The implementation gap in conservation biology: is math contributing to the problem? (guest post)

Note from Jeremy: This is a guest post from Allison Moody, a postdoc in conservation biology at the University of Maine. She’s a regular reader and commenter here, and has also corresponded with me on topics relating to my recent E. O. Wilson post and the blog more generally. Her comments are always very thoughtful, and so I invited her to share some of her thoughts in the form of a guest post. So here’s Allison, on the “implementation gap” in conservation biology, whether mathematical modeling is part of the problem (!), and how it could be part of the solution.

*************************************************

Conservation biology as an academic discipline is a relatively new field. The two major conservation biology journals, Biological Conservation and Conservation Biology (not the most creative bunch when it comes to naming) were founded in 1969 and 1986. Obviously, the published work in conservation biology has increase significantly since then (Figure 1). Conservation biology has a unique place in ecology because it’s a combination of theoretical science and applied science, although other ecological fields also have this combination (e.g. wildlife management and invasion biology to name two). It has also often been referenced as a ‘crisis-driven’ field; one that addresses problems after they occur and less frequently, tries to prevent problems from happening in the first place.

moody figure

Fig. 1: Web of Science search for topic = conservation biology, showing published items in each year. Timescale runs from 1935-2013. Yep, it doesn’t go back farther than that.

All along there have been cries of What about implementation?! The research-implementation gap in conservation biology has been noted since the early days of the field. A survey of authors published in Conservation Biology between issues 1 and 12 (1986-1998) showed that of the 223 respondents, 78% (n = 173) had included management recommendations but of these, only 54% (n = 164) believed their recommendations were being used (Flashpohler, Bub and Kaplin 2000). A more recent survey of conservation assessments published between 1998 and 2002 indicated about a third (n = 29, total n = 88) of conservation assessments led to any implementation, and of these, only 14 were “highly effective”, whereas 21 were “poorly effective” or “ineffective” (Knight et al. 2008).

So, my question is, do the mathematics and statistics published in conservation biology serve as one of the barriers for getting research used by actual conservation practitioners?

Backing up, there is a difference between conservation biologists and conservation practitioners although sometimes these are the same people. Conservation biologists are generally employed by universities as academics and they have the general pressures of academics (teaching, tenure, publishing (publishing, publishing)). Conservation practitioners, on the other hand, are a much broader group that includes non-profit organizations, land managers, politicians, private landowners, etc. If you bought some land and decided you’d like to encourage more Eastern Bluebirds, congratulations, you’re a conservation practitioner! Conservation practitioners have little time to read the literature and often no access, especially in developing countries. They may have undergrad degrees or be many years out of school which means they may not have a thorough, up to date grounding in math and stats.

So, conservation practitioner, what are you going to preserve and how? The conservation biology literature is full of methods for prioritizing species, areas, and actions. Complex models have been developed that will tell you exactly where you need to conserve to preserve a specific suite of animals at some degree of redundancy. But, for you on the ground, you know that one specific patch that is really important for the conservation objective is right in the middle of a planned development and its cost has gone up 10 times since the paper was published. Sure, you could contact the authors of the paper and request they integrate a cost function, or you could just buy this other patch over there that you can afford. Conservation in practice isn’t just about ecology, it involves economics, politics and sociology. The math and statistics involved in modelling are just a couple tools among many that conservation biologists must have to be effective.

It’s cliché, but still my favourite saying – all models are wrong, but some are useful – but useful for whom? In conservation biology, as in any applied ecological field, it has to be useful for the practitioners, not just the people who generate the models. Remember Flaspohler, Bub and Kaplin’s paper above? The #3 reason (behind “agency initiative” and “agency recognition of problem”) authors felt their recommendations were being used, was “recommendations were easily understood” (12%, n = 24). Conservation decisions are always being made by practitioners and it’s important that their decisions are supported by science but to a large degree, it is the conservation biologists who must ensure the models or decision support tools they develop help practitioners directly.

What are the keys to making sure you’re making a useful model?

  1. Make sure you’re asking the right question. Academia and publishing have different objectives for a study (e.g. novelty, broad focus) than developing tools to help conservation practitioners. Conservation practitioners may be more interested in well-tested decision support tools and a local focus (although not always; Shaw, Wilson and Richardson 2010). To ensure you’re asking the right questions, practitioners have to be involved early in the process and have their input actually guide the process rather than just be for show.
  2. Model transparency. Explaining the biases and assumptions in a model is also key to having practitioners accept your model outcomes. To convince someone to make a decision using your decision support tool, the process must be transparent and you must be open to defending your process. Transparency also helps with buy-in because people know what they are agreeing to.
  3. Training. Getting practitioners in early to help develop the question(s) and the methodology for analyzing the problems also helps you explain the process. Remember that these people may not have taken a stats class in 10 years or ever. Technical terminology can be especially dense to practitioners. But conservation biologists also need to accept they need training to work with practitioners. There is a lot of literature out there about expert elicitation, decision theory, and risk analysis; all of which can be important aspects of conservation.

I think among the different ecology fields, the best, most interesting and cleverest modelling is happening in wildlife ecology and conservation biology (I predict this statement might be controversial). Most of my colleagues and I are math and/or stats nerds who are jumping at the chance to stretch our brains into multiple dimensions. We just need to remember that these fields are defined as ‘applied’ and that we need to take our models into the real world.

I want to end by thanking Jeremy for giving me the opportunity to write a guest post. I come from a different background than the front page posters here at the blog, and instead of Jeremy just saying ‘Well, we don’t do that so we won’t cover that’, he responded by asking me to share my point(s) of view. Obviously this post was spurred by the Wilson vs. Fox debate but this issue is something that I’ve been thinking about for a long time now. My first research was on animal behaviour and I’ve done some lab biology before switching to a more applied field. I’ve always been interested in politics and policy and making the switch to applied ecology, and specifically conservation biology, was a really fun and interesting jump. Even though my PhD is in conservation biology or wildlife ecology depending on who you ask (me or my PhD granting university, respectively), I am definitely still learning about the field and what is above is obviously my opinion.

References:

Flaspohler, DJ, BR Bub, and BA Kaplin. 2000. Application of conservation biology research to management. Conservation Biology 14: 1898-1902.

Knight, AT, RM Cowling, M Rouget, A Balmford, AT Lombard, and BM Campbell. 2008. Knowing but not doing: selecting priority conservation areas and the research-implementation gap. Conservation Biology 22: 610-617.

Shaw, JD, JRU Wilson, and DM Richardson. 2010. Initiating dialogue between scientists and managers of biological invasions. Biological Invasions 12:4077-4083.

16 thoughts on “The implementation gap in conservation biology: is math contributing to the problem? (guest post)

  1. I think the main arguments of Alison is interesting and relevant, but even if managers decide to read and understand the models produced by scientists, they are of no use for politicians.
    Politics don’t read papers. And worst than that, scientific data does not bring any vote for a candidate. If you are a politician and are looking for votes (particularly in the 3rd world) and you have to choose: (1) inaugurate a soccer field where it will be attended by 1,000 people or (2) inaugurate a new protected park that will be attended by 10 people. So, I don’t believe that the major problem in conservation is the models or the math.

    • Politicians are a whole other issue definitely. But I’m not just talking about getting new areas conserved; I’m talking about managing what you have in the best manner possible. That means that who you’re dealing with is less likely to be a politician or even elected, and more likely to be a career manager/scientist.

      Managing politicians is probably best done using personal relationships and flattery, not math

    • I really don’t intend to pick a fight: but I think you are being overly pessimistic.

      Even if politicians studied at the Orwellian School of Leadership (which I don’t believe is always true), there is always the NGO and private sectors that can implement evidence-based conservation.

      And let’s not forget, it’s a great political power-play to show science in action, especially if you’re the Minister of Science and Technology (or Environmental Affairs) and your political party’s logo covers all the leaflets, bill boards, advertisements etc.

  2. Academic researchers who tend more to the “applied” end of the spectrum also face the challenge / problem of publishing. Sure, managers of Reserve A would like to know the effect of introduced rats on their plants, mammals, birds, herps, etc., but unless the effect is fantastically catastrophic (or the affected species rare or endangered), good luck getting it published anywhere by the Conservation Journal of Eastern-Central Saskatchewan (Series B) – in other words, not in a “top” conservation or ecological journal, and not likely to be of much benefit to someone trying to land a faculty job or get tenure (depending on a number of things, I know).

    So the result, is that theory and modelling outpace practice, no matter the location, because those are the only kinds of papers that appear in Conservation Biology, Biological Conservation, Biological Invasions, etc., and those are the kinds of journals looked on favourably by granting agencies, hiring committees, and in tenure and promotion dossiers.

    And this in turn drives many of us, who are inspired by Leopold, or Thoreau to know, understand, and conserve a system, farther and farther away. True, there are some journals that specialise in this type of research (Conservation Evidence comes to mind), but by and large, we’re forced to couch this important, but localised research in pseudo-broader frameworks.

    The effect of introduced rats on a variety of taxa has been well documented, but that doesn’t diminish the importance to the conservation practitioner of what rats are doing to the flora and fauna s/he is trying to conserve.

  3. I really enjoyed this post and agreed with almost everything.

    We shouldn’t see the implementation gap as some sort of hazard that separates applied and fundamental fields… it’s an amazing opportunity to unite both branches of ecology! I am certain tenure committees, grant agencies and journal editors will be very impressed by someone who manages to implement complicated theory in the real world. Similarly, such people will have no difficulties finding a job as a conservation practitioner.

    I think among the different ecology fields, the best, most interesting and cleverest modelling is happening in wildlife ecology and conservation biology (I predict this statement might be controversial). Most of my colleagues and I are math and/or stats nerds who are jumping at the chance to stretch our brains into multiple dimensions. We just need to remember that these fields are defined as ‘applied’ and that we need to take our models into the real world.

    I don’t necessarily think that models need to be simplified to be useful in the real world: and by simplified I mean less informative (i.e. dumbed-down). In fact, I think the exact opposite. An elegant and widely-useful model is MUCH harder to produce than a convoluted (although technically correct) one. Once people acknowledge this, then we will no longer be faced with the false dichotomy of advancing our academic careers instead of implementing on-the-ground initiatives.

    • Unfortunately, I think working with someone else’s elegant and useful model in a new system is much less publishable (and therefore beneficial to you personally in terms of your career) than creating a new model for the new system. And I think this is how the math gets more and more complex.

      • I am not a conservation biologist, but from my experience I would say the opposite holds for most subdisciplines in ecology. Application of simple generic models to your system would be of interest to a broader audience than the development of a new model. It is quite a time commitment to read through a paper presenting a new model (how does it work? are the assumptions reasonable? etc.). So for someone to invest time to go through and read the paper they must be very interested in the system-question combination the model was designed for. On the other hand, when you used someone else’s elegant and useful model in a new system people who aren’t interested in your specific system may also be interested, and reading your paper requires on average less “activation energy” as more people are already familiar with the model you used.

        As for model complexity, I would wager there is a strong inverse relationship between number of parameters used in a modeling paper and the number of citations it receives.

  4. Pingback: The implementation gap – not just a problem for conservation biology | The Lab and Field

  5. Nice post Allison.

    A question: you remark in passing that conservation practitioners aren’t always interested in a local focus. Can you elaborate? And is there greater potential for mathematical theory (esp. the sort of general-purpose theory that leading academic journals publish) to contribute to “non-local” conservation practice?

    One possible example might be the IUCN Red List criteria, which have been adopted in only slightly-modified form as the criteria for designating species under Canada’s Species At Risk Act. It’s my understanding that those criteria are based in substantial part on simple, general-purpose models of stochastic population dynamics developed by Russ Lande. So perhaps this is one role of general-purpose models not tied to any specific system: to suggest general “rules of thumb” on which we rely in the absence of more system-specific information?

    Another question: do methods for prioritizing areas, species, etc. get applied higher up in the organizational hierarchy at big NGOs like the Nature Conservancy or WWF? So that models help to steer the broad priorities and budget allocation at some high level, but don’t matter so much when it comes to the nitty-gritty of local NGO branches negotiating individual land purchases or conservation easements or whatever? Hopefully my question here is not too naive…

    • A question: you remark in passing that conservation practitioners aren’t always interested in a local focus. Can you elaborate? And is there greater potential for mathematical theory (esp. the sort of general-purpose theory that leading academic journals publish) to contribute to “non-local” conservation practice?

      In this particular case, the authors surveyed conservation practitioners at a conference that had some applied and some theoretical talks. These were the labels that the scientists had given their own talks. The practitioners felt the applied talks were too particular to another area (too local) to be useful to them. The theory talks spurred their thinking about broader issues. I don’t think it was the mathematical theory that particularly interested the practitioners though.

      perhaps this is one role of general-purpose models not tied to any specific system: to suggest general “rules of thumb” on which we rely in the absence of more system-specific information?

      General models are great and necessary for conservation biology. I think sometimes we as scientists get too focused on the differences between systems. Things like PVAs are really useful for figuring out how big to make reserve patches (number of individuals needed for a sustaining population * territory size) but scientists get hung up on what a sustaining population means or how to measure territory size and never think about what’s ‘good enough’. Or, given the risks, what direction you want to err in.

      [D]o methods for prioritizing areas, species, etc. get applied higher up in the organizational hierarchy at big NGOs like the Nature Conservancy or WWF? So that models help to steer the broad priorities and budget allocation at some high level, but don’t matter so much when it comes to the nitty-gritty of local NGO branches negotiating individual land purchases or conservation easements or whatever?

      Yes, definitely. Things like the Important Bird Areas and Myers et al (2000) Biodiversity Hotspots have allowed large organizations to coordinate conservation efforts. In the US, the bird-centred joint ventures and the more general Landscape Conservation Cooperatives also work at large scales with governmental and NGOs. Downscaling these efforts is a lot harder so I’m not sure they are terribly useful to local efforts. I think at large scales you can produce useful decision support models without integrating a lot of political, social or economic information. You can just ignore political boundaries and land costs. But once you’re on the ground, those issues become the most important restrictions on what you can do.

      • “I think at large scales you can produce useful decision support models without integrating a lot of political, social or economic information. You can just ignore political boundaries and land costs.”

        I think it’s precisely the opposite. The key drivers of conservation priority (biodiversity, land cost, threat rates, political feasibility) vary more dramatically at international scales than within nations. If you were deciding where to prioritise conservation action within the Americas, you couldn’t ignore the fact that land in Mexico is cheaper than land in the south-west United States. Or that Venezuela has a higher corruption index than French Guiana. Or that El Salvador has a higher population density than Honduras. These differences dwarf the inter-state or inter-county variation you’d observe in the USA.

        “Conservation in practice isn’t just about ecology, it involves economics, politics and sociology. The math and statistics involved in modelling are just a couple tools among many that conservation biologists must have to be effective.”

        I definitely agree with this first point, but I don’t quite see how the next point follows. Surely the only way to coherently and transparently integrate ecology, economics, politics and dynamics is through modelling? And the only way to parameterise these models, or sort the good models from the bad is by statistical methods?

  6. Thanks for a very interesting post.

    Though I would like to believe the gap stems from the problems you discuss, I think that differing objectives between research and application may play a much larger role. I suspect that scientific papers that are most useful and influential for conservation practitioners and policymakers are those which confirm what they already believe, or whatever the interests opposed to them least want to hear. Let’s call these “Cassandra” papers, since in this context they usually forecast disaster. For the practitioner it may matters little whether the math is simple or complex, clearly explained or impenetrable, or even right or not so right. Worm et al 2006 paper which the media quickly decided predicted the end of global fisheries within 50 years is perhaps a good example.

    Okay, so beyond bolstering arguments already being made by those who propose, implement or legislate conservation against their opposition, there are certainly unknowns that they might turn to research to answer. Resource allocation might be an example of this; e.g. do we prioritize purchasing pristine areas that are not likely to be threatened or less pristine areas in more immediate danger (a la Pfaff). Let’s call these “rule of thumb” papers, where the conclusion is an easily applied guide-line. It seems doubtful that the practitioners would be inhibited by their access and understanding of the math in this case, since they want the research to provide an answer they can trust, and not worry so much about what math justifies it. They are more likely to use proxies of quality (journal, researcher, affiliation, popularity of the method), then working through the assumptions to see if they like them; no?

    So there is a third case in which the conclusion is of the sort “apply my method and it will tell you what to do”, as opposed to “here’s what to do”. I think only this case falls at risk to the mathematics being a barrier, though when accompanied by user-friendly software tools perhaps that can be dismissed as well. These “methods” papers are probably the favorite option of many researchers, as they seem the most rigorous, accurate way, reflecting the details of the problem at hand. Scientists are probably least fond of the first example, where even the paper’s authors may feel the conclusions are being overstated, while others feel such work is wrong and counterproductive. I’d guess many researchers are lukewarm towards the middle case, as better than a coin flip. I imagine from the conservation practitioner’s ranking is reversed. To what extent would you agree with this classification? If so, how is the conservation literature distributed across these categories, and how might we want it to be distributed? Do we indeed have the greatest impact writing Cassandra papers rather than writing nice clear methods, and if so, what are the implications?

    • I suspect that scientific papers that are most useful and influential for conservation practitioners and policymakers are those which confirm what they already believe

      I might agree with you in regards to policymakers but I don’t really know enough about them to say. But the conservation practitioners I know are people trained in science (although they may not with PhDs) and often their differences in opinion about what papers are important actually come down to their differing objectives, i.e. conserving birds while keeping the deer population high. I don’t think conservation scientists should mandate what is important science just because we think it’s correct; again, we really need to listen to the practitioners to figure out what they want.

      I’m a unclear about your three cases so I’m not going to address your questions directly. I think that telling practitioners what to do is exactly the wrong solution if you want your science to impact conservation on the ground. People generally don’t like being told what to do and I don’t think conservation practitioners are any different (to bring another cliche into things – horse, water, drink, etc.).

      But your general point about writing papers I think misses the point, and is something people above bring up – publishing is skewed towards certain types of papers which differ from papers useful to conservation practitioners. Also you have to assume that conservation practitioners aren’t scanning the literature regularly. Both of these factors mean that scientists have to actively engage with practitioners.

      • Very good points about engagement going beyond writing papers; I only focused on them from the examples. I don’t have much experience in conservation practice, coming from the far-removed mathematical modeling background myself; so I appreciate the discussion here!

        From your own experience, would you suggest that “the gap” is driven more because academic researchers pursue largely the “right” questions, but fail to communicate due to math and stats barriers, or that the gap is more due because we are not listening to practitioners to figure out what they want in the first place?

  7. This is in reply to Mike Bode’s comment above – I couldn’t get a reply button.

    There is a benefit to ignoring political or social boundaries at the international level. A lot of these large scale prioritization efforts like the hot spots of conservation were to draw attention to broad issues like species biodiversity and threat levels. It lumps a lot of different rankings into one measure which smooths out differences.

    Another issue is there are few organizations actually making these trade-offs at a large scale. WWF is probably one of the few organizations that actually can pick between conservation efforts in Venezuela or French Guiana. And there again, overall there might be more corruption in French Guiana (I know nothing about this so sorry about insulting entire countries) but if you have a large landowner in FG who you can work with, you still might be better off prioritizing conservation in FG rather than in Venezuela.

    Which leads to your next question, yes there are methods of integrating social and political issues into making these conservation plans. Prioritization in marine areas is especially concerned with these issues and is probably leading the way; Possingham’s group in Australia is also working on these issues in the terrestrial realm. We definitely need to get more sociologists and economists directly working on these issues with us instead of trying to model them ourselves.

  8. Pingback: Friday links: zombie ideas in evolution and psychology, a world without statistics, and more | Dynamic Ecology

Leave a Comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s