Building resilient landscapes in a semi-arid watershed: Anthropogenic and natural burning histories in Late Holocene Tesuque Creek,northern New Mexico

Abstract

As mega-fires have swept the North American West in recent decades, studies of past fire events have gained academic interest. Deep-time perspectives are necessary to better understand the periodicity of fire events and to identify basic drivers of frequent fire episodes through time, including potential anthropogenic fire use. Our study contributes to the growing field of archeological fire ecology in the Southwest with an examination of a high-resolution geobotanical dataset for Holocene fire regimes as indicated within sediments from an alluvial section from Tesuque Creek in northern New Mexico. Using phytolith, geochemical, and sedimentological data, we reconstruct indices for burn cycle seasonality and destabilization of riparian vegetation communities that suggest potential anthropogenic land use and possible patterns distinguishing human burning from lightning ignitions. Through this study, we consider the potential role of hunter-gatherer behavior in producing fire landscapes as part of their strategies for managing risks in the fragile environments of the Middle-Late Holocene in northern New Mexico. We suggest that combining geo and botanical datasets from alluvial sections with archeological evidence will allow researchers to fine-tune the resolution of local climate sequences and their environmental impacts on small landscape scales. When such small-scale studies can be aggregated, they may further contribute to testing models of regional patch-burning, either natural or anthropogenic. We further posit that human relationships with fire in the past should be considered as collaborative efforts with lightning-ignited burns, to manage the risks associated with both too much and too little fire in fragile dryland ecosystems.

Keywords

Archaic environmental archeology fire ecology geoarchaeology New Mexico phytoliths pyrodiversity Southwest

Introduction

There is perhaps no more iconic symbol of the current environmental vulnerability in the North American West than wildfire. For the past several decades, global climate change has profoundly impacted regional drought cycles. Decreasing precipitation has aridified landscapes and desiccated woodlands and grasslands, creating more fuel for wildfires. In turn, greater wildfire severity and scale have disrupted seasonal and interannual cycles of water availability and vegetation growth by killing off unprecedented amounts of old growth vegetation (Bowman et al., 2020; Higuera and Abatzoglou, 2021; Lian et al., 2021). The destruction of old growth vegetation communities increases erosion and water seepage and has seriously impacted regional biodiversity (Jiménez-Moreno et al., 2008; Lian et al., 2021). This has consequences for ecosystem services, modern economic and recreational land use, and the preservation of fragile cultural heritage sites (Keppel et al., 2012; White and White, 1996). Unplanned fires are risky to human and non-human populations alike in that they have adverse effects on otherwise controlled anthropogenic spaces, such as neighborhoods, recreation areas, grazing land, wildland-urban interfaces, and infrastructure.

The recent increase in scale and frequency of wildfires has been documented via paleoecological and geological proxies in a variety of environmental contexts around the world (e.g. Bird et al., 2018; Bowman et al., 2009; Cuthrell et al., 2016; Mori and Lertzman, 2011; Swetnam and Baisan, 2003; Trauernicht et al., 2015). Most modern landscape burning is caused accidentally by humans (e.g. failed electric infrastructure, smoking, or outdoor cooking accidents) and occurs outside of planned and controlled anthropogenic burning programs (Short, 2017). Modern planning for wildfire has tended to be reactive to recent fire damage, which has produced a focus on fire suppression rather than long-term controlled fire management (Hessburg et al., 2021; Mourao and Martinho, 2019). A more sustainable approach requires an understanding of the relationships of humans to landscape fires throughout time.

This is a critical modern problem with ancient roots. Increasingly, researchers and policy makers acknowledge that in the past, planned burning by ancestral Indigenous communities played an important role in managing fire and protecting biodiversity in the North American West and elsewhere (Cuthrell et al., 2016; Liebmann et al., 2016; Lightfoot and Cuthrell, 2015; Lightfoot et al., 2013; Roos and Guiterman, 2021; Snitker et al., 2022). Intentional burning was integral to complex human relationships with fire, vegetation communities, water, and land because burning sustained biodiversity and provided ecological niches important for hunting and plant propagation (Scherjon et al., 2015; Trauernicht et al., 2015). Anthropogenic burning interacted with seasonal lightning ignited burns, rather than working to suppress burning altogether. Fire promoted landscape resilience and supported Indigenous ways of life within that landscape. By applying strategic burning in concert with natural fire regimes, Indigenous communities understood that some loss of vegetation and land was necessary in the short term to ensure its recovery and usefulness in the long term. Resilience, in this sense, encompassed temporary loss as a mechanism for sustained ecological balance, renewal, and continuity (Cajete et al., 2018; Johnson et al., 2016).

Understanding deep-time patterns of human-fire interactions is critical in the contemporary context of a changing global climate (Brewer et al., 2014; Herman, 2016). Setting local baselines for past fire activity is crucial to implementing successful regional risk management strategies for increasingly fire-prone, aridifying landscapes. Fire ecology studies have demonstrated that local environmental variables such as soil composition, humidity, topography, and vegetation cover impact broader patterns of fire risk (Mori and Lertzman, 2011; Trauernicht et al., 2015). Similarly, studies of Indigenous burning practices have shown that local, context-specific human knowledge significantly impacts the way in which such risks are perceived and managed (e.g. Kimmerer and Lake, 2001; Roos, 2008; Roos et al., 2018; Trauernicht et al., 2015).

Our research into Mid-Late Holocene local fire histories in northern New Mexico contributes to the growing field of fire ecology in the North American Southwest by reconstructing past fire events and possible proxies for fire-human interaction in northern New Mexico. This study focuses on the Tesuque Valley catchment, which feeds an area of the Rio Grande floodplain in the Tewa Basin between Santa Fe and Española (Figure 1). In this study, we applied geoarchaeological and paleobotanical methods to identify locally discrete fire events as indicated within sediments from an alluvial section in Tesuque Creek, a small tributary to the Rio Grande.

Figure 1.

Map and picture of the Tesuque Creek East Section including C14 dates and geomorphic units identified.

The alluvial section presented here was deposited within the Late Archaic (3500–1500 BP). In this paper, we first establish the environmental and cultural context of this section in the Late Archaic as known from previous studies. We then present the results of our geoarchaeological and microbotanical analyses, which strongly suggest human land use increased in this area during the Late Archaic, including some of the burning activity evidenced. We then discuss the implications of those results for improving our understanding of the potential role of hunter-gatherers in producing and managing fire landscapes as a strategy for managing fire risk during regional Middle-Late Holocene aridification episodes. Approaches that combine geochemical and botanical datasets from alluvial sections with archeological evidence offer the potential to refine reconstructions of local, place-based environmental responses to changes in hydroclimate and human landscape management. We conclude that assembling such local datasets will contribute to testing models of both natural and anthropogenic regional burning. This will allow researchers to determine how burning patterns relate to ways that people buffered the risks of fire regimes and climate change in the past.

Regional landscape and climate history

Past climates are significant drivers of fire frequency due to shifts in vegetation composition and density, stream flow and soil stability. Understanding the dynamically shifting temperature and precipitation regimes through paleoenvironmental proxies provides insight into human-environmental interactions and the decisions that people made in manipulating their environmental settings over time (e.g. Asmerom et al., 2010; Haynes, 2008; Holliday and Meltzer, 2010; Polyak et al., 2001).

In the North American Southwest, the Younger Dryas (YD) was a brief (~12,900–11,700 years BP) cold and moist period following the Bølling-Allerød warm oscillation (Ballenger et al., 2011; Hall, 2017; Holliday and Meltzer, 2010). Wetter conditions resulted in high water tables and would have introduced new springs, wetlands, and marshes in many localities. Pollen studies indicate that the range of alpine forest vegetation migrated downslope in response to the cooling climate in Colorado during this time (Reasoner and Jodry, 2000), although this is not yet demonstrated for the New Mexico or Arizona mountains. Artemisia steppe vegetation dominated the Southwestern landscape (Jiménez-Moreno et al., 2008). Some pluvial-lake records, such as Lake Estancia, New Mexico (Allen and Anderson, 2000) and Lake Bonneville, Utah (Oviatt, 2015) have a higher YD shoreline, indicating a significant environmental change during even this brief cold-wet climate. Wetland deposits from the YD, associated with Paleoindian sites during the YD, suggest that early groups used these wetland environments for subsistence (Duke et al., 2022).

Initially, the early Holocene was cooler and wetter than today, but a gradual warming trend reached present-day conditions by at least 9000 cal BP. Drier conditions increased into the middle Holocene (Asmerom et al., 2010; Polyak et al., 2004). Pluvial lake levels lowered during the early Holocene (Allen and Anderson, 2000; Hall, 2017; Oster et al., 2019). The number of springs and the formation of thick, organic deposits (black mats) decreased at certain sites as the water table fell. The current vegetation models show that sagebrush steppe diminished after 11,000 cal BP, moving northward to higher ground and eventually being replaced by shortgrass prairie and desert grassland (Hall and Penner, 2013). The thick soils that formed across the late Pleistocene landscape began to erode due to drought, decreased vegetation cover, and possibly the effects of fire (Hall and Penner, 2013). The prehistoric record shifts from Paleoindian to Archaic occupation with the increasingly arid environment (Hall, 2017; Holliday and Meltzer, 2010). Research indicates various environmental responses to regional aridity beginning ~7000 BP. Soils across the Southwest became more dry and more alkaline (Hall, 2017), and alluvial sedimentary records are sparse due to the lack of sediment deposition and arroyo formation. Salt-tolerant shrubs, such as saltbush, became more abundant, invading desert grasslands (Davis and Shafer, 1992; Hall and Penner, 2013). Much of the Archaic hunter-gather period is documented under these types of climatic conditions.

During the Late-Holocene, climate proxies indicate a shift to generally wetter and cooler conditions. Broad regional studies tend to date this “Late-Holocene Wet Period” slightly earlier, c. 4500–1000 cal BP, perhaps related to increased precipitation from the North American Monsoon pattern during this time (Asmerom et al., 2007; Hall, 2017). Such increased precipitation would have led to greater runoff into the valley alluvial systems, as well as higher water tables. This period also coincides with the Late Archaic human occupation of the region and the eventual transition to the early farming practices that characterize the Basketmaker traditions (Hall and Penner, 2013; Vierra, 2017). Indeed, speleothem records and cultigen data from a rock shelter from the Guadalupe and Sacramento Mountains in southern New Mexico suggest that the earliest evidence of maize cultivation is concurrent with the onset of the “Late-Holocene Wet Period” (Polyak et al., 2001; Tagg, 1996).

The Late Archaic Landscape of Tesuque Creek and its Environs

The Late Archaic Period coincides with the “Late-Holocene Wet Period” which is characterized by increased affective moisture in the region. In northern New Mexico the Late Archaic is generally dated from c. 3500 to 1500 BP and is typically defined by the seasonal aggregation of larger communities (possibly 30–50 individuals; Irwin-Williams et al., 1973; Vierra and Foxx, 2009), the introduction of maize (McBrinn and Vierra, 2017; Vierra and Ford, 2006), and an increased diversity of projectile points in use (Vierra, 2013). Although cultivated maize was likely introduced to the northern Rio Grande at around 3000 BP, it appears only as a supplement to broader foraging patterns in this area until c. 1250 BP. There is no evidence for local village aggregation during this time (McBrinn and Vierra, 2017; Vierra, 2018). Late Archaic foragers in this area continued to practice mobile lifeways between the uplands and the lowlands to take advantage of the abundant seasonal diversity of plant food sources across these rapidly changing ecological niches. Late Archaic sites show already sophisticated management and low-level cultivation of various wild taxa in both upland and lowland niches, perhaps most significantly at Jemez Cave (Vierra, 2013) and the Chama Alcove site (McBrinn and Vierra, 2017; Vierra and Ford, 2006). Riverine settings also appear to have been used for winter campsites, and researchers note that water-table fed maize fit easily into a diverse subsistence base in niches that were already being cultivated with wild seeds (Ford, 2013; McBrinn and Vierra, 2017; Moore, 2018; Post, 2013; Vierra and Foxx, 2009).

Researchers are identifying Late Archaic sites within the Tewa Basin with increasing frequency (Moore, 2018; Vierra, 2018, 2020). These finds suggest potentially far greater Archaic human impact on the landscapes of the northern Rio Grande region than was previously understood. A number of the recently identified sites are close to the Tesuque Creek alluvial section presented here. Of particular significance to this study, LA 6562 was exposed in the east edge of the U.S. 84/285 road cut within 1 km of our study site and contained Late Archaic horizons below a Late Developmental Period site component (Moore et al., 2003). Skinner et al. (1980) also identified Late Archaic features within another exposed section of the Tesuque Creek arroyo are located approximately 12 km South-Southeast of our study site, and from a site near Nambé Falls, c. 10 km northeast of our study site (See also Moore, 2018). The earliest documented Late Archaic material from Tesuque Creek itself was identified by Miller and Wendorf (1958) in hearths within an exposed cutbank of Tesuque Creek very similar to our own study site, 14 km to the southeast.

Deeply buried Archaic occupations are present throughout these Rio Grande tributary valley settings surrounding our study site (Moore, 2018). Moore extensively analyzed a multi-component Late Archaic site within LA111333, at the juncture of U.S. 84 and 591 (the “Tesuque Y”) along Tesuque Creek on Tesuque Pueblo land (Moore, 2008, 2018, personal communication). This site returned calibrated dates from both cultural features and paleosols, ranging from 3390 to 2970 cal BP. In the Rio Oso Valley in the northern Jemez Mountains, Hall and Periman recorded a three-meter-deep sequence of paleosols that resembles those recorded at Tesuque by Moore (Hall and Periman, 2007). Other recent identifications include LA 51912, a Late Archaic site near San Ildefonso Pueblo, c. 9 km east of our study site, which returned dates suggesting an occupation between 2490 ± 70 and 1840 ± 80 cal BP (Lent, 1991). The site is interpreted as a longer-occupation cold weather camp based on floral remains (Lent, 1991: 64–65). Nearby, LA 65006 consisted of a stratified Archaic site dated between ca. 3379 and 2750 cal BP (Moore, 2001).

Survey results also confirm that there was abundant Late Archaic activity in the study region. Sixty-four sites with Archaic components were documented in the southwest Tewa Basin (Post, 2002). Survey and testing along NM 502 in Los Alamos Canyon at the western edge of the Tewa Basin documented at least six Archaic site components (Moore, 2001).

Cumulatively, this archeological record indicates a well traversed and potentially managed Late Archaic landscape in the Tewa Basin and immediately around Tesuque Creek. Evidence for long- and short-term seasonal camps between the uplands and lowlands are increasingly common finds as researchers improve their methods for identifying them. Around these sites, in the immediate and adjacent areas of the northern Rio Grande, the paleobotanical remains give evidence for both seasonality of occupations and for low-level cultivation of wild taxa and occasionally domesticated maize (Vierra, 2020). Vierra (2017) has suggested that maize agriculture was not fully adopted in this area until much later than elsewhere in New Mexico because of the abundance of high value plant and animal resources accessible between closely linked upland and lowland niches. It is possible that patterns of movement and seasonal use of different vegetation communities were valuable risk-buffering strategies for foragers. Multiple, diverse sources of nutrition and the social and cultural activities rooted in seasonal mobility protected against food insecurity if one or more resources became temporarily unavailable. Seen in this light, a transition to sedentary agriculture was perhaps perceived by Archaic peoples as posing significant risks. Land management, including burning, allowed for Archaic populations to sustain their mobile and semi-mobile lifeways, which may have been perceived as less risky than a transition to sedentism.

The Late Archaic managed cultural landscape provides a human context for our case study of an exposed alluvial section in the Tesuque Creek watershed. For this study, we sought to determine how the landscape developed and how its vegetation communities changed over time in this specific area, with particular attention to the role that fire played as Archaic sites began to appear in the surrounding area. We consider the role of fire both as a risk to foragers and as a tool for maintaining their resilience in the face of other perceived risks.

Study site

The headwater of the Rio Tesuque is located along the western slopes of the Sangre de Cristo mountains. As it drains into the Pojoaque River (a tributary of the Rio Grande), it is flanked by alluvial terraces and incised arroyos at lower elevations. The modern channel of Tesuque Creek generally supports an ephemeral stream with some limited perennial flow throughout most of the year. Today, this series of arroyos and drainages is deeply incised by strong flash flooding after storms, but the flow can no longer reach bankfull conditions or develop a floodplain in our study area.

The alluvial section presented in this study, Tesuque Creek East (TCE) (35°51′29.2″N 106°0′02.9″W; elev. 1860 masl) is an approximately 5-m deep sediment sequence exposed by recent erosion along Tesuque Creek (Rio Tesuque). The TCE Section is located on the Pojoaque Pueblo land grant, in the southernmost part of the Española Basin, along the lower catchment area of Tesuque Creek as it passes Pojoaque Pueblo and the ancestral Tewa village of Cuyumungué (Figure 1).

The modern channel is about ~35 m across, with steep cut banks whose exposed sequences represent the depositional history of a restricted area of the ancient floodplain of Tesuque Creek. The modern vegetation is composed of semi-arid sagebrush scrubland with a mosaic of cacti, and forbs, as well as patches of bare, rocky soil. Today, precipitation in the study area averages around 350 mm per annum, with the heaviest rainfall occurring during the summer months, and most of the streamflow driven by spring and summer snowpack runoff.

Methods

Field methods

Damick identified the Tesuque Creek East Section (TCE) during a pedestrian survey in Summer 2018 while working with the Cuyumungué Project, a collaboration with Scott Ortman (University of Colorado at Boulder), Bruce Bernstein (Tribal Heritage Preservation Officer at Pojoaque Pueblo), and Pojoaque Pueblo. In the following year (2019), Damick and Rosen returned to the locality to describe, sample, and interpret the in-situ sediments from the exposed section. Our goal was to assess human impact on vegetation patterns within this alluvial floodplain. We collected 30 sediment samples sealed in plastic airtight bags and shipped them to the Environmental Archaeology Lab in the Anthropology Department at the University of Texas at Austin for laboratory analyses.

Lab methods

At the University of Texas at Austin, Damick conducted phytolith analyses to reconstruct vegetation communities throughout the sediment profile, and sediment analyses to help us understand the sediment depositional environment and relate that to fire history and environmental change. Krause ran XRF determinations to aid in the reconstruction of natural pedogenic processes and help identify the source of the sediments. We also sent the samples for radiocarbon dating which revealed that the middle to upper portions of the sections was deposited from about 3700 to 2600 cal BP. Lab methods are described briefly here and in detail in the Supplemental Information.

Phytolith methods

Phytoliths are silt-sized particles of opaline silica that are formed in and around the intracellular and extracellular spaces in plant bodies when those plants take up soluble silica from the ground water. The deposited silica creates durable inorganic silica “casts” of the plants’ cells. This process is genetically and environmentally determined. Grasses, sedges and palms (monocotyledons) readily produce phytoliths, often distinctive to plant family, genus and more rarely, species. Woody trees and other herbaceous dicots also produce phytoliths, although far fewer and with more irregular forms.

Phytoliths were extracted from each sample according to Rosen’s lab protocol (Rosen, 1999a, 1999b; see details in Supplemental Information), which employs a series of techniques to remove carbonates, clays and organics, and separating the phytoliths from other mineral content in the sediment. The phytolith slides were counted at 400× magnification using a transmitted-light microscope (Nikon Eclipse E200). Morphotype descriptions adhere as closely as possible to the International Code for Phytolith Nomenclature 2.0 (Neumann et al., 2019). The concentrations per sample (number per gm sediment) for each phytolith type were calculated using equations described in detail in the Supplemental Information (Ramsey and Rosen, 2016; Rosen, 1999b).

Phytolith analysis allowed us assess changes in vegetation communities and land stability over time in relation to the sedimentological data, particularly relative to the burn events. Applying phytolith data to fire ecology research is a novel technique for this region and requires far more data from different sediment profiles before it can be used to model past burning patterns in different times and places. Therefore, these results are presented preliminarily, and to demonstrate the importance of advancing phytolith studies alongside sediment analyses in future paleo-fire research to gather sufficient data to support or modify these initial observations.

XRF methods

Subsamples were analyzed using a handheld (X-ray fluorescence) XRF analyzer (reported in weight percent) to measure the concentrations for elements within the sediment profile, which provides a better understanding of natural pedogenic processes and autochthonous and allochthonous inputs to the system. Samples were homogenized, powdered, and sieved through a 150 μm mesh screen, and then placed into cells that were then run in triplicates using the Olympus handheld XRF in the Soils and Geoarchaeology laboratory at the University of Texas at Austin. There are limitations to consider when using pXRF instruments. For example, the lighter elements within the spectrum detectable by pXRF (such as Mg or P) have higher detection limits in comparison to heavier elements, and the analyzer can only detect their presence when their concentration is relatively high, and elements lighter than Mg are grouped as light elements (Coronel et al., 2014; Newlander et al., 2015; Šmejda et al., 2018). We use XRF here as a secondary proxy to better understand natural changes within the system that could change the elemental makeup of the sediment at different times throughout the Holocene. Such changes can reflect shifts from stability to instability within the system, or anthropogenic changes such as agricultural use or fire maintenance of landscapes. For a detailed review of the use of sediment geochemistry in archeology, see Holcomb and Karkanas (2019) as well as Bintliff and Degryse (2022).

Results

Results of sediment analyses, including sample depths, grain size analysis, organic content from Loss on Ignition, and XRF results are illustrated in Figure 2. Histograms and diagrams representing the relevant phytolith patterns under discussion can be seen in Figure 3; complete phytolith concentration data can be found in the Supplemental Information.

Figure 2.

Results of the analysis of sediment samples for grain size, organic content, and XRF weight percent.

Figure 3.

Histograms and diagrams of relevant phytolith analyses, including (a) ecozone groups, (b) comparisons of overall monocots to dicots, (c) comparisons of elongate dendritics to elongate psilates, and (d) comparison of spheroid echinates to bulliform flabellates.

Sediment descriptions

Sediment descriptions are based on field observation as well as grain size analysis results. The lowermost portion of the TCE section consists primarily of sandy silts with some laminated bedding, indicating deposits from an aggrading floodplain of a primarily perennial stream, with some overbank levee deposition (see Figure 2 for depths and grain size analysis quantifications). In the upper 3 m of the section there are sets of organic-rich clayey silts and silty clays indicating ponding and seasonal marsh deposits on these ancient floodplains. We identified eight distinct units within the profile which reflect changes in the depositional regime, including marshy paleosol deposits (Units 1, 3, and 5), near-channel overbank sediments (Units 2, 4, 6, and 8), and one unit with sets of both types of depositional microenvironments (Unit 7). Small well-sorted and rounded channel gravels appear at intervals throughout the section indicating periods of well-sustained perennial flow. Throughout the section there are laterally extensive lenses of ash and charcoal indicating burned surfaces. These burned lenses are commonly associated with the marshy facies of these deposits, which often contain the imprints of abandoned root channels (Units 1, 3, 5, and more rarely 7). One burn level appears in Unit 6, within the bedded silt deposits.

XRF results

XRF results illustrate the bulk geochemistry within the different stratigraphic units in the section. The bulk elemental composition of the sediments within our study area have resulted from complex pedogenic processes controlled by the local natural geology, aeolian inputs and local geomorphology processes such as soil erosion, deposition, and weathering, detectable within the limits of the pXRF. Within the analyzed samples, light elements represent the bulk of the elemental composition at concentrations greater than 70%. Al, Si, Ca, and Fe are all present within the sediment at different concentrations throughout the profile between 2% and 15% (Figure 2). All the other elements detected within these samples were generally detected in amounts of less than 2%, many less by several orders of magnitude. Of note, Ca, Al, and Fe concentrations generally are correlated in the column, and Si is generally inversely correlated to the others. This may be due to differential erosion, leaching carbonate, and different sedimentation rates over time within different environmental regimes. The XRF provided a clearer understanding of the geochemical makeup of the sediments within the study area. The shifts in the elemental profile across the different units illustrate pedogenic processes and enhance our understanding of landscape processes during periods of burning. Although none of the samples yielded a geochemical signature that could be directly linked to human fire on the landscape, the elemental changes in association with phytolith concentrations (discussed below), support the interpretation of significant changes to the environment and land stability over time. This lends important context to the conditions before, during, an after the burning events.

Phytolith results

The full results of all phytolith morphotypes counted are provided in the Supporting Information as absolute numbers of phytoliths per gram of sediment, also referred to as phytolith concentrations. Only single-cell phytoliths were counted; there were a very small number of multi-cell structures. Single-cell phytoliths generally do not indicate specific genera, but are proxies for general vegetation types (i.e. grasses, shrubs, trees, etc), broader vegetation communities (i.e. marshlands, grass/shrubland, forests), and environmental conditions (i.e. seasonally or perennially wet, arid, sometimes warm/cool conditions). Therefore, it is more meaningful to discuss the results as an assemblage, rather than summarizing results for individual taxa as is common for other paleobotanical proxies.

Generally, we found substantial quantities of phytoliths in all samples deriving from all depositional layers with high organic content, including those formed by ponding, well developed wetland deposits, and the burn lenses. The massive alluvial silt layers and channel gravels contained relatively few phytoliths. This is expected, as these layers reflect more rapid depositional environments with few or no stable episodes for sustained plant growth. Throughout the section, the phytolith concentrations for silica bodies that form within the woody parts of dicotyledonous trees and shrubs (dicots) are close to or higher than those for monocotyledons (monocots) (grasses, sedges, etc.). We also found large quantities of calcium oxalate raphids from succulents in several the samples, particularly dominant in the lower units. In the lowermost silty units this may reflect the dominance of perennials versus annuals on more rapidly aggrading surfaces. Other implications of these results as assemblages are presented in the Discussion below.

Analysis and discussion

The units with ashy deposits suggest intensified human land use, including potential low-level cultivation and seasonal burning by Late Archaic populations inhabiting the Tesuque Creek area between c. 4600 and c. 2600 cal BP. This land use coincides with changes in the depositional regime and vegetation community of the restricted riparian floodplain represented by the TCE section. In the mid-late fourth millennium BP, the environment around Tesuque Creek transitioned from an arid or semi-arid landscape dominated by cactus and other succulents to a more mesic landscape defined by a slowly aggrading floodplain hosting a much more dense, mature riparian vegetation community. During this time, throughout Units 7 and 6, only rare, minor indications of burning are recorded. In the late fourth millennium BP, the alluvial regime shifted and a series of marshes and pond levels appeared, frequently capped by distinct, charcoal-rich burn horizons. These conditions formed the horizons of our Units 5–3, which coincided temporally with the appearance of Late Archaic encampments in the surrounding area. The vegetation also changed at this time to a riparian wetland dominated by abundant monocots (sedges, reeds, some other grasses) and dense shrubs. The non-endemic but cold and fire-hardy palmetto (Sabal sp.) is also introduced at this time. Significantly, there is no phytolith evidence for grass inflorescences throughout these Units. After c. 2600 BP, at the top of the section, the results indicated that conditions returned to slow floodplain aggradation and the dominance of a mature, dicot-rich riparian vegetation community with similarities to Unit 6. These data allow us to track the developmental history of this part of the Tesuque riparian floodplain and suggest direct associations with increased human land use in this locality during the Late Archaic, including low-level cultivation and seasonal burning that interrupted and changed the depositional environments recorded in the section.

Depositional environments and landscape change over time

The interpretation of Late Archaic human land use, including burning, in the vicinity of the TCE section is supported by multiple lines of evidence but is most clearly revealed in the phytolith data. To gain insight into landscape development at TCE, we assessed the phytolith data in two steps. First, we analyzed the phytoliths throughout the section by eco-zone vegetation community groups. For this, we grouped phytoliths from the TCE section into representative ecozones according to the types of vegetation communities for which different morphotypes act as proxies (Table 1; Figure 3a). The three main categories (woodland, wetland, and grassland) are grouped generally following the method defined by Ramsey and Rosen (2016), excluding the multi-cell types they used because multi-cells do not appear in representative numbers in samples from the TCE section. Wetland morphotypes include bulliform flabellates, rods, and both culm and achene-type cones after Murungi and Bamford (2020). For the final group (dryland), calcium oxalate “raphids” from cactus and succulants are the only proxies used. The ecozone analysis illustrates which vegetation assemblages were most represented at different points in the depositional history of the section.

Table 1.

Phytolith morphotypes used to model ecozones, modified from Ramsey and Rosen (2016).

Ecozone type	Phytolith morphotype used as proxies
Woodland	Platelets/tabular forms, polyhedral verrucates, blocks, spheroid psilates, tracheids
Wetland	Cyperaceae “cones” (elongate conical types, short acicular cones and cone bases) and rods (elongate cylinders), bulliform flabellates (reeds)
Grassland	Elongate dendritics, papillae, all short cells
Dryland	Raphids and druses (calcium oxalates from succulents/cacti)

Units 8–6: Transition from dryland to riparian vegetation on a slowly aggrading floodplain

The lowest portion of the TCE Section, Units 8–6, is made up of massive, fine-grained sandy silt deposited by overbank flow from a slow, consistent stream. These sediments represent the slowly aggrading floodplain near the ancient creek channel. The vegetation community is made up of generally similar taxa producing low concentrations of phytoliths in all ecozone categories. Significantly, however, these units contain the highest concentration levels for calcium oxalate raphids from succulents and/or cacti, and relatively larger numbers of amorphous silica aggregates from shrubs. This suggests an overall arid scrub-shrub environment during this early long, slow aggradation period.

There are only two burn lenses across all of Units 8–6. These lenses appear as thin light gray ashy bands within the fine sediments, suggesting low intensity burns at very low frequencies. Within Units 6 and 7, the phytoliths show the presence of both slower-growing dicots and the full growth cycle of monocot taxa, including elongate dendritics from the inflorescences of grasses. The low intensity burns and the presence of grass inflorescences in these lower Units prompted us to investigate the possibility for a previously stable soil horizon at this time, which may no longer be macroscopically visible due to taphonomic processes (Fredlund and Tieszen, 1997; Golyeva and Svirida, 2017). The elemental data supports this possibility, as there is also an increase in Sr, Ca, and Mn and a decrease in Si and Ti in these Units which could represent accumulation and leaching of minerals from a previous topsoil into lower horizons. These proxies tentatively suggest that although the only visible terrestrial soil in the section is that in Unit 1, the modern topsoil horizon, an additional buried paleosol may have existed in the lower part of the section as well. If that is the case, it gives us a possible indication for a more stable landscape that preceded intensified land management by Late Archaic communities in the later units, where burning, ponding, and rapid sedimentation increases are indicated (i.e. Unit 5 and above).

Units 5–3: Mature riparian vegetation community with frequent burns

In upper Unit 6 and lower Unit 5, the depositional environment changed. Units 5–3 show much more rapid deposition of sediment horizons and greater variability in overall sediment composition (Figure 2). Gleyed silty clay deposits with blocky structure, indicating relatively stable ponding or marshy facies (Units 5 and 3) alternate with massive sandy silt deposits (Units 4 and 2). The elemental data supports the interpretation that this was a much more rapidly changing and generally wetter environment. In Unit 3 in particular, an increase of Mn and Fe within the bulk geochemistry of the sediment could indicate that the sediment underwent wet–dry conditions indicative of a more wetland environment. These indicate alternating phases between wetter and dryer conditions, which correlates with climate proxies for the Late-Holocene Wet Period. From the bottom of Unit 5 to the top of Unit 3, we identified at least 7 distinct layers of dark black charred horizons with macroscopic charcoal, all of which lay directly on top of organic and clay rich marshy facies. We can link these to localized burn events either in-situ or deriving from nearby upstream in the watershed. They appear to be more intense burns occurring at a higher frequency than those in Units 6 and 7.

The phytolith assemblage indicates that the vegetation community transitioned from arid scrub-shrub to a dense riparian ecozone in late Unit 6 or early Unit 5. When left to grow to maturity, the overall riparian forest and shrubland environment of the northern Rio Grande should be dominated by woody dicots, similar to the communities of juniper, cottonwood, willow, and a number of woody shrubs including seepwillows (Baccharis sp.), and thinleaf alders (Alnus sp.) that still grow around rivers and drainage systems throughout northern New Mexico. In fact, forested and scrub-shrub vegetation make up the greatest percentage of native riparian vegetation communities recorded across northern New Mexico by the New Mexico Natural Heritage Program at the University of New Mexico (Muldavin et al., 2000).

Within the TCE section, the concentrations of phytoliths from dicots representing woody vegetation are generally high throughout Units 5–2. They reach extremely high levels in relation to monocots, however, only in the upper (Units 1–2) and lower (Unit 6) of the section (Figure 3b). Given the production discrepancy between monocots and dicots, this is a significant indicator for a dicot-dominated environment along the river during these phases. There are also high concentrations in these Units of phytolith proxies for wetland reeds, grasses and forbs, such as horsetail (Equisetum sp.), reeds (Phragmites sp.), and bulrush sedges (Scirpoides sp.). These are consistent with the known taxa in the riverine environments of northern New Mexico.

The phytolith proxies for water-loving taxa give good indications for relative water availability in the immediate area. For instance, when the spheroid echinate phytoliths (representing palmettos (Damick et al., 2021) and bulliform flabellates (typical of reed grasses) are graphed together, it is clear that their relative concentration levels increase and decrease during the same depositional sequences throughout all Units (Figure 3d). Increases in density of these phytolith types correspond with the marshy sediments indicating wetter depositional environments. As streamflow is related most strongly to runoff from the mountains, which is itself driven by cool season rain and snowfall, the relative wetness of these lowland depositional environments may lend some precision to modeling moisture availability over time in future studies. Cumulatively, the sediment and phytolith data for Units 5–3 indicate a robust riparian ecozone with dense trees, shrubs, and wetland grasses.

Units 2–1: Return to a slowly aggrading floodplain and recent soil development

Unit 2 returned to the slowly aggrading floodplain conditions seen in the earlier Units 6 and 7, made up of massive, fine sandy silts and overall low organic and phytolith content. Unit 1 contains a developing topsoil that reflects the modern vegetation and geomorphic regime, and includes an ashy lens likely related to modern fire on the landscape.

Late-Holocene burning in relation to human land use

In order to link the burn events in the TCE section (as described above) to indications for human land use, we compared the combined sediment and phytolith data from the burn facies and related deposits in the different depositional environments. We found that frequent wet-season burn events on low-flammability wetland vegetation communities occurred within the time frame represented by the middle of the alluvial section, Units 5–3. This time frame, dated between c. 3600 and 2600 cal BP, coincides with the dates of nearby Late Archaic camp sites, as currently available in the archeological record. As discussed above, macrofloral data from many of these sites indicate fall or winter occupation (Moore, 2018; Skinner et al., 1980). Such seasonal occupation patterns would predict that land management, including possible burning, should occur during those cooler seasons as well.

Importantly, the fall and winter cooler seasons are outside most grass flowering seasons. We can use the relationship between elongate psilate and elongate dendritic phytolith morphotypes in burn levels to interpret the season when burning occurred (Rosen, 2008). Elongate morphotypes appear in all monocots, particularly Gramineae including wetland and non-wetland grasses. Elongate psilates form within leaves and stems of many different taxa. Elongate dendritics (previously called dendritic long cells) are formed within the inflorescence of Gramineae, and therefore can act as proxies for the flowering seasons where these vegetation communities are present (usually late Spring and Summer).

In the TCE section, elongate dendritics are absent from samples from burn lenses or the organic horizons directly below them. Elongate dendritics occur only at the top of the section, in the most recent paleosol in Unit 1, and in the bottom of the section, in the massive silts in the lower part of Unit 6 and the upper part of Unit 7 (Figure 3c). The middle Units (5–3), which contain frequent burn lenses on top of organic-rich marsh surface levels and elemental indices for buried wetlands, contained no elongate dendritics from inflorescences. This absence indicates that the highest frequency of burn events occurred during seasons when flowering plants, likely grasses and forbs, were still dormant, or after these plants had sprouted but before they flowered. In other words, for the entire period represented by Units 5–3, the monocots represented by our samples never appear to have reached the flowering stage in this locality.

In the lower elevations of northern New Mexico, the flowering seasons of grasses generally range between late Spring and early Fall (Allred and Ivey, 2012). Since the burn facies themselves contain no elongate dendritics, the burns are likely to have occurred between the winter and early Spring, and so would coincide with the longer-term winter camps along riverine systems, common to the Late Archaic in the vicinity. In northern New Mexico, the natural wildfire season is typically most intense during Summer and early Fall after the Summer monsoons. The numerous out-of-season burns represented in our TCE sediment profile suggest humans were most likely involved in igniting those fires (Margolis et al., 2017).

The absence of elongate dendritic phytoliths from inflorescences only in the middle of the sediment section can be a powerful indicator of seasonality. Since we base part of our argument for off-season burns by humans on these phytolith morphotypes, it was important to determine if there were a taphonomic explanation for this pattern. In some experimental cases, inflorescence phytoliths appear generally less stable than leaf/stem phytoliths, and this instability can be enhanced by burning (Cabanes et al., 2011). However, these studies also showed other morphotypes (such as rondels and elongate echinates) exhibited similar instability as the dendritic forms. However, those morphotypes are unaffected in our samples, suggesting that burning was not the reason for fewer elongate dendritics. Additionally, if burning were causing the degradation of elongate dendritics differentially, we would expect elongate dendritics to appear in the non-burned levels of Units 5 and 2 and to be absent only in the burn lenses. However, there are no dendritic phytoliths in either burned or unburned horizons between Units 5 and 2. Therefore, while relative instability may account for some loss, it cannot explain the complete absence of elongate dendritics in Units 5–2 when they are present in Units 1, 6, and 7. Grazing animals present an additional potential impact on vegetation communities in many areas, that could produce differences in the phytolith assemblages (e.g. Painter et al., 1993; Wendt et al., 2022). However, as the TCE section represents a restricted riparian floodplain and not an open grassland where large grazing herds would pass through, the impact would likely be minimal.

We suggest that the burn events in these units provide the explanation for the pattern of elongate dendritics in this section: the vegetation growing in the wetlands that developed on the floodplain (wetland grasses, reeds, sedges, or forbs) were frequently killed off by burning before they reached the flowering stage. Again, this corresponds with the archeological data for the appearance of cool-season encampments of Late Archaic populations in this immediate area. Data previously analyzed from the TCE section also suggested management of this area by Archaic populations through the introduction of Sabal palmettos, during the same time period in which the frequent burn lenses appear. Palmettos are not native to this region and were likely introduced by early Archaic foragers as they provided an important alternative starchy food source and important fibrous weaving material. These resources were previously provided by succulents, which disappeared when palmettos were introduced (Figure 3a; see Damick et al., 2021). Since we published the evidence for low-level cultivation of Sabal palmetto elsewhere, we will not repeat it in detail here. For this discussion, it is important to note that the appearance of palmettos corresponds with the intensification of burn levels, the disappearance of grass inflorescences, and the evidence for cool-season Late Archaic encampments in the area. In order to maintain a habitat for water-loving palmettos, the use of fire would have been beneficial to control the growth of vegetation communities that might crowd out competitors. Palmettos are very fire-resistant as well as cold-tolerant, and would have been well suited to this management technique (Haynes and McLaughlin, 2000; Zona, 1990).

All of these lines of evidence converge to give the picture of significant human management introduced into the TCE landscape as Late Archaic people appear in the area. The interpretation of human production of at least some of the burn levels is supported by multiple lines of evidence. First, the burns in Units 5–4 correspond with (a) dates for Late Archaic settlements within walking distance of the section site (as discussed above), (b) overall increase in intensity and frequency of burn events; (c) evidence in the phytolith data for the introduction of non-endemic palmetto phytoliths (Damick et al., 2021), and (d) phytolith indices for cool-season burning and burning of low-flammability wetland vegetation communities.

If similar patterns in the relationship between burn events and the presence or absence of inflorescence proxies can be seen to repeat in other sections, this observation should provide evidence that aligns with the ethnographic literature and archeological evidence for patterns of anthropogenic burning during moist seasons in other areas as well.

Implications of this study for Paleo-fire and risk management research

Much recent work in ancient fire studies demonstrates that small-scale, localized patch-burning took place far more often in the past than large-scale, basin-wide anthropogenic burning (Bird et al., 2012, 2016; Bowman et al., 2020; Swetnam et al., 2016; Lynch et al., 2018). Small-scale patch-burning serves to mitigate both the risk of having either too much fire, when wildfires get out of control, and the risk of not having enough fuel to produce fire for a range of burning purposes including ceremonial, communicative, and for resource-gathering. For all these burning needs, the scale and timing of fire must be controlled. Burned patches interrupt the spread of lightning-ignited fires during dry seasons, and thus modify their impact and their pattern in the geo-records. Patch-burning also ensures that areas with sufficient fuel for burning are always available, when burning is called for by the community.

In order to understand fire regimes more fully, it’s also important to compare these small-scale seasonal cycles of moist/dry episodes to larger-scale cycles of climate change. Armour (2002); Hall and Penner (2013) and Hall (2017) conducted paleoclimatic analyses on speleothem and pollen data from this region. These analyses point to short, intermittent warm/dry episodes within the otherwise cool/wet Late-Holocene. The longer cool/wet conditions would naturally increase the density of vegetation and therefore regenerate fuel. The dry snaps would have increased natural fire ignitions. Since the cool/wet phases were the dominant ones throughout this period leading to denser stands of vegetation, fires during the droughts could have been catastrophic in scale in an unmanaged landscape. Patch-burning would have been a logical risk management strategy given the frequency of camp sites in this area during the Late Archaic.

It is also essential to remember that Late Archaic populations in this region were mobile and semi-mobile throughout the year on a strongly seasonal basis. This means that they had fixed places in the landscape that they moved between and returned to annually or over the course of several years. They would be stationary at a given location for one or more seasons depending on factors like resource availability, other groups in the area, and cultural or ceremonial needs (McBrinn, 2010; Vierra, 2013). They needed to have long term risk management strategies in place to remain resilient in the face of changing climates that impacted those factors. This was particularly true during episodes of aridification that would increase natural burning on landscapes beyond their campsites. Archaic peoples may have conducted strategic burning practices in different parts of the landscape beyond their habitats knowing that they would return to those localities at a different season. This strategy would maintain rhythms of seasonality and scales of resilient habitat development that provide a wider variety of options in an unpredictable environment. Piecing together these dynamics and strategies over time will help us understand how Archaic peoples managed both their need for fire, and their need to control it, in relation to the complex eco-dynamics of the Mid-Late Holocene world in which they lived (Kimmerer and Lake, 2001).

The evidence from the TCE Section contributes to previous work on fire risk management by Archaic populations in Northern New Mexico. Vierra and Ford, for instance, hypothesize that when lightning-ignited fires were more frequent during periods of drought, upland settings such as Jemez Cave may have served as critical sources for cheno-ams that would occupy recently burned areas in the late summer (Vierra and Ford, 2007: 120). Around Tesuque Creek, in the lowland riparian context, rhizomes from wetland grasses, which populated the area after the burns, may have served a similar role, alongside the introduced palmetto resources. These plants would have provided excellent, quickly available resources for local Archaic communities. There is also significant work on vegetation community succession after burns for animal resource management (Whelan, 1995.; Foxx, 1996). This research shows that after a fire, shrubs sprout and plant species that have roots or underground stems regenerate quickly, thus attracting large game such as elk and deer. Recently burned areas not only protected those zones from further immediate fire infiltration, but they provided for ecosystem regeneration that was useful for, and undoubtedly well understood by, human occupants of the area.

Ethnographic observations from all over the world point to seasonal patterns of landscape burning that may be significant to the New Mexico case study, when considered critically within local contexts. Broadly speaking, the ethnographic evidence suggests that hunter-gatherer populations in many parts of the world seem to conduct landscape burning activities more frequently during moister, cooler times of the year, and use those burned patches to buffer and control for natural fires that occur in warmer, dryer times of year (Kimmerer and Lake, 2001; Laris et al., 2018; Murphy et al., 2013). In tropical regions, anthropogenic burning therefore tends to take place in the early dry season, whereas in temperate regions it will occur more frequently in both the spring and autumn (Bilbao et al., 2010; Kimmerer and Lake, 2001; Laris, 2005; Laris et al., 2018; Murphy et al., 2013; Russell-Smith et al., 2003). Lightning-driven ignitions, on the other hand, are more frequent in the late dry seasons of a given region. Extended dry seasons due to climate change will increase the frequency of lightning-driven ignitions but will not change the overall pattern of burning (Fill et al., 2019). In fact, in cases where modern mega-fires exceed the expectation for ignitions based on number of lightning strikes, they are almost exclusively related to human ignitions and increased winds within human-modified landscapes, as seen in Southern California (Bendix and Hartnett, 2018).

Ethnographers have also documented the management of wild plant resources via strategic burning among the Zuni (Ford, 1985, 1999), the Athapaskan (Buskirk, 1986), and Piman (Dobyns, 1981) groups in the North American Southwest. These groups use fire to stimulate the growth of wild grasses (Ford, 1999), and other select species of plants that thrive in post-fire settings, such as cheno-ams and leafy greens (Adams, 2004; Foxx, 1996). Researchers using archeological data infer such fire use in the past, as well. For instance, Sullivan and Forste (2014) suggest that in the Grand Canyon area of Arizona, Pueblo II period populations used fire to increase local wild plant resources for economic gain, due to the poor potential for cultivating maize in that region. In this model, the use of fire removes competition for seed-producing annuals such as edible cheno-ams and may even additionally enhance the yield of piñon crops (Roos et al., 2010; Sullivan and Forste, 2014; Sullivan and Mink, 2018). Bohrer has suggested similar strategies for burning select shrub fields to promote the long, straight shoots necessary for making stick figurines and baskets (Bohrer, 1991, 1992), and Karuk scholar Carolyn Smith has documented the management of riverine burning to control for basket weaving materials in California, as well (Smith, 2016).

The sediment and phytolith evidence from the TCE section seems to suggest such risk management practices in the Late Archaic landscape around Tesuque Creek. First, we have significant indicators in the geobotanical record at TCE for repeated burning of open rangeland and wetlands (some of the least flammable vegetation communities). Second, these burns seem to occur during cooler months, outside the common flowering season for most plants in the area. Third, we see the interruption of the development of stable soils during the period dominated by burn sequences. Finally, this data correlates with the introduction of the non-autochthonous palmetto to the area. Given the density of Late Archaic occupation in the immediate environs of our study section, we include this portion of Tesuque Creek as part of the known, traveled, and utilized landscape of people during this period. Importantly, our evidence for the seasonality of the burns coincides with the archeological indicators of longer-term winter camp sites in the area, as well.

A note on distinguishing fire “types”

The bulk of this study is concerned with reconstructing the periodicity of burning and landscape development in relation to Late Archaic settlement patterns, and therefore often makes the distinction between “natural” (i.e. lightning-ignited) and “cultural” (i.e. intentionally human-ignited) fires. It is important to note, however, that the distinction between “natural” and “cultural” fire may not be, or may not have been nearly as meaningful in many societies as it is considered to be by much of academic science. In many cases in pre-contact New Mexico, people engaged with their landscape in terms that emphasized the cooperative labor of non-humans (Alberti and Fowles, 2018; Fowles and Alberti, 2021; Roos, 2017; Scherjon et al., 2015; White, 1932). We should likely consider anthropogenic burning in relation to lightning regimes, intended to complement the work of “natural” fire rather than entirely distinct from it.

Conclusions and future directions

The data presented here suggest intensified cool-season patch-burning by Late Archaic communities in the Tesuque Creek area between c. 3600 and c. 2600 cal BP, coinciding with changes to the depositional environment and vegetation communities in the area. In the beginning of the fourth millennium BP the environment around Tesuque Creek transitioned from an arid or semi-arid landscape dominated by cactus and other succulents to a more mesic landscape defined by a stable, slowly aggrading floodplain hosting a much more dense, mature riparian vegetation community with flowering monocots. Sometime around 3600 BP the alluvial regime shifted to a series of marshes and pond levels typically capped by burn lenses. These deposits indicate an interruption to former landscape conditions and frequent transitions between short wet and dry phases. This formed the lower horizons of our Unit 5 which coincided temporally with the early Late Archaic encampments in the surrounding area. The vegetation also changed simultaneously with the appearance of more abundant monocots (grasses, sedges, reeds) and with the non-endemic palmetto (Sabal sp.). Phytolith evidence for grass inflorescences was also absent during this time. After c. 2600 BP the top of the section indicated that slowly aggrading floodplain conditions returned, as did the dicot-dominated vegetation community seen in Unit 6. These data allow us to track the developmental history of this part of the Tesuque floodplain and suggest direct associations with increased human land use in the area in the Late Archaic, including low-level cultivation and seasonal burning that interrupted and changed the depositional environments recorded in the section.

We suggest that phytolith studies such as the one presented here offer a unique and important way forward for ancient fire studies. To improve robusticity of the data and a better understanding of patterns over time, we need to record and map many more deep-time geomorphological sequences within walking distance of known forager occupation sites. Fire studies could benefit from a record of sequential burning events over time and the documentation of concurrent changes in vegetation communities and other successional habitat indices through as much comparative proxy data as possible. While we do not yet fully understand all the variables and dynamics at play in the Tesuque Creek catchment, the phytolith evidence certainly demonstrates the potential for this type of study. Further analysis of phytolith data within geoarchaeological research in the North American West will help to clarify the patterns in vegetation communities over time in relation to their geomorphic environmental contexts. These patterns, in turn, may offer the means for detecting more precise localized expressions of human land use in the past, and the signatures of both human and lightning-ignited fires on the landscape. Only then can we begin to understand the ways in which past peoples perceived and managed the risks of fire itself, as well as how they used fire to manage other risks in their environments.

Supplemental Material

sj-docx-1-hol-10.1177_09596836221121771 – Supplemental material for Building resilient landscapes in a semi-arid watershed: Anthropogenic and natural burning histories in Late Holocene Tesuque Creek, northern New Mexico

Supplemental material, sj-docx-1-hol-10.1177_09596836221121771 for Building resilient landscapes in a semi-arid watershed: Anthropogenic and natural burning histories in Late Holocene Tesuque Creek, northern New Mexico by Alison Damick, Samantha Krause and Arlene Rosen in The Holocene

Footnotes

Acknowledgements

We are very grateful to The Pueblo of Pojoaque for their generous hospitality and collaboration for this research process. We would particularly like to thank Fermin Lopez, Adam Duran, and Gaby Montoya. We thank Scott Ortman, the members of the Cuyumungué Archaeological Project and Bruce Bernstein for inviting us to participate in their ongoing collaborative research project with the Pueblo, and to the University of Colorado-Boulder archeology field school of 2018 for all their hard work. This project could not have been completed without support from the Environmental Archaeology Lab at UT-Austin (PI Rosen) and the Soils and Geoarchaeology lab at UT-Austin (PIs Beach and Luzzadder-Beach) and funding from the National Science Foundation. We thank Kaitlyn Davis for help processing the samples. We also thank the anonymous reviewers for their time and effort to improve this paper.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Damick’s NSF SBE Postdoctoral Fellowship (Award number 1810563) for the research and authorship of this article.

ORCID iDs

Alison Damick

Arlene Rosen

Supplemental material

Supplemental material for this article is available online.

References

Adams

(2004) Anthropogenic ecology of the North American southwest. In: Minnis PE (ed.) People and Plants in Ancient Western North America. Washington, DC: Smithsonian Books, pp.167–204.

Alberti

Fowles

(2018) Ecologies of rock and art in northern New Mexico. In: Birch, Pilaar, Suzanne, E (ed.) Multispecies Archaeology. New York, NY: Routledge, pp.133–153.

Allen

Anderson

(2000) A continuous, high-resolution record of late Pleistocene climate variability from the Estancia basin, New Mexico. Geological Society of America Bulletin 112(9): 1444–1458.

Allred

Ivey

(2012) Flora Neomexicana. Raleigh, NC: Lulu.

Armour

(2002) 15 ky paleoclimatic and glacial record from northern New Mexico. Geology 30(8): 723–726.

Asmerom

Polyak

Burns

, et al. (2007) Solar forcing of Holocene climate: New insights from a speleothem record, southwestern United States. Geology 35(1): 1–4.

Asmerom

Polyak

Burns

(2010) Variable winter moisture in the southwestern United States linked to rapid glacial climate shifts. Nature Geoscience 3(2): 114–117.

Ballenger

JAM

Holliday

Kowler

, et al. (2011) Evidence for Younger Dryas global climate oscillation and human response in the American Southwest. Quaternary International 242(2): 502–519.

Bendix

Hartnett

(2018) Asynchronous lightning and Santa Ana winds highlight human role in southern California fire regimes. Environmental Research Letters 13(7): 074024.

10.

Bilbao

Leal

Méndez

(2010) Indigenous use of fire and forest loss in Canaima National Park, Venezuela. Assessment of and tools for alternative strategies of fire management in Pemón indigenous lands. Human Ecology 38(5): 663–673.

11.

Bintliff

Degryse

(2022) A review of soil geochemistry in archaeology. Journal of Archaeological Science Reports 43: 103419.

12.

Bird

Bliege Bird

Codding

(2016) Pyrodiversity and the anthropocene: The role of fire in the broad spectrum revolution. Evolutionary Anthropology Issues News and Reviews 25(3): 105–116.

13.

Bird

Fernandez

, et al. (2018) Aboriginal burning promotes fine-scale pyrodiversity and native predators in Australia’s Western Desert. Biological Conservation 219: 110–118.

14.

Bird

Codding

Kauhanen

, et al. (2012) Aboriginal hunting buffers climate-driven fire-size variability in Australia’s spinifex grasslands. Proceedings of the National Academy of Sciences 109(26): 10287–10292.

15.

Bohrer

(1991) Recently recognized cultivated and encouraged plants among the Hohokam. KIVA 56(3): 227–235.

16.

Bohrer

(1992) New Life from Ashes II: A Tale of Burnt Brush. Desert Plants 10(3): 122–125.

17.

Bowman

Balch

Artaxo

, et al. (2009) Fire in the Earth system. Science 324(5926): 481–484.

18.

Bowman

DMJS

Kolden

Abatzoglou

, et al. (2020) Vegetation fires in the Anthropocene. Nature Reviews Earth & Environment 1(10): 500–515.

19.

Brewer

Joseph

Warner

EAK

(2014) Protecting Indigenous knowledge in the age of climate change. International Environmental Law Review 27: 585.

20.

Buskirk

(1986) The Western Apache: Living With the Land Before 1950. Norman, OK: University of Oklahoma Press.

21.

Cabanes

Weiner

Shahack-Gross

(2011) Stability of phytoliths in the archaeological record: A dissolution study of modern and fossil phytoliths. Journal of Archaeological Science 38(9): 2480–2490.

22.

Cajete

Nelson

Shilling

(2018) Native science and sustaining Indigenous communities. In: Nelson MK and Shilling D (eds) Traditional Ecological Knowledge: Learning From Indigenous Practices for Environmental Sustainability. Cambridge: Cambridge University Press, pp.15–26.

23.

Coronel

Bair

Brown

, et al. (2014) Utility and limitations of portable X-ray fluorescence and field laboratory conditions on the geochemical analysis of soils and floors at areas of known human activities. Soil Science 179(5): 258–271.

24.

Cuthrell

Striplen

Hylkema

, et al. (2016) A land of fire: Anthropogenic burning on the central coast of California. In: Contemporary Issues in California Archaeology. New York, NY: Routledge, pp.153–172.

25.

Damick

Rosen

Ortman

(2021) Palm springs on the Rio Grande: Insight into Archaic forager plant use from phytoliths recovered from a Late Holocene alluvial section in northern New Mexico. PLoS One 16(10): e0258231.

26.

Davis

Shafer

(1992) A Holocene climatic record for the Sonoran Desert from pollen analysis of Montezuma Well, Arizona, USA. Palaeogeography Palaeoclimatology Palaeoecology 92(1-2): 107–119.

27.

Dobyns

(1981) From Fire to Flood: Historic Human Destruction of Sonoran Desert Riverine Oases. Socorro, NM: Ballena Press Socorro.

28.

Duke

Wohlgemuth

Adams

, et al. (2022) Earliest evidence for human use of tobacco in the Pleistocene Americas. Nature Human Behaviour 6(2): 183–192.

29.

Fill

Davis

Crandall

(2019) Climate change lengthens southeastern USA lightning-ignited fire seasons. Global Change Biology 25(10): 3562–3569.

30.

Ford

(2013) The cultural ecology of Jemez Cave. In: Vierra

(ed.) From Mountaintop to Valley Bottom. Salt Lake City, Utah: University of Utah Press, pp.69–79.

31.

Ford

(1985) Prehistoric Food Production in North America. Ann Arbor, MI: University of Michigan Press.

32.

Ford

(1999) Ethnoecology Serving the Community. Tucson, AZ: University of Arizona Press.

33.

Fowles

Alberti

(2021) Foreword: What Was an Image, There and Then? Ontologies of Rock Art. New York, NY: Routledge, pp.1–10.

34.

Foxx

(1996) Vegetation succession after the La Mesa fire. In: Fire effects in southwestern forests: Proceedings of the second La Mesa fire symposium, Los Alamos, NM, 29–31 March 1994. USDA Forest Service, Rocky Mountain Forest and Range Experiment Station.

35.

Fredlund

Tieszen

(1997) Phytolith and carbon isotope evidence for late Quaternary vegetation and climate change in the southern Black Hills, South Dakota. Quaternary Research 47(2): 206–217.

36.

Golyeva

Svirida

(2017) Quantitative distribution of phytoliths as reliable diagnostical criteria of ancient arable lands. Quaternary International 434: 51–57.

37.

Hall

Periman

(2007) Unusual Holocene alluvial record from Rio Del Osos, Jemez Mountains, New Mexico: Paleoclimatic and archaeologic significance., New Mexico Geological Society. In: Kues

Lueth

(eds) Geology of the Jemez Mountains Region II; 58th Field conference Proceedings. Socorro, NM: New Mexico Bureau of Geology and Mine Resources, pp.459-468.

38.

Hall

(2017) Paleoenvironments of the American Southwest. In: Vierra

(ed.) The Archaic Southwest. Foragers in an Arid Land. Salt Lake City, Utah: University of Utah Press, pp.16–28.

39.

Hall

Penner

(2013) Stable carbon isotopes, C3–C4 vegetation, and 12,800 years of climate change in central New Mexico, USA. Palaeogeography Palaeoclimatology Palaeoecology 369: 272–281.

40.

Haynes

Jr (2008) Younger Dryas “black mats” and the Rancholabrean termination in North America. Proceedings of the National Academy of Sciences of the United States of America 105(18): 6520–6525.

41.

Haynes

McLaughlin

(2000) Edible palms and their uses. Gainesville: University of Florida, pp.4–6.

42.

Herman

RDK

(2016) Traditional knowledge in a time of crisis: Climate change, culture and communication. Sustainability Science 11(1): 163–176.

43.

Hessburg

Prichard

Hagmann

, et al. (2021) Wildfire and climate change adaptation of western North American forests: A case for intentional management. Ecological Applications 31(8): e02432.

44.

Higuera

Abatzoglou

(2021) Record-setting climate enabled the extraordinary 2020 fire season in the western United States. Global Change Biology 27(1): 1–2.

45.

Holcomb

Karkanas

(2019) Elemental mapping of micromorphological block samples using portable X-ray fluorescence spectrometry (pXRF): Integrating a geochemical line of evidence. Geoarchaeology 34(5): 613–624.

46.

Holliday

Meltzer

(2010) The 12.9-ka ET impact hypothesis and North American Paleoindians. Current Anthropology 51(5): 575–607.

47.

Irwin-Williams

Irwin

Agogino

, et al. (1973). Hell gap: Paleo-Indian occupation on the high plains. Plains Anthropologist 18(59): 240–253.

48.

Jiménez-Moreno

Fawcett

Scott Anderson

(2008) Millennial- and centennial-scale vegetation and climate changes during the late Pleistocene and Holocene from northern New Mexico (USA). Quaternary Science Reviews 27(13-14): 1442–1452.

49.

Johnson

Howitt

Cajete

, et al. (2016) Weaving Indigenous and sustainability sciences to diversify our methods. Sustainability Science 11(1): 1–11.

50.

Keppel

Morrison

Watling

, et al. (2012). Conservation in tropical Pacific Island countries: Why most current approaches are failing. Conservation Letters 5(4): 256–265.

51.

Kimmerer

Lake

(2001) The role of indigenous burning in land management. Journal of Forestry 99(11): 36–41.

52.

Laris

Wechsler

(2018) Effects of landscape pattern and vegetation type on the fire regime of a mesic savanna in Mali. Journal of Environmental Management 227: 134–145.

53.

Laris

(2005) Spatiotemporal problems with detecting and mapping mosaic fire regimes with coarse-resolution satellite data in savanna environments. Remote Sensing of Environment 99(4): 412–424.

54.

Lent

(1991) The Excavation of a Late Archaic Pit Structure (LA 51912) near Otowi, San Ildefonso Pueblo, New Mexico.

55.

Lian

Piao

Chen

, et al. (2021) Multifaceted characteristics of dryland aridity changes in a warming world. Nature Reviews Earth & Environment 2(4): 232–250.

56.

Liebmann

Farella

Roos

, et al. (2016 ) Native American depopulation, reforestation, and fire regimes in the Southwest United States, 1492–1900 CE. Proceedings of the National Academy of Sciences of the United States of America 113(6): E696–E704.

57.

Lightfoot

Cuthrell

(2015) Anthropogenic burning and the Anthropocene in late-Holocene California. The Holocene 25(10): 1581–1587.

58.

Lightfoot

Cuthrell

Boone

, et al. (2013) Anthropogenic burning on the central California coast in Late Holocene and early historical times: Findings, implications, and future directions. California Archaeology 5(2): 371–390.

59.

Lynch

Ross

Carter

(2018) Indigenous Guidance in Australian Environmental Management. Australasian Journal of Environmental Management 25(3): 253–257.

60.

Margolis

Woodhouse

Swetnam

(2017) Drought, multi-seasonal climate, and wildfire in northern New Mexico. Climatic Change 142(3-4): 433–446.

61.

McBrinn

(2010) Everything old is new again: Recent approaches to research on the Archaic period in the western United States. Journal of Archaeological Research 18(3): 289–329.

62.

McBrinn

Vierra

(2017) The Southwest Archaic. In: Mills BJ and Fowles S (eds) The Oxford Handbook of Southwest Archaeology. Oxford: Oxford University Press, pp.231–245.

63.

Miller

Wendorf

(1958) Alluvial chronology of the Tesuque valley, New Mexico. The Journal of Geology 66(2): 177–194.

64.

Moore

(2001) Prehistoric and historic occupation of Los Alamos and Guaje Canyons: Data recovery at three sites near the Pueblo of San Ildefonso. Archaeology Notes 244. Santa Fe: Office of Archaeological Studies, Museum of New Mexico.

65.

Moore

(2018) Land Use, Settlement, and Community in the Southern Tewa Basin, Vol. 3: The Prehistoric Sites and Site Components. Manuscript on file, Office of Archaeological Studies, Museum of New Mexico, Santa Fe.

66.

Moore

Boyer

Wolcott Toll

(2003) Archaeological Testing Results and a Plan for Data Recovery at LA 111333, U.S. 84/285 Santa Fe to Pojoaque Corridor, Santa Fe County, New Mexico.

67.

Mori

Lertzman

(2011) Historic variability in fire-generated landscape heterogeneity of subalpine forests in the Canadian Rockies. Journal of Vegetation Science 22(1): 45–58.

68.

Mourao

Martinho

(2019) Forest fire legislation: Reactive or proactive? Ecological Indicators 104: 137–144.

69.

Muldavin

Durkin

Bradley

, et al. (2000) Handbook of Wetland Vegetation Communities of New Mexico. Albuquerque, NM: New Mexico Heritage Program.

70.

Murphy

Bradstock

Boer

, et al. (2013) Fire regimes of Australia: A pyrogeographic model system. Journal of Biogeography 40(6): 1048–1058.

71.

Murungi

Bamford

(2020) Revised taxonomic interpretations of Cyperaceae phytoliths for (paleo)botanical studies with some notes on terminology. Review of Palaeobotany and Palynology 275: 104189.

72.

Neumann

Strömberg

Ball

, et al. (2019) International code for phytolith nomenclature (ICPN) 2.0. Annals of Botany.

73.

Newlander

Goodale

Jones

, et al. (2015) Empirical study of the effect of count time on the precision and accuracy of pXRF data. Journal of Archaeological Science Reports 3: 534–548.

74.

Oster

Warken

Sekhon

, et al. (2019) Speleothem paleoclimatology for the Caribbean, Central America, and north America. Quaternary 2(1): 5.

75.

Oviatt

(2015) Chronology of Lake Bonneville, 30,000 to 10,000 yr BP. Quaternary Science Reviews 110: 166–171.

76.

Painter

Detling

Steingraeber

(1993) Plant morphology and grazing history. Vegetatio 106(1): 37–62.

77.

Polyak

Cokendolpher

Norton

, et al. (2001) Wetter and cooler late Holocene climate in the southwestern United States from mites preserved in stalagmites. Geology 29(7): 643–646.

78.

Polyak

Rasmussen

JBT

Asmerom

(2004) Prolonged wet period in the southwestern United States through the Younger Dryas. Geology 32(1): 5–8.

79.

Post

(2013) Transitional archaic and emergent agricultural settlement in the lowland-upland settings of the northern Rio Grande, New Mexico. In: Vierra

(ed.) FroM Mountaintop to Valley BottoM. Salt Lake City: University of Utah Press, pp.80–94.

80.

Post

(2002) Emerging from the shadows: The Archaic period in the Northern Rio Grande. In: Schlanger

(ed.) Traditions, Transitions and Technologies: Themes in Southwestern Archaeology. Boulder, CO: University Press of Colorado, pp.33–48.

81.

Ramsey

Rosen

(2016) Wedded to wetlands: Exploring Late Pleistocene plant-use in the Eastern Levant. Quaternary International 396: 5–19.

82.

Reasoner

Jodry

(2000) Rapid response of alpine timberline vegetation to the Younger Dryas climate oscillation in the Colorado Rocky Mountains, USA. Geology 28(1): 51–54.

83.

Roos

(2008) Fire, climate, and social-ecological systems in the ancient Southwest: Alluvial geoarchaeology and applied historical ecology. Dissertations, The University of Arizona, Tucson.

84.

Roos

(2017) Anthropogenic landscapes. In: Mills BJ and Fowles S (eds) The Oxford handbook of Southwest archaeology. Oxford: Oxford University Press, pp.683–696.

85.

Roos

Guiterman

(2021) Dating the origins of persistent oak shrubfields in northern New Mexico using soil charcoal and dendrochronology. The Holocene 31: 1212–1220.

86.

Roos

Sullivan

III McNamee

(2010) Paleoecological evidence for systematic indigenous burning in the upland southwest. In: Dean RM (ed.) The Archaeology of Anthropogenic Environments. Carbondale, IL: Center for Archaeological Investigations, pp.142–171.

87.

Roos

Zedeño

Hollenback

Erlick

(2018) Indigenous impacts on North American Great Plains fire regimes of the past millennium. Proceedings of the National Academy of Sciences 115(32): 8143–8148.

88.

Rosen

(1999a) Phytolith Extraction Protocol. Unpublished Lab Protocol in possession of the authors.

89.

Rosen

(1999b) Phytolith Analysis in Near Eastern Archaeology. The Practical Impact of Science on Aegean and Near Eastern Archaeology. London: Archetype Press, pp.86–92.

90.

Rosen

(2008) Phytolith analysis. In: Pearsall

(ed.) Encyclopedia of Archaeology. New York: Academic Press, pp.1818–1822.

91.

Russell-Smith

Yates

Edwards

, et al. (2003) Contemporary fire regimes of northern Australia, 1997 - 2001: Change since Aboriginal occupancy, challenges for sustainable management. International Journal of Wildland Fire 12(4): 283–297.

92.

Scherjon

Bakels

MacDonald

, et al. (2015) Burning the land: An ethnographic study of off-site fire use by current and historically documented foragers and implications for the interpretation of past fire practices in the landscape. Current Anthropology 56(3): 299–326.

93.

Short

(2017). Spatial wildfire occurrence data for the United States, 1992-2015 [FPA_FOD_20170508]. 4th Edition. Fort Collins, CO: Forest Service Research Data Archive. https://doi.org/10.2737/RDS-2013-0009.4

94.

Skinner

Shaw

Carter

, et al. (1980) Archaeological investigation at Nambe Falls. Unpublished report.

95.

Šmejda

Hejcman

Horák

, et al. (2018) Multi-element mapping of anthropogenically modified soils and sediments at the bronze to iron Ages site of Tel Burna in the southern levant. Quaternary International 483: 111–123.

96.

Smith

(2016) Weaving pikyav (to-fix-it): Karuk Basket Weaving in-Relation-with the Everyday World. UC Berkeley.

97.

Snitker

Roos

Sullivan

, et al. (2022) A collaborative agenda for archaeology and fire science. Nature Ecology & Evolution 6: 835–839.

98.

Sullivan

Forste

(2014) Fire-reliant subsistence economies and anthropogenic coniferous ecosystems in the Pre-Columbian northern American Southwest. Vegetation History and Archaeobotany 23(1): 135–151.

99.

Sullivan

Mink

(2018) Theoretical and socioecological consequences of fire foodways. American Antiquity 83(4): 619–638.

100.

Swetnam

Baisan

(2003) Tree-ring reconstructions of fire and climate history in the Sierra Nevada and southwestern United States. In: Veblen

Baker

Montenegro

, et al (eds) Fire and Climatic Change in Temperate Ecosystems of the Western Americas. New York, NY: Springer, pp.158–195.

101.

Swetnam

Farella

Roos

, et al. (2016) Multiscale perspectives of fire, climate and humans in western North America and the Jemez Mountains, USA. Philosophical Transactions of the Royal Society B Biological Sciences 371(1696): 20150168.

102.

Tagg

(1996) Early cultigens from Fresnal shelter, southeastern New Mexico. American Antiquity 61(2): 311–324.

103.

Trauernicht

Brook

Murphy

, et al. (2015) Local and global pyrogeographic evidence that indigenous fire management creates pyrodiversity. Evolutionary Ecology 5(9): 1908–1918.

104.

Vierra

(2018) The Archaic Southwest: Foragers in an Arid Land. Tucson, AZ: The University of Arizona Press.

105.

Vierra

(2020) Some Thoughts on the history of Archaic research in New Mexico. In: Walth

(ed.) The Archaic of New Mexico. Albuqurque, NM: New Mexico Archaeological Council, pp.1–16.

106.

Vierra

(2013) From Mountain Top to Valley Bottom: Understanding Past Land Use in the Northern Rio Grande Valley, New Mexico. Salt Lake City, UT: University of Utah Press.

107.

Vierra

Ford

(2006) Early maize agriculture in the northern Rio Grande valley, New Mexico. In: Histories of Maize. New York, NY: Routledge, pp.497–510.

108.

Vierra

Ford

(2007) Foragers and farmers in the northern Rio Grande valley, New Mexico. Kiva 73(2): 117–130.

109.

Vierra

Foxx

(2009) Archaic Upland Resource Use: The View from the Pajarito Plateau. Between the Mountains, Beyond the Mountains: Papers in Honor of Paul R. Williams, pp.154–166.

110.

Vierra

(2017) The Southwest Archaic. In: Fowles

BMaS

(ed.) The Oxford Handbook of Southwest Archaeology. Oxford: Oxford University Press, pp.231–245.

111.

Wendt

JAF

McWethy

Widga

, et al. (2022) Large-scale climatic drivers of bison distribution and abundance in North America since the Last Glacial Maximum. Quaternary Science Reviews 284: 107472.

112.

Whelan

(1995) The Ecology of Fire. Cambridge: Cambridge University Press.

113.

White

(1932) The Acoma Indians. Santa Fe, NM: Rio Grande Press.

114.

White

(1996) Old-growth oak and oak-hickory forests. In: Davis

(ed.) Eastern Old-Growth Forests. New York, NY: Island Press, pp.178–198.

115.

Zona

(1990) A monograph of Sabal (Arecaceae: Coryphoideae). Aliso: A Journal of Systematic and Floristic Botany 12(4): 583–666.

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

0.16 MB