Over the course of the last year, much of the journalism about climate change has shifted from explanatory science stories to solutions-oriented politics and business stories. Scientific “consensus” on one fundamental point-that humans are causing global warming-encouraged the shift, but there are still many unresolved questions about the impact of warming.
What happens, though, when two studies, published within a week of each other, come to completely opposite conclusions? Does a lack of scientific consensus translate into a lack of journalistic consensus about coverage? A study by British researchers published yesterday in the journal Nature concluded that warmer waters in the Atlantic Ocean increase hurricane frequency and intensity. Last week, however, scientists from the National Oceanic and Atmospheric Administration published findings in Geophysical Research Letters that suggested the opposite-that warming oceans, by increasing scissor-like vertical wind shear, may decrease the number of Atlantic hurricanes that hit the United States.
Neither study got a tremendous amount of coverage. The Miami Herald and the South Florida Sun-Sentinel each dedicated an article to the NOAA research, predicting fewer hurricanes, which seems to make sense given Florida’s vulnerability to hurricanes and sea-level rise. But neither paper picked up the British study, predicting greater frequency and intensity. In fact, there was not a lot of original reporting on the later research, although an article by the Associated Press got fairly widespread play. The author of that piece, Seth Borenstein, appears to be the only reporter who covered both studies, with back-to- back articles.
The contrast in the coverage-a mixed bag of original reporting and stories grabbed from the wire-is somewhat striking and raises a question: with the plethora of scientific “discoveries” every day, how do newspapers choose which ones to cover? Most climate journalists know that no single study is going to provide a real “breakthrough,” and the hypotheses underpinning the two hurricane studies this week are certainly not new. So how does the press judge which individual studies to report? The strength of the science? The attractiveness of the findings? The potential for conflict (The NOAA study received a second day of AP coverage after it provoked a verbal dispute at the annual American Meteorological Society meeting)?
Or have journalists simply grown dependant on the existence (or at least the impression) of scientific consensus? The mainstream media seem less willing to go out on a limb to cover smaller, individual studies than to write broader pieces on breaking news from institutions like the Intergovernmental Panel on Climate Change. When the IPCC released the first part of its influential assessment report in February 2007 (which produced the fundamental finding that humans play a role in global warming), papers from The New York Times to the Omaha World-Herald in Nebraska dedicated space to it.
That aligns with how Nils Bruzelius, science editor at The Washington Post, decides what to cover. He said that his paper only writes about single studies when the results are new and different:
In the case of the hurricane thing … there have been dozens of studies on this topic. When there’s sort of an ongoing controversy, that makes us less likely to write it as a daily story. When we’ve accumulated enough material, we’ll do a broader story.
For wire services like the AP, whose mission is to provide round-the-clock news to thousands of member publications, a strategy focusing on more in-depth pieces won’t fly. With an increasing demand for science news, particularly online, the AP is writing more science stories than ever, said health and science editor Kit Frieden. Who publishes a study is occasionally a factor in whether the wire service covers it or not, but it is far from the only factor. Frieden explains:
In general, in choosing which studies to write, we try to stick to research from the best peer-reviewed journals, but have no specific rules about that. We make our judgments based on the subject matter and how relevant it is to AP’s broad readership and the quality of the research. There is tremendous interest in research on hurricanes and climate change.
It appears to me that many science journalists simply assume that whatever is published in the leading journals reflects the consensus view. A reasonable shortcut for normal science, but a perilous one for hotly contested subjects. Still, what is the real life alternative?
Posted by Joao T. da Costa
on Fri 1 Feb 2008 at 02:13 PM
Let me suggest one reason why this pair of studies didn't garner more coverage: There is less duel here than meets the eye, although you have to wade through the weeds in each paper to figure this out.
Team No. 1 looked at the impact of warming waters in the tropical Atlantic have on hurricane trends. Warm water fuels hurricanes. The British researchers crunched the numbers and estimated that a warming trend in the tropical Atlantic Ocean accounts for about 40 percent of the increase in hurricane activity from 1996 to 2005. (That leaves a whopping 60 percent of the cause unidentified, at least in this study. Could be natural variability?) The upshot: Some studies implicate global warming in heating up the Atlantic. If waters continue to warm, one might expect to see a continued long-term increase in hurricane activity. The scientists involved stopped short of making that call, at least in the research paper they published. But the team's bottom line -- that waters have warmed and appear to have led to increased hurricane activity -- is not news. Several studies over the past three years have come to similar conclusions. And those studies have received a lot of attention. On the topic broadly rendered, this adds nothing fresh to the discussion.
So did Team 1 waste its time? Certainly not. First, the scientists are adding their increment to a body of knowledge. That's how science generally works. It's incremental, even glacial, in some cases. (Although global warming may redefine the speed of "glacial"!) Second, they are trying to quantifying the relative contribution warming waters make to trends in hurricane activity. Assuming other teams come up with similar results using independent tests, the work should help modelers give the proper weight to changing sea-surface temperatures as they try to refine their projections of global warming's effect on tropical cyclones.
Enter Team 2, which also looked at rising temperatures in tropical oceans -- including the Atlantic. But this duo was interested in how those changes affect broad wind patterns over the Atlantic's hurricane-forming region. (As a lay reader, I have to say this paper wanders too much and is far less clearly argued.) The team explains that when water warms in the tropical Atlantic, it encourages regional wind conditions that coddle hurricanes. But conditions in the tropical Indian Ocean and eastern tropical Pacific, off South America, have their say as well. So the team looked at those regions, too. They note that increasing water temperatures in these far-flung regions coincide with wind patterns over the tropical Atlantic that stifle hurricane formation. No news there, either; that's how El Nino works. El Nino sends a warm pool of water in the western tropical Pacific east toward South America. That eastward-ho migration has a long reach: It changes atmospheric circulation patterns over the Atlantic hurricane-forming region in ways that stunt hurricane growth. This phenomenon has come up widely and frequently in stories about hurricanes, hurricane forecasting, and climate for years. The team's analysis indicates that those far-flung changes tend to win out over the more 'cane-friendly wind conditions a warming tropical Atlantic would bring. So Team 2 suggests that long-term trends in hurricane activity probably will be determined in large part by the relative impact global warming has on these three different ocean regions. The implication: If global warming affects the Indian and eastern Pacific oceans more than it does the tropical Atlantic, it could set up persistent El Nino-like conditions that would allow fewer Atlantic hurricanes to form.
This pair had the added "burden" of working for the National Oceanic and Atmospheric Administration's Atlantic Oceanographic and Meteorological Laboratory in Miami -- that purported hot-bed of dissent on the hurricane/climate issue. (Although when you talk to several of the researchers at length, their stances based on current evidence are more nuanced than they often are portrayed.)
Ah ha! Conflict between the two! Well, not really. Why? Team 1 acknowledges that it intentionally subtracts wind shear from its calculations so it can isolate the contribution warm waters make. And it notes that during El Nino years, hurricane activity would fall well below what it might otherwise be if warming water was all that mattered. This essentially acknowledges the broad effect that conditions in the eastern Pacific can have (at least on El Nino time scales). And Team 2 acknowledges that other factors, including other atmospheric circulation changes, as well as the increased water vapor that rises from warmer oceans, also are important and not fully understood.
Neither paper gives a best estimate of how these two factors -- wind shear and water temperatures locally and globally -- play out in tandem and into the future. Undoubtedly, that is fodder for later research. But that vacuum, perhaps, is what sucked some into seeing a contradiction.
Interestingly, the true disagreement between these two groups deals less with the future and more with the present: Whether Atlantic hurricane activity has been increasing, and if it has, how much of the change should be attributed to global warming and how much to natural variability. Team 1 notes that sea-surface warming in the Atlantic is consistent with model projections of global warming's impact there. That would suggest that 40 percent of the increase in hurricane activity has global warming's fingerprints all over it. But trends in Atlantic hurricane activity, let alone tropical cyclone activity in other ocean basins, trigger heated discussions among researchers. Team 2 says it sees a slight decrease in the number of land-falling Atlantic hurricanes, and so that should apply to all Atlantic hurricanes. That approach also has elicited yelps. But debates over the appropriate way to gauge trends in hurricane activity aren't new. And neither paper adds much, if anything, to the broader debate of whether or not global warming's fingerprints stand out in current hurricane trends.
So in terms of the broad conclusions, the two groups don't disagree. They are a bit like the proverbial blind men groping at different parts of the same elephant. Through the science journals, they bring to the subject bits of additional evidence or the initial results of a slightly more inclusive look at the problem.
This is a complex subject, as my science-writing colleagues realize. We need to keep readers informed about developments. But we also run the risk of falling into a trap that has sometimes snared medical coverage -- x reduces the risk of or cures y one week, based on the latest study. Then, a few weeks later, x doesn't reduce the risk, or has too many nasty side effects. And three weeks later.... Medical writers and researchers have expressed concerns about such coverage and its potential to confuse even well-informed readers who may desperately be seeking cures. The same might be said for the hurricane/global warming issue, where dueling groups of scientists trying in their own way to sort through a difficult subject stand at 20 paces and fling peer-reviewed studies and Powerpoint slides at each other. From a coverage perspective, in some cases a more-prudent course may well be fewer stories with more context, rather than quick hits on individual results -- unless they significantly add to the discussion. This certainly was the case in 2005, when Kerry Emanuel at MIT published a paper on global warming and what he detected as a decades-long trend toward more-powerful tropical cyclones. Up to that point, as one of the leading experts in the tropical-cyclone field, he'd been a self-described agnostic on the issue. But his outlook changed when he took a fresh look at the available data.
Posted by Pete Spotts
on Sun 10 Feb 2008 at 08:17 PM