Dynamic Baselines

The fundamental goal of conservation is to protect natural systems from the deleterious effects of human activities (see Chapter 2). This remains unchanged under climate change. Indeed, it is imperative that we do not lose sight of this goal (Maxwell et al. 2016). However, the historical baselines that have generally been used to define the natural state are no longer tenable. As temperatures increase, ecosystems will adjust accordingly, which means there will be no going back to the way things were in the past, even in principle. An ecological baseline that incorporates climate change is needed instead. Establishing this baseline requires us to revisit the concept of “natural” that we developed in Chapter 7.

On the one hand, there is a clear scientific consensus that the present episode of warming is anthropogenic in origin, resulting from our release of CO₂ and other greenhouse gases (Cook et al. 2016). Because climate change is not natural (in this instance), it should logically be considered an anthropogenic threat that needs to be countered. On the other hand, once CO₂ and other greenhouse gases are released, they become indistinguishable components of the atmosphere. The threat they pose is amorphous and global in scope, beyond the purview of local conservation managers. Moreover, the process is irreversible on timescales relevant to biodiversity management (Archer and Brovkin 2008). As a global society we can (in principle) control how much CO₂ we release, but we have no control over changes in climate that arise from the CO₂ already added to the atmosphere.

The pragmatic solution is to deal with climate change on two levels: climate mitigation and climate adaptation. The idea is to differentiate the release of greenhouse gases from the climatic and ecological effects they produce. Climate mitigation refers to preventative efforts, most of which are focused on reducing the release of greenhouse gases (Hansen et al. 2013). These efforts need to be global in scope to be effective. It does not matter if the CO₂ comes out of the tailpipe of a car in Hamilton or the smokestack of a power plant in China—it all pools together. This requires international government cooperation and policies that influence the actions of businesses and individuals.

Currently, the main international effort to curb greenhouse gas emissions is the Paris Climate Agreement, to which Canada is a signatory. This agreement sets national targets for reducing greenhouse gas emissions, with the aim of keeping the global rise in temperature below 2°C (UN 2015). Unfortunately, global emission reductions to date have been uneven and collectively fall substantially short of what is required to achieve the Paris Agreement goal (Raftery et al. 2017).

The climatic changes that result from greenhouse gas emissions demand a different response. What is needed is adaptation rather than mitigation. Climatic changes cannot be reversed, and efforts to prevent ecological systems from responding to changing conditions would be counterproductive and ultimately futile (Dunlop et al. 2013; Hamann and Aitken 2013). Biodiversity is best served by treating all forms of climatic change as natural phenomena and helping species adapt to these changes.

Because the climate is changing, the ecological baselines we use to define working objectives for conservation need to become dynamic, describing the natural trajectory of change rather than the natural historical state (Simberloff 2015). The natural trajectory is what we would observe within a large, pristine protected area over time (Murcia et al. 2014). Indeed, monitoring the climate-induced transitions that occur within protected areas is one of the approaches that can be used to characterize dynamic baselines.

In summary, climate change presents a threat to biodiversity that must be addressed proactively, through the control of emissions, rather than after the fact. Climatic changes that occur despite preventative efforts are essentially irreversible and should be accommodated rather than resisted. This can be accomplished by defining a dynamic ecological baseline and focusing on facilitating climate adaptation, rather than resisting ecological change. The aim is to ensure that climate-induced ecological transitions unfold as they would in undisturbed systems. For the most part, this means attending to conventional threats using conventional methods. We will examine dynamic benchmarks and related conservation methods in more detail later in the chapter, in the context of specific conservation applications.

A remaining challenge with climate adaptation involves the social dimension of conservation. Unmoored from the perceived objectivity of the preindustrial landscape as a benchmark, support for conservation may waver. Some voices are already beginning to question whether any systems can be considered natural, and thus worth conserving, in a world of constant change (Murcia et al. 2014). Harris et al. (2006) write:

We must tread very carefully. A consequence of rapid climate change may be the loss of public interest in conservation and restoration goals. Inured to the change, the idea of supporting painstaking restoration goals will give way to functional, emergent, and designer ecosystems. (p. 175)

A designer ecosystem is a system engineered to provide specific ecosystem services (Higgs 2017). The concept arose in the context of restoring highly degraded sites, but it is now being discussed in the context of the novel ecosystems produced by climate change (Higgs 2017; Backstrom et al. 2018). This raises serious concerns, as expressed by Murcia et al. (2014):

What is at stake is whether we decide to protect, maintain, and restore ecosystems wherever possible or else adopt a different overall strategy, driven by a vision of a ‘domesticated’ earth, and use a hubristic, managerial mindset. (p. 552)

While it is true we can no longer keep natural systems exactly as they were in the past, this is not grounds for abandoning the fundamental tenets of conservation. The value of biodiversity does not diminish in a warmer world, and protecting species from the deleterious effects of human activities is no less important. After all, species are not changing, just their location. Conservation practitioners need to help the public and decision makers understand these distinctions and ensure that fundamental concepts about biodiversity and conservation are not abandoned.

Climate Scenarios

Another step in adapting conservation to climate change is incorporating climate projections into the planning processes we have discussed in previous chapters. Conservation practitioners should have a basic understanding of how these projections are created and how they can be obtained and utilized. It is also important to understand their limitations.

The process begins with projections of future greenhouse gas emissions. These emissions depend on global population growth, economic growth, land use, technological innovations, and most importantly, social awareness, concern, and willingness to respond (Van Vuuren et al. 2011). Because of the wide range of possibilities, emission trajectories are not amenable to quantitative modelling. Instead, the Intergovernmental Panel on Climate Change has developed a suite of four emission scenarios, termed Representative Concentration Pathways (Van Vuuren et al. 2011). As discussed in Chapter 7, scenarios allow us to explore plausible alternative futures without committing to them as forecasts (West et al. 2009).

The second stage in projecting the future climate involves simulating global climatic processes on the basis of fundamental physical principles. These climate models, formally referred to as general circulation models, have been developed by several teams around the globe. By using the same set of emission scenarios as inputs, comparisons can be made among models and among emission scenarios.

Climate models are evolving rapidly, but many climatic processes, especially those involving feedback loops, are still only partially understood. As a result, the climate models developed by the various modelling teams differ in important ways and produce different results under the same emission scenarios. This variance among models is an expression of modelling uncertainty. As a rule, the farther into the future that projections are made, the higher the level of uncertainty. The year 2100 is used as the limit for most management applications.

The climate projections from more than 20 international models are readily available to conservation practitioners (Wang et al. 2016b). It is not data acquisition that presents a challenge, but data overload. Working with the temporal projections from all available models and all four emission scenarios is not practical.

The obvious solution is to focus on the most reliable models; however, reliability is not easily determined. The available comparative studies tend to focus on specific climatic processes, such as the simulation of clouds, rather than overall performance (Jiang et al. 2012). In any case, there is really no gold standard to test against. Examining how well the models replicate past climatic patterns is helpful, but it is not a dependable guide to their reliability in future periods when CO₂ concentrations will be much different.

A better approach is to treat the entire gamut of climate projections as scenarios, rather than predictions. We can then select a subset of these scenarios to represent the full spectrum of potential climate outcomes. This is illustrated in Fig. 9.8, which shows the temperature and precipitation projections to 2080 under high- and low-emission scenarios as predicted by a suite of climate models. The four peripheral circles are candidates for representing extreme scenarios (i.e., coolest, hottest, wettest, and driest). The central circle would be an appropriate choice to represent an intermediate climate scenario. The coolest scenario has special relevance for management because it carries a high level of certainty—there is complete agreement among all models that the climate will become at least this warm.

Fig. 9.8. The projected increase in mean annual temperature and precipitation to 2080 for 18 climate models. The grey diamonds represent a low-emission scenario, and the black squares represent a high-emission scenario. The coloured circles are candidate climate scenarios for use in planning applications. Data are for the province of Alberta, adapted from Schneider 2013.

An alternative approach is to pool the projections from all models into an ensemble mean. This approach is less complicated but provides no insight into the range of outcomes possible. Consequently, it is not appropriate for planning horizons beyond mid-century, when climate projections begin to diverge significantly.

The last step is to link the climate projections to the models used to support management decisions. Many of these models include climate-sensitive parameters that can be dynamically adjusted on the basis of the projected future climate. For example, Battin et al. (2007) incorporated climate change into a population model for Chinook salmon via parameters for water temperature and stream flow. This enabled them to explore the effects of warming on habitat restoration efforts (Fig. 9.9). Other climate-sensitive parameters commonly used in decision support models include winter survival, soil moisture, and the rate of natural disturbances.

Climate projections can also be used in combination with bioclimatic envelope models to predict changes in species distributions and changes in habitat conditions, as previously discussed. This information can be integrated into decision support models or used to directly inform decisions. The main caveat is that the bioclimatic envelope projections represent equilibrium conditions and do not account for the time needed to achieve these conditions. Furthermore, our ability to predict changes under the hottest scenarios is limited. This is uncharted territory, and our statistical models may break down under such conditions.

Robust Decision Making

Having discussed dynamic baselines and the incorporation of climate change into decision support models, we now turn to the decision-making process itself. Because the baseline is no longer fixed, greater consideration has to be given to the long-term repercussions of conservation actions. The best course of action in the near term may not be optimal over the longer term, setting up the potential for trade-offs among time periods.

Planning efforts must also grapple with the added uncertainty that climate change presents. In conventional planning, we use models or expert opinion to forecast outcomes under alternative management approaches, and then select the option that best achieves the stated objectives. It is understood that these forecasts are subject to uncertainty, but we assume they are reliable enough to differentiate the performance of the management alternatives under consideration. To backstop this assumption, efforts are made to identify points of uncertainty and to address these uncertainties through additional research. In addition, the state of the system is monitored over time, and variances between the plan and actual outcomes are addressed through periodic replanning.

For short-term planning (i.e., time horizons under ~20 years), conventional planning approaches remain viable, despite the added uncertainty from climate change. Climate model projections are reasonably consistent at this early stage (Fig. 9.10). Moreover, the amount of warming is relatively subdued, compared with later periods, and is unlikely to result in unpredictable ecological outcomes.

Fig. 9.10. The range of potential future climates diverges significantly in the latter half of the century because emission scenarios become more distinct and climate modeling uncertainties increase. Adapted from Schneider 2013.

Climate change presents a much greater challenge for long-term planning. The range of potential future climates expands significantly in the latter half of the century (Fig. 9.10), forcing us to reconsider the meaning of optimality. The performance of management approaches—and hence our assessment of which is best—is unlikely to be the same under distinctly different climates (Fig. 9.11). The relative rankings may change under different climate scenarios, and there may be no management approach that is consistently optimal under all conditions.

For such situations, a “robust” or “no regrets” approach to decision making may be most appropriate (Millar et al. 2007; Kunreuther et al. 2013). The basic idea is to select a management option by how well it performs across all potential futures, rather than just one. The simplest method is to select the management option with the highest mean score across all options (this would be Option C in Fig. 9.11). Alternatively, priority might be given to the management option that has the least variance or the best worst-case outcome (Option B in Fig. 9.11). There are also more sophisticated mathematical approaches, such as the minimax-regret method, that permit the preferential weighting of scenarios by their perceived likelihood (Kunreuther et al. 2013).

Conservation approaches that perform well across varying conditions come in various forms. One approach is to enhance a system’s overall resilience (Seidl 2014). Efforts to reduce the intensity of industrial impacts and to limit cumulative effects fall into this category.

Another way of achieving robust performance is bet-hedging (Millar et al. 2007). Bet-hedging addresses deep uncertainty by simultaneously applying different strategies across the landscape. This is basically a risk-spreading strategy, which is useful for avoiding widespread management failure when the outcome of potential strategies can not be predicted in advance. The bet-hedging approach can also serve as an effective method of increasing knowledge, especially for new approaches that have not yet been adequately field tested. There is overlap here with the concept of adaptive management, which we will discuss in Chapter 10.

A variant of bet hedging is the optimal portfolio approach (Crowe and Parker 2008; Ando and Mallory 2012). In this case, instead of applying different actions in different places, a single multipronged strategy is applied throughout the planning area. For example, in the context of reforestation, diverse seed stock might be used to maximize genetic and species diversity across the landscape, in the hope that some genotypes and species will thrive regardless of how the climate changes.

Institutional Support

Adaptation to climate change can be enhanced or hindered by institutional structures and norms (Williamson et al. 2012). Though there is now widespread awareness of the need for adaptation, there is still great uncertainty about what should be done and there are various barriers to implementation (see Box 9.2). Initial efforts have focused on information gathering and dissemination, vulnerability assessments, and research.

Structural changes to decision-making systems will eventually be needed. These are only now being contemplated and will take time to be realized. The challenge is to develop a system that embraces flexibility while safeguarding against activities that are inconsistent with the aims of conservation and abuse by actors seeking to avoid environmental regulation (Craig 2010). Furthermore, while we must accept that management outcomes are less predictable under climate change, companies and government agencies must still be held accountable for the decisions they make and the actions they take (Hagerman et al. 2010a). It is as yet unclear how this might be accomplished.

Because climate change is occurring at scales much larger than even the largest planning regions, successful adaptation will require collaboration among jurisdictions (Heller and Zavaleta 2009; West et al. 2009). Additional funding and staff will also be needed. At present, managers wishing to implement adaptation programs usually have to scavenge funds from other programs, which is not a viable long-term solution. Finally, there is a need for enhanced monitoring and additional research. Pilot projects are a promising approach, serving as laboratories for identifying and solving the many practical issues that must be addressed.