Science Friday: The Problem of Reference Conditions, and Alternative Management Approaches

American Chestnut planting in Vermont

For many years, scientists and others have been talking about the problems of using HRV and reference conditions as management targets.  Jon and I have had many mind-numbing discussions about it (as placed in the 2012 Planning Rule)  on this very blog. So I thought on Science Friday, I’d start exploring the work of other scientists who have expressed concerns about this. Sadly for me the 2012 Rule is water under the bridge, but the Feinstein bill is not. Let’s see how that bill talks about using reference conditions in the large landscape projects:

“2) evaluates ecological integrity and reference conditions for the landscape;
3) identifies areas that have departed from reference conditions;
4) identifies criteria for determining appropriate restoration treatments;
5) are based on the best available scientific information, including, where applicable, high-resolution imagery and LiDAR; and
6) identifies priority restoration strategies.
o Restoration actions shall 1) emphasize the reintroduction of characteristic fire; 2) for any proposed mechanical treatments, seek to restore reference conditions and the establishment..

Now, you don’t need a Ph.D. in vegetation ecology (or plant evolution, or wildlife biology) to suggest that restoring (say, American Chestnut) may be difficult or impossible to achieve due to a) other changes that have happened since, like other species having taken over  b) climate change, c) invasive species, d) human population impacts (pollution, domestic animals and pets and so on)  e) lack of Native American practices and so on. So why does this idea have such a hold on the imagination? Why has it been argued that this is a more “scientific” approach, when many scientists disagree? And while we can discuss that reference conditions aren’t HRV, well, then aren’t they just conditions that some group finds desirable, and not any more “scientific” than any other conditions?

This article is by Dr. Connie Millar of the Pacific Southwest Research Station, and she uses her climate studies to come up with a different conclusion about what to do- not exactly what many mean by “restoration” but helping systems/organisms be resilient and adapt to future change. It seems like a very different, and possibly more realistic and less expensive, paradigm (to me). The paper also includes a history of the concept as used in forest management on page s30. Here are a few quotes:

From the standpoint of this review, I focus on an underlying assumption of stationarity that continues to emerge in the HRV literature even where discussions address historic variability. Further, two elements in the discussion of HRV remain unconfronted and problematic: First, what historic time period is most relevant to current and future conditions (and, in contrast, what periods are inappropriate); and second, whether “approximating historic conditions” of any historic time period is a wise approach to managing for functional ecosystems of the future

I agree that the latter deserves much more discussion. Especially since it could be argued that generalized “restoration to reference conditions” is a “nice to have” and strategic fuel treatments are a “need to have”; I’d put my tax dollars on the latter rather than the former.

For the many ecosystems not severely degraded, historically informed strategies focus on (a) removing barriers that impede inherent ecological capacities to respond to change, and (b) assisting species and communities to transform in ways most compatible with their inherent capacities and with social goals. In regard to the first, we learn from historic retrospect that the truly novel conditions at present and increasing in the future are not so much about the magnitude or even pace of climate change, rather the overwhelming transformation modern humans have imposed on Earth. The Anthropocene era (Ruddiman, 2003) is characterized by nonanalog conditions for species survival, and the accelerated pace of extinction shows that many species have not been able to use their inherent capacities to respond to change in the face of such barriers. Functional restoration thus can emphasize, to the extent possible, removal or mitigation of impacting barriers derived from land development; fragmentation; air, land, and water pollution and contamination—carbon dioxide being among the worst offenders; land-use changes, invasive species, and many others (Millar, Stephenson, & Stephens, 2007).

Assisting ecosystems to transform in ways most compatible with their inherent capacities involves exploiting species tendencies to move geographically and to adapt genetically in the face of change. In regard to the former, if barriers to dispersal cannot effectively be removed, assisted translocation, either of species beyond their current range limits or of genotypes beyond their current provenances, might be effective. Understanding changes that are already underway will enable restorationists to ease transitions to future states, often with less extreme variability or outcomes than socially tolerable. Assisting genetic adaptation can occur in many ways informed by natural selection—such as experimenting with seed diversity in restoration mixes, taking advantage of the significant opportunities that insect-mediated mortality events can have to ratchet population adaptedness forward, and setting goals at bioregional not local scales.

Goal Setting
While much can be learned from reconstructions of historic responses to climate change, and especially from historic periods similar to what might be expected in the future (e.g., the Medieval interval or the middle Holocene), historic conditions generally, and especially those from nonanalogous historic periods (such as the Little Ice Age) make inappropriate reference conditions for the future. In such a case, target and goal setting for functional restoration must pioneer additional approaches. The return to more utilitarian goals and targets recognizes both the reality of the Anthropocene and that resource-management has, in fact, always been a human-directed and conceived endeavor. Emphasis on ecosystem services—including essential utilitarian functions such as clean air and water, landscapes for recreation, production of fiber and meat, and sustenance of biodiversity desired by humans and provided by healthy ecosystems—will guide the next generation of management strategies. Achieving these goals will be greatly benefited by historically, as well as ecologically informed management.

19 thoughts on “Science Friday: The Problem of Reference Conditions, and Alternative Management Approaches”

  1. Thanks for posting this Sharon. I have been thinking about the use of reference conditions in relation to our changed climate. While the 2012 Planning Rule is water under the bridge, it may warrant considering modifications to that rule.

    Reply
    • I propose the FS fund the National Forest Foundation (or someone else) to convene a diverse group (like a not-FACA committee) to put out a request for online public comment… on
      1) how well the Rule has worked for everyone, stakeholders and FS employees, and 2) ideas for improvement. The comments would be easily searchable, and professional facilitators would facilitate the online dialogue.
      Then the collaborative group could make recommendations to the FS for potential modifications to the Rule.

      Reply
    • You might be interested in this one as well..https://esajournals.onlinelibrary.wiley.com/doi/abs/10.1890/06-1715.1, CLIMATE CHANGE AND FORESTS OF THE FUTURE: MANAGING IN THE FACE OF UNCERTAINTY. Around 2010, the FS required forests to develop climate change plans.. I was involved in this effort as our Regional Rep on the National Climate Team. We used this kind of thinking in developing those plans. It seemed to me at the time that there was a disconnect between the FS work on climate and the NRV requirements under the 2012 Rule. My views were not greeted with great enthusiasm in planning circles at the time.

      Reply
  2. The whole idea of picking a particular historical state to “restore” a landscape to just seems bizarre and arbitrary to me. Landscapes change over time. That’s just a fact of life. No landscape on Earth is the same as it was 1000 years let alone a million years ago. What makes the state of the landscape at one particular time more valuable than others?

    I agree that instead of looking backward at some arbitrary time in history, we should be looking forward to how we want the landscape to be in the future, which requires us to figure out what is both most useful to humans and healthiest for ecosystems humans consider desirable.

    As that article said, this is the anthropocene and humans have been shaping landscapes for millennia. It’s been that way as long as humans have existed, and certainly didn’t start when white “colonizers” showed up. Native Americans did plenty of landscape shaping of their own long before Europeans arrived. Like it or not, humans are in charge of nature now and the landscape will be whatever we want it to be. We should own up to that instead of futility trying to revert landscapes to some imaginary pre-columbian perfect state of nature which never actually existed. Which is what I suspect a lot of the groups pushing HRV are really trying to do.

    Reply
  3. Our many “mind-numbing discussions” may have led you to miss this key point: the 2012 Planning Rule does not require that desired conditions = historic conditions. Desired conditions = NRV and NRV is developed in the Planning Handbook as follows (§23.11a):

    “3. If past conditions relative to the natural range of variation are not appropriate, practical, possible, or desirable approaches:
    a. The Interdisciplinary Team should design plan components based on a general scientific and ecological understanding of the conditions that would sustain key ecosystem characteristics and sustain at-risk species using factors such as: representativeness, redundancy, habitat associations of particular species, disturbance dynamics, or observed conditions in reference areas. (FSH 1909.12, ch. 10, sec. 12.14b); and
    b. The Responsible Official should briefly explain in the plan decision document the rationale for NOT basing the design of the plan components on those conditions that were common in the past relative to the natural range of variation.”

    NRV as defined by the Forest Service fully comports with the stated point of this paper: “Achieving these goals will be greatly benefited by historically, as well as ecologically informed management.”

    (I am a little concerned about the use in this legislation of the term “reference conditions” instead of NRV or “desired conditions” because that term is conventionally used to refer to current conditions in “undisturbed” areas rather as a than future-looking consideration.)

    Reply
    • The way the Handbook reads is that “the Responsible Official should briefly explain in the plan decision document the rationale for NOT basing the design of the plan components on those conditions that were common in the past relative to the natural range of variation.”

      NRV seems to have a predisposition for being “the right way”, otherwise why would the RO have to explain if they don’t use it -“those conditions that were common in the past.” My argument is that the default (NRV) should not be the default, with people having to explain why they didn’t use it.

      Reply
      • Ok, that’s a reasonable position to take, especially going forward into a different climate (and it seems to narrow our differences on this to not very much). But keep in mind that whatever basis is used for establishing desired conditions will have to be explained, and conflicting scientific views will have to be discussed (even Millar’s article acknowledges the continuing relevance of historic conditions). I think this “default” remains a reasonable approach to get that result.

        Reply
    • Sharon is right in the larger context of the planning handbook, the NRV is presented, explicitly, as the necessary and default, even if it not a sole requirement (in the sense that you can justify not using NRV).

      In addition to Sharon’s citations:
      12.14a: “When assessing whether an ecosystem has integrity, the Interdisciplinary Team should use the natural range of variation as the ecological reference model, unless the past information regarding the selected key ecosystem characteristic is lacking, or the system is no longer capable of sustaining key ecosystem characteristics identified as common in the past based upon likely future environmental conditions”

      23.11a: “In general, where appropriate, the Interdisciplinary Team should design plan components aimed at maintaining or restoring the natural range of variation of specific key ecosystem characteristics needed to promote ecosystem integrity in the plan area.”

      So I do think the practical question posed here does remain, and remains in a way that gives a bit of a dilemma: are reference conditions mostly NRV, and is NRV preferable as the core basis of plan components that form desired conditions? If so, why is that preferable and on what basis? (while I’m aware of the larger arguments around NRV, not going into it here, but see why its a potentially open and worthwhile question). Reference conditions as a key baseline for desired conditions invite a host of philosophical questions, and I think it requires real work to make sure we’re not being arbitrary in assigning preference to one factor over others.

      (again, not saying NRV is arbitrary, but it is far from the sole factor in determining what makes a condition desirable, so why is it given the default? how prescriptive could/should the planning rule be with how these things are stacked alongside/against one another? is the somewhat inherent tendency for folks, myself included, to fall prey to the genetic and naturalistic fallacies at play in the default towards NRV?)

      There may be good reasons, of course, but it does seem that looking to the past can run the risk of curtailing creative options for climate change adaptation. Speculating wildly, of course, but still, we’re firmly in the anthropocene, so any pretense of claiming the need is removing human influence seems naive.

      Reply
      • See my response to Sharon above. We can disagree about the best approach, but I don’t think we can say the Forest Service was wrong to make historic conditions the starting point for discussing ecological integrity. The history of this is intertwined with the development of the “coarse filter” approach to conserving species (and there was plenty of discussion of this when the Rule was adopted), and range of variation is a way of managing for the coarse filter. The question is what “natural” should mean in this context.

        Here is the “smoking gun” for the Forest Service decision to adopt “natural range of variation” as its definition of ecological integrity (its definition of “ecological integrity” was cited in the EIS for the Planning Rule): https://academic.oup.com/bioscience/article/53/9/851/311604?login=true

        I like their proposal of “acceptable range of variation” instead of “natural,” which in the forest planning context would be the “desired conditions” (which, according to the Planning Rule would have to be the same as the “natural range of variation”). The article is pretty clear that there should be reservations about using historic data alone. I am not aware of any documentation of why the Planning Handbook opted for that as a default. The academic debate could continue, but I don’t see a case being made that there is anything wrong with that as a planning strategy.

        Reply
  4. Is there any effort at a landscape level to develop resilient, non-“native” forest components anywhere in the US? Where I was in school in Germany (not so many years ago) the state forest service was introducing Italian poplar varieties and (US derived) Douglas fir into their forests to hedge against climate change, sometimes over protests from environmentalists who wanted a ‘purely native’ forest- which of course is also a slippery thing considering that before the last ice age there were Douglas firs in Germany… It’s a fascinating question.

    Reply
    • Jimmy.. here’s my take on these efforts.. as a forest geneticist I know a bit of what we don’t know.
      1. We don’t know how climate change affects microclimates as experienced by forest trees.
      2. We don’t know how sensitive to those unknown changes individual trees will be.
      3. We don’t know how sensitive to those unknown changes their offspring will be.
      Basically we don’t know squat about forest tree adaptation; but they’re not shrinking violets (so to speak) so they did manage to recolonize North America after the Ice Age, so… I think Millar points this out in one of her papers.

      For the western US, if I were planting, I would try to make sure I had relatively the same mix of tree species as was there before the fire/harvest entry. Some of them do fine naturally, others not. Given that, I might mix some in from adjacent elevation or seed zones. Other than that… I wouldn’t do anything differently.

      Reply
      • Your last paragraph is exactly what you just argued against. Default to historic conditions unless you’ve got a reason to do something different.

        Reply
        • Great question. But no, not historic.
          I’m thinking of current conditions (which trees grow there now) not (which trees grew there before fire suppression, or before European settlers, or before glaciation) and future conditions. As a vegetation person, I’d say that many “future climate” interventions are expensive (unless you sell trees and that’s its own can of worms, as we know) and may not work (like planting species that don’t grow there today). So I’d be more centered in “what works today” with some attention to “what might work better in the future.”

          Reply
          • The question is how should you decide about (desired) future conditions, and the role that historic conditions should play in that decision. It sounds like you prefer current conditions to historic conditions for this purpose. That would not make sense where current conditions are artificially maintained, such as by fire suppression. And I’m not seeing any science that says to ignore historic conditions.

            Reply
            • I think you’re making my point. .. how relevant is “history without fire suppression” to “dealing with desired conditions in a changing climate and where fire suppression is a fact of life?”

              I didn’t say, and Millar is not saying to “ignore” them. She is a scientist (as am I) and the title of her paper is “Historic Variability: Informing Restoration Strategies, Not Prescribing Targets”
              https://www.fs.fed.us/psw/publications/millar/psw_2014_millar004.pdf . I’m completely in agreement with her on this.

              Reply
              • “Where fire suppression is a fact of life.” I would distinguish that from where fire is playing its “natural role” in ecosystems (whatever “natural” means), which is what a lot of people argue for (including the Forest Service). Identifying which area is where is a planning question, and must be addressed within the overall framework of ecological integrity.

                Reply
  5. Just a couple of observations after reading the whole thing. I wondered why she had used the phrase “prescribing targets” in the title. I see that her only references to Forest Service planning are to a 1995 document when historic range of variation was being given much more weight than being used as a “default” in the 2012 Planning Rule. So this 2014 article is not a commentary on the 2012 Rule.

    To the larger point – what time period, if any, should be considered a good starting point (default) for national forest ecological integrity – I point out her Mono Lake example: “The long road toward functional restoration of the Mono Lake ecosystem began with setting goals that were informed by recent historic conditions, that is, the known predisturbance ecosystem diversity and function. These were further informed by projections about future climates—especially temperature, precipitation, snowpack, and evapotranspiration.” I don’t see a lot of difference between this and the use of HRV as a starting point for NRV in forest planning. Also, the Planning Handbook does not specify what time period should be used for “historic,” and “recent” (post-Little Ice Age) would be consistent with this planning guidance (and maybe with your concept of “current”).

    I agree with what I think is her main conclusion: “Assisting ecosystems to transform in ways most compatible with their inherent capacities involves exploiting species tendencies to move geographically and to adapt genetically in the face of change.” Ecological integrity should put a heavy emphasis on connectivity (which is recognized by the Planning Rule).

    Reply
  6. I alluded to the use of the “coarse filter” approach to managing for wildlife viability above. I happened to see this in the Ochoco Black Mountain Project record, and it sums up the idea pretty well (MIS are found in the 1982 planning regulations, and existing forest plans).

    “Viability of Management Indicator Species (MIS) is being
    assessed using the historical range of variability (HRV)
    concept; comparing current amounts and distribution of
    habitat to historical conditions (Wisdom et al. 2000;
    Suring et al. 2011). Scientists assume that species are
    more likely to persist into the future under the conditions
    that remain most similar to the conditions that they
    persisted in during the past (Landres et al. 1999;
    Samson et al. 2003). By managing habitat within HRV it
    is assumed that adequate habitat would be provided
    because species survived those levels of habitat in the
    past to be present today.

    If the goal is to preserve existing diversity, it seems clear that historic conditions should be the starting point for desired conditions for the forest plan. To be accompanied by an analysis and discussion of whether they are achievable, and if not, then what.

    Reply

Leave a Reply to Patrick McKay Cancel reply