Ancestral Pueblo and Historic Ute Rock Art, and Euro-American Inscriptions in the Canyons of the Ancients National Monument, Colorado, USA

Abstract

In the central Mesa Verde region, rock art occurs on canyon walls and on boulders that are frequently associated with other archaeological remains. Moreover, rock art, together with architecture and pottery, is actually a primary source of archaeological information about the presence of various cultures in the area. It includes paintings and petroglyphs of ancestral Pueblo farming communities, images and inscriptions made by post-contact Ute and possibly Diné (Navajo) people as well as historical inscriptions of the early Euro-Americans in this area. This paper presents the results of archaeological investigations at four large rock art sites from Sandstone Canyon, southwestern Colorado, within the Canyons of the Ancients National Monument (CANM). Methods of rock art recording included advanced digital photography, photogrammetry, terrestrial laser scanning (TLS), hand tracing, and consultations with members of indigenous societies and rock art scholars. Geophysics and sondage excavations were conducted at one site revealed important information about archaeology, environment, and geology of the area. Analysis of rock art and other material evidence aims to help reconstruct and understand the mechanisms and nature of cultural changes, migrations, and human–environmental interactions and later cross-cultural contacts between indigenous peoples and Anglo-American ranchers and settlers in pre-contact southwestern Colorado and the US southwest.

1. Introduction

Rock art (paintings/pictographs and petroglyphs), apart from ceramics and the well-known Ancestral Pueblo sites in the Mesa Verde National Park and Canyons of the Ancients National Monument/CANM and Hovenweep National Monument, is one of the most visible traces of the varied groups that previously occupied the central Mesa Verde region of present-day southwestern Colorado and southeastern Utah. Although there are no studies that address the entirety of this rock art, extended or partial discussions of it may be found in several significant publications and unpublished materials (e.g., Castleton 1978; Charles and Cole 2006, pp. 187–200; Cole 2004, 2008, 2009; Palonka 2021, 2022, pp. 107–12, 158–63; Schaafsma 1980, 1994, 2022). Moreover, in cases where historic tribes (mostly Utes, but possibly also Navajos) left only very little archaeological material evidence, rock art is a primary source of information about their presence. On the other hand, Euro-American inscriptions and modern graffiti carved close to or directly on the Native American rock art by early Euro-American settlers and travelers (e.g., Blackburn and Williamson 2004; Geary 2004) that include cartoon-like figures, initials, names, dates, and livestock brands, are also part of the history of this area. Together with indigenous rock art they cover canyon walls, boulders and rock alcoves constituting a distinctive chronicle of this region.

The rock art sites presented in this paper are located in the lower part of Sandstone Canyon (Figure 1), one of the largest canyons of southwestern Colorado. Sandstone Canyon is situated in Montezuma County within the Canyons of the Ancients National Monument (CANM) created in 2000 by the declaration of President Bill Clinton. The Monument is administered by the Bureau of Land Management, U.S. Department of the Interior. It contains the highest known archaeological site density in the United States (with around 30 thousand archaeological and historical sites), with rich and well-preserved evidence of Indigenous cultures and Anglo-Spanish settlers (e.g., Horn 2004; Lipe et al. 1999; Steinbrecher and Hopkins 2019). Part of the area of today’s Monument was, in the past or even recently, bought or leased by the BLM from private landowners and cattle ranchers and in many places is still an undiscovered, unspoiled area in the heart of the Mesa Verde region, and (as in the case of Sandstone Canyon) it has previously received only cursory archaeological analysis.

During the research and reconnaissance of the lower part of Sandstone Canyon we managed to record four sites with large rock art panels, mostly petroglyphs (Palonka et al. 2018, 2019, 2023a, 2023b; Ciomek and Palonka 2024; Schaafsma 2022) as well as (to a much lesser extent) sandstone architecture; initially we also noted a few more rock art sites to the north and south, up to a distance of about 4 km. Of the four initially recorded sites, only one, 5MT13288 (Painted Hand Petroglyph Panel/Strawman Panel), had been known to archaeologists. The site consists of a 129 m long sandstone wall with abundant petroglyphic depictions and panels, initially dated to different time periods (Bowman 1998; Cole 2005; Palonka 2014, 2015a, 2017; Palonka et al. 2023a, 2023b; Schaafsma 2022). The other three sites under discussion are 5MT22149 (Sandal Panel), 5MT22150, and 5MT22208. There are also several additional panels to the north which have not been recorded officially. Rock art from the Basketmaker and Pueblo periods occurs in all four recorded sites, and three of the four sites have figurative rock art from the historic period that is most probably Ute in origin. These historic figures clearly dominate at the Strawman Panel and are a significant component of sites 5MT22149 and 5MT22208, in addition to numerous Euro-American inscriptions of names, dates, and livestock brands. All of these suggest that the accumulation of rock art in the area may be even greater than in other parts of the Mesa Verde region.

This research project has been conducted by the Department of New World Archaeology, Institute of Archaeology, Jagiellonian University in Kraków (Poland) as a part of a larger project in southwestern Colorado focusing on socio-cultural changes in the Ancestral Pueblo communities (e.g., Palonka 2019a, 2019b, 2021; Palonka et al. 2020) and with strong collaboration with American institutions, mainly CANM, the Crow Canyon Archaeological Center (CCAC) and several American scholars and Native American collaborators.

2. Environment, Geography, and History of Research in Sandstone Canyon

Sandstone Canyon is a major tributary to Yellow Jacket Canyon, which is the largest canyon in the area of the McElmo Canyon drainage (Figure 2) that, in turn, is a major northern tributary of the San Juan River. Other nearby drainages in this canyon system include Burro Canyon and Woods Canyon. Sandstone Canyon and nearby canyons run northeast–southwest from the upland Dolores River Canyon and its surroundings located to the northeast of the project research area to the McElmo Canyon. The lower Sandstone Canyon area is located at ca. 1640–1650 m above sea level. It is also worth noting that the Dolores River and the San Juan River are the two major rivers in the region and both are confluences of the Colorado River in Utah (ca. 150 km to the west).

In the portion in which our investigations were conducted, Sandstone Canyon has a relatively wide bottom (ca. 100–200 m and more), which continues southwards and is now quite heavily overgrown with various grasses and shrubs, including greasewood and big sagebrush as well as prickly pear/cacti and other species, including datura (Datura L.); datura was noted at a few sites in Sandstone Canyon and also between sites.

Massive soil depletion in the uplands, possibly due to heavy rains, floods, and over-grazing, among other reasons, have subjected the sites under consideration to various types of damage. Over a period of time, many rock art panels on these friable sandstone surfaces were directly and negatively affected by water, wind erosion, as well as cattle taking refuge under sandstone overhangs. Basketmaker and ancestral Pueblo rock art on the lower portions of the cliffs has been scrubbed and eroded or otherwise impacted by these events, and some images are undoubtedly still buried under deep soil deposits below the modern ground surface. Historic floods from the area are known from written records and oral histories (e.g., Cline 2010; Follansbee and Sawyer 1948, pp. 132–39; Head 2017). The most devastating of such floods in southwestern Colorado took place in 1911. However, research conducted by our team confirms those catastrophic events even for the earlier periods.

One of the two Ute reservations in southwestern Colorado, the Ute Mountain Ute Tribe Reservation, is located ca. 12 km directly south of the mouth of the Sandstone Canyon. At the turn of the nineteenth and twenty centuries the area of investigation was still the home for at least some Ute families. In late historic times, from at least the second half of the nineteenth century and continuing into the twentieth century, this area has served, for example, as winter pastures for cattle and sheep, severely impacting the natural environment. Heavy grazing of the uplands is thought to have been initiated around the 1870/80s, lasting through to at least the 1930s (Horn 2004, pp. 11–39).

The first known recordings of the rock art of the area included only the Strawman Panel (5MT13288) in 1998 by Douglas W. Bowman for the Four Corners School and Bureau of Land Management (Bowman 1998) and in 2005 by Sally J. Cole and the Elderhostel volunteers, a project sponsored by the Canyons of the Ancient National Monument, Bureau of Land Management and McElmo Canyon Research Institute (Cole 2005). It is clear from these records that new inscriptions (graffiti)—as we should call it in lieu of “vandalism“—appeared between 1998 and 2005. Additionally, during our research (after 2014) we noted carved “additions” at a few more sites in the vicinity. Strawman Panel site is now protected from not only vandals but also from animals, including cows, by a locked gate and metal fence that runs across the whole sandstone wall at the distance of ca. 3 m. The nearby site, 5MT22208, is now also protected by a fence and two gates.

Our investigations in Sandstone Canyon (digital documentation, hand tracings, geophysics research, and test-pit excavations) began with the initial visit at the Strawman Panel in 2013 with Vincent M. MacMillan, the chief archaeologist in the CANM. In 2014 digital documentation of the Sandstone Canyon sites was initiated by our team consisting of scientists and students from the Department of New World Archaeology, Institute of Archaeology at the Jagiellonian University in Kraków, Poland. Field work continued during several seasons between 2014 and 2019 and again after the break caused by the COVID-19 pandemic in 2022 and 2023. Consultations were also conducted with local residents and rock art scholars as well as native consultants and this is still an ongoing process. In 2022 we started consultations with the elders and other members of the Ute Mountain Ute Tribe, which were conducted together with researchers from CCAC and CANM.

This text is the first comprehensive publication of the analysis of rock art sites from Sandstone Canyon, though the initial results of Jagiellonian University’s research of rock art from the Sandstone Canyon have been presented at several conferences in Europe and the United States (e.g., Ciomek et al. 2023; Palonka et al. 2018, 2019) as well in some publications that focus on the methodology of digital recordation (e.g., Palonka and Zych 2022; Palonka et al. 2022, 2023a, 2023b) and in some unpublished reports (Palonka 2014, 2015a; Schaafsma 2022). Since 2015, the photo exhibit has been presented in several museums and universities in Poland and twice in the US, in 2018–2019 at the Canyons of the Ancients Visitor Center & Museum, Dolores (Colorado) and in 2020–2021 at the Edge of the Cedars State Park Museum, Blanding (Utah). A small catalogue from the exhibition was also published bilingually (Polish and English) with the cooperation of the US Consulate General in Krakow (Palonka 2015b). Three-dimensional models, photos, and tracings from the Strawman Panel, taken since 2024, are placed within the permanent digital exhibit in the form of an interactive kiosk at the Canyons of the Ancients Visitor Center & Museum. However, this paper is the first to describe and analyze all of the investigated sites in Sandstone Canyon, placing them in a wider context and with analogies to other sites, located beyond the research area.

3. Methodology

3.1. Digital Photography, Photogrammetry, and Laser Scanning

The main methods we used for rock art documentation, besides, to some extent, hand tracings, were digital photography and photogrammetry (Figure 3). Photography of entire panels with numerous depictions, (e.g., narrative scenes) as well as individual motifs, were captured using scales and calibrated color checkers. One of the challenges of taking digital photography in the exposed sites (like 5MT13288 and 5MT22149) is strong sunlight (Palonka et al. 2023a, 2023b). Photos were also taken from a UAV/drone. Additionally, we took panoramic photographs using GigaPan EPIC Pro equipment (Portland, OR) and software (see for example Mark and Billo 2011). These photographs are especially useful for an analysis of the relationships between rock art and landscape and for presentation/publication of the data in a digital format to facilitate further study and allow access for other researchers, etc. via the internet (the sites are closed to public). The panoramas were taken at the two abovementioned sites and they are available at http://www.gigapan.com/gigapans/5cb2afa7d45e97293bcdab019c655d76 (accessed on 12 December 2024) and http://www.gigapan.com/gigapans/c7c587962ea666dcc0bbcc4b35924cef (accessed on 10 December 2024).

Close-range photogrammetry and structure-from-motion (SfM) (a photogrammetric process and connection between computer vision and photogrammetry) were conducted at three of four sites (excluding 5MT22208). Photogrammetry was chosen for its accuracy, reasonable costs, and its relatively fast process for collecting data, which was important given the limited field time available. Similar issues were also raised in other field projects (Bourdier et al. 2015; Jalandoni and May 2020; Palonka et al. 2022, 2023b; Simek et al. 2022). The image rectification photogrammetry method (Historic England 2018, pp. 9–11; Remondino and Campana 2014) was chosen first in 2014. For this documentation, for example, we positioned 918 control points/markers on the canyon wall and between petroglyphs at 5MT13288/Strawman Panel; every control point was recorded with X,Y,Z coordinates using an electronic tachymeter and subsequent processing in AutoCAD Photoplan software (a detailed description of this method is in Palonka and Zych 2022; Palonka et al. 2023a, 2023b). Subsequently, in 2023 we were able to conduct close-range and UAV multiple image/ortho-image photogrammetry (mainly at 5MT13288 and 5MT22150) and create 3D models using new digital cameras and software (mostly Reality Capture 1.3 but also Agisoft Metashape 2.0.1).

The three-dimensional terrestrial laser scanning (TLS) (e.g., Craig et al. 2015; Historic England 2018; Jaillet et al. 2017; Remondino 2011) of rock art sites analyzed in this paper was conducted using different equipment at the 5MT13288 and 5MT22149 sites. The aim was also to compare 3D scanning with photogrammetry and develop the best digital methods for recording rock art, especially rock that is not easily accessible or for large panels with many overlapping images (both cases are present in the analyzed sites). Part of the results from 3D scanning is published at https://sketchfab.com/radek.palonka (accessed on 10 December 2024).

3.2. Advanced Digital Photography and Enhancements: DStretch and RTI

In recent years distinctive methods for digital documentation and enhancement of rock art have been in use, including analysis of digital images in Adobe Photoshop and Adobe Lightroom software. Nowadays, probably the most popular and most used method, especially for paintings (with a focus, though not a singular one, on the color red) is DStretch/Decorrelation Stretch (e.g., Harman 2020). Besides paintings, we used DStretch for analyzing the petroglyphs with very promising results. Our research also employed a more difficult to use and less popular field method, reflectance transformation imaging (RTI), which has great potential, especially for faint and deeply patinated depictions and surfaces. This method is based on the set of images taken in multi-lightning conditions from a fixed camera position (e.g., CHI 2020; Earl et al. 2010; Malzbender et al. 2000).

3.3. Geophysics Research and Sondage Excavations

At 5MT13288, geophysics research, including ground penetrating radar (GPR) and electrical resistivity, as well as observations of soil deposits below some of the petroglyphs suggested the possibility of the existence of structures below the modern ground surface and subsequent test-pit excavations with boreholes revealed evidence that there are traces of anthropogenic activity; for example, architectural features 1.6–2.0 m below the modern ground (Palonka 2014; Palonka et al. 2019). All of this confirms the presence of a large accumulation of soil, possibly from catastrophic flooding episodes. Rock art also suggests buried deposits, as some of the petroglyphs are located on the very lowest section of the sandstone wall and are even, in some cases, partly underground. In other analyzed sites there are also traces of periodic soil accumulation and removal (i.e., mud and soil in rock crevices on the canyon wall at the 5MT22149 site, or depression and traces of destroyed sandstone architecture at the 5MT22150, along again with mud deposits in rock crevices and niches). More geological and environmental study needs to be conducted at other sites in Sandstone Canyon.

4. Rock Art Sites in Sandstone Canyon

The four analyzed rock art sites from the lower part of Sandstone Canyon contain a great density of petroglyphs representing distinct cultural and chronological periods.

4.1. 5MT13288 (Painted Hand Petroglyph Panel/Strawman Panel)

Site 5MT13288, the largest site documented and analyzed in this paper, is a 129 m long and 26 m high canyon wall in the form of shallow alcove, the ground-level of which is situated at the elevation of 1640 m asl (Figure 4a). The present ground surface is relatively flat but is clearly filled with recent wind or water-borne soil deposits that have buried the oldest petroglyphs, some of which are partially visible at ground level or just above it. The lowest section of the wall have endured rough treatment, possibly by wind erosion and cattle. As mentioned previously, the entire length of the panel is currently protected behind a chain-link fence with a locked gate for protection. There are no associated structures visible on the ground and only a few artifacts were collected: there is information about two cores found in and close to the site (Cole 2005), and in 2022 our team collected one ancestral Pueblo small painted pottery sherd and a fragment of historic glaze pot, possibly of Spanish-American origin. This is not surprising given the results from test excavations, indicating that the ground surface below the canyon wall buried both petroglyphs and other traces of human activity such as artifacts and architecture. On the small “plateau” above the wall with petroglyphs, there is lithic debitage, mostly from quartzite, indicating that this location may have been a flintknapping/manufacturing place.

Among the Pueblo petroglyphs at this site is a pecked human stick figure whose upraised arms and head reach above the current ground surface. Another figure, fully above the ground level, has five-digit hands and toes that are roughly incised. There is also a pecked zigzag line, one possible depiction of a flute player, and some abstract and geometric motifs; pecked handprints could be attributed to either ancestral Pueblo or historic Ute or both. The grooves typical at the late thirteenth century sites nearby (see below, for example at sites 5MT22149 and 5MT22150) appear to be lacking at the 5MT13288 site, although without excavations and exposing the lower wall, one cannot discount the possibility of their former existence. Other ancestral Pueblo/Basketmaker petroglyphs or paintings may well be buried underground. With these few pre-Contact exceptions, the remainder of the representational content at this site is primarily Ute, Hispanic, and Anglo in origin (Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9).

The Ute rock art, notable for its large numbers of riders and animal herds (bison, deer, elk, bighorn sheep), horses, and riders, is clustered into roughly six or seven large panels, suggesting that these assemblages also represent different episodes of petroglyph-making at this site, although no major chronological differences are apparent (maybe besides different types of weapon, including bows and arrows and guns and a distinction between pedestrian and horseback warriors). Most figures are solidly carved, although there are a few bison in outline (see for example Figure 5c,f). Altogether, at least 27 bison are represented, along with 44 elk/deer and 17 bighorn sheep as well as what appear to be dogs. Stick figure horses and riders are in large numbers, especially near the base of the central and right parts of the site. There are 34 or more horseback riders in total, as well as several scattered pedestrians that are integral parts of these scenes. These complexes appear to document events (hunting and fights, including gunfights) and thus seem narrative in intent (Figure 6 and Figure 7). At least two or three human figures with a decoration of their heads and holding a stick in a hand appear in the middle of herds of animals (mostly deer or bighorn sheep); this strongly suggests possible shamanistic interpretations (for example, shamans helping in hunting). In this case the potentially symbolic values ascribed to bison, for example, are downplayed. Weapons are rarely depicted and include bows and arrows, as well as rifle or guns in a few cases. Scattered among the petroglyphs at either end of the most central panel are seven bear tracks, with eleven or twelve more to the left and right. While their cultural origins are somewhat ambiguous and they are a subject that is common to both the ancestral Pueblo and the Utes, their presence here among the proliferation of historical Ute petroglyphs, carved well above the soil deposition at the base of the cliff, argues for their being of Ute origin.

In the central part of the site there are two armless figures that may represent a birthing scene, as they appear to be wrapped in blankets, one with a baby or a small child inside and one with a child below (see Figure 4b and Figure 6b). To the left is a depiction of a man with a beard and two crosses in his hands, possibly representing a missionary. In the most eastern part of the site there are some larger Ute cartoon-like people and animals including a seated fat cowboy with a domed hat pointing a rifle at a mountain lion (or a horse) which, in turn, is facing a deer (Figure 9). We suggest that these may represent the Late Historical style of Ute rock (ca. 1830–1880 A.D.) or might even be related to the Ute Representational style (1880–1950 A.D.) known from Mancos Canyon located ca. 50 km southwest from Sandstone Canyon (Cole 1990, pp. 235–51). Additionally, there is a two-story house with a pitched roof and a few other depictions of houses or barns. A packed burro is represented with “HAY” inscribed on its pack. Two large-scale realistic horses, one of which is embellished with a fancy saddle, bit and reins, are defined in outline by sparse pecking; nearby, a large water bird and two chicks are similarly rendered, probably by the same artist. Elsewhere in this panel, an indication of a Ute origin for the art can be found in the large-saddled horse, due to its stylistic similarity to other Ute petroglyphs in the vicinity of Bluff, Utah and on the Uncompahgre Plateau and on Uintah and Ouray Indian Reservations in Utah (Cole 1987, 1988; Schaafsma 2022).

In addition, numerous names, dates, and cattle brands are inscribed throughout, and these comprise a written document on stone of the late history of this place. Many dates are from the late nineteenth into the early twentieth century, for example, 1861 (?), 1892, 1895, 1897, 1905, and 1913. Notable among the names is “Menefee 2/12/1895”, a recognized local name today. “Joe Ismay 1941” relates to the Ismay trading post established in the early twentieth century on the Colorado/Utah border at the junction of Yellow Jacket and McElmo canyons. A few dates are a reference to Mel (Melvin) Forest, a “U.S. trapper.” Other notable names include Dolores, Rex, and Don Tozer, a prominent family name in the area. A number of early twentieth century inscriptions are Hispanic, along with the names of their villages of origin such as “Juan Maestas, Enero 1947, Parkview, n.m.” and “ Mayo 7 1924, J. R. Gomez, Lumberton NM.” JA Vijil (sic) dated his inscription as Abril 19 1939. Other names are associated with Rosa, New Mexico and Nambe Pueblo, the latter likely Pueblo in origin. The Hispanic names document what were probably seasonal migrations of New Mexican sheepherders to new locations in southwestern Colorado known for their winter pastures. As such they are of historical interest in terms of land use.

As for much of the pictorial content, sometimes it is difficult to discriminate between the work of Ute, Anglo and Hispanic cowboys and sheep herders as they are all participants in a common “cowboy culture” (see Horn 2004 for a summary of the history of late nineteenth to early twentieth century land use in this region). There is at least one signature (Ira C.), date (1936), and a few associated images possibly left by Ira Cuthair—a famous Ute rodeo cowboy and fluteplayer. Initial interviews were conducted by our team at the site with his two daughters and a granddaughter in the fall of 2024 (analyses are ongoing along with other consultations with Ute tribal members and local ranchers families).

4.2. 5MT22149 (Sandal Panel)

The Sandal Panel site is located 800 m northeast of Strawman Panel. It consists of several main concentrations of indigenous rock art and graffiti and historical inscriptions (Figure 10). Around six meters above the modern ground surface, a large panel ca. three meters in length portrays a herd of a dozen or so large, pecked bison and additional animals. A second distinct panel, three to four meters below, consists mostly of Ancestral Pueblo depictions and a few historical (very late) inscriptions.

The upper panel (Figure 10b-1 and Figure 11) is well defined by two natural cracks which meet at the left lower corner. The upper crack forms a hill pattern that at least one bison is climbing (this is a similar manner of depiction as one of the panels at Strawman Panel, see Figure 7). In this panel depictions of at least eleven or twelve bison, and additionally at least ten possible elk and what may be bighorn sheep and/or antelope are represented (these are deeply incised—even gouged—among the bison and to the right of them). The bison strongly indicate Ute origins, and there are no historic inscriptions and graffiti in association with this panel. Three or four pedestrian human figures (two fighting each other, possibly with knives or other weapons) and probably dogs surround the herd seemingly keeping it together; there is one wavy line below the largest bison and this line may represent a snake. The absence of horses and riders in this scene is notable.

Below, in the second group, is a heavily eroded upper part of a Basketmaker anthropomorph with a necklace as well as possibly some plant images (two also resemble elongated bird tracks) (Figure 10b-2 and Figure 12). Below and to the left and right from the observer there are several much later Ancestral Pueblo petroglyphs. These include, among other things, a panel with nine elaborately made footprints and sandal depictions that are represented both singly and in sets (Figure 13 and Figure 14). Roughly rectangular or slightly tapered, the edges are marked with heavy grooves. Toes, represented by cupules, confirm that the more ambiguous rectangular elements are, in fact, sandals. At least three have lightly incised interior patterns, suggesting the geometric designs found on actual woven sandals. Another has a bird track inside. Notable are the deeply cut wide grooves that define the sides of some of these images, suggesting that the grooves themselves contribute meaning to the icon. One sandal image is especially notable, in that it looks like the representation of a sandal divided into four sections, each containing a geometric pattern. However, after further examination aided by digital enhancements it is clear that this large sandal contains four small sandals inside it: two placed horizontally and other two vertically; it is one of the most beautifully and skillfully designed sandal petroglyphs known (for some comparisons of sandal images, see Schaafsma 2016; Bellorado 2018, 2020). Bellorado (2020) has suggested that the patterned sandal imagery in post-Chaco rock art sites in the Four Corners region were made to mark social or territorial identities.

At the right side of the site (as viewed from the observer) are some vertically placed carved recesses, hand and toeholds, possibly Pueblo in origin. To the left from this panel there are numerous sets of vertical grooves, horizontal grooves, natural holes modified by abrasion, drilled holes, niches and cupules, and a modified deep crack; there are also possibly rectilinear stick human figures with loosely attached three-digit appendages. At the very base of the cliff a likely spiral and other pecked lines are barely visible, having been nearly eroded into oblivion (Figure 15).

As for historical inscriptions/modern graffiti, the most visible in the left-central part of the site are deeply incised initials “JP” in association with a skillfully carved horse head and, besides the sandal glyphs and in another hand, the name BRENNA, below which the letter “W” and initials “BRW” are deeply and sharply cut (Figure 13b, right). “W” has a curved line beneath, a common element in local ranchers’ brands. Ash-colored soil and charred plant remains indicate recent fires at this location. It seems that those initials are very recent additions to the site. More historic inscriptions are located to the left in the southern part of the site (including initials, dates, names, and even one exact address, from Phoenix, Arizona).

Similarly to 5MT13288, no associated structures or artifacts were found at this site. This is perhaps not surprising, given the unstable nature of the local topography and historically known floods. The Pueblo period petroglyphs and the much more ancient Basketmaker period figures have endured some rough treatment, having been being covered by alluvium that was subsequently washed away by flooding as evidenced, for example, by residual mud deposits in pockets and crevices in the cliff wall near the petroglyphs. In summary, the vertical stratigraphy of the petroglyph panels on this cliff face is further indication of major fluctuations in the ground level at this site: as for these topographic changes, at the time the panel with bison was made, there was a very deep soil deposit in this spot that provided access to the upper cliff, temporarily burying the ancestral Pueblo and Basketmaker petroglyphs below. Subsequent floods removed these temporary deposits, exposing the earlier work, which appears to have sustained some abrasion damage in the process.

4.3. 5MT22150 (The Face)

Across the canyon from 5MT13288 and facing north-northwest is 5MT22150; its unofficial name is for a peculiar erosional geological phenomenon that consists of two large elongate side-by-side recesses under an alcove within a looming sandstone cliff (Figure 16). At the upper part of each of these recesses is a deep, dark, roundish cavity that give the appearance of eyes. Between the recesses is a tall narrow band of rock that extends downward and resembles a nose; this giant feature is framed within an overhanging section of cliff, and the entire complex has the look of a monstrous ominous face (this very intriguing idea came from Polly Schaafsma’s observations). White bird dropping on ledges within the “eyes” indicate that this locality sees substantial use by large birds, possibly owls and/or hawks (see Figure 16b). Today directly below this natural complex, a mix of mud and soil, with remnants of masonry stones, rests against the cliff, deposited there possibly by a past flood. We noticed Pueblo II-III period pottery sherds (Mesa Verde Black-on-White and possibly Deadman’s Black-on-Red, the latter of which implies contacts with southeastern Utah) and a stone axe at the site, offering further indication of a late Ancestral Pueblo occupation (most probably twelfth and thirteenth century A.D.).

Across a wide section of heavily worked cliff to the left of the “Face” are between 50 and 60 pits and small niches or cupules, exceeding in number those at any other site in the general vicinity (Figure 17). Grooves also occur in abundance as well as the usual sets of randomly incised marks. Those that intersect at right angles suggest incomplete ‘boxes’ or other motifs; some may indicate sandals. Among these markings are incised bird tracks and a rectangular-bodied Pueblo II-III type human figure surrounded by pits and short grooves. There is also a somewhat rectangular design loosely defined with incised double lines, inside of which are several niches and grooves; a small, incised circle; and other patterning. There is at least one spiral (Figure 17a,d) incised just below the “Face” and smaller pencil-width holes. The rock face also includes a large section of pitted surface but it is not clear if this surface is due to natural weathering, and if so, whether it may have been subject to modification by the Ancestral Puebloans to some degree.

Further to the left of the “Face,” around seven to eight meters above the modern ground surface (possibly floods in the past washed away a significant portion of the site) there are at least three detectable Basketmaker anthropomorphs and nearby on the same rock face is a collection of grooves, niches, pits and abraded markings (Figure 18). The Basketmaker figures (two incised and one pecked) are faint and partially erased by weathering or abrasion and are located within a now mud-streaked area. Most details are indeterminate, but their necklaces are clear. Two have facial features defined with small holes in the rock (slightly resembling the natural face of rock formation mentioned above); the torso of one of the anthropomorphs (most left one) is also with pecked holes. Remnants of soil deposits, from what must have been a past flood of historic proportions, remain on the ledges above this cliff face and within cliff cavities as reminders that, at one point, this entire set of petroglyphs had been buried.

Its regional importance is also supported by the fact that this is one of the most extensive Pueblo rock art sites in the entire area. Pueblo ethnographic data that confirm such relationships between landforms and resident spiritual entities are understood to exist via traditional narratives and rock art production. There is no known Ute or other historic rock art at this site, meaning that it is rather unique for at least this part of Sandstone Canyon.

4.4. 5MT22208

Site 5MT22208 is located around 1 km southwest of the Strawman Panel, almost at the confluence of the Sandstone and Yellow Jacket canyons. The site overlooks a wide canyon bottom, which continues southwards and is now quite heavily overgrown with various grasses and shrubs, including greasewood and big sagebrush. The site consists of remnants of two masonry rooms in the northern part, some sparse artifact scatters and distinctive rock art. The site has been highly vandalized, as evidenced by a large pit, possibly made by looters, in the central part of the site and with indications of intensive grazing in the locality. It is possible that much of the stone masonry within the alcove was removed early in the twentieth century (Ciomek and Palonka 2024, p. 3). Most of the rock art, ancestral Pueblo in origin, is located in the central part of the back wall of the alcove. This suggests that at least some of the figures may have been contemporary with the rooms, and for example could originally have been placed at the back walls of the sandstone rooms then located within the site (Figure 19) (Chmelir and MacMillan 2015; Ciomek and Palonka 2024).

Basketmaker and/or early ancestral Pueblo petroglyphs at 5MT22208 include at least four ducks (Figure 20) deeply pecked and shown in profile located ca. 2–2.5 m above the ground (however, no broad-shouldered anthropomorphs Basketmaker are found at this site). Very weathered plants forms are also depicted. (Figure 21). Incised elements include a set of textile/pottery designs and miscellaneous other figures and four pecked hands–three right and one left. Bird tracks, rows of dots, and an asterisk are all also within the ancestral Pueblo inventory. Although abstract textile/pottery patterns are common elsewhere on the Colorado Plateau after around 1100 A.D., they are unusual in the CANM sites documented here (but see above 5MT22150). Two groups of schematic pecked anthropomorphic stick-figures with upraised hands, in a few cases shown with tails, are located in the central and southwestern parts of the site, and constitute a further group of representations from the Pueblo period (Figure 22).

Grouped in some panels in the south-central part of the site is a series of human and animal images (10–20/30 cm in height), which include horses and riders fighting or hunting. These representations were made using a technique of delicate incising and are shown in a schematic, minimalistic linear form. Possibly from the early historical period, possibly ca. 1600–1830 in the Early Historic style (Cole 1990, pp. 225–35; Schaafsma 2022; Jim Keyser personal communication 2018, 2024), these figures include horseback riders with typical plains outfits and gear that includes long feathered headdresses and a long feathered staff; one horse has a bridle as well as delicate radiating lines around the mouth, which is also a typical Northern Great Plains feature (Figure 23). A similar depiction is present at 5MT13288 (see Figure 5b). There is one scene in which two fallen men lie to the back and above a rider (Figure 23-3). In addition, there is a pecked bear or bison depicted with claws or feet with several digits, and not hooves. While it is tempting to attribute this set of images to the Utes, typologically they do not resemble Ute horses and paraphernalia as depicted elsewhere in the Four Corners region. Temporal differences and/or different cultural associations could account for such disparities.

The northern and southern parts of the alcove features a large collection of so-called historic inscriptions or contemporary graffiti in the form of signatures, dates, human and horses, place names and ranching marks/stock brands, which are less frequent throughout the entire site (Figure 24). The graffiti includes, for example, a signature from “Heather Tozer 1991” and a few brands of the different members of the Tozer family, an inscription of interest in that the Tozer family has been in the vicinity for at least several decades (Horn 2004, pp. 29, 90, 97, also personal communication and initial identification of some brands with Don Tozer, personal communication 2024).

5. Discussion

Within the greater context of the Canyons of the Ancients National Monument of the central Mesa Verde region, the rock art sites in the lower part of Sandstone Canyon are most notable as documents of the recent historical use of this northern region of the San Juan drainage by both Indigenous Utes as well as White settlers. Secondly, the pre-Hispanic/pre-Columbian ancestral Pueblo rock art, viewed in conjunction with other sites in the Four Corners region, raises new questions about regional relationships during the thirteenth century A.D. and earlier that may be seen through the detailed analyses and interpretation of rock art.

Particularly significant among the Sandstone Canyon sites is the previously described site 5MT22150, where an array of incising and other modes of surface manipulation of the sandstone cliff face occur below a natural “face” formed by geological processes. This location is likely to have had religious significance for hundreds of years, if not millennia. It is proposed that this geological feature prompted the incisions and other rock art nearby. This suggestion is supported by similar markings at other thirteenth century sites in the Canyons of the Ancients National Monument area where cupules and incised features seem to coordinate with various types of natural irregularities in the rock. Ethnographic accounts from Zuni (e.g., Stevenson 1904; Young 1988) describe a similar association between particular cliff features and rock formations thought to embody natural spiritual entities that, in turn, can be approached, petitioned, and likewise manipulated via offerings. This historical example of interaction with features of the rock face provides a basis for the proposed interpretation or hypothesis of ancient sites, opening a route of investigation for sites in Sandstone Canyon and elsewhere in the entire Mesa Verde region (Schaafsma 2022).

Furthermore, the presence of repeated shapes and specific representational icons such as bird tracks defined by grooves and incised lines argues against the practical necessity of only “tool-sharpening” as a satisfactory or primary explanation for these markings; in many cases the grooves are awkwardly placed and not conducive to this type of activity. Other sites, composed of primarily grooves and smaller incisions, have been documented from elsewhere in western/southwestern Colorado, beyond traditional Ancestral Pueblo territories (e.g., Doery 2024). An investigation of a widespread tradition involving rock-grooving practices linked to both practical and spiritual needs may have existed, unconfined by specific cultural boundaries and time frames. More research on the possibility of a regional tradition of this type of marking is merited.

The rock art sites in the lower part of Sandstone Canyon are most notable, however, as documents of the recent historical use of this northern region of the San Juan drainage by both Indigenous Utes as well as White settlers. As described previously, three of the Sandstone Canyon sites have large panels of figurative rock art—among which are horses and cattle—from the late historic period, in addition to numerous inscriptions of names, dates, and brands. The inscriptions and figures carved on these canyon walls document the use of this area in regard to grazing and animal husbandry as well as population movement and as a route of travel to neighboring regions, factors of interest to historians and archaeologists alike.

The bison, one the most frequently depicted animals from these sites, were either known locally or from expeditions from further afield. Nevertheless, without exception all of the bison representations are thought to be Ute in origin. Although some of these figures could date back to the seventeenth or eighteenth century, it is likely that many are nineteenth century in origin. While there are no known representations of bison in Ancestral Pueblo rock art in the Four Corners region, there is also a little evidence of bison in the area of southwestern Colorado during the late pre-Hispanic and historic period (see, for example, Anderson 1966; Martin et al. 2017). More evidence in archaeological remains and historic written records are from much further northwestern and eastern Colorado (e.g., Bailey 2016; Espinosa 1939; Meaney and Van Vuren 1993).

Environmental damage is evident at several of the Sandstone Canyon sites. With the introduction of livestock by Native people as well as White settlers, the natural environment of this area, serving as winter pasture for cattle and sheep, has been severely impacted by erosion. Heavy rains causing massive soil depletion in the uplands or floods due to the effects of overgrazing upstream, have subjected all of these rock art sites to heavy flooding or the effects of being temporarily buried by deep soil deposits left in the canyons as the run-off slowed down. The perpetrators of this disaster left their names and pictures on the cliffs. “Severe overgrazing by enormous numbers of cattle and sheep in the 1880s and early 1890s, and a 10-year-long drought culminating in 1896 resulted in starvation of animals on the range in southeastern Utah and, presumably, in southwestern Colorado” (see Horn 2004, p. 11, who notes that the overgrazing problems lasted through the 1930s). Meanwhile severe floods also occurred in 1911 and during 1916–1917. The remains of deep mud deposits, later removed by subsequent flooding, are evident today in the soil remnants lodged in pockets and recesses, high on the cliffs. At sites in Sandstone Canyon, Basketmaker and Pueblo period rock art on the lower portions of the cliffs have been scrubbed and damaged or otherwise impacted by these large-scale events.

6. Summary and Conclusions

Within the Canyons of the Ancients National Monument, the analyzed rock art sites from the lower part of Sandstone Canyon, alone within the Canyons of the Ancients National Monument, raise many questions about the human history of the wider central Mesa Verde region (today southwestern Colorado and southeastern Utah) and beyond. The rock art described here consists of large panels of abstract, geometric, and figurative Indigenous imagery (mostly petroglyphs) from both the pre-Hispanic and historic periods in addition to numerous Spanish and American historical inscriptions of initials, names, place names, dates, and cattle/horse brands.

Based on the initial results of our research, and for practical purposes, the last 2000 years of the human history of this canyon can be divided into three major periods or components (Palonka et al. 2023b; Schaafsma 2022): (1) pre-Hispanic rock art made by sedentary Ancestral Pueblo societies: (200 B.C.–1280/1300 A.D.); (2) Historic Ute rock art and inscriptions: mostly Ute tribes, possibly also Diné in a few cases (ca. 15th/17th–early 20th c.); and (3) Late historical inscriptions (ca. 1870s.–20th c.): Spanish and American missionaries, travelers, cowboys, shepherds, and ranchers. Regarding the indigenous rock art, historic Ute depictions, including narrative scenes of hunting on deer/elk, bighorn sheep, and bison, but also fights, prevail in at least two sites with abundant Ute petroglyphs.

Sandstone Canyon, together with the nearby Burro and Yellow Jacket canyons, as well as other canyons in the area, formed natural travel routes that were certainly used in pre-Hispanic times, by historic Native Americans and by early cattle ranchers, shepherds, and travelers. They formed the shortest travel distance between the Dolores River and the San Juan River, the two biggest rivers in the area (both rivers have confluences with the Colorado River in Utah) and the routes were accompanied by a variety of associated sites both, habitation and ritual, but today the most frequent archaeological evidence of human presence there is rock art. The accumulation of rock art in the area may be even greater than in other parts of the Mesa Verde region and may be connected with the location of the large rock art sites in or close to the confluence of the Sandstone and Yellow Jacket canyons and with specific landform features (the latter being the case in 5MT22150, but also possibly also for 5MT22149).

As noted previously, parts of the sites are underground, probably due to catastrophic floods from historic (and possibly also pre-Hispanic) times; some of these events are known from historic sources and oral histories (e.g., Head 2017). This environmental impact on the local landscapes and human–environmental relations could be seen through the detailed analyses of rock art sites and its context, including the location of some petroglyphs underground and, as mentioned above, through the traces of clay and other materials found in crevices between particular panels and petroglyphs, which act as evidence of past floods and other episodes.

This research was facilitated by multiple types of advanced technologies. The use of high-resolution photographic documentation and 3D digitization via photogrammetry, terrestrial laser scanning (TLS), reflectance transformation imaging (RTI), and DStretch software and other analyses allowed us to capture and document rock art from the lower Sandstone Canyon to produce a detailed inventory of this cultural heritage. One of the benefits of extensive digitization of sites with abundant rock art images is the relatively short amount of time devoted to the documentation in the field. In our case this was important, because of the limited number of times we were able to use the private road leading to the sites and other factors that limited our time in the field. Digital documentation also offers different possibilities for the documentation of rock art sites in the context of the surrounding landscape (for example, using panoramic photography) and analysis of individual objects and entire panels virtually. Finally, the analyzing and processing of data in a virtual environment may be a solution for other documentation techniques, whose application in the field may be difficult (for example hard-to-reach and large areas and areas affected by weather conditions and/or with limited access).