Tankersley et al. (2022) recently published a brief paper in Nature Scientific Reports, arguing that the Hopewell culture of the first half of the first millennium CE in the North American Midwest was ended, or at least severely helped along toward its demise, by a comet fragment airburst event. The paper is getting quite a bit of (unfortunately uncritical) media coverage, including in very visible outlets such as the Washington Post and in the Smithsonian Magazine. Catastrophes make good headlines of course, but there are some serious questions about the claims and conclusions in this one.

The Claim

The authors report finding iron and silica rich microspherules, elevated iridium and platinum levels, and burned surfaces in Hopewell sites, all consistent, they say, with an airburst event. They say that one of the large earthen mounds for which the Hopewell are famous, is in fact comet shaped, and would have been built to commemorate the event, and that a number of stories that are part of indigenous knowledge detail cosmic events and could be a reflection of this airburst.

Tankersley is part of the Comet Research Group, a team of researchers who have published a number of recent papers arguing that comet fragment airbursts have had a significant impact on human history. They notably support the Younger Dryas Impact Hypothesis, which claims that the sudden and severe cold snap during the last deglaciation about 12kya, and the end of the Clovis culture were the result of an airburst event. In another headline grabbing article, they recently argued (also in Nature Scientific Reports) that a Middle Bronze Age airburst event over the Jordan Valley, destroyed a number of communities, and could have been recorded in the bible as the story of Sodom and Gomorrah (Bunch et al. 2021).

These papers are part of a long conversation, in archaeology and in the historical disciplines in general, between catastrophist and uniformitarian views of change, and between internalist and externalist explanations of history. To simplify and put it briefly, catastrophists, as the name implies, tend to look for catastrophic and sudden events to explain change, while uniformitarians tend to invoke regular, everyday processes at work around us all the time.

Externalists tend to look for causes of change that are external to a system, such as comet impacts, and internalists, if I can call them that, first look at internal dynamics of system change and collapse, such as economic and social strains, for example. There is a lot more to say on this, and I will do post on it soon. For now, I will just say that catastrophists tend to be externalists, and the Tankersley paper, and the Comet Research Group’s work in general, qualify as both.

As someone who more naturally tends toward internalist and uniformitarian views, I will preface my critique of Tankersley et al. by acknowledging that yes, catastrophes do occur, and they can have significant impacts on us and on our societies. It would be difficult to argue otherwise in the middle of a global pandemic, when a massive submarine volcanic eruption has just upended the lives of an entire nation. But even when one generally acknowledges catastrophe as one of the drivers of archaeological change, one must still archaeologically demonstrate a causal link between a particular catastrophe and a particular sequence of change.

The problems

The main problem with Tankersley et al.  is that while the evidence presented could be consistent with an airburst event, it is also consistent with a number of other, simpler, more likely explanations, and with processes that we know were at work at the time and place of the claimed airburst.

There is definitely meteoric material at some of the Hopewell sites investigated by Tankersley et al. There are elevated levels of platinum group elements (mainly iridium and platinum) in some of the sites, which are consistent with meteorites. The soil no doubt does contain micro-fragments of pallasite, a rare type of meteorite. There are, as the authors record, charchoal enriched layers at the sites, and they probably do contain iron-rich microspherules, all of which, the authors argue, proves an airburst event. In fact, none of this is surprising, and none of it requires an airburst event.

Burnt, charcoal enriched layers are familiar to all archaeologists who excavate settlements. As for meteorite fragments, there happens to be a rare, very large pallasite that fell on nearby Kansas about twenty thousand years ago, and that we know provided raw material highly prized in the entire Hopewell cultural sphere for the manufacture of tools and ornaments. About four tons of material from the Brenham meteorite are known today, and we know (and the authors acknowledge) that this material was used, worked, and transformed by the Hopewell people at some of the sites they investigated.

Because the material was worked there, it would be very surprising not to find pallasite contamination in at least some soils from Hopewell sites, even absent any airburst event. The regional distribution of the platinum group anomalies (Figure 1), that the authors argue is evidence of the location and direction of the airburst, is not surprising either, if there was a centre of production of pallasite objects at the time period the authors are examining. It looks like they may have found one of those centres, which in itself would be an important contribution.

Figure 1: Figure 23 from Tankersley et al showing the distribution of iridium anomalies at some Hopewell sites

The authors don’t give the Pt and Ir anomaly values exactly, but we can estimate them from the figures they provide. I put my estimates in Table 1 below. It shows that there is a very strong concentration of anomalous values in a very small area, and the surrounding values are very flat, suggesting they may even be simply environmental background. I would like to see this discussed and ruled out. I don’t see much of a fall-off over the whole area.

Table 1: My estimate of Pt and Ir anomalies at some of the Hopewell sites, based on figures shown in Tankersley et al.

There is another nagging potential explanation for the platinum group element anomalies that should be ruled out. I mention it only in passing because it is not very likely, but it is not impossible that the relatively high values at the Turner site are due to modern environmental contamination from catalytic converters. Turner is very near Cincinnati, which is the largest urban centre near any of the investigated sites. There has been a growing literature on platinum group element contamination from cat converters, especially in and around urban centres (for example, see Ravindra et al. 2004). I would like to see that possibility at least evaluated.

The authors make much of the presence of iron rich microspherules in soil samples from the sites. These have become a bone of contention as an indicator of airburst events between proponents and opponents of the Younger Dryas Impact Hypothesis. The consensus at the moment, if such a thing is possible in this debate, seems to be that while spherules are present in many soil samples, they are related to earthly processes rather than cosmic ones.

They seem to be the result of rapid heating and quenching of soil, which happens on a regular basis at any modern campsite, and no doubt happened on what are now archaeological sites. I have often seen iron rich “macrospherules” in ancient hearth soils, but I admit I have never looked for the micro variety.

Tankersley et al. also mention a regional change in vegetation at the time of their claimed airburst. They identify shifts in the stable carbon isotope signature of local biomass, which of course, is not surprising, and is even expected, in an area with evolving patterns of agriculture, land clearing, and changing population densities over time.

So yes, changes in stable carbon isotope signatures could result from an airburst that causes widespread devastation, but it could also be due to other processes that we know were operating in that time and place. And that’s the main story of this paper and many like it. The evidence presented could be the result of the process claimed, but it could also be the result of other processes known to be operating, and sufficient to produce it.

The conclusion

In other words, there could have been an airburst over Fort Turner 1700 years ago, and none of the evidence presented argues against it. But an airburst is not required to explain the pattern of evidence Tankersley et al. present.

Then there is the comet-shaped earthwork, and there are the stories of cosmic events, which are an important part of indigenous knowledge, and which the authors link to the claimed airburst. Again, absent a Hopewell airburst 1700 years ago, it would be very surprising, even stunning, if indigenous knowledge did not in some way record and account for cosmic events. Because cosmic events do happen, and they do affect us, sometimes quite severely. Plus, they’re visually impressive. On that basis alone I would expect them to be accounted for in stories and in symbology, including possibly architecture.

The airburst claimed by Tankersley et al would rank in magnitude somewhere between the 1908 Tunguska event, estimated at about 15k kilotons of TNT, and the 2013 Chelyabinsk event, estimated at 500 kilotons, and which caused quite a bit of devastation and disruption. So these kinds of things aren’t even that infrequent or unusual, on an archaeological scale.

In fact, a cosmic event, the Brenham meteorite impact twenty thousand years ago, had a profound influence on the Hopewell nineteen thousand years later. They used the material from that impact to organize and channel part of their symbolic, economic, and material lives.

Finally, the paper argues that the airburst could have led to the demise of the Hopewell. Yes, it could have, if airburst there was, but there are several qualifiers needed here. When archaeologists talk about the “end” of a culture, or a period, they really mean that a certain diagnostic type of artefact stopped being made, or that a particular settlement pattern changed, or that an exchange network was reoriented. They don’t necessarily mean literal destruction, although that does also happen, of course.

The relationship between what archaeologists call the end of a culture or period, and what is actually going on in people’s lives at that time and place is often quite unknown. What did the end of what we identify as Hopewell life mean to the people we call Hopewell?

There is some significant archaeologically visible change in the region over the couple of hundred years following the claimed airburst, but that doesn’t necessarily mean that there is a collapse to explain, and so no airburst needed to explain it. Not to mention, coming back to our conversation between the catastrophists and the uniformitarians, and the externalists and internalists, that if there was an actual collapse, there are plenty of processes and dynamics internal to a society that could have led to it.

So for this paper and the other recent comet airburst papers, the most I can say is that the airbursts they are trying to document are not needed to produce they evidence they are trying to explain.


Bunch, T.E., et al. 2021. A Tunguska sized airburst destroyed Tall el-Hammam a Middle Bronze Age city in the Jordan Valley near the Dead Sea. Scientific Reports 11(1): 1–64. https://www.nature.com/articles/s41598-021-97778-3

Ravindra K, L Bencs, R Van Grieken 2004. Platinum group elements in the environment and their health risk, Science of The Total Environment, 318:1-43.

Tankersley, K.B., Meyers, S.D., Meyers, S.A. et al. The Hopewell airburst event, 1699–1567 years ago (252–383 CE). Sci Rep 12, 1706 (2022). https://doi.org/10.1038/s41598-022-05758-y

2 thoughts on “Did a comet airburst destroy the Hopewell? Comment on The Hopewell Airburst Event, 1699-1567 Years Ago (252-383 CE), by Tankersley et al. (2022).

  1. Thank you for this.

    One thing that struck me was those beautifully-drawn and precise-looking contour lines (your Fig 1; their Fig 23). What data, exactly, was used to draw them? As far as I can tell, it’s just from their few points from excavations at important Hopewell sites. Then they drew/fitted some fancy curve, and then added the contour lines. Talk about making the most out of limited data.

    Shouldn’t they also have sampled a huge set of grid of points throughout that entire region of Ohio and Kentucky. If there truly were an airburst I would think that that is what I’d try to do to establish it, and then go back to see its impact on the Hopewell. Not trying to prove anything by going in the reverse direction.


    1. You are absolutely correct. They don’t give a lot of details in the article, but it looks to me like the contours are produced by a kernel density function using only the values for the sampled sites. Adding even one or two more values could potentially change the contours quite a bit.

      A spatially regular grid of core samples would give much better results.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s