Figure 7b from Thorne and others (2013). This is the conceptual model from their study, showing two mantle piles moving towards each other and combining, eventually creating a new hotspot plume. Present day is marked in Time 2, when the two piles are just beginning to interact. Time 3 represents when a new plume would be initiated. Image: Thorne and others (2013), Earth and Planetary Science Letters. I read a lot of science media. Probably too much and I spend a lot of time and words trying to educate people on where the media goes astray when covering either volcanic events or geologic research ... but sometimes it feels like a losing battle. There is only so much time in the day to deal with every piece of slipshod science journalism I run across. Thankfully, some science media is great, many times coming from journalists and bloggers who deeply care about getting the science right and conveying thoughtfully. However, sometimes, the media coverage of a study just jumps that shark of volcanic hyperbole and leaves me wondering how I can ever keep up with dispelling scientific myths and misdeeds on these interwebs. So, as they say, I will "teach a man to fish." I've decided to disassemble a media report in io9 on a recent research article that has garnered some, ahem, questionable coverage. Now, I don't particularly mean to pick on io9 and the author of said piece, George Dvorsky ... oh wait, no I actually do. It suffers from a few of the classic blunders of science media reporting of research: a lack of understanding of the basics of the science research in question, relying too heavily on a press release, a penchant for hyperbole and inappropriate illustrations. This io9 piece is no by no means the worst of the coverage -- NPR gets that trophy -- but it, in fact, used that NPR story/interview as one of its sources. How to begin this autopsy? When I read an article in the mainstream media (internet or traditional), the first thing I do is IGNORE THE HEADLINE. Few things, especially in this age of pageview-hunting, are as misleading and sensationalist as the headline. For this io9 article, the chosen headline was "A Massive New Volcano May Be Forming in the Pacific." So, at first glance, you're thinking that this is a something happening RIGHT NOW. Maybe tomorrow. For fun, let's compare it to the title of the research study by Michael Thorne and others in
Earth and Planetary Science Letters
that started this whole news: "Mega ultra low velocity zone and mantle flow." Not exactly the best title ever but notice the absolute and utter lack of any mention of massive new volcanoes? Here we are, less than a headline into the story and we're already heading off into some doomsday-fueled wasteland. This lead me to another key issue with many science media article -- can we actually, please, just look at the original study? I don't care if you can critique it for its merits, but just look at it. Even as a layperson, you can look at the title, read the abstract and look at the figures. Do you notice a lot of talk about supervolcanoes in the abstract and figures? Any mentions of doomsday for all of civilization? No? Well, maybe that shouldn't be how you lead off the story, then? Of course, the press release put together by the media arm of the University of Utah didn't exactly help dispel the doomsday scenario, but that is why good science journalists don't rely on merely the press release to understand the study in question. Press release are exactly that -- releases meant to garner press for the institution in question. That very nature makes it problematic to solely base your media writing on the press release rather than at least looking at the original study, possibly calling the author (note: as a scientist, I can say, I do not mind journalists calling me to actually get my research right!) or even calling other geoscientists to get a comment on the study. And there should always be a direct link to the original study. If you're too lazy to find that, then we already know to find your story suspect. Back to the io9 article: Once you've ignored the headline, you can move into the meat of the story. The second sentence in reads "This process, says geologist Michael Thorne, could eventually lead to a cataclysmic eruption that could "cause very massive destruction on Earth." At the very first exit we could take in the story, we get off at "Apocalypse Avenue" on our way to "Armageddonton" (I think that's in Alberta). Sure, the lead author didn't do himself any favors by mentioning massive destruction but remember, these quotes are coming from the NPR piece or the press release. Who knows the context! This is followed by: "But don't panic quite yet. His research suggests that this super-volcano-in-the-making may not erupt for another 100 to 200 million years." These two sentences, on their own, betray the exact goal of the author of the io9 story: I want to talk sexypanicdestruction. Let's get this straight right now. No self-respecting scientist writes an academic paper whose main conclusion is WE'RE ALL DOOMED!!!!! In fact, if we look at the conclusions of the EPSL study in question, Thorne and others say "
Infrequent massive eruptions that define large igneous provinces (LIPs) may be a natural consequence of periodically merging lowermost mantle thermo- chemical piles
." So, maybe, we might get a large eruption to occur. Maybe. If their models are correct. Nowhere does it say "which will kill all the humans." You have to wonder if the io9 author has actually laid eyes on the study ("The new study, which is set for publication in this month's issue..."). He goes on to give some fuzzy explanation, which itself is a fuzzier version of what is in the press release and/or NPR story. That is okay considering that maybe we're not all seismologists and geophysicist, but as I mentioned above, even a layperson can glean something from the original research article if you bother to find it. After ignoring the secondary header ("Pick Your Poison"), you run into the point where the io9 story goes off the rails. Thorne and others (2013) mention hotspots because one of their main conclusions is that the collision of mantle piles might cause a new hotspot plume to form. The problem is that if you read the io9 story, you would think that the research paper spent pages on Yellowstone and flood basalts and death. However, no where in the article is there a SINGLE MENTION of Yellowstone, Deccan Traps or extinction (complete with dead dinosaurs). Trust me, I searched. But wait, the University of Utah press release sure mentions this stuff! So, instead of looking at what the study actually says, the io9 piece decides to tell us about the press release, not the actual research. Do we see the disconnect here? We're now reporting on reporting on the research? To top this off, the io9 piece, along with the Utah press release, gets the science wrong! The eruption of Yellowstone 2 million years ago might have been related to the hotspot, but both pieces make it sound like a specific event called a "hotspot plume". There are lots of hotspots out there, including places like Hawaii. The plume drives volcanism for millions of years and isn't a singular event. Even the second scenario listed in the io9 and Utah pieces, a flood basalt province, is also hotspot-related! So, when you read extractions of extractions like this, you can usually tell because the descriptions and explanations just sound vague. Even a basic look at a geology text can tell you that Yellowstone and the Deccan Traps are likely both related to hotspots. The io9 pieces goes on: "Needless to say, these eruptions are extremely disruptive to ecosystems, and may be tied to some extinction events; ash and volcanic gases make life difficult for organisms (including massive die-offs in the oceans caused by oxygen loss)." Again, the io9 author decides he NEEDS to make the connect with mass extinction and offers a vague and underdeveloped explanation. This is where you, the reader, can look at the original study and wonder from where is this doomsday-talk is arising.
How io9 choose to represent their story on Thorne and others (2013). The study must be terrifying, right? Image:io9 Before we get too far, one thing that should stand out in this io9 piece are the visuals. There are two images shows (1) a still from "Supervolcano" showing a hypothetical eruption of Yellowstone (see above) and (2) poor dead dinosaurs trying to drink from a lava fountain. Neither of these have anything to do with the study. Nothing. The third visual is actually from Thorne and others (2013), but is small, out of context and lacks any caption. If you want to come up with ways to have visuals add nothing to your story, this io9 pieces is a great example. Don't be fooled by visuals in science news articles -- if they don't have a caption or aren't referred to in context, they likely are only there for effect rather than understanding. We've now survived four paragraphs. At this point, you're either wondering "is he actually going to talk about this study" or "this study must jump around a lot based on this description" or "where's my sandwich?" After ignoring the second secondary header ("Spongy molten blobs of doom" ... tonight on Sick Sad World), you suddenly realize that the io9 article hasn't even started talking about the study! Yup, you're now spent all this time and you've only read an article about stuff from a press release and another article about the study. Of course, the main author of the study (Michael Thorne) isn't doing himself any favors -- especially when you're only allowed to get snippets of what they have said about their study (note to scientists: if you blog about your research, then you can offer all the context you want, the way you want it). The io9 piece grabs onto the quote "squeezing this huge molten blob at the middle of it like some kind of balloon, and it is going on right underneath us." Now, if that isn't vague I don't know what is -- we're venturing into Philip J. Fry "it's like a balloon and something bad happens" terrible analogy territory. This is where, we, the audience, demand clarification! What the heck does Thorne mean by this? More clarity, more depth! Help us understand. It is a journalists job to make that happen by delving deeper or talking to sources. In fact, it is the last two paragraphs of the study that get at the real conclusions of Thorne and others (2013): computer models of seismic data suggest that the collision of these two mantle regions could cause a new hotspot plume to initiate. To top it off, in the dregs of the io9 story, and we finally get "It's by no means guaranteed". Yes, we've spent half the article on all the ways that this new study is the end of us all (note: interestingly, most media reports on the study mention this 100-200 million year timescale for a new plume -- this isn't in the article either, but again just from the press release and NPR interview), and now we find out that its (a) based on a model and (b) it may or may not happen. The io9 piece goes onto proclaim that the study used "data extracted from over 50 earthquakes", which we are lead to believe is a lot (and it really isn't). However, no context is given for any of this and without context, we're left wondering what, if anything to get from this. If anything, these sorts of caveats should be up front when discussing the study, rather than saved for the last paragraphs. So, to compare the conclusions of the Thorne and others (2013) study versus the io9 story: Thorne and others (2013): "The lower mantle beneath the Pacific Ocean being composed of two or more individual thermochemical piles that are currently in the process of merging. As individual piles merge, the thermal boundary layer along the tops of the merging thermochemical piles is disrupted, often resulting in anomalously large plumes that can entrain pile and ULVZ material." Translation: They see evidence for merging mantle components that, based on their model, could produce a new hotspot plume. io9: SUPERVOLCANO DOOM WILL KILL US ALL. Translation: SUPERVOLCANO DOOM WILL KILL US ALL (made you look!) When you run into articles like this that make you think that something doesn't seem quite right or the conclusions, according to the media story, seem outlandish, then you should do your own digging. Find the article -- even if you don't have access to the whole thing, the abstract and some of the figures are usually available. Does it seem like the media article and the research paper are telling similar stories? Maybe you can find the press release on the research. Does the media article merely repeat what is in the press release rather than talk directly to the author or offer additional insight? It is your responsibility, as a reader, to be able to identify problematic science media, but all that takes is the ability to know sensationalism and lazy journalism when you see it. If you're tired of bad articles, make sure to seek out the good science journalists -- the Alex Witzes, Charles Petits, Ed Yongs, and Brian Switeks of the science media world -- and read their pieces. Use them as a tool to help identify when other sources take the easy way out to increase their pageviews. Demand that science media is written to increase your understanding, not watered down to create "homeopathic news" that lacks any real substance ... and you deserve better than that.
