Restoration + climate change = cognitive dissonance

Brandon Bestelmeyer's picture

Last week, an agricultural researcher from the Monte Desert in Argentina emailed me to put him in contact with a seed supplier in the US. They wanted seeds of the perennial grasses blue grama and Lehmann lovegrass for some revegetation trials. I wanted to hit reply with: ¡Por favor, no lo hagas! (don’t do it).


Blue grama, a native grass in the western US (the state grass of New Mexico, in fact) I know and love, but it can be highly dominant where it occurs and it excludes other plant species. Lehmann lovegrass is an African grass that was planted in degraded rangelands to increase forage production, also used in roadside stabilization, that has become invasive in the southwestern US. Several papers indicate potential negative effects of Lehmann on biodiversity due to its aggressive spread and negative relationships between its abundance and those of certain native species. The Monte Desert has a wonderful assemblage of native herbaceous species, so why would you introduce either of these exotic plants into that ecosystem, given what is possible (blue grama) and what we have already seen happen (Lehmann)?


Last week I also read the latest IPCC report on climate change effects. One passage hit home: “The paleoecological record and models provide high confidence that it will be difficult or impossible to maintain many ecological systems in their current states if global warming exceeds 2 to 3°C, raising questions about the long-term viability of some current protected areas and conservation schemes, particularly where the objective is to maintain present-day species mixtures”. The report also states that there is “limited evidence and low confidence” that warm, arid grasslands and shrublands, like the Monte, will experience a loss of biomass production (that is, less of everything).


I came to experience what psychologists refer to as cognitive dissonance: the discomfort in holding two contradictory beliefs at the same time. On the one hand I am concerned about a potentially invasive species disrupting a plant community, on the other hand I know it’s going to be disrupted anyway and that it probably doesn’t matter. Or maybe having a few more perennial grasses in the game could increase “climate resilience” and the likelihood that any perennial grass will persist in the Monte through the coming decades. An argument can be made that Lehmann lovegrass provides this function even now to parts of Arizona that experienced grass loss and soil degradation long ago.


Restoration efforts in southwestern New Mexico create a parallel source of discomfort. The removal of woody plants by federal agencies, using chemical or mechanical means, is explicitly directed at restoring degraded areas to a grassland structure that existed before 1900. In many cases we are getting some perennial grasses back. But what if desert shrublands are the best we can hope for in the hot, dry years of the 2080s? What if the limited grassland recovery we have seen to date is reversed by prolonged droughts and the woody plants do not recover either? In restoring elements of the past to the present, do we cause desertification in the future?


We are not prepared psychologically, intellectually, or institutionally to respond to such questions. The community of land users, managers, and policymakers still believes that the historical range of variation is our goal for the future—that belief is codified in government policy. There has been a great deal of writing to shake this foundation, most notably papers by Richard Hobbs and colleagues suggesting that we should consider more open-ended goals in management, including the value of “novel ecosystems” that have no historical precedent. But these ideas are hard to put into practice because of the great uncertainty involved in creating alternative goals (such as by letting invasive species persist) and our ignorance of what climate resilience actually entails (resilience of what exactly, and, assuming we know what that is, to what degree of change in climate?). As Nicole Heller and Hobbs wrote in a recent paper, “Management rhetoric seems paradoxically to ask that managers allow for change so that ecosystems can adapt but also that they not permit change so that systems can remain intact (i.e., not damaged or impaired).” This paradox, plus the uncertainty, leads to inertia. We continue to manage for what we can measure and understand.


I will go out on a limb to suggest alternative, measurable, resilience-based goals for the Monte and Chihuahuan Deserts. Maintain the ecosystem’s ability to produce biomass and stabilize soils in the face of weather variability. Conserve, to the extent possible, species diversity and endemic species of the region. Identify options for maintaining human livelihoods, including adaptation (such as a shift in livestock breeds) and transformation (such as a shift to mesquite bean farming and honey production). The devil is in the details of how much of what happens where, but nothing here is very new or unreasonable. Confronting those goals with data-driven, logical, and imaginative appraisals of invasive species and woody plants, on the other hand, might lead to radically new management practices, greater comfort with some controversial ones, and less comfort with others.


Nice post, Dr. B.
Certainly there is an infomercial personality out there telling us that if everyone simply changed their way of thinking, life will become nothing but rainbows and unicorns. (Unfortunately, rainbows resist carbon storage while unicorns do nothing to reverse desertification or global climate change.)
I agree with you Dr. B: Society is no more well prepared to deal with the complexities of climate change than we are with Gliese 710. Collectively, we apparently have little sense of scale or time. Why else would humans be so willing, at least superficially, ready to grab at unicorn reins with the intent of trampling off to set the World Thermostat to preindustrial temperatures? (For some reason when I read this last back to myself, it sounds more absurd than it usually does.)
The real question is, if a quarter of the neighborhood is on fire and we have one fire engine, where should we commit our effort and resources?
Attempting to turn back the clock from a vegetation perspective is probably not such a good investment when the climate conditions which gave rise to that combination of vegetation are no longer there to sustain the community we think belongs there. Does not the changing of climate conditions imply the transformation of one of the key elements which define an ecological site? In other words, if an overriding abiotic factor changes enough has not the vegetation potential of the land changed? The redirection of resources in attempts to break a positive feedback loop perpetuated and accentuated by a catalyst we cannot directly counter seems futile as well as expensive.
We have seen the same results in regard to when watersheds fall apart due to increased discharge rates and restoration efforts focus on riparian zones and then wonder why gullies continue to advance and the water table continues to drop. In order to understand the potential long range outcomes, we have to be able to see the big picture from a hydrologic perspective rather than a localized vegetative perspective. The same goes for climate change and landscape potential –we are just a little further behind the curve when it comes to the big picture.
Successes stories in watershed management may provide some examples for climate change adaptation strategists to examine. Hopefully, we can select key buildings to hose down thus keeping much of the remaining neighborhood intact.

Brandon Bestelmeyer's picture

First do no harm. I think that maxim is key in cases where its not clear we can attain or sustain a restoration objective. The tricky part is knowing what the potential harm is.

That requires knowing how the landscape functions, defining clear objectives, and prioritizing, as you said. Doing nothing about upland parts of watersheds, and diverting resources from restoring them, while focusing on the riparian 'symptom' can be harmful if the goal ultimately fails (maybe a pretty obvious case).

But the same goes for spending resources to restore a grassland that may not persist while we miss other opportunities and potentially cause harm to certain attributes (not such an obvious case). I'm not saying I know how the risks and potential rewards balance out in the specific situations discussed above (would be a nice scenario analysis to do), but we (federal agencies, NGOs, NRCS, Conservation Effects Assessment Program) are not really talking about restoration, conservation practices, and response to invasives in this way. How can that discussion be started within the relevant agencies, funders, and NGOs?

[Part of the problem is, and has been, that if a field application fails 99 times out of 100, there will be at least 101 applications of the same method.   (The 2011 rangeland CEAP implies this in at least a couple of the chapters.)  Of course this is because as humans we tend to under-analyze or entirely dismiss the majority of failures and over-emphasize the successful exception which perpetuates redundancy of this inefficiency.  On a personal note, I keep a picture hidden somewhere of me affectionately clinging to my sweet, sweet confirmation bias.]
If I understand correctly, you are talking about a sea change in how society approaches prioritization which would transcend federal departmental firewalls.  Now we are talking about politics, which is sort of what I was trying to avoid in my response but ultimately is the destination of any conversation involving prioritization.  As you well know, political clout within federal agencies is limited in various ways but ultimately stops at the department level. 
If this is the case, third party influences are likely to be more effective since departmental mandates are not self-induced for the most part.  Even a cursory examination of an existing process or group of processes would help in development of some sort of proposed action that a group or a bunch of groups could get behind and influence the appropriate people. 
Since we are talking about complexity of factors far beyond anything federal agencies have ever uniformly addressed, I again point out that both success and shortfalls in watershed restoration efforts should provide a collective test case/model for strategists to consider.  I picked the watershed restoration scenario because the flaws in approach are obvious yet we still have problems, the root of which symptomatic of many similar efforts but with the outcomes being less transient and more easily tracked than other efforts of similar complexity.
I’m not a strategist, but I would suspect that any model considered would require some analysis.  What works? What doesn’t work?  What barriers consistently arise?  What are the fundamental characteristics of every success?  What is the collective outcome of the effort?  What is the price of efficiency verses the cost of inefficiency?  What specific actions would need to take place to create leaps in efficiency?  …and so on. 
As far as how to start, I believe that your foot is upon the threshold.

Brandon Bestelmeyer's picture

I sure hope we are at the threshold of change. You are right that it may be overly optimistic to expect agencies to consider climate futures in applying restoration practices when theres not much consideration of the 'climate present' either. Its no secret that the current bureaucratic structure for dispensing restoration/conservation support on private and public lands does not favor measurement or learning about effects. That will have to change first. But even when effects on vegetation/soils/wildlife are monitored, my concern is that early interpretations of the data (less shrubs, more short-lived perennial grass) would indicate a favorable response, yet these changes might ultimately compromise resource values if the future climate doesnt support the altered vegetation. I think agencies and NGOs like the Nature Conservancy should start to consider these scenarios in working groups, using inventory data and simulation models to inform the possibilities. Ranchers and farmers are on the front line as far as having an impression of the trends, so they would be a big part of this effort. An even simpler place to start is with conceptual models (including general state and transition type models) to get individuals seeing these issues with new eyes. 
Maybe others have ideas?