on artiodactyl populations in the Great Basin (Broughton et al., 2008). The generally warmer conditions seen during the MCA may have tempered the effects of wetter than average winters, improving conditions in wintering grounds and reducing artiodactyl mortality. A region-wide increase in summer precipitation that made maize farming a more attractive economic strategy during the MCA over the Fremont area remains an untested but plausible hypothesis.

In addition to the paleoenvironmental question of whether an increase in growing season precipitation occurred during the Fremont period, it is also unclear whether such a change would actually make maize farming a more attractive economic strategy. It has long been assumed that increased growing season precipitation should make maize farming more attractive but this argument ignores potentially important energetic tradeoffs in maize farming and foraging. Barlow (1997, 2002) provided a particularly compelling treatment of the rise of agriculture among the Fremont using the logic of evolutionary ecology and foraging theory that explores this tradeoff. She placed agriculture within an optimal diet framework (MacArthur and Pianka, 1966; Winterhalder, 1981) by combining ethnographic and experimental data on energetic return rates for maize farming and foraging for wild foods found in the Fremont region. Barlow concluded that contrary to the assumption that farming is a major innovation improving energetic production, farming – especially intensive farming – can be a very low-return subsistence strategy compared to foraging for wild foods. If energetic return rates are the principal motivation, farming should only be the best economic strategy when the energetic returns from foraging for wild foods are low. For people who simultaneously or serially practice foraging and farming, an increase in summer precipitation might have the effect of making farming a less attractive production strategy by making foraging for wild foods more productive.

We address the question of whether there was a region-wide increase in summer precipitation during the Fremont period in Utah and focus on the Fremont site complex at Range Creek Canyon (RCC) in Utah’s Tavaputs Plateau region (Figure 1). We use sediment-based, paleoecological proxy data and a nearby tree-ring sequence spanning the past 2000 years to investigate climate, especially temperature, precipitation and precipitation seasonality before, during and after the Fremont occupation of the area.

Figure 1.

Map of Range Creek Canyon showing the location of Billy Slope Meadow and localities mentioned in the text.

Site description

The sediment-based paleoecological record comes from Billy Slope Meadow (BSM), a spring-fed wet meadow at the confluence of RCC and Billy Slope Canyon at an elevation of 6100’ above sea level (Figure 1). The site is one of several small perennial springs found in the canyon, and was chosen for this study for its central location in the canyon and its proximity to a number of Fremont archaeological sites. Of the over 450 prehistoric sites recorded in RCC, 23 are found within 1 km of the BSM. The nearby sites include all the prehistoric site types found in RCC: multi-dwelling villages, single dwelling residences, storage sites and rock art sites. Virtually all of the prehistoric sites with diagnostic artifacts or rock art in RCC are associated with the Fremont archaeological complex (Boomgarden et al., 2014). The presence of the full range of prehistoric sites nearby suggests Fremont agricultural activities likely occurred in and around BSM, perhaps relying on the higher soil moisture found near the spring to grow maize, and/or taking advantage of the spring as a water source for irrigation.

Dominant vegetation communities near BSM today consist of stands of Douglas fir (Pseudotsuga menziesii), two-needle pinyon pine (Pinus edulis), and Utah juniper (Juniperus osteosperma) on the canyon slopes. The gallery forest surrounding Range Creek and BSM today consists primarily of box elder (Acer negundo) and narrow-leaf cottonwood (Populus angustifolia). The well-drained sediments near BSM support big sage (Artemisia tridentata) and rubber rabbitbrush (Ericameria nauseosa), while the meadow itself is covered in grasses, various herb and wildflower species and alfalfa (Medicago sativa, introduced by cattle ranchers in the late 19th century).

Methods

Coring

We cored BSM in June 2009 using a 5 cm Livingstone piston corer to retrieve a 509 cm sediment core for analysis. We chose an area of shallow (~5 cm) standing water approximately 10m downstream from the BSM spring’s main discharge point as the core site. The core was collected in six non-overlapping drives which were extracted on site and wrapped for transport in plastic wrap and aluminum foil for transport to the RED lab. The core was then sliced in the lab into 1 cm samples which were stored at 1°C in the refrigerator at the RED lab until needed for analysis.

Chronological Control

Chronological control for this study is provided by three radiocarbon dates on aggregated pollen samples recovered from the core. Pollen was aggregated by chemical digestion following Faegri et al. (1989), with the exception that acetolysis was replaced by a 3-min bath in Schulze reagent (KClO₃ + HNO₃) to avoid contaminate of materials by acetic acid. Samples were sieved at 150 micron before treatment and not again. Dates are provided in Supplemental Material Table I, available online. An age-depth model was created with the three pollen radiocarbon dates using smooth-spline interpolation in the R age-depth modeling package BACON using its default settings with a post-bomb curve (Blaauw and Christen, 2011). We assigned the top of the core (0–1 cm) the age of −59 ± 10 cal BP, equivalent to the calendar year 2009, the year the core was collected. Calibration used the IntCal13 calibration curve (Reimer et al., 2013). The model shows a nearly linear age-depth relationship, and indicates the 5 m record spans the past 8000 years, with a mean resolution of 15 years per centimeter (Supplemental Material Figure 1, available online). The study presented here focuses on the top 150 cm of the core which spans the past 3000 years and encompasses the periods immediately before, during and after the Fremont occupation of RCC.

Stable carbon isotope analysis of sediments

We analyzed the stable carbon isotope composition of bulk sediment samples from the core for evidence of maize agriculture during the Fremont occupation of RCC. Variation in the contribution to the sediment organic matter of BSM by plants using a C4 photosynthetic pathway should be detectable in sediments as enrichment in δ¹³C values (Webb et al., 2007). We initially analyzed one sample per 8 cm over the entire length of the core. After identifying a peak in δ¹³C at 62–63 cm, we analyzed one sample per cm between 50 and 68 cm in depth.

Macroscopic charcoal analysis

Charcoal analysis was carried out to construct a local fire history. Samples of 1 or 5 cc for each 1 cm core increment were screened using 125 μm sieves and counted (volumes were adjusted down for core sections with exceedingly high counts). The >125 μm charcoal particles were counted under a dissecting microscope at 10–40× following Whitlock and Larsen (2002). Charcoal influx data were analyzed with CHAR analysis (Higuera, 2009). Background charcoal influx was calculated using a 500-year moving average. Fire events were defined in the record as events where charcoal influx exceeds the 500-year mean (background) value by two standard deviations.

Magnetic Susceptibility

Magnetic Susceptibility (MagSus) provides an estimate of the amount of iron-bearing allochthonous inorganic sediment being deposited over time in a closed-basin depositional setting (Thompson et al., 1975). While BSM is not a closed basin, its topographic setting and the nature of deposition approximate a closed-basin for the purposes of interpretation. Peaks in MagSus may result from increases in the amount of clastic sediment (high MagSus) relative to organic matter (low MagSus) deposited at a site. Peaks can occur as a result of the loss of upslope vegetative cover resulting from drought or fire, or from increases in precipitation resulting in the acceleration of alluvial and colluvial processes. We measured MagSus at 1 cm increments with 8 cc sediment samples using a Bartington MS2 cup Magnetic Susceptibility System.

Loss on Ignition

Loss on Ignition (LOI) was done at 550°C and 900°C in 1-cm increments to measure total inorganic carbonate, carbonate and non-carbonate sediment (Dean, 1974). In a closed-basin lacustrine setting, the carbonate content of sediments provided by LOI allows a relatively robust estimate of water temperature, depth, and internal productivity over time (Bischoff et al., 1997; Oviatt, 1997; Patrickson et al., 2010; Wetzel, 2001). Because BSM is an open-system, LOI is most useful as a measure of sedimentation rates and vegetative productivity.

Pollen

Pollen was processed using standard chemical digestion following Faegri et al. (1989), using 1cc samples. We used spores of the exotic Lycopodium as a tracer for influx calculation. Samples were counted at 500× magnification to 300 terrestrial pollen grains or 300 Lycopodium spores, whichever was reached first. Pollen was identified using a dichotomous key (Kapp et al., 2000) and by comparison with the western North American pollen collection housed at the University of Utah Department of Geography Records of Environment and Disturbance (RED) laboratory. A list of all pollen taxa identified from BSM along with examples of each pollen taxon found in RCC is provided in Supplemental Material Table V, available online. A total of 24 pollen samples were examined from the core from its upper 144 cm. Pollen was analyzed at 1 cm increments for the top 6 cm, and at 8 cm increments from 16 to 144 cm. The samples at 128 and 136 cm had very poor preservation and were not included in this analysis. Pollen counts are provided in Supplemental Material Table VI, available online.

Pollen sums and indices

We used pollen counts to calculate several sums and indices relevant to summer precipitation and maize agriculture during the Fremont period in RCC. We chose taxa relevant to reconstructing temperature, effective precipitation and precipitation seasonality. Sums were normalized to a zero to one scale by dividing all sums by the maximum value. Indices were calculated using the formula (a−b)/(a+b), resulting in values from −1 to +1. Positive values indicate higher relative abundance of taxon a, and negative values indicate dominance of taxon b. Changes over time in the relative abundance of one taxon compared to another should indicate climatic changes favoring one taxon over the other. A brief description of each sum or index is provided for reference in Table 1.

Table 1.

Pollen sums and ratios used in this study.

Name	Abbreviation	Proxy target
Total pollen influx	TPI	Total vegetative productivity
Amaranthaceae + Poaceae + Artemisia	APA	Total annual precipitation
Pinus: Juniperus	P:J	Precipitation and temperature
Acer negundo: Populus	A:P	Summer high temperature
Pinus: Douglas fir	P:DF	Precipitation seasonality

TPI (terrestrial productivity)

Total pollen influx (TPI) was used as a measure of total terrestrial vegetative productivity over time. TPI is a measure of the number of terrestrial pollen grains per year per unit area falling at a site. It is calculated here as (total pollen cm⁻² year⁻¹).

APA (summer precipitation)

We use the sum of Ambrosia, Poaceae and Amaranthaceae (APA) as a proxy for summer precipitation based on the climate space and modern pollen abundances for these taxa. In western North America, pollen abundances for these taxa are higher in sites where most of the annual precipitation falls in summer (Brunelle et al., in revision; Minckley et al., 2008; Williams, 2006)

P:J (overall moisture)

The abundance of Pinus compared to Juniperus pollen (P:J) has been used as a proxy for mesic/xeric conditions in Great Basin settings (Brunelle et al., in revision; Louderback and Rhode, 2009). Pines generally require cooler and/or wetter conditions than junipers. Two-needle pinyon and Utah juniper are consistent with this generalization, and we use the index as a mesic/xeric index. Higher P:J values indicate more pinyon and relatively mesic conditions, while lower values indicate more juniper and more xeric conditions.

Nearly all pine pollen grains in the core were identified as two-needle pinyon pine using the key provided by Jacobs (1985). This key identifies pollen grains of haploxylon type less than 70 μm in total length as P. edulis/P. monophylla and haploxylon grains larger than 70 μm as P. flexilis/P. strobiformis. Only one diploxylon pine grain was identified, a grain of ponderosa pine (P. ponderosa) in the 6–7 cm sample.

Pollen of the genus Juniperus is only identifiable to the family level (Cupressaceae). Three species of Juniperus occur in RCC, Utah juniper (J. osteosperma), Rocky Mountain juniper (J. scopulorum) and common juniper (J. communis). Utah juniper is by far the most abundant in the vicinity of BSM, so the family-level identification is sufficient for goal of the P:J index. The index was calculated as (Pinus − Cupressaceae)/(Pinus + Cupressaceae).

A:P (annual average temperature)

We use A:P, an index of the abundance of box elder compared to cottonwood (Populus spp.), as a proxy for summer temperature based on the climate zones these trees occupy today (Williams, 2006). The gallery forest surrounding BSM today consists of a mix of box elder and narrowleaf cottonwood trees. Pollen percentages in the BSM core show below, however, that the composition of the gallery forest has changed over time from dominance by one or the other taxon for periods of several 100 years.

The temperature ranges that both tree species occupy today, along with quaking aspen (P. tremuloides, see discussion next paragraph) can be found in Supplemental Material Table III, available online. While narrowleaf cottonwood and box elder do have considerable overlap in the temperature variables we analyzed, box elder occurs in areas with higher minimum, maximum, and mean annual temperatures. The average annual minimum, maximum, and mean temperatures over the geographic range of narrowleaf cottonwood are −1.84°C, 12.81°C, and 5.48°C. The corresponding values for the geographic range of box elder are 1.64°C, 16.85°C, and 9.25°C, several degrees warmer for each value. Changes in the abundance of box elder relative to narrowleaf cottonwood pollen in the BSM core should therefore indicate changes in annual average temperature over time, dominance by box elder indicating warm periods, and dominance by Populus indicating cooler intervals.

Pollen of box elder is identifiable at the species level, whereas Populus is only identifiable at the genus level. The only species of Populus within several miles of BSM today is narrowleaf cottonwood. Fremont cottonwood (P. fremontii) occurs several miles to the south at lower elevations in RCC. Quaking aspen occurs in RCC at elevations above 8000’ and has notoriously short-travelling and poorly-preserving pollen grains (Sangster and Dale, 1964). Quaking aspen therefore likely never migrated close enough to BSM to contribute pollen to the core. However, quaking aspen occupies a slightly cooler temperature range than does narrowleaf cottonwood (Supplemental Material Table III, available online). Its inclusion in A:P index therefore would only strengthen our interpretation of that index with respect to temperature.

A:P is calculated as (A. negundo − Populus)/(A. negundo + Populus). Values closer to +1 indicate warmer periods and values closer to −1 indicate cooler. In this study, the A:P values range from −1 (samples with only Populus) to +1 (samples with only box elder), indicating periods during which the gallery forest surrounding BSM was composed nearly entirely of one taxon or the other.

P:DF (precipitation seasonality)

The relative abundance of pinyon compared to Douglas fir is used here as an index of precipitation seasonality. Two-needle pinyon pine is adapted to areas receiving abundant summer precipitation. Compared to the single-needle species found in the Great Basin (P. monophylla), the two-needle pine has several adaptations requiring considerable summer precipitation. The two-needle per fascicle configuration increases the surface area of needles which allows for a greater amount of transpiration during the day, but requires higher soil moisture, especially in summer. Two-needle pinyon also produces smaller seeds than its single-needle counterpart, also necessitating higher soil moisture throughout the critical first summer growing season (Petersen, 1988, 1994). Douglas fir, however, is an excellent indicator of winter snowpack. Knight et al. (2010: Figure 3), for example, show that winter snowpack on the Tavaputs Plateau is the most important factor in Douglas fir annual ring widths. The relative abundance of Douglas fir pollen in sediments from RCC should therefore be a good indicator of the amount of winter precipitation over time. Further discussion of the geographic ranges and climatic signals for Douglas fir and two-needle pinyon pine are provided in Supplemental Material Section 3, available online.

Given the distinct precipitation preferences for two-needle pinyon and Douglas fir, the P:DF index should provide a measure of precipitation seasonality over time in the BSM sediments. We calculated P:DF as (Pinus − Pseudotsuga)/(Pinus + Pseudotsuga). Positive values indicate periods with more summer than winter precipitation, while negative values indicate periods with greater winter precipitation.

Results and discussion

We present results here for the suite of analyses conducted on BSM sediments but focus our discussion on the proxies which could help determine whether there was an increase in summer precipitation during the Fremont period which in turn could have contributed to the adoption of maize farming by the occupants of Range Creek Canyon from AD 800 to 1150 (1150–800 cal BP). There are several environmental and climatic factors which may affect maize productivity in dry farming and irrigation farming settings. The most significant effects are those of temperature and water. Temperature can be measured in accumulated heat units (often calculated as Growing Degree Days, GDD; Shaw, 1998), and can be affect by topography as well as climate. For example cold air drainage in canyon settings may negatively affect the accumulation of GDD. This may be a significant factor in canyon settings such as RCC. The analysis conducted here is not capable of reconstructing topographic effects such as cold air drainage, and we focus instead on the more broadly defined temperature and precipitation variables.

Dry farming maize is not possible in RCC today due to insufficient growing season precipitation in areas where temperature is adequate (Boomgarden, 2015; Boomgarden et al., 2019). Precipitation in the canyon is a function of elevation, and where sufficient rainfall occurs (6"+ during the growing season), the growing season is too short and accumulated GDD too few to grow maize. Conversely, where the growing season is sufficient in length and GDD accumulate to a sufficient amount for maize growth, precipitation alone is insufficient and supplemental water is required. A substantial increase in summer precipitation would be necessary to grow maize in any section of RCC with dry farming methods. We hypothesize that if increased summer precipitation is in fact what drove the rise of maize agriculture in RCC, then the Fremont portion of our record should indicate wetter summers than modern, and wetter summers than before or after the Fremont period. Conversely if we find no evidence of increased summer precipitation during the Fremont period, it should indicate that dry farming was not practiced to a large extent in RCC during the Fremont period.

Macroscopic charcoal

Mean fire return interval varies over the BSM record from 150 to 450 years (Figure 2). For most of the record, fire frequency and magnitude appear to be controlled by drought. Comparing the record with the Harmon Canyon Douglas fir ring-width based precipitation reconstruction from Knight et al. (2010), the highest magnitude fires occurred at transition points between very wet intervals and extreme droughts. The two highest magnitude fires occur at around 1850 cal BP (AD 100) and 1450 cal BP (AD 500). These two periods mark the transition from extended wet intervals to the two most severe droughts of the past 2000 years on the Tavaputs Plateau.

Figure 2.

Harmon Canyon precipitation reconstruction (31 year running mean, from Knight et al. (2010), Figure 5), fire history, magnetic susceptibility, total organic Carbon, total carbonate, stable Carbon isotope ratio values for BSM sediments. Zones 1–5 refers to pollen zones – see pollen discussion. Shaded background curve shows Range Creek Canyon summed probability distribution for calibrated radiocarbon dates from archaeological materials.

The Fremont period of the BSM fire history is marked by a near absence of charcoal peaks. This is not unique in the record, there are reductions in apparent fire activity during other intervals. However, if summer precipitation had increased during the Fremont period, total biomass in the area would have increased, and the resulting fuel buildup should have left the site ripe for a high severity fire event (Roos and Swetnam, 2012). The AD 1100s drought should therefore have had a larger effect on the fire record. A modest charcoal peak does occur beginning near 820 cal BP (AD 1130), coincident with the onset of the AD 1100s drought (Knight et al., 2010). That peak however is an order of magnitude smaller than the peak associated with the AD 500s drought and two orders of magnitude smaller than the peak associated with the AD 100s drought.

The archaeological SPD shown in Figure 2 and Supplemental Material Section 2, available online indicates that the Fremont occupation of RCC was bookended by two moderate fire events, with none between. It could be the case that the area around BSM was burned by people for agricultural purposes, and was kept clear with frequent maintenance burns. A similar pattern was found for Chery Meadow, a spring site down canyon from BSM, investigated and reported by Morris (2010) and Hart et al. (2011). This may have kept fuel from accumulating near the meadow and prevented fires large enough to contribute charcoal to the BSM record from occurring. Without several more records from different parts of the canyon it is difficult to determine if the absence of fire during the Fremont period at BSM has anything to do with human activity. But it is interesting to note that while for much of the record drought appears to drive fire activity, during the Fremont period, fire and climate appear to be decoupled and there is no evidence for an increase in summer precipitation during that interval in any of the proxies evaluated here.

Magnetic Susceptibility (MagSus) and Loss on Ignition (LOI)

Results for MagSus and LOI are presented in Figure 2. Interpreting the geomorphologic implications of MagSus and LOI is not as straightforward as other proxies, particularly due to the hybrid nature of the BSM site; it is neither a stream channel nor a lake. Variations in stream channel degradation and aggradation depend on a variety of geomorphic characteristics such as sediment grain size and composition, vegetative cover, slope and aspect, etc. in addition to precipitation, discharge rates and flood frequency (Miller et al., 2004). In an upland wet meadow context at the margin of a colluvial slope such as BSM, we might expect elevated levels in MagSus with increased summer precipitation due to increased sediment from upslope being introduced due to more frequent slopewash events. Or, with higher summer precipitation we might also expect increases in total organic carbon of sediments due to increased productivity at the site. In fact no real pattern is evident in these proxies during the Fremont period at BSM. These lines of evidence offer no support for or against increased summer precipitation at RCC during the Fremont period.

Stable carbon isotope chemistry of BSM sediments

Figure 3 shows δ¹³C values for a variety of native Utah plants, maize from experimental, ethnographic and archaeological settings, and for sediment organic remains from the BSM core. As the figure shows, C3 plants (most of the native plants of Utah) have δ¹³C values between −29 and −22‰. Most wild C4 plants produce δ¹³C values of between −16 and −11‰. Maize analyzed by Coltrain and Leavitt (2002) produced δ¹³C values of between −12 and −10‰, slightly enriched over native Utah C4 plants. Maize grown at the Range Creek Field Station in 2014 as part of Boomgarden’s (2015) research produced values between −12.7 and −11.0‰ with a mean value of −11.7 ± 0.4‰. The enriched δ¹³C values of BSM sediments at 62 and 66 cm in depth (dating to ca. 1150 and 1270 cal BP, respectively) indicate the presence of the remains of maize plants in these sediments, evidence that maize was grown at BSM during the Fremont period and thus bolstering this context as one clearly relevant to evaluating the effects of precipitation variation on maize agriculture. A similar enrichment in δ¹³C was found in sediments from Cherry Meadows (core RCCM07B; Hart et al., 2011), and these results combined with the pollen evidence presented below, which do not show increases in the native C4 plants, support the utility of this method for finding evidence of maize agriculture in sediment organic chemistry.

Figure 3.

Stable carbon isotope ratio values for native Utah C3 and C4 plants, archaeological and experimentally grown maize, and sediment samples from the 2009 BSM core (circles). Values for individual plant species come from plants collected in RCC.

Pollen percentages, sums, and indices

Pollen percentages for taxa that compose more than 1% of the identified pollen in any sample are presented in Figure 4. Taxa which do not compose more than 1% are not included in Figure 4 but were used to calculate total pollen influx. Pollen sums and indices are presented in Figure 5. We used Tilia’s stratigraphically constrained cluster analysis function (CONISS) to identify major pollen zones (Grimm, 1987, 1990). We interpret the resulting clusters to show five stratigraphic zones, which were used to organize Figures 2, 4, and 5. Zone 3 is the chronological zone from 1300 to 850 cal BP, which encompasses the Fremont occupation of RCC and will be the focus of our discussion here.

Figure 4.

Pollen percentages for all taxa making up more than 1% of any sample. Taxa with low percentages are exaggerated 2× with shaded curves. Shaded gray background curve is the SPD of Fremont archaeological radiocarbon dates from Range Creek Canyon.

Figure 5.

Pollen sums and ratios for the past 2350 calibrated radiocarbon years. TPI: total pollen influx (terrestrial productivity proxy); APA: The sum of Amaranthaceae + Poaceae + Artemisia (summer precipitation proxy); P:J: Pinyon:Juniper (overall moisture proxy); A:P: Acer:Populus (summer temperature proxy); P:DF – pinyon:Douglas fir (precipitation seasonality proxy).

The most noteworthy result of the pollen analysis is the presence of direct evidence of maize farming occurring at BSM. A grain of maize pollen was identified in pollen zone 3 at 56 cm. Using the age-depth model discussed earlier, this sample dates to between 930 and 1190 cal BP, with a weighted mean date of 1060 cal BP (AD 890). As noted above, the stable carbon evidence also indicates the presence of maize at a similar depth. Eleven maize cobs from RCC have been dated using radiocarbon (Boomgarden, 2015). These 11 cobs have calibrated radiocarbon median ages ranging from 1020 to 850 cal BP (AD 930–1100; Supplemental Material Table II, available online), consistent with the age of evidence of maize agriculture at BSM during the Fremont occupation of RCC.

After identifying the grain of maize pollen and stable carbon isotope evidence of maize, an additional effort to identify maize pollen was undertaken. At intervals of 1 cm, a twelve centimeter section of the core, from 66 to 54 cm was processed for maize pollen detection. 5-cc samples were screened at 35 and 150 microns (retaining the 53–150 micron size fraction for analysis) and processed via standard chemical digestion (Faegri et al., 1989). Samples were then scanned at 100× magnification. No additional maize pollen was found.

The Fremont period at BSM is marked by a reduction in most pollen taxa, and increases only in xeric-oriented taxa such as Amaranthaceae (goosefoot) and Cupressaceae. Aquatic types disappear completely and most upland herbs decline. Within the tree taxa, mesic species decline (pinyon, Douglas fir and Populus), while Cupressaceae are higher. A relatively low P:DF (precipitation seasonality) index indicates moderately dry summers and wet winters (Figure 5). TPI (terrestrial productivity), APA (summer moisture) and P:J (overall moisture) are very low during this period, consistent with warm, dry summers. A:P (summer temp) starts the zone relatively low and transitions to very high toward the end of zone 3, consistent with warming summers throughout the zone. These proxies together indicate generally warm and dry conditions until approximately 700 cal BP (AD 1250) followed by a cooler, dry period from approximately 700 to 150 cal BP (AD 1250–1800) in the Tavaputs Plateau area.

A dominant climate driving mechanism during the Fremont period and MCA more broadly may have been the presence of persistent La Niña-like conditions in the northern hemisphere (Cohen et al., 2012; Mann et al., 2009; Metcalfe et al., 2015). La Niña conditions are associated with generally dry winters in the Southwest and wet winters in the Northwest. The Tavaputs Plateau, however, lies near the dipole zone between these regions, and the effects of the El Niño/Southern Oscillation (ENSO) are less predictable than to the north or south (Wise, 2010). Knight et al. (2010) found a significant negative correlation between ENSO and Douglas fir ring widths in Harmon Canyon, indicating El Niño years on average bring more winter precipitation to the region. Table 2 presents the average summer and winter precipitation values for Bruin Point at the north end of RCC, for El Niño, La Niña and ENSO neutral years since 1950 (based on historic southern oscillation index values from the NOAA climate prediction center, http://www.cpc.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml). Precipitation data for Table 2 are from the PRISM Climate Group (2004).

Table 2.

Average seasonal precipitation (mm) at Bruin Point for ENSO phases since 1950.

	E Niño avg	La Niña avg	ENSO neutral
DJFMAM	238.82	221.48	229.96
JJASON	302.98	314.41	341.2

Table 2 shows that while El Niño years do result on average in slightly heavier winter precipitation at Bruin Point (consistent with the findings of Knight et al., 2010), La Niña years do not differ significantly from ENSO neutral years with respect to winter precipitation. Both El Niño and La Niña years have reduced average summer precipitation at Bruin Point. All else being equal, a persistent La Niña climate state during the Fremont period would result in slightly lower average snowpack and a larger reduction in average summer precipitation. These expectations are consistent with the pollen record from BSM, with reduced Pinus and Pseudotsuga pollen, lower than average P:DF (winter-dominated precipitation seasonality), lower P:J (moisture), low TPI (productivity) and low APA (summer precipitation).

Implications for human foraging efficiency and the adoption of maize agriculture in RCC

The only ethnobotanically significant wild taxon found in the core is pinyon pine. Pinyon pine is an important resource ethnographically and is taken by people virtually everywhere it is encountered. Pinyon is generally accepted as one of the highest ranked plant based resources in terms of energetic returns (Barlow and Metcalfe, 1996; Simms, 1985). Because of its high rank, decreases in its abundance likely resulted in decreases in the average rates of return for full-time foragers as well as foragers that farmed. The transition between pollen zones 3 and 4 in the BSM record shows a decline in pinyon pine abundance, as well as the several pollen indices and sums related to summer precipitation and overall productivity. These data show warm, dry conditions during the Fremont period, indicating low overall terrestrial productivity, and that return rates available to foragers for plant foods would likely have declined during the Fremont period in RCC.

Although we do not present data here from archaeological investigations of Fremont dietary change in RCC, taken together, these data suggest that the adoption of maize farming in RCC during the Fremont period was coincident with an overall reduction in human foraging efficiency. No evidence can be found in the data presented here for an increase in summer or annual precipitation in the BSM record during the Fremont period. The adoption of maize agriculture in RCC may instead have been a response to deteriorating environmental conditions and declining foraging efficiency.

Declining human foraging efficiency is seen elsewhere in the Southwest in areas and times where prehistoric people made the transition from wild food hunting and gathering to agricultural production (e.g. Cannon, 2000, 2001). These observations produce a number of interesting implications and avenues for further research in the Fremont region. Chief among them regards the observation that human populations, as measured by radiocarbon date frequencies, were reaching their prehistoric peak across the state of Utah at this time (Louderback et al., 2010; Massimino and Metcalfe, 1999). This pattern is also seen in the broader Southwest in other chronologies such as those based on tree-cutting dates for architectural timbers in the Ancestral Puebloan region (Benson et al., 2007; Berry, 1982; Berry and Berry, 2003). So if environmental deterioration and declining foraging efficiency played a role in driving people to adopt agriculture during Fremont times and Ancestral Puebloan times, why would human populations respond by growing?

We assume that the increase in date frequency is not simply a function of longer site residence times and increased archaeological visibility, but this may not be the case. Archaeological visibility is a function of site residence time, and it could be that the increased sedentism associated with agricultural societies simply made them more archaeologically visible, with more dateable material, Estimates of actual human population size over time using genetic diversity of southwestern skeletal materials could in the future rule out taphonomic factors as contributing to the observed spike in the radiocarbon date frequency curve. This approach however may be impractical given modern attitudes regarding destructive analysis of human remains. But research with modern human populations may provide valuable insight as well. For example, Page et al. (2016, 2018) argue that the increased morbidity, infant mortality and overall negative health effects resulting from increased sedentism in agricultural populations are offset by increased fertility and reduced inter-birth intervals when compared with mobile hunter-gatherer populations in the Philippines. Analysis of Fremont and Ancestral Puebloan skeletal remains have provided results consistent with these observations. Skeletal pathologies consistent with dietary stress are common in Fremont and Ancestral Puebloan burials (Bright and Loveland, 1999; Coltrain and Leavitt, 2002; Fawcett and Simms, 1993; Owsley et al., 1996; Roberts, 1991), as are age profiles showing high degrees of infant mortality (Roberts, 1991). The observed increases in human populations after the adoption of agriculture might be the product of ecological pressures favoring reduced investment in more offspring for humans, resulting in human population growth despite signs of reduced health overall.

In addition to the reduced availability of wild foods (declining foraging efficiency) during the Fremont period in RCC, the observed reduction in summer precipitation would have made maize farming impossible without irrigation (Boomgarden, 2015). Because of this, maize productivity in RCC would have been a function of several variables including annual snowpack and groundwater holding potential of the Tavaputs Plateau. On a local scale in the canyon, it would have been a function of the amount of available water in the creek, the amount of irrigable land at varying elevations, and the costs of building and maintaining irrigation networks. The costs of irrigating may further be broken down into building dams and maintaining ditches, both of which may also be influenced by empirical variables relating to available water flow magnitude and variance. The frequency and intensity of precipitation events during the growing season may create a tradeoff between building dams and maintaining ditches; more frequent monsoons would cause frequent dam failure of weak dams while building more robust dams would cause ditches to fill with sediments from flood events. A better understanding of the costs of building and maintaining dams and ditches would thus contribute significantly to our understanding of the Fremont occupation of RCC and of prehistoric maize cultivation, and of the transition to agriculture more broadly (Boomgarden, 2015, 2016; Boomgarden et al., 2019; Kuehn, 2014).

Conclusion

We have presented a multiproxy paleoenvironmental reconstruction for the past 3000 years at BSM in RCC. Enriched δ¹³C values and maize pollen show that the meadow was used as a maize field at some point between 930 and 1190 cal BP (AD 760–1020). These results support the utility of stable carbon isotope analysis of sediment organics for identifying prehistoric maize fields. Our pollen-climate reconstruction is in agreement with local, regional and hemispheric climate records. The BSM record shows a cool, wet period preceding the Fremont occupation of RCC, a warm interval with especially dry summers, a cool dry and a historic period since AD 1850 similar to modern conditions. These observations contradict the hypothesis that an increase in summer precipitation drove the rise of maize agriculture in RCC during Fremont times. It may instead have been a predictable snowpack and with relatively dry summers during Fremont times which allowed the florescence of maize agriculture across the Fremont homeland. This interpretation is complicated by records from sites such as Homestead Cave in Utah’s West Desert, where faunal remains indicate a generally wet period with rising lake levels occurring at ca. AD 1000. Further paleoenvironmental investigations paying special attention to precipitation seasonality might help show whether the pattern seen in RCC is typical or anomalous for Fremont times.

Analysis of oxygen isotopes of archaeological maize cobs in RCC may also contribute in this regard in the future. Williams et al. (2005), show that maize grown in the Southwest can be identified as having been grown from predominantly winter (snowmelt and perennial creeks) or monsoon-based precipitation based on the relative abundance of two stable isotopes of oxygen, ¹⁸O and ¹⁶O. Identifying the climatic source of the water used to grow maize in RCC and across the Fremont and Ancestral Puebloan regions would contribute significantly to our understanding of the tradeoffs prehistoric maize farmers in the Southwest faced.

Interpretations of the paleoenvironmental proxy data presented in this paper would be strengthened by more thorough analysis of available tree ring datasets. For example, the Douglas fir chronology from Harmon Canyon (Knight et al., 2010) has not been analyzed for early wood/late wood thicknesses or stable oxygen isotopes. Such analyses would provide a better understanding of precipitation seasonality on the Tavaputs Plateau for the past two millennia. Given the climatic relationships between Douglas fir, two-needle pinyon pine and precipitation seasonality a chronology of pinyon pine ring widths from RCC with special attention paid to stand age to identify periods of high and low seedling recruitment could also improve interpretation of the pollen data presented here.

Supplemental Material

sj-pdf-1-hol-10.1177_0959683620972767 – Supplemental material for Evidence for a winter-snowpack derived water source for the Fremont maize farmers of Range Creek Canyon, Utah, USA

Supplemental material, sj-pdf-1-hol-10.1177_0959683620972767 for Evidence for a winter-snowpack derived water source for the Fremont maize farmers of Range Creek Canyon, Utah, USA by Isaac Alfred Hart, Joan Brenner-Coltrain, Shannon Boomgarden, Andrea Brunelle, Larry Coats, Duncan Metcalfe and Michael Lewis in The Holocene

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Isaac Alfred Hart

Supplemental material

Supplemental material for this article is available online.

References

Barlow

(1997) Foragers that farm: A behavioral ecology approach to the economics of corn farming for the Fremont case. PhD Thesis, University of Utah, Salt Lake City/University Microfilms, Ann Arbor.

Barlow

(2002) Predicting maize agriculture among the Fremont: An economic comparison of farming and foraging in the American Southwest. American Antiquity 67(1): 65–88.

Barlow

Metcalfe

(1996) Plant utility indices: Two Great Basin examples. Journal of Archaeological Science 23(3): 351–371.

Benson

Petersen

Stein

(2006) Anasazi (pre-Colombian Native American) migrations during the middle-12th and late-13th centuries–were they drought induced? Climatic Change 83: 187–213.

Benson

Berry

Jolie

, et al. (2007) Possible impacts of early-11th, middle-12th, and late-13th-century droughts on western Native Americans and the Mississippian Cahokians. Quaternary Science Reviews 26(3–4): 336–350.

Benson

Berry

(2009) Climate change and cultural response in the prehistoric American Southwest. Kiva 75(1): 87–117.

Benson

(2011) Factors controlling pre-Columbian and early historic maize productivity in the American southwest, Part 2: The Chaco Halo, Mesa Verde, Pajarito Plateau/Bandelier, and Zuni archaeological regions. Journal of Archaeological Method and Theory 18(1): 61–109.

Bird

O’Connell

(2006) Behavioral ecology and archaeology. Journal of Archaeological Research 14(2): 143–188.

Bischoff

Fitts

Fitzpatrick

(1997) Responses of sediment geochemistry to climate change in Owens Lake sediment: An 800-k.y. record of saline/fresh cycles in core OL-92. Geological Society of America Special Paper 317: 37–48.

10.

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6(3): 457–474.

11.

Boomgarden

Springer

Metcalfe

, et al. (2014) Report of the Archaeological Investigations in Range Creek Canyon, Utah: University of Utah Archaeological Field School 2013. Manuscript on file at the Natural History Museum of Utah and the Antiquities Section of the Utah Division of State History, Salt Lake City Utah.

12.

Boomgarden

(2015) Experimental maize farming in Range Creek Canyon. PhD Thesis, The University of Utah, Utah.

13.

Boomgarden

(2016) Maize Farming Experiments in Range Creek Canyon, Utah. In: Poster presented at the 81st Society for American Archaeology conference, Orlando, Florida, April 7 2016. Salt Lake City: The University of Utah.

14.

Boomgarden

Metcalfe

Simons

(2019) An optimal irrigation model: Theory, results and implications for future research. American Antiquity 84(2): 252–273.

15.

Broughton

Madsen

Quade

(2000) Fish remains from Homestead Cave and lake levels of the past 13,000 years in the Bonneville Basin. Quaternary Research 53(3): 392–401.

16.

Broughton

Byers

Bryson

, et al. (2008) Did climatic seasonality control late Quaternary artiodactyl densities in western North America? Quaternary Science Reviews 27(19–20): 1916–1937.

17.

Broughton

Cannon

(eds.) (2010) Evolutionary ecology and Archaeology: Applications to Problems in Human Evolution and Prehistory. Salt Lake City, Utah: University of Utah Press.

18.

Broughton

Cannon

Bartelink

(2010) Evolutionary ecology, resource depression, and niche construction theory: Applications to central California hunter-gatherers and Mimbres-Mogollon agriculturalists. Journal of Archaeological Method and Theory 17(4): 371–421.

19.

Broughton

Smith

(2016) The fishes of Lake Bonneville: Implications for drainage history, biogeography, and lake levels. In: Developments in Earth Surface Processes, vol. 20. Amsterdam: Elsevier, pp.292–351.

20.

Brunelle

Eckerle

Petersen

, et al. (in revision) A high-resolution Holocene paleoenvironmental history of Lake Bonneville/Great Salt Lake. In: Rosen

Starratt

(eds.) Geological Society of America Special Papers (2019).

21.

Byers

Broughton

(2004) Holocene environmental change, artiodactyl abundances, and human hunting strategies in the Great Basin. American Antiquity 69(2): 235–255.

22.

Cannon

(2000) Large mammal relative abundance in Pithouse and Pueblo period archaeofaunas from southwestern New Mexico: Resource depression among the Mimbres-Mogollon? Journal of Anthropological Archaeology 19(3): 317–47.

23.

Cannon

(2001) Large mammal resource depression and agricultural intensification: An empirical test in the Mimbres Valley. PhD Thesis, University of Washington, New Mexico

24.

Cohen

Nanson

Jansen

, et al. (2012) A pluvial episode identified in arid Australia during the Medieval Climatic Anomaly. Quaternary Science Reviews 56: 167–171.

25.

Coltrain

Leavitt

(2002) Climate and diet in Fremont prehistory: Economic variability and abandonment of maize agriculture in the Great Salt Lake Basin. American Antiquity 67(3): 453–485.

26.

Dean

Jr (1974) Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: Comparison with other methods. Journal of Sedimentary Petrology 44: 242–248.

27.

Faegri

Kaland

Krzywinski

(1989) Textbook of Pollen Analysis. Chichester, UK: John Wiley and Sons Ltd.

28.

Fawcett

Simms

(1993) Archaeological Test Excavations in the Great Salt Lake Wetlands and Associated Analyses. Contributions to Anthropology No. 14. Logan, UT: Utah State University.

29.

Graybill

(1990) Mammoth Creek, Utah, Tree Ring Record. Boulder, CO: International Tree Ring Data Bank. GBP PAGES/World Data Center-A for Paleoclimatology, NOAA/NGDC Paleoclimatology Program.

30.

Grayson

(2000) The Homestead Cave mammals. In: Madsen

(ed.) Late Quaternary Paleoecology in the Great Basin, Utah Geological Survey Bulletin No. 130. Salt Lake City, Utah: Utah Geological Survey, pp.67–89.

31.

Grimm

(1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers and Geosciences 13(1): 13–35.

32.

Grimm

(1990) TILIA and TILIAGRAPH. PC spreadsheet and graphics software for pollen data. INQUA Working Group on Data Handling Methods Newsletter 4: 5–7.

33.

Hart

Brunelle

Coltrain

(2011) A 4,000 year pollen record from Range Creek Canyon, Utah. Paper presented at the 76th Society for American Archaeology conference, Sacramento, CA.

34.

Higuera

(2009) CharAnalysis 0.9: Diagnostic and Analytical Tools for Sediment Charcoal Analysis. User’s Guide. Bozeman, MT: Montana State University.

35.

Jacobs

(1985) Identification of pine pollen from the southwestern United States. AASP Contribution Series 16: 155–168.

36.

Kapp

King

Davis

(2000) Ronald O. Kapp’s Pollen and Spores. College Station, TX: American Association of Stratigraphic Palynologists Foundation Publication.

37.

Kennett

Winterhalder

(eds.) (2006) Behavioral Ecology and the Transition to Agriculture, vol. 1. Berkeley, California: University of California Press.

38.

Knight

Meko

Baisan

(2010) A bimillenial-length tree-ring reconstruction of precipitation for the Tavaputs Plateau, Northeastern Utah. Quaternary Research 73(1): 107–117.

39.

Kuehn

(2014) The agricultural economics of Fremont irrigation: A case study from south-central Utah. M.S. Thesis, Utah State University, Logan, Utah.

40.

Leavitt

(1994) Major wet interval in White Mountains medieval warm period evidenced in δ¹³C of bristlecone pine tree rings. Climatic Change 26(2–3): 299–307.

41.

Louderback

Grayson

Llobera

(2010) Middle-Holocene climates and human population densities in the Great Basin, western USA. The Holocene 21 (2): 366–373

42.

Louderback

Rhode

(2009) 15,000 years of vegetation change in the Bonneville Basin: The Blue Lake pollen record. Quaternary Science Reviews 28(3): 308–326.

43.

MacArthur

Pianka

(1966) On optimal use of a patchy environment. The American Naturalist 100(916): 603–609.

44.

Madsen

Simms

(1998) The Fremont complex: A behavioral perspective. Journal of World Prehistory 12(3): 255–336.

45.

Mann

Zhang

Rutherford

, et al. (2009) Global signatures and dynamical origins of the Little Ice Age and Medieval Climate Anomaly. Science 326(5957): 1256–1260.

46.

Massimino

Metcalfe

(1999) New form for the formative. Utah Archaeology 12(1): e16.

47.

Matson

Lipe

Haase

(1988) Adaptational continuities and occupational discontinuities: The Cedar Mesa Anasazi. Journal of Field Archaeology 15(3): 245–263.

48.

Metcalfe

Barron

Davies

(2015) The Holocene history of the North American Monsoon: ‘Known knowns’ and ‘known unknowns’ in understanding its spatial and temporal complexity. Quaternary Science Reviews 120: 1–27.

49.

Miller

House

Germanoski

, et al. (2004) Fluvial geomorphic responses to holocene climate change. In: Chambers

Miller

(eds.) Great Basin Riparian Ecosystems: Ecology, Management and Restoration. Washington, DC: Island Press.

50.

Minckley

Bartlein

Whitlock

, et al. (2008) Associations among modern pollen, vegetation, and climate in western North America. Quaternary Science Reviews 27 (21–22): 1962–1991.

51.

Morris

(2010) After the Fremont: Fire and vegetation history of prehistoric abandonment and historic occupation of Range Creek Canyon, Utah. M.S. Thesis, The University of Utah, Salt Lake City, UT.

52.

Owsley

Novak

Hamzavi

(1996) Examination of Burials from Around the Great Salt Lake, Utah. Manuscript on File, Salt Lake City: Department ofAnthropology, University of Utah.

53.

Oviatt

(1997) Lake Bonneville fluctuations and global climate change. Geology 25: 155–158.

54.

Page

Viguier

Dyble

, et al. (2016) Reproductive trade-offs in extant hunter-gatherers suggest adaptive mechanism for the Neolithic expansion. Proceedings of the National Academy of Sciences 113(17): 4694–4699.

55.

Page

Minter

Viguier

, et al. (2018) Hunter-gatherer health and development policy: How the promotion of sedentism worsens the Agta’s health outcomes. Social Science & Medicine 197: 39–48.

56.

Patrickson

Sack

Brunelle

, et al. (2010) Late Pleistocene to early Holocene lake level and paleoclimate insights from Stansbury Island, Bonneville basin, Utah. Quaternary Research 73(2): 237–246.

57.

Petersen

(1988) Climate and the Dolores River Anasazi: A Paleoenvironmental Reconstruction from a 10,000-year Pollen Record, La Plata Mountains, Southwestern Colorado. University of Utah Anthropological Papers No. 113. Salt Lake City: University of Utah Press.

58.

Petersen

(1994) A warm and wet Little Climatic Optimum and a cold and dry Little Ice Age in the southern Rocky Mountains, USA. Climatic Change 26 (2–3): 243–269.

59.

Reimer

Bard

Bayliss

, et al. (2013) IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP. Radiocarbon 55(4): 1869–1887.

60.

Roberts

(1991) A comparative analysis of human skeletal remains from Parowan Fremont, Virgin Anasazi, and Kayenta Anasazi archaeological sites. M.A. Thesis, University of Nevada, Las Vegas.

61.

Roos

Swetnam

(2012) A 1416-year reconstruction of annual, multidecadal, and centennial variability in area burned for ponderosa pine forests of the southern Colorado Plateau region, Southwest USA. The Holocene 22(3): 281–290.

62.

Sangster

Dale

(1964) Pollen grain preservation of underrepresented species in fossil spectra. Canadian Journal of Botany 42: 437–449.

63.

Shaw

(1988) Climate requirement. Agronomy Monographs 18: 609–638.

64.

Simms

(1985) Acquisition cost and nutritional data on Great Basin resources. Journal of California and Great Basin Anthropology 7(1): 117–126.

65.

Stine

(1990) Late Holocene fluctuations of Mono Lake, eastern California. Palaeogeography, Palaeoclimatology, Palaeoecology 78 (3–4): 333–381.

66.

Thompson

Battarbee

O’Sulliuan

, et al. (1975) Magnetic susceptibility of lake sediments. Limnology and Oceanography 20(5): 687–698.

67.

Webb

Schwarcz

Jensen

, et al. (2007) Stable Carbon isotope signature of ancient maize agriculture in the soils of Motul de San José, Guatemala. Geoarchaeology 22(3): 291–312.

68.

Wetzel

(2001) Limnology: Lake and River Ecosystems. Academic Press, San Diego.

69.

Whitlock

Larsen

(2002) Charcoal as a fire proxy. In: Last

Smol

(eds) Tracking Environmental Change Using Lake Sediments. Dordrecht: Springer, pp. 75–97.

70.

Williams

Coltrain

Lott

, et al. (2005) Oxygen isotopes in cellulose identify source water for archaeological maize in the American Southwest. Journal of Archaeological Science 32(6): 931–939.

71.

Williams

(2006) An Atlas of Pollen-Vegetation-Climate Relationships for the United States and Canada (No. 43). American Association of Stratigraphic Palynologists Foundation.

72.

Winterhalder

(1981) Optimal foraging strategies and hunter-gatherer research in anthropology: Theory and models. In: Hunter-Gatherer Foraging Strategies: Ethnographic and Archaeological Analyses. Chicago, University of Chicago Press.

73.

Wise

(2010) Spatiotemporal variability of the precipitation dipole transition zone in the western United States. Geophysical Research Letters 37(7).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.04 MB