Let's get one thing straight about the current rash of news about Yellowstone and its eruptions: really, nothing much has changed. Sure, a lot of news media took the angle that a shocking new study shows that Yellowstone erupts more often than we might have thought and, yes, we're all doomed and such. However, if you look at the new findings (and pause to think about it), very little has fundamentally changed about our understanding of the giant caldera in western Wyoming. The new study in question by Ben Ellis and others (in June's Quaternary Geology) takes apart the largest eruption from the current caldera, the Huckleberry Ridge Tuff (HRT). They used Argon-Argon isotope dating to carefully refine the ages of the deposits from the ~2500 km^3 Huckleberry Ridge Tuff - and instead of getting three ages that were indistinguishable within error of the analyses, the three members of the HRT come out to have slightly different ages that are outside of the error of the analyses. So, whereas before all three members of the HRT could only be resolved to be about 2.06 million years old, the new Ar/Ar ages show that the HRT A unit (~1340 km^3) is 2.135 ± 0.006 Ma, the HRT B unit (~820 km^3) is 2.131 ± 0.008 Ma and the HRT C unit (~290 km^3) is 2.113 ± 0.004 Ma. Using these new dates, it looks like the HRT took at least two eruptions, possibly separated by thousands of years, to be deposited. This is in contrast to the idea that the entire 2,500 km^3 was erupted in one monstrous event. This is not to say that the subdivided HRT is small - they are each bigger than pretty much every other eruption during recorded human history, but ~1,340 km^3 is a lot less than 2,500 km^3, volcanically speaking. Now, this isn't really a revelation by any means. If you remember one of my posts from the 2011 AGU Meeting, I discussed a great talk I saw by Colin Wilson that looked at just this issue - how long does it take to deposit some of these so-called "supereruptions"? He looked at giant events in New Zealand, Long Valley and, yes, Yellowstone, but instead of using refined dating, he looked for physical features in the deposits that tell us about time. Does the ash look like there was time for it to be moved by wind or rain before the next layer? Is there enough time for soils to develop during the course of the deposit? He concluded that many of these large eruptions took weeks to hundreds of years (or more) to be deposits and many large deposits are likely multiple events spaced over time. This new study by Ellis and others just confirms some of these ideas through isotopic age dating. For all intents and purposes (geologically speaking), a single HRT is no different than three smaller eruptions to form the HRT. However, Dr. Wilson did bring up, back at AGU in December 2011, the idea that if a massive eruption is actually spread over long periods of time (anthropologically speaking), how might that affect how we think about the potential hazards. If the first part of the HRT was erupted and then 4,000 years passed (all of recorded history) before the next piece of the HRT, would that have a larger or smaller effect on potential recovery of civilization from such massive eruptions? That, I think, is the real interesting aspect to studies like this - it isn't the fearmongering of places like Gizmodo or the Daily Mail that a Yellowstone eruption might be that much closer (it isn't ... at all), but rather how could civilization deal with these events when they come separately rather than in bulk.