What gets left out…and why (High Plains hydration paper)
So again dipping into the shameless self-promotion file, GG is lead on a new paper in Geology (sorry, it lives behind a paywall but we’ll figure something out; the GSA press release blurb can be found here). The paper argues for the possibility that large parts of North America–most obviously the High Plains–were elevated by water reaching the lower crust, reacting with garnet to produce a new mineral assemblage lower in density. (You can see a little animation summarizing this idea here and the CU press release is here).
In this case, we have a new and different idea for how continental elevations can be changed and we want to share it with the broadest possible audience. Why is that important? Geoscientists only develop their experiments and tests with a specific set of hypotheses in mind. Although we hope to pursue this further ourselves, we’d like to see this notion be part of what other scientists consider as possible when they go about making new measurements. Somebody else might well have a toolkit that can test this better than we can. A short paper, where we can air the idea enough to make it a plausible hypothesis, is one that is far more inviting than a 20 page pulse-deadener in JGR or GJI. So if we could make the paper fit without dropping large chunks into supplemental material, publishing in a letter journal made sense.
Publishing in a letter journal carries the risky opportunity of press coverage. The press watch higher profile journals like Science and Nature and, increasingly it seems, Geology. Well, they watch the press releases associated with these journals, and the journals have increasingly sought to get coverage for their papers. Why is this risky? It is easy to discover when you read the news story that you have solved all the problems mentioned in your paper; this can get to the point where colleagues are upset that you could possibly say such twaddle. Press releases and newspaper reporters would prefer to start with a line like “A report today in the journal Geology solves the long-standing problem of …”–which is utter rot. In this case, we are adding to a rather lengthy list of possibilities for uplift; it would be disproving some of these that would come closer to solving anything. But there is an opportunity–getting science in front of the public is a plus if you can keep the message on target. It just takes some real careful discussion with the press people in the university and any reporter you talk to. And, frankly, this is a topic where the people who pay for us to do science might actually be interested in some of what we have found.
One reason we could go to a letter journal was that there was precious little new data in this paper. We didn’t collect or analyze any of the seismic data and the xenolith data was already published, so the paper was really just connecting existing dots in a different way.
But what of the downsides? These can be considerable. Consider the beginning of the first paragraph of our paper:
The diversity of tectonic and magmatic events in the western United States has spawned many explanations for the modern high elevation of the region in whole or in part, including dynamic topography, crustal thickening, magmatic underplating, mantle depletion, lower crustal flow, vertically non-uniform extension, delamination and/or foundering, mantle hydration, and lithospheric erosion.
Note there is not a single citation in the pile. Is this because we ourselves are making these suggestions? Not at all. There is no citation because, simply, there was no way we could burn nearly a quarter of a page on adding all the appropriate citations. Similarly, we don’t try to careful rebut all of these ideas; that would be a much longer paper and is beside our point (we do provide a quick summary of some objections to many of these ideas). We really want to just add an idea to the mix in this paper, not pretend it alone is the ultimate answer. We make a few cursory notes of why these existing ideas are unsatisfactory at some level before moving on to provide some justification for our kooky idea.
Reviewers kind of wanted us to do more to consider the relative role of hydration of the mantle as well as of the crust, and undoubtably some readers will disagree with our terse dismissal of their favorite ideas. Although we tried to make clear that we couldn’t rule out a contribution from mantle hydration, spending too much space on that (already published) idea would just dilute our main message that we should look to the crust as well as a possible source of changing buoyancy. Again, it isn’t just the place to evaluate the relative roles of all possibilities (it is possible, had we gotten some different reviewers, that we might have been asked to clarify our dismissal of lower crustal flow); in not engaging in that discussion, we risk some readers feeling that we had erred in dismissing some other hypotheses.
And of course there is the (un)cited literature problem. A lot of relevant papers were not cited; there simply wasn’t room. A reviewer suggested several papers that were relevant but we chose not to add that information (reaching the level of citation the reviewer wanted would have made us want to add several additional citations to the paper). This was kind of painful: we had to cite some papers for very specific points, but there were many others bearing on issues important to the paper where we had to decide which to cite and which to overlook. And, of course, in not citing some papers, we risk readers thinking we were unaware of issues brought up in those papers (no doubt in some cases, we are not, but in many we probably are). If the goal was to convince readers that an alternative hypothesis was untenable, these constraints would be fatal; here, though, we are just introducing an idea, so the risks of misunderstanding are more acceptable. Perhaps the paper that was most deserving of citation was one by Paul Morgan published in 2003 (Colorado Plateau and southern Rocky Mountains uplift and erosion, in Raynolds, R. G., and Flores, R. M., eds., Cenozoic Systems of the Rocky Mountain Region, Rocky Mountain SEPM, p. 1-31; a preprint lives here) that is one of the few prior papers to even consider a phase change cause for uplift of the Colorado Plateau (we only found that paper after we had pretty much built out our version of the hypothesis, and its focus on the Colorado Plateau made it awkward to cite in this context, but GG wishes we had found a way to cite it).
So that’s why we did what we did. Do we believe we have “solved” the mystery of the High Plains? No. We have precious little reliable information about the lower crust under the Plains. Arguably the pattern of elevation correlating with crustal thickness in the High Plains made in earlier papers doesn’t stand up in Earthscope based studies (though its been a bit hard to see; most of the maps out there only get to eastern Colorado, not far enough to really know if things break down horribly, and you need the crustal velocity structure as well as Moho depth to really test this out). We haven’t dated hydration in xenoliths from Wyoming at all, and only dated it in one xenolith from the Colorado Plateau (and some might dispute the monazite dates or their interpretation–that manuscript has yet to be submitted). the paper is just to open a door on a new possibility, not close off others.