Incorrect representation of Barrier Canyon rock art site’s history and other factors invalidate reported dates

Recently, Pederson et al. reported on optically stimulated luminescence (OSL) and radiocarbon (RC) dating of a fallen rock in Horseshoe Canyon, Utah, that contains a pictograph (1). Unfortunately, the authors inexplicably imply that this site is pristine.

Many thousands have accessed and even camped at the Great Gallery site, arriving by foot, horseback, and four-wheel drive, including cowboys, oilmen, and wildcatters with access to dynamite. In 1928, a road was cut into Horseshoe Canyon (to access oil property) at a point about 2.5 miles downstream, and the 1950s uranium boom led to many roads being created in the adjacent Maze area, also giving access from the south and east. For example, a four-wheel drive road traversed Horseshoe Canyon upstream from the site. It was not until 1971 that the canyon was added as a detached unit to Canyonlands National Park, and ranger and volunteer supervision began in the 1990s. The site was never fenced off. Silver (2) quotes a 1992 Park Service site report: “Visitors … have impacted the surfaces and edges of the rocks and boulders in front of the panels.” It is quite likely that many of the rocks were turned painted side down to protect the art from further damage. Recently, a pictograph site about 25 miles from the Great Gallery was vandalized.

In 1940, E. J. Bird led a Work Projects Administration expedition of seven artists who documented and produced murals of the site (including the Holy Ghost panel). Bird (3) reported that a local rancher (Lee Tidwell) told him he came upon oilmen dynamiting this panel and that on this, and many succeeding visits, he saw the painted surfaces of the fallen rocks. Because the artists were documenting perhaps the finest of all Native American paintings in North America, it is inconceivable that they would not have turned over and examined any rocks with downward-facing painted surfaces. Fig. 1 shows the actual rock used for the OSL measurements, not discernable in figure 3 of ref. 1.

Fig. 1.

Photographs of front (Left) and back (Right) views of the rock, taken before movement for OSL measurements on it. Reliable OSL measurements are on quartz and feldspar grains from areas on the lower surface of the rock not exposed to light. Clocking of the time interval for accumulation of the trapped electronic charge build-up initiates at the time of burial and continues until exposure to light. Notice that this rock, if not moved after its fall from the panel above it, landed in a tilted position. Other rocks are below it and now covered with sand, indicating that this rock fell on top of previously fallen rocks and suggesting that all of its sides were originally exposed to light. Aeolian sand and perhaps even the leaf are most likely to have filled in some of the vacant regions below the rock sometime after its fall. Even at the time of this picture there remain regions at the corners of the lower side that are exposed to light (Right). Patina along the rock edges has been removed, either from abrasion, scraping, or past turning over or other movements of the rock. The rock is in close vicinity to an area shown in the left photograph, which has been a gathering spot for visitors for at least 90 y. Note the extreme abrasion of the edges of those three rocks. Other rocks in the immediate area show clear evidence of rearrangement over time. The above evidence supports the premise that this rock also has been moved or turned over (perhaps many times) before OSL measurement.

The leaf, RC dated to ∼900 y ago, could have been introduced during the many historical period visits or prehistorically. Because of its large surface-to-volume ratio, a leaf is not considered an ideal RC dating specimen, and if RC dated, very stringent storage and preparation measures, not mentioned in Pederson et al. (1), must be used (ref. 4 and references therein) to avoid, for example, younger microorganism carbon contamination.

The “agreement” of Pederson et al.’s (1) date with that of Watchman (5) on a specimen (obtained by R.R. and N.S.) is meaningless, because the Wood’s Hole accelerator mass spectrometer facility did not receive an application with the sample stating that a very minute (<20 μg) amount of incidentally incorporated C was to be dated, and consequently used preparation procedures designed for much larger C content, ∼120–300 μg, (5). In fact, a preliminary date obtained by Watchman on the Black Dragon Barrier site (with higher C content) was thousands of years older.