Let’s Discuss: The Levine et al. Paper on High-Severity Fire and Industrial Forests in California

Many thanks to Anonymous for sending a link to this paper. I wrote to the primary author and (quickly, thank you, Jacob!) received a copy, attached here.

What the paper does is correlate observations of high severity fire with landownership patters. When we talk about top-down versus bottom-up research, this is definitely a top-down. And for many of us bottom-up types, it can be great fun to hypothesize the mechanisms for these correlations, whether our own observations support these findings or not, compare with similar research (although the authors did that) and possibly dream up studies that have not been done to test the hypotheses.

So, bottom line, more acres of high severity fire on private industrial than private non-industrial and USG land (Forest Service, Park Service, BLM combined).
I can’t say anything about the technical details of their analysis, but they have a number of caveats in the discussion and conclusions section on page 5.

What would you do with this paper?

Based on the places I’ve formerly lived and worked, I’d want to look at each fire and the industrial forest landowners’ practices. My first thought was perhaps the private folks ended up with the most productive (able to grow high biomass) of the area. That would explain the fires reaching into nearby landowners’ properties, because nearby properties would have the same high biomass.

It also seems a bit counterintuitive.. in many places I’ve worked, jackstrawed dead trees are on the forest floor, conceivably leading to high severity impacts, while industrial landowners tend to remove dead trees, so you’d think there might be less fuel, at least on the ground.

In my experience, industrial landowners vary widely in their practices (think Basin and Range (NE Calif) versus the Coast Range). Some may be barely distinguishable from FS practices. So I think if you looked at each landowners’ practices, and looked at the ground where the high severity fire occurred, you might be able to explain for each landowner/fire combination, based on, I guess, four factors I can think of:

For each fire:

1. Did the industrial landowners (ILs) tend to have a certain location (topographical/soil) differences?
2. Did the ILs tend to have different age classes or other historical effects?
3. What IL practices may have influenced fire intensity?
4. Was the IL forest in the wrong place at the wrong time?

Other ideas?

It seems to me that building up from each fire, although labor intensive, might yield a great deal more information. Including how much sense it makes to combine over fires, ecoregions and industrial landowners. For a mere $300k, I’d do it; I see lots of fun interviews and field trips. Anyway, I know there are many Californians in TSW ranks and I’m curious what you think.

31 thoughts on “Let’s Discuss: The Levine et al. Paper on High-Severity Fire and Industrial Forests in California”

  1. The biggest difference between industrial tree farming and industrial crop farming is that in crop farming, with rare exceptions, they get a measure of harvest value each year and optimize their growing plan the following year based on that feedback. They also use weather forecasts for the upcoming growing season to optimize the investment in their probable yield. As crop farmers often explain: every decision they make creates both benefits and drawbacks and figuring out that puzzle defines if their farm continues to be profitable or not.

    Contrast that with tree farming, where the feedback from harvest yields takes many decades to realize and the people who plan the next rotation of trees to be grown rarely, if ever, are the same people who planted and grew that site’s previous rotation. This means that rather than site-specific knowledge of growth rates, tree farmers depend on text book knowledge / dogma to plan their next harvest rotation.

    Specifically, no tree farmer in California planted their tree farm decades ago knowing that the weather forecast for the early 2020’s was going to be a mega drought that’s the worst documented in 1,200 years. Tree farmers decades ago also didn’t know a whole body of new science directly correlating a huge increase in fire severity is due to climate change, which in California’s case has led to one 1 of every 8 acres of California burning in past decade: https://www.mercurynews.com/2021/09/29/top-10-california-wildfires-megafires-map/

    If tree farmers had known these details when planting their next rotation decades ago, there’s huge areas of land they would of not risked the cost of investment to grow trees on.

    This goes to the heart of why us enviros are so hated by the timber industry, In a word it’s called “feedback.”

    A crop farmer will fail and go out of business in a matter of years if they don’t analyse feedback on growth rate to successfully keep their yields profitable. But if a tree farmer fails it could be decades before they realize they failed on their investments and the only feedback they get are based on outdated harvests rates / text books about how to grow trees, as well as us enviros telling them they’re doing it wrong.

    So when us enviros explain to them that they aren’t paying attention to the slowly approaching demise of the profitability of their operation, it’s easy to shoot the messenger and treat us like we’re the ones who are wrong and that the text books are guiding them on path of certainty towards maximum sustained yield.

    And so now here we are… The chickens come home to roost and tree farming is fast becoming an unworthy investment compared to other more profitable types of land use and we have an abundance of data to prove it. Perhaps the most significant data point is seedlings sold for reforestation. Quite simply tree farmers are giving up based on the severity of fire they’ve experience in recent years: https://www.registerguard.com/story/news/environment/2022/02/19/replanting-oregon-forests-wildfire-seedling-demand-supply-unexpected/6786316001/

    • Deane, I think we’re talking about Washington State here. Do you have any specifics about genocide of indigenous people in Washington State by the timber industry? It would be odd, because indigenous people actually have their own timber industry in Washington. For example, the Colville Tribes https://www.colvilletribes.com/forestry and the Yakama https://www.colvilletribes.com/forestry. And the Yakama are certified to SFI..so…
      Please be more specific as to the location of the genocide.

      I did read the Propublica article and I have been following the California effort for some time. However, I don’t agree that in the absence of carbon credits for trees that companies would stop producing carbon. They might move out of California to a more desirable regulatory environment. Or they might simply buy other forms of carbon credits. I actually would prefer more direct air capture than dealing with trees, but I don’t have a problem with standards and verification. And the people who are doing the verification.

      • Hi Sharon, you got your comments mixed up…

        Your reply here is for my comment on the DNR post in my home state of Washington. But you replied to my comment on high fire severity in California instead?

        In the DNR post I was referring to forests certification globally as an example of why 3rd party forest certification is not independent or honest and and when that dishonest process is most harmful.

        If you want I could go on at length about residual old growth trees currently being logged on Washington state lands and how the DNR protection of older trees announcement is more window dressing than meaningful protection.

    • I’ve watched an entire cycle now (80s) of “gee we want to plant trees, how can we scale up? What’s the right technology? Seed sources, nursery practices, tree coolers, vermiculite slurries and so on. OSU had its Fundamental FIR program; and we all finally figured it out. But then the FS decided that we should use naturals (which made sense) and the operations shrank. I remember when nurseries (federal, state and private) bloomed on the landscape. If we could crank up then, we can certainly crank up now. But as with all cycles, they happen on a scale that people who remember the previous one are retired. And so it goes..

      • I enjoyed working on the ‘Lift and Pack’, at the Placerville Nursery. It was valuable experience to see how trees were supplied to requesting ‘customers’. It was a happening place in the 80’s and early 90’s, when clearcuts had to be planted (and treated).

        • Growing and planting trees is something we can agree on Larry! So much time and research has been put into how to rapidly grow each variety. So many different techniques are required to do this quickly and efficiently… I hope someday we do that for all kinds of native plants that need help to reclaim lost territory.

          • There are some very odd plantations out there, when seed stock for ponderosa pine was lumped together. Seeds from California, Idaho, South Dakota and Oregon were used in plantations, resulting in trees with very strange forms. I’ve had to thin out such plantations, picking the best trees out of all those ‘off-site’ specimens.

            At the Nursery, we also had ‘specialty’ trees, like aspen and giant sequoia. It’s interesting to see which trees are going where, across California.

            • It’s been recently found that aspen trees photosynthesise through their bark in winter and continue to respirate/sequester carbon even in freezing weather. Also Aspen don’t burn in wildfires very well, in some cases even redirecting wildfires around them. Of course tree farmers have used a huge amount of herbicide to replace aspen with conifers and that has made fire severity way worse, especially in British Columbia.

              What’s more, planting non-native trees is what most of the world considers reforestation to mean. Specifically Palm Oil, Eucalyptus and Pine. If you were one of those trees you’d be assured world dominance on almost every continent for centuries to come.

              Meanwhile my friends at the Conifer society could tell you about thousands of varieties that we’ve barely begun to even study. I’m sure someday people will look back on our mono-crop tree farming days the same way we look back on history before there were books and photography.

            • Sharon, what you’re referencing is not site-specific enough and not at the scale and quantity we need them in. And while container sized natives are relatively well funded by mitigations requirements at construction sites these day, their cost of production is still prohibitive for large scale ecological restoration which wherever possible needs to be bare root oriented not container oriented.

              As I said in a Twitter discussion recently, the cultivation knowledge for each type of native plant is still a knowledge that’s not widely shared in the way tree farm tree cultivation is. And that one of the best solutions for this is making sure every new employee hired by the forest service or other agency has an opportunity to propagate native plants for nearby restoration projects as a side hustle. As in we need the next generation of forest workers to further advance native plant propagation into an entirely new land use now that tree farms aren’t a good long term investment any more. As in creating new economies!

              • Hi Deane: I’m not sure where you are located, but I can say that several nurseries in NW Oregon are doing well with native plant propagation. Not sure why you prefer bare roots over containerized plants — or seeds or cuttings — but planting success can vary by species, depending on their individual responses to each method. I do agree with your concern regarding specific locations, though. Here is what I have been doing with camas in recent years, for example: http://www.orww.org/Native_Plants/Camas/

        • Larry, you and I were there (at Placerville Nursery) at the same time in the 80’s. I was the director of a genetics lab located in the seed extractory building. We may have eaten lunch in the same break room back then!

          • I, too, worked at the Placerville nursery in the 1980s, on the lift crew and sometimes trimming roots and packing seedlings — many of which I later planted.

          • I was close friends with a number of people working there. Holly Stewart worked for you and said you were the best supervisor she ever had. Emil Obrey worked with the cone collection folks. Julie Roberts monitored the contracted lift crew. Gina King was the best personnel clerk, ever! I worked for one winter, doing the sorting and packing.

          • Hi Sharon: I didn’t know that you were doing genetics research! My response to Deane above includes a link to about 30 different camas populations in western Oregon that I have been trying to get analyzed for several years. For reasons that Deane states, I think this type of analysis of key native landscape species might be really helpful. Any leads or suggestions? Better to discuss specifics by email?

      • Sharron, planting a tree farm and growing it all the way to harvest and having the money you put into the effort cost way less than the money you make at harvest once was a very lucrative investment that was better than all other long term investments for most forests landscapes.

        But with the uptick in fire severity, as well as shorter and shorter rotation ages (unsustainable forestry) the numbers no longer add up into a profitable long-term investment.

        And not just from fire severity, but along the I-5 north of Portland, Oregon on June 28th last summer it reached 120′ F and older slow growing trees not part of a tree farm normally died back their older leaves and kept growing their new growth, albeit more slowly.

        But young doug fir tree farms that have been hybridized to grow as fast as possible consistently died back from the tips of their branches and not only lost a couple years of growth, but the trees will lose their ability to grow at their previous rate for several years while each branch works past that die back. And what happens if gets up to 120′ F again this Summer or soon after?

        Point being USFS post wildfire recovery plans are no longer heavily investing in replanting mono-crop tree farms because bringing a stand to harvest in 40-60 years is not an investment they can afford to pay for anymore, especially if it’s going to likely be lost to fire. So portions of the landscape in current planning documents is finally going to return to a more naturalized and diverse ecosystem based regrowth process.

        Furthermore, as tree farmers on private land look at their own investment plans, they’re having the same realization. As in Tree Farming was once the most profitable way to manage a landscape, but with wildfire and climate change getting worse and worse it no longer looks like a reliable long term investment and that’s why tree seedling orders are way, way lower than first projected after the Labor day fire that burned more acres in the PNW than all fires that burned in the past 35 years combined.

        In the long run, this is a big win for ecosystem health and biodiversity and a well deserved huge loss for mono-crop tree farmers that are no longer going to be able to operate profitably.

  2. “…jackstrawed dead trees are on the forest floor, conceivably leading to high severity impacts, while industrial landowners tend to remove dead trees, so you’d think there might be less fuel, at least on the ground.”

    It is a different story and metric, despite the use otherwise by some, but the soil burn severity for many of the named fires within the paper tell a different story.

    • Anon.. I wish I understood the different measures of fire severity and how they are measured, for sure. I wish there was a glossary easily accessible. Perhaps you know of one or could write one for TSW? Anyway, maybe that’s another idea for a further paper… let’s look at all the measures and see how they do or do not line up!

  3. My very astute colleague indicated that their extended assessment might have biased their findings towards a higher likelihood of high severity on industrial lands, due to the fact that salvage logged areas have a tendency towards registering as high severity using CBI. Salvage logging moves very fast on private lands in California, and is implemented across a range of burn severities. I respect these authors, and they clearly note this in discussion. However, this won’t stop some from waving this around to make a case for ending intensive forestry in CA. Gotta love the weaponization of science.

    • I also respect the authors and they were very careful about their findings. To me, though, with all due respect to them, it’s not particularly helpful. If high intensity is bad (which I agree with), what physical conditions lead to it? In other words, how can industrial, private or government landowners manage their forests to reduce the likelihood of high fire intensity?
      If this were a codesigned coproduced research effort, we might start with…
      (1) Defining bad intensive fire effects (soils? wildlife? dead trees?)
      (2) Observing where they occur on each fire
      (3) Parsing out why.. perhaps using the knowledge and models of FBANs- Fire Behavior Analysts?
      (4) comparing across fires, are they the same or different? Why would that be?
      (5) Generating hypotheses about “why that would be” and testing them against new fires (regardless of ownership)
      Then at some point making recommendations to landowners of management to reduce the risk of bad effects. Acknowledging that one person’s bad effects (dead trees) is not the same as others’ (woodpeckers). And that to get HRV you need some intense effects to achieve pyrodiversity. But anyway, whatever managers wanted, you would have an idea of management to get there.

    • The authors seemed to address the issue of CBI: “To investigate this potential bias, we examined the aforementioned 2007 Moonlight Fire, for which extensive salvage logging occurred on private industrial forestland. We found a slight increase in the probability of a pixel burning at high severity
      in the extended versus initial assessment: 0.017. However, this proportion was small relative to the observed difference between private industrial and public land throughout our entire dataset (0.017 versus 0.14) and fell within the reported classification error for the dataset (Miller et al. 2009; see WebPanel 2).”

  4. The findings of this paper are consistent with Zald and Dunn (2018, https://esajournals.onlinelibrary.wiley.com/doi/full/10.1002/eap.1710), which also found industrial timber lands were more of a high severity fire hazard than public lands in southwestern Oregon.

    For me, the findings of this study confirm the duh hypothesis. Plantations are composed of uniformly-spaced high densities of smaller trees, with lower canopy base height, and thin bark. This is a recipe for high severity fire and running crown fires. When these conditions occur at a landscape scale, they bleed into neighboring properties, and contribute to generating plume-dominated fires.

    I have routinely seen SPI point a finger at the Forest Service as needing to do something about the fire risk USFS land’s pose to their industrial lands, when it has always been clear to me that the condition of SPI lands poses as much, if not more, of a threat to the landscape than public lands.

    Industrial timber owners like SPI started installing fuel break networks on their lands about 5 years ago (they are very proud of themselves), but these fuel breaks are often only about 300 feet wide, and will only be effective for suppression under more moderate fire weather, but the trend is for a lot more sever fire weather, so this is not a viable landscape strategy.

    I also want to call attention to the USFS’s playbook on post-fire “restoration”. Following high severity fire in the Sierra, the standard practice for the USFS is to mimic an industrial forest owner…following salvage, they nuke large areas with herbicides and install 300 tpa at an even spacing. Once installed, many USFS foresters are loathsome to the idea of reducing plantation productively in the name of fuels management, partly because the regional silviculturalist is stuck in the glory days of the1980s. They cross their fingers and pray they can get to that first commercial harvest at year 30. This is not a strategy likely to work in 2022 and beyond. We need to use a new playbook for these new conditions.

    • Again, that’s precisely the point of field visits. What were the physical characteristics of the high-severity areas?
      I’m thinking say 5 or 10 years old plantations, not so much… you have grass and not that much woody material, and the fire would burn right through the grass.
      20-30 years.. are they thinned? I don’t have an idea of what the stand looks like structurally nor the understory. And what is a “crown fire” of a 30 foot tall thinned stand?

      And the FS does plantations too, although now in clumps rather than evenly spaced, but how exactly over the last 40 years or so.. could have been evenly spaced. So again the physical characteristics between industrial and the FS plantations for some age classes would not be that different. That’s why I think the key to understanding is to take one fire and look at the specific physical characteristics of the different areas with high intensity fire.

    • Zald and Dunn placed a lot more emphasis on fire weather than simply ownership, and while I have some issues with assumptions made in how they determined fire weather, it’s peer reviewed and acceptable.
      More to the point, you reference “300 TPA” on USFS, but I’m less certain that that is a requirement? I point this out as 300 TPA is a mandatory requirement for private landowners under the California Forest Practice Rules, along with substantial retention requirements for non-evenaged management types. Perhaps that is where change really needs to begin, first. Similarly, on private timberland in California, there is no regulatory requirement for either reforestation or minimum stocking levels – as I understand it, SPI has in fact been replanting burned and salvaged areas at far lower tree densities that before.
      Food for thought. Perhaps this paper justifies exactly that.

      • Reforestation spacing on higher quality USFS sites is routinely 250-300 tpa, this I know from experience. Due to limited funds these days, the USFS starts with the highest quality sites for reforestation and may not get to the others. Starting with Rim and King, mid and upper slope positions started to be planted less densely, and higher quality sites had the option of some lower density planting, but still allowed up to 300 tpa. These changes were a result of pressure from PSW scientists pushing GTR 220 ideas and watching plantations go up in smoke in both of these fires.

        To capture and control a site with trees in the Sierra requires planting 250-300 tpa, and foresters are control freaks. This also gives the foresters something predictable to manage. When there are less than 250-300 tpa, there’s growing space available for shrubs. Either way, it’s a fire hazards by year 10 and needs fuels management. Planting lower density does not require a PCT and prescribed fire can and should be used in lower density plantations around year 6-8 to control shrubs instead of herbicides. This puts these sites on the path to fire use as a maintenance tool. If prescribed fire is done in lower density plantations in the fall when the buds are hardened, P pine mortality will be low.

        • I was a reforestation contractor for more than 20 years and personally planted more than 2.3 million seedlings and my crews performed more than 85,000 acres of reforestation contracts from 1970 to 1990 — mostly in western Oregon and SW Washington, but also Idaho and northern California. I’m not aware of a single acre of this work ever burning in a wildfire. A few probably did, but many have been clearcut and planted again and wildfire has not been a major problem.

          During this time we planted through ancient huckleberry fields, oak savannas, camas meadows, and grassy prairies, as well as clearcuts, pioneer apple orchards, and major burns. What USFS apparently refers to as “TPA” and having something to do with PSW and GTR 220 varied by regulation and ownership. Most important was seedling quality, site preparation, and planting location. Requirements usually varied from 300 to 640 trees per acre, depending on ownership and seedling quality, but many areas should have never been planted, and most others did best at 150 to 180 trees per acre, planted in “micro-site densities” rather than a grid. Then the spotted owl flew in and “low bid” contracts with migrant crews changed everything. And not for the better, in most instances.

          I am highly suspicious of anyone using “severity” measures. They are mostly subjective and largely meaningless. “Mortality” has always worked fine, and almost everybody knows what it is. Douglas fir naturally grow in even-aged stands almost exclusively containing just this one species, and hardly any “biodiversity.” The “monoculture” complaint is largely unfounded — that’s how Douglas fir, alder, Sitka spruce, and hemlock often grow “naturally.” Especially when seeding in areas that previously held few, if any, trees. Finally, tree farmers don’t learn from “books” or latest literature. They know their land better than anyone, what it grows, and how. And contrary to other statements, these lands in western Oregon (private tree farms) are rarely ever subjected to wildfires — unless they are coming from nearby federal lands. Documented facts, with photos.

  5. Dang! I just responded to this paper on the “New Topics” page. I haven’t waded through the Comments yet, but I strongly agree with you, Sharon, that far better analysis of these fires is needed. I’ve been trying to get interest (“funding”) for this type of research on the Oregon 2020 Labor Day Fires, but no luck so far. Past efforts for this type of study in both western Oregon and northern California have been as fruitless. I can speculate why, but that is a topic for another forum.


Leave a Comment