A brevia paper in Science last week titled "Post-Wildfire Logging Hinders Regeneration and Increases Fire Risk" [PDF] touched off a storm of controversy in my neck of the woods. The topic of this paper is very political - as acknowledged by the authors in their first sentence: "Recent increases in wildfire activity in the United States have intensified controversies surrounding the management of public forests after large fires." A few years ago, President Bush pushed forward his "healthy forest initiative" aimed at timber harvest as a fire preventative measure and reducing the obstacles to post-fire salvage logging.
The paper, based on a M.S. thesis at Oregon State, looked at the effects of salvage logging in areas burned by the Biscuit Fire in the summer of 2002. The authors measured seedling density and fuel loads in 17 plots during spring 2004 and 2005. They found that seedling densities were 71% lower in areas that were burned and then salvaged than those that were only burned. They also found that both fine and coarse woody fuels were significantly higher in the logged areas than in the unburned and burned/not logged areas.
Based on their data, the authors make several broad conclusions with management implications: "Our data show that postfire logging, by removing naturally seeded conifers and increasing surface fuel loads, can be counter-productive to goals of forest regeneration and fuel reduction. In addition, forest regeneration is not neccessarily in crisis across all burned forest landscapes."
The controversy began when the paper was posted on-line a few weeks before its print publication. Several faculty members in the College of Forestry at Oregon State University immediately began to criticize the study, suggesting that its conclusions and implications were too broad based on the dataset. As reported in the Oregonian: "[Nine] professors and scientists from OSU, along with the Forest Service, took the unusual step of asking the Science editors to delay publication of the study until it addresses their criticisms. Alternatively, they asked that their concerns be included in a letter accompanying the study." Several of these scientists receive significant funding from the timber industry, and the dean of the college (not among the letter writers) has testified before Congress in favor of salvage logging.
Science declined to delay publication of the article nor did it oblige to put a letter of criticism accompanying the article. However, one controversial sentence ("The results presented here suggest that postfire logging may conflict with ecosystem recovery goals.") was removed by the authors between on-line and print publication (with no public explanation given).
So here we have some (though by no means, all) members of a university faculty publicly airing their internal dirty laundry and suggesting censorship rather than open debate. The customary method of science is that if you don't believe someone else's conclusions, you try to find data to support your position. That way, down the road, a third party can read all the relevant articles, evaluate the data for themselves, and draw their own conclusions. By custom then, the dissenting professors were totally on the wrong side of the fence.
But at a deeper level, it's more complex. First off, Science is one of the top 3 journals in all of science. This means that its rare for a topic like salvage logging to appear on its hallowed pages. And when a paper, like this one, does make the cut, it will be much more widely read and discussed than a paper appearing in some run-of-the-mill forestry journal. It'll have a higher impact - and it may end up being a more highly regarded paper whatever the relative quality of the science. So perhaps, the dissenters were right to be concerned that what they viewed as poor conclusions were going to get major attention and that their rebutting papers were not. Is that fair to readers outside of that specialty? Or to policy makers making decisions on limited information?
A second issue, in my opinion, is the <1 page nature of the paper that appeared in Science. I've now read it 4 times and I am still a bit uncertain about the details of the study. With more space, I could have seen a map of the sites and gotten more information about the methods. This information is available as a supporting document on-line, but not very many readers are likely to go look at it. In a traditional, longer format, details of methods, etc. would have been included along with the data and conclusions, allowing the reader to immediately decide for themselves whether the scope of the conclusions was warranted. How do journals like Science decide the correct balance between number of articles and depth of details presented in print? How does this effect the understanding of readers specific to the discipline and to those who are exposed to an article only because it catches their attention in a high-impact journal (or resulting media coverage)?
Finally, I feel for the grad student at the center of this firestorm. His advisor is on the paper, but other members of his department are publicly criticizing his research. Should there be standards of conduct for treatment of students and colleagues when research becomes controversial? Should it matter that the criticism is coming from his own institution? How does the instituional administration deal with issues like this? (No comment from OSU).
As for me, I'm glad that the paper got printed with its sweeping conclusions. Not only does it renew the debate about how to manage burned forests, but it also opens up discussion about academic ethics. And both of those conversations are ones that we need to be having.
For more discussion: