“Fuel Treatments Can Address Wildfire Severity”

The Sierra Nevada Adaptive Management Project (SNAMP) is a joint effort by the University of California, state and federal agencies, and the public formed in 2004 to assess how treatments designed by the USDA Forest Service to prevent severe wildfires affect fire risk, sensitive wildlife populations, forest health and water resources. SNAMP is in year five of an ambitious 7-year experiment to evaluate the effectiveness of management strategies to modify fire behavior across the landscape.

SNAMP has examined real-world fires and developed computer models to evaluate wildfire severity and environmental impacts in response to fuel-reduction treatments looking 30 years to the future. In its Northern Sierra project covering roughly 30,000 acres, SNAMP evaluated three different treatment scenarios. In each case, fuels were reduced across approximately one-third of the study area, and all treatments showed substantial reductions in high-intensity wildfire across the landscape, not just treated areas for 20 years after implementation.

This is from California Forests Magazine, and this issue is full of articles about severe wildfires. The whole article is here. The picture is one of mine from the Lassen National Forest’s 1987 Lost Fire.

18 thoughts on ““Fuel Treatments Can Address Wildfire Severity””

  1. The article claims that these treatments benefit fisher, goshawk, and spotted owls, all species that prefer dense fuel-rich forests. Fuel is habitat and fuel reduction is habitat reduction.

    Modelling studies like this typically do not provide an meaningful assessment of the risk to habitat from logging plus fire versus fire alone. Instead they assume that wildfire occurs with 100% certainty in the exact place where fuels were reduced and at the most opportune time after the fuel is reduced. These are not realistic assumptions and strongly undermines claims of habitat benefits from fuel reduction.

    In the real world no one can predict the location, timing, or severity of wildfire. Treatments must be widespread and many acres treated won’t encounter fire, so habitat will be degraded unnecessarily. No one can say that fuel reduction benefits wildlife unless these probabilistic factors are accounted for.

    The only study that really gets at these issues is in the arena of fuel reduction carbon storage, but the analogy is strong, because goshawks, fisher and spotted owls live in carbon rich forests. See Mitchell, Harmon, O’Connell. 2009. Forest fuel reduction alters fire severity and long-term carbon storage in three Pacific Northwest ecosystems. Ecological Applications. 19(3), 2009, pp. 643-655 http://ecoinformatics.oregonstate.edu/new/FuelRedux_FS_CStorage_Revision2.pdf

    • Canopy cover is retained, while ladder fuels are removed from below. All three species are considered and adjusted for. Of course, nesting sites are preserved, and with goshawks and owls being territorial, there is only so much habitat to go around. Thinned forests make excellent foraging habitats, as well as growing into future nesting habitat. In fact, we build in clumps of untouched old growth in chunks up to 1 acre, within our cutting units. We are “sculpting” crowded forests into more diverse and resilient ecosystem components.

      We have already seen what uncontrolled fire does in the Sierra Nevada. Trading endangered species habitats for snag and brush patches doesn’t seem like a good idea, to me.

  2. I was in grad school with John Battles, glad to see him forging ahead on this. But i cannot open the link to the magazine with his article, connection not good here, please email it to me. Thanks

      • Hello: I don’t make any judgement about this one way or another, but California Forests Magazine is produced by the California Forestry Association. According to their website, the California Forestry Association is a non-profit trade association representing California’s forest products profession, dedicated to assuring an adequate and sustainable supply of forest products at an affordable cost while enhancing forest healthy and safety.

        • This is Matt’s more cultured way of saying “the usual suspects”, using guilt by association, instead of actually reading and addressing the actual text, itself.

          • No, it’s actually not that Larry, but thanks. I would assume that if this same report was in the Sierra Club’s glossy magazine that that simple fact would have been shared when it was posted…or certainly would have been pointed out by subsequent comments.

            • I agree with Matthew on this… the context of how something comes to our attention is a piece of the puzzle so it is worth sharing.

  3. Tree- I agree with you generally about modelling because modelling is so complex that many assumptions are required, and they can’t all be tested using sensitivity analysis.

    On the other hand, we can’t throw the approach out either. We can just be careful.

    A brief look at the Mitchell paper shows that they used stand models and some mathematical projections, while the Collins et al. paper used a bunch of actual data from the forest, plus models.

    I looked at the original article, and it doesn’t appear that they assumed wildlife occurred with 100% certainty. Here’s what I found on page 80 of the paper here.http://cnr.berkeley.edu/stephens-lab/Publications/CollinsEt_ForSci_11.pdf

    For each scenario and time step, we simulated 5,000 randomly placed ignitions, burning for 240 minutes (one 4-hour burn period). This burn period duration was selected such that simulated fire sizes (for one burn period) approximated large spread events (daily) observed in actual fires that occurred near the study area (Ager et al. 2010).

    Now, that would be a really great student project to compare the two studies, if in fact they reached different conclusions, where did their assumptions differ? It seems to me like they could just be in areas that are so different from each other, given that I have worked in California and Oregon. Seems like that would be a profitable project for graduate seminar in fuels and fire modeling. They could present their findings and discuss on this blog (as well as other papers on this topic). I like tables that compare things, myself. You could have rows on the tables for things like “estimates of fuel loadings-how derived” and “assumptions about frequency of fire starts”.
    Takers? I wonder if JSFP would fund such a thing..

    I clicked on Tree’s link for the Harmon et al. paper, and it didn’t work for me, but here is one that does:

  4. “5,000 randomly placed ignitions” is a lot like assuming 100% chance of wildfire. What they did not do is calculate the effects of treatments based on the actual fire return interval as influenced by fire suppression.

    Also, Larry said “Thinned forests make excellent foraging habitats,” This is not well supported by the evidence. Forest Service EAs often assert that logging opens the canopy allowing goshawks to fly more easily and access prey. On the contrary, telemetry studies show that goshawks are associated with complex forest structure, while avoiding open areas.

    Greenwald, Crocker-Bedford, Broberg, Suckling, and Tibbitts. 2005. A review of Northern goshawk habitat selection in the home range and implications for forest management in the western United States. Wildlife Society Bulletin 33(1):120-129. http://www.fs.usda.gov/Internet/FSE_DOCUMENTS/fsbdev3_021122.pdf

    Beier, P., E. C. Rogan, M. F. Ingraldi, and S. S. Rosenstock. 2008. Does forest structure affect reproduction of northern goshawks in ponderosa pine forests? Journal of Applied Ecology 45:342–350. http://oak.ucc.nau.edu/pb1/vitae/Beier_etal.2008.Goshawk.JAE.pdf

    • Tree: They can’t base their numbers on “actual fire return interval” because there is no such thing — unless regulated by people. If I build a fire every day at noon to cook a hot dog for lunch, then the “fire return interval” is 24 hours. If I do the same thing, but in different locations on some days, then the “fire return interval” varies at each location and according to human decision.

      There is no evidence anywhere I know of on any portion of this planet that grows vegetation on land that has a non-human-related “fire return interval” during the past 10,000 years. It is a theoretical number (in the absence of people) that has very little basis in documented reality. And it is a very expensive number, too, despite its dubious origins.

      This kind of science (“modelling”) is often based on theoretical numbers, unstated assumptions, personal biases, and all sorts of other confounding parts so that the final “prediction” rarely ever comes true (growth models are an exception). Looking at Hansen’s predictions of the “next 20 years” of Global Warming that he was making in 1989 and 1990 is one example. They are (predictably) wrong. Again. I’m guessing similar luck for current AGW predictions for the next 20 (and 40) years as well, and for the same reasons.

      How many angels can dance on the head of a pin? You have to believe in angels first, to tackle that one. What is the fire return interval for this “natural” forest, grassland, etc.? You’ll have to ask James Agee or Jerry Franklin on that one — I can’t find any evidence of accuracy in this regard when it comes to actual documentation. This is the type of questionable data that is “actually” a form of rudimentary tree ring analysis. Or tea leaves.

      • It’s not the historic fire return interval that matters here. It’s the real interval as influenced by fire suppression. The shorter the de facto FRI, the smaller the ratio of acres-treated-and-not-burned vs acres-treated-and-burned, and the more likely the fuel treatments might yield some benefits. The longer the FRI, the larger the ratio, the more acres treated unnecessarily, and the less likely that fuel treatments would yield net benefits to habitat for species that prefer dense forests.

    • Tree said, “This is not well supported by the evidence. Forest Service EAs often assert that logging opens the canopy allowing goshawks to fly more easily and access prey. On the contrary, telemetry studies show that goshawks are associated with complex forest structure, while avoiding open areas.”

      Since goshawk young are so dependent on their parents for so long, don’t you think that the reason for them spending so much time around their nests could be their offspring?!? There is a LOT more prey in those thinned areas, and they require both kinds of habitats to survive.

      With already-existing protections still in place, the biggest danger to nesting habitats is wildfire. With the ban on logging large trees still in force, habitats are very unlikely to be impacted by cutting 10-18″ dbh trees. Adequate canopy cover is built into each thinning project in the Sierra Nevada.

      • Larry, You are building arguments to shore up your preconceptions. You obviously did not read the goshawk paper I provided a link to. The study relied on 12 telemetry studies that looked at habitat selection “outside nest stand.” The study concluded that “goshawks are selecting forests for their structure … Goshawks did not select stands with the greatest prey abundance.”

        • As always, you are behind the curve, Tree. There is a difference between “home range” and “nest stand”. A nest stand is the immediate area surrounding the nest, which is completely protected. A “home range” includes the nest stand, and the “best habitat” acres adjacent to the nest stand. Remember, there are always multiple nest stands, usually within the “home range”.

          Also, you seem to refuse to address the difference between nesting and foraging habitats, which goshawks (and owls) need both of. Obviously, if historical logging didn’t kill off the goshawks, what makes people think that modern forest planning and thinning projects, where appropriate, will impact the birds?!? The biggest threat to them continues to be stand replacement wildfires.

          • Another extremely important fact about owls and goshawks is that they are territorial. There is only so much nesting habitat to go around, and owls and goshawks must share it (despite goshawks eating owls!) The young of both birds are not guaranteed to find nesting habitat outside of the parent’s “home range”. This is especially so, here in California. The parents always chase them out. It’s kind of sad, hearing a juvenile begging for food, all alone.

            Also, goshawks tend to “make do” with the less dense forests of the Black Hills (where I surveyed for them). Remember, a nesting pair needs a network of nests to have offspring every year.

    • Tree- I find this to be an interesting discussion and I think we can all learn a lot.
      However, I think when we get to the scientific details it would be helpful if you would cite, not only the paper, but the paragraphs or pages you are using as part of your argument.

      For example, I dutifully looked up Beier et al and found this paragraph.


      Like Krüger & Lindström (2001), we observed that some goshawk breeding areas consistently produced more fledglings than others. The contrary findings of McClaren et al. (2002) may reflect their use of fledglings per active nest as the response variable, ignoring differences in occupancy rates among breeding areas. However, if forest structure does not drive productivity, what does? Plausible hypotheses include prey abundance, disturbance, weather patterns and parenting ability of goshawk breeders.

      Italics mine..

      Maybe you and Larry could help explain to the non-Goshawkian how people think about these things. We are talking about bird biology and not particle physics so regular folks should be able to understand the concepts..

      Another point about the ignitions, sort of before Bob’s point in the logic path (that there is no such thing as a past number), is that I don’t think we’re interested in past ignitions, aren’t we interested in what the results would be given likely future ignition scenarios? Because we want to protect birds under today’s and tomorrow’s conditions. We might think that ignitions would be different because of more people wandering the Sierra, or due to different weather patterns, or due to differences in spark arrestor technology, I don’t know all the possibilities, but the past doesn’t seem relevant to this question.

      Finally, I will try to find out from the author why they picked 5,000.


Leave a Comment