It looks like the 23ky old human footprints at White Sands are solid

Two years ago, Bennet et al. (2021) made a lot of headlines with a claim of (quite amazing) 23 000 year old human footprints on the margin of paleolake Otero, at White Sands National Park in New Mexico. If this was true, it would be the oldest securely dated archaeological site in the Americas by a wide margin. It would be solid evidence of human activity during the Last Glacial Maximum (LGM) and would push the boundaries of archaeology in the Americas by close to ten thousand years.

The question of when and how humans came to the Americas would then be wide-open. If we have solid LGM archaeological sites, then we may have very much older sites as well, since humans would have had a hard time reaching paleolake Otero at the height of a glaciation. That would indicate they were here before. Perhaps much before. But let’s not get ahead of ourselves.

Problem with the initial claim

The claim was very promising, but there was one significant unanswered question. The dating of the site was done on Ruppia cirrhosa, an aquatic plant whose seeds were found in the same layer as the footprints, and sometimes embedded in them. As an aquatic plant, Ruppia can take up old carbon from the water in which it grows, sometimes leading to C14 ages that are older than the actual seeds or plants themselves. In other words, the aquatic plants could have made the footprints, and the entire site, look older than they actually are (Madsen et al. 2022, Pigati et al. 2022).

Pigati et al. (2023) have now published a new study in which they directly take on this critique of their original report. They bring in three additional lines of evidence that are independent of the original carbon dates taken from the Ruppia seeds.

First, they isolated the pollen of terrestrial plants found in the same layers as the footprints and the Ruppia seeds, which actually provides two kinds of evidence: Carbon dates on the pollen, and species identification and frequencies. Second, they managed to get Optically Stimulated Luminescence (OSL) dates on quartz grains in those same layers. Both of those are difficult to achieve and to get right.

The results in brief

In brief, both the OSL dates on quartz and the carbon dates on the terrestrial pollen are consistent with the dates originally taken from the Ruppia seeds. They all come in at between 21k and 23k years old, which would indicate that the original dates on Ruppia did not suffer from a major old carbon effect.

The terrestrial pollen, including pine as a dominant tree species, is consistent with the colder climate of the LGM, and denotes an ecology quite different from the post-glacial environment of the region. If the dates of the tree pollen and the aquatic plant pollen had been similar, but the trees had been warm weather species, that would still have been a major problem.

The association

One key problem in archaeology is that we are often not getting a date (radiometric date, for example) on the objects we are directly interested in. In this case, archaeologists are not dating the footprints directly. We often date objects that are found in association with the things in which we are really interested. In this case, the archaeologists are dating seeds, pollen, and quartz grains found in or near the footprints.

Figure 3 from Bennet et al 2023 showing the location of some of the dated samples

It is therefore important to establish an association between all the things we are dating, and the things we are interested in. Here, the associations seem solid. The maps and diagrams provided with both the original report and this updated one show that the aquatic seeds, the terrestrial pollen, and the quartz grains all come from the same layers as, and in close proximity to, the footprints of interest.

Are there remaining questions?

At this point, it becomes incumbent on potential critics to propose either mechanisms that could bias all four lines of evidence in the same direction and by the same magnitude (aquatic pollen, terrestrial pollen, OSL dating, and plant species distribution), or to show methodological errors in all three dating processes, or to show that the association between all four elements (and the footprints) are faulty.

This is getting to be a tall order. The carbon dating of the terrestrial and aquatic plant remains respond to different biasing mechanisms, and the OSL dating works on completely separate basic principles from carbon dating. The plant species distribution that looks more like a glacial than post glacial environment is just a nice bonus at this point.

Potentially the weakest link in the argument here is the association between all those remains and the footprints, but as far as I can tell from the published material, it looks good. It certainly doesn’t look horrible.

The various specialists in carbon dating, in OSL, and in pollen recovery and identification will chime in over time, and I will read their reactions carefully. But at this point, the overall assessment has to be that we most likely have a solid 23k year old archaeological site in New Mexico. And if that’s true, then there is probably much older stuff to be found.

References

Bennett et al. 2021. Evidence of humans in North America during the Last Glacial Maximum. Science 373, 1528–1531 https://www.science.org/doi/full/10.1126/science.abg7586

Madsen et al. 2022. Comment on “Evidence of humans in North America during the Last Glacial Maximum”. Science 374 https://www.science.org/doi/10.1126/science.abm4678

Pigati et al 2022 Response to Comment on “Evidence of humans in North America during the Last Glacial Maximum”. Science 375 https://www.science.org/doi/10.1126/science.abm6987

Pigati et al. 2023. Independent age estimates resolve the controversy of ancient human footprints at White Sands, Science 382, 73–75. https://www.science.org/doi/10.1126/science.adh5007