Conservation palaeobiology and the shape of things to come

Abstract

Conservation decision-making is a forward-looking process that involves choices among alternative images of how the future will unfold. Scenarios, easily understood as stories about plausible futures, are emerging as a powerful approach used by the conservation community to define a range of socio-ecological futures when standard, predictive modelling approaches to decision-making are inappropriate, providing a framework for making robust decisions under uncertainties. Conservation palaeobiologists can help the conservation community imagine the future. The utility of the past centres on orienting us to the present—grounding the future in the realm of what is plausible—by providing context against which to think about future scenarios, which may help stakeholders and decision-makers to develop a new mental map of a conservation problem, inspiring our intentions and moving us purposefully toward a desirable tomorrow.

This article is part of a discussion meeting issue ‘The past is a foreign country: how much can the fossil record actually inform conservation?’

1. Introduction

The great predicament of the human race is that all experiences are of the past but all our decisions are about the future. Unless we at least think we know something about the future, decisions are impossible, for all decisions involve choices among images of alternative futures. K. Boulding [1, p. 1].

In the face of declining biodiversity and ecosystem services, decision-makers are increasingly looking to conservation scientists to solve the complex and politically sensitive conservation and environmental problems facing society. The predictive capacity of science is appealing to decision-makers because it promises to define a rational course of action, reducing the need for divisive debate based on differences in individual values and beliefs [2,3]. The allure of this potential is strong.

This need for policy-relevant scientific information is often cited as a motivation for conservation palaeobiologists—not to mention conservation biologists more broadly [4–6]—to develop predictive models to inform conservation decisions, everything from predicting the vulnerability of species to extinction to where species are likely to move as the climate warms (e.g. [7–9]). As conservation palaeobiologists venture into the necessary (albeit frequently disappointing) ‘prediction enterprise’ [10], however, it is sensible to remind ourselves of the appropriateness (and hidden dangers) of predictive modelling for imagining an uncertain future.

My view is that it is desirable, possible and even necessary to anticipate the future to some extent, but, instead of making specific predictions to meet the increased demand for policy-relevant scientific information in facilitating action, conservation palaeobiologists should strive to help imagine plausible scenarios of the way our shared future may unfold.¹ Here, the questions I have in mind are: When is prediction appropriate? Are there other ways in which conservation palaeobiologists can contribute when it is not? And, why do scenarios of the future need conservation palaeobiologists?

2. The probable and the plausible

(a). The probable

Prediction means different things to different people [10]. I use ‘prediction’ to refer to the use of a quantitative model to predict, with some degree of uncertainty, the state of any variable of interest at a given point in the future, conditional on initial conditions, assumptions about drivers, measured probability distributions of parameters and choice of model [12–14]. Prediction methods for envisioning the most likely future state of the material world are appropriate when drivers are relatively well-understood and controllable, complexity is low, time frames are short and uncertainties are low ([15,16]; figure 1). In cases of high, irreducible uncertainties, however, reliable prediction may be an impossible task, seriously limiting a predict-and-prescribe strategy ([13,15,16,18]; but see [19]). It is often assumed that the poor performance of a predictive model results from knowledge gaps. To improve model predictions, all we need to do is better estimate model parameters and more precisely specify initial conditions [2,18,20].

Figure 1.

Conditions under which the use of scenario planning is more appropriate than predictive modelling as a decision support tool to envision the future (modified from [17]).

Several factors, however, strongly limit the potential for model improvement. Chief among these factors is that natural systems are always open [20–22], model input parameters and assumptions are almost always poorly known [23,24] and context and initial conditions (i.e. contingency) matter [12,24]. For example, only using geological analogues from the deep-time geological record (the ‘universal mentor’ of Bjornerud [25]) as a guide to predict change, now or in the future, is highly likely to be limited in practice and incomplete in principle because the future dynamics of socio-ecological systems are contingent on interactions among drivers that are outside the domain of the past [26]. Although the immanent processes (e.g. competition, feedback, selection and adaptation among other universal phenomena [27]) apply in all times and places, helping us to explain the general conditions of the unknown (constraining the possibilities of what the future holds), the current contingent (i.e. unique configurational; sensu [28]) circumstances of the Anthropocene world—whether we like it or not, and whether we recognize it or not—limit the possibility of predicting the specifics of the future from the past alone. ‘Even if a model result is consistent with present and past observational data, there is no guarantee that the model will perform at an equal level when used to predict the future’ [21, p. 643]. And, paradoxically, sometimes the more we learn the less we know; that is, sometimes efforts to reduce uncertainty have the opposite effect of expanding uncertainties as we learn more about the complexities of a natural system [20].

The reflexivity of human behavior further challenges our ability to make accurate predictions, shrouding the future in uncertainty (e.g. [15,26,29–33]). By making predictions about something, we can alter the actions of others, thus altering the situation and our mental map² (see also Karl Popper's [36, p. 13] discussion of how a prediction might influence an event that is predicted—either to ‘bring about’ or ‘prevent’ the event—a phenomenon he called the ‘Oepidus effect’). Reflexive interactions make predictions about natural systems at any scale that are affected by human activities and choices (either individually or collectively)—basically everywhere on the planet today—highly uncertain. In reflexive settings where human values and choices matter, a true (here, meaning accurate) statement about the future is not possible [29,37]—we can only make a provisional and heuristic statement, a proposal about the future. The standard tools used to support decision-making under uncertainty are unreliable in such situations [37]. Still, conservation palaeobiologists commonly contend that advances in theory and data collection will provide increasingly accurate and actionable predictions (e.g. [38]).

Getting things right matters when the stakes are high in conservation decisions because illusions of validity (sensu [39]; or trustworthiness), which too much predictive machinery easily can suggest, are an important source of bad decisions. The illusion that we understand the past encourages overconfidence in our ability to predict the future [39]. Decision-makers also may use predictions with little understanding of their accuracy [40,41]. If predictions are wrong, the credibility of the community that generated them is undermined, resulting in a loss of support [22,23]. Conservation palaeobiologists may be blamed for faulty predictions, even if limitations of a model are communicated [40]. Deep uncertainties may also lead to policy paralysis [13].

How do we come to terms with this situation? In my opinion, the first step is to dispel any hope of ever actually predicting the specifics of the future (except perhaps on near-time scales, daily to decadal; [42]). Most model predictions will be wrong. This is inescapable. The bigger the prediction—the more ambitious it is in time, space and the complexity of the natural system involved—the more chances there are for it to be wrong (figure 2). Thinking otherwise is a set-up for failure. ‘All our decisions are about imaginary futures and are made in the imagination, but if these imaginations are unrealistic and ill informed, decisions based on them are all too likely to be disastrous' [1, p. 1].

Figure 2.

Broadening range of plausible future states over time as surprise events and different decision options change trajectories (modified from [17]). (Online version in colour.)

This troubling reality, however, does not mean that we need to stop thinking about the future. Personally, I take a middle ground between predictive hubris and supine skepticism [43]. My view is that, rather than attempt to predict the most likely state of a variable at a given point in the future, conservation palaeobiologists could help develop ‘what if’ scenarios of the future that highlight, for decision-makers and other interested stakeholders, a range of possibilities and uncertainties of how the future may unfold. To do this, however, we may need to view the future more like the way a historian views the past [29,44] and less like the way a scientist would in striving for accurate predictions.

(b). The plausible

A scenario is not a prediction; it is a story that describes a plausible situation—futures that could be rather than futures that will be³ [15]. Unlike technically driven, predictive models, scenarios acknowledge the uncertainty, contingency and reflexivity inherent in socio-ecological systems and therefore do not attempt to predict the future; that is, no level of confidence is asserted that any specific change or event will occur.⁴ With the scenario method, we can incorporate a variety of quantitative and qualitative information to yield an informed assessment, factoring in well-known trends, multiple drivers and uncertainties about the future, of the plausible outcomes of our choices [15,16,29,46]. If prediction is a decisive statement of what the future will be, then scenarios—mental maps of the future—are heuristic statements that explore plausible futures, providing a framework for making robust decisions no matter what happens [29]. Scenarios inspire action; great uncertainties of the future do not paralyse efforts to make decisions today [46], providing support against the ‘agnostic decree … that we generally know too little to sacrifice the known for the unknown’ ([52, p. 30]; see also [53]).

Equipped with a prediction, a decision-maker—embedded in the context of larger social and environmental systems with many stakeholders—acts with confidence but sacrifices flexibility should surprises emerge [29]. Guided by a scenario (a ‘rehearsal for the future’ [46]), a decision-maker acts with caution (for the desire for accurate prediction should not overwhelm the desire of attaining ‘mere knowledge of possibilities’⁵ [52, p. 29]), but with the knowledge that multiple alternatives have been considered [29]. With scenarios, decisions are adaptive; flexibility to deal with surprises is maintained. Scenarios are also vehicles to communicate complex information about the dynamics of socio-ecological systems to a diverse stakeholder community, which presents an opportunity for people to apply this information to shape their future or adapt to changing conditions. Ideally, scenarios are also co-developed by a diverse group of people [15,46]. If scenarios are to serve the purpose of helping people prepare for change, it is important that the perspectives of the people whose futures are affected by a decision are included in developing the scenario.

Recognizing that many of today's conservation problems cannot be solved via predictive modelling approaches [15,16,55], scenarios are emerging in the conservation community as an alternative way to imagine the future. The Millennium Ecosystem Assessment [56] is perhaps one of the most well-known examples, but there are many others (e.g. [17]). For example, several long-term ecological research (LTER) sites around the USA are actively engaged in scenario planning efforts ‘to explore possible as well as desirable and sustainable future states of their respective regions' [16, p. 368]. The long-term (i.e. decadal- to centennial-scale) perspective on ecosystem dynamics provided by LTER-based science already has proved invaluable in developing scenarios for some of these sites [16]. Palaeoecological data could complement long-term ecological data by providing an even longer-term context for imagining the future. To my knowledge, however, conservation palaeobiologists have rarely participated in these discussions even though their science also can be applied to imagine the future (box 1).

Box 1. Case study: Using palaeoecology to develop scenarios of future ecosystem dynamics.

Ecosystems in the Rocky Mountain region of the USA are threatened by climate-warming induced outbreaks of native tree-killing insects, wildfires and severe drought, habitat fragmentation from development and large-scale fossil-fuel extraction and invasive species [57]. Using palaeoecological data derived from the fossil record of the past 15 000 years of ecosystem dynamics in the Rocky Mountain region, Jackson et al. [57] developed plausible and credible scenarios of environmental change in the face of an uncertain future. Their scenarios considered two drivers associated with a high degree of uncertainty—drought frequency and burned area extent—that varied along a continuum of conditions, ranging from those that are similar to observed historical records to those that are drastically different from today (but having past precedent). Beyond a business as usual scenario in which conditions were assumed to remain the same as they are today (which may lead to resource managers being caught off guard if change does occur), three scenarios were imagined based on an understanding of past ecosystem dynamics. Patches and fragments: frequent fires create a complex mosaic dominated by pioneer plant communities; novel ecosystems: frequent droughts, large fires and invasive species lead to new ecosystems that have never been recorded in the region; and inevitable surprises: extended periods of status quo are punctuated with rapid turnover of ecosystems in response to large fires. Palaeoecological data served to orient the past to the present and provide insights that grounded the future of ecosystem dynamics in the Rocky Mountain region within the realm of what is plausible. Guided by these scenarios, ‘stakeholders and decision-makers are more likely to consider multidecadal droughts, landscape altering fires and rapid species invasions as possibilities for the future because they have happened in the past’ [57, p. 82].

Image source: U.S. Department of Agriculture.

Scenarios of the future may also demand information that can only be provided by conservation paleaobiologists. For instance, geological analogues from the distant past (e.g. the Paleocene–Eocene Thermal Maximum (PETM) climatic event that resulted in the world's global temperatures rising on average by as much as 5°C [58]), though more challenging to comprehend than data from the familiar past where the species and habitats are the same as those found today, possess great potential as banks of information on drivers of ecological change to help decision-makers and other stakeholders imagine alternative scenarios of the future. It is important, however, to not overlook the fact that analogues are always imperfect because they are comparisons of different situations (e.g. rapid global warming in the coming decades is expected to occur at a much faster rate than the PETM climatic event and across a more substantially fragmented landscape, among other differences [58]). As L.P. Hartley [59, p. 1] reminds us in his novel ‘The Go-between’, ‘The past is a foreign country; they do things differently there’. What Hartley meant by this is that the past and present should not be considered on the same terms [60]. The same applies for the future relative to the present, or the past. If we only see how two situations are similar with each other we might be blindsided by the differences—the small and unknowable differences in starting points—in those situations [61,62].

In my opinion, the usefulness of the past in scenario development centres on providing an orientation (sensu [63]) to the present, helping us appreciate how the past and present have been different and how future changes may be similar to but different from those of the past or present (see also [60]). Consideration of the past—particularly of environmental contexts of biotic response outside of our experience (but toward which we might be headed)—provides valuable information that can ‘stretch’ our thinking about future changes by widening the range of alternatives considered plausible [60], helping stakeholders and decision-makers alike develop a fundamentally new mental map of a conservation problem. We cannot develop a scenario of how the future may be different without first orienting ourselves to the present using the past because—as Bradfield et al. [60] aptly stated—our consideration of the future would be ‘rudderless’.

3. A usable image of tomorrow

Because we are always uncertain as to the precise shape of things to come, the ‘truth’ value of a scenario is less important than its usefulness as a guide in helping people think about and make decisions in the present [29]—that is, informing our mental map of the future (the ‘image’ of Boulding [35]). Because the truthfulness of scenarios about the future—a domain that is beyond our direct experience—cannot be tested, what criteria can we use to judge their validity? I agree with historian David Staley [29] in thinking of the issue in terms of utility, i.e. something that allows us to accomplish a task or purpose. In other words, one can think of a geohistorically informed scenario of the future as ‘the act of seeking meaning, of telling a story that allows a group to act, to provide purpose and direction’ [29, p. 150].

4. The shape of things to come

Decision-making is always a forward-looking process, which must include an image of what a desired future will look like as far as this is possible [41]. If the past cannot predict the particulars of the future, then what is the value of conservation palaeobiology? Admitting that the past will not tell us what will happen does not mean that conservation palaeobiology cannot contribute to the prediction enterprise. Rather, the value of conservation palaeobiology in relation to the shape of things to come centres on it providing not predictive accuracy but insight and orientation. By broadening society's peripheral image of what can happen instead of what will happen, conservation palaeobiology can help us to act with wisdom, balanced perspective and good judgment to whatever does happen.

Competing interests

I declare I have no competing interests.

Funding

I received no funding for this study.