Incorporating climate change into systematic conservation planning

Hunter et al. (1988) first suggested a strategy to address climate change by conserving a diversity of landscape units defined by topography and soils. Since that time, several conservation plans have relied on combinations of soil and topographic variables as surrogates for biodiversity features for regions that lack biological inventories to underpin regional planning exercises (see Table 1 in Beier and Brost 2010). Given the uncertainties associated with projections of future climate changes and their spatial expression, the use of geophysical variables as planning elements has resurfaced as a practical alternative to conservation planning approaches that rely on modeling of potential climate change impacts.

At its core, this approach involves focusing conservation efforts on the underlying physical environment—the metaphorical stage—instead of the species or the actors. A recent analysis by Anderson and Ferree (2010) in the northeastern United States provides strong evidence for the merits of this “saving the stage” strategy. They demonstrated that the number of species found in 14 northeastern states and adjacent provinces can be accurately predicted from the number of geologic classes, the elevation range, the latitude, and the amount of limestone bedrock (Fig. 1). If geophysical diversity maintains species diversity, then conserving geophysical settings offers an approach to conservation that conserves diversity under both current and future climates, although the species constituting the diversity may change through time.

Beier and Brost (2010) advocate using recurring landscape units as conservation features. These units, which they call land facets, are defined on the basis of geology, soil, and topography and are similar to those used by Anderson and Ferree (2010). Based on findings from several previous studies, they argue that such units can serve as useful surrogates for today’s biodiversity and tomorrow’s climate-driven range shifts, and help conserve ecological and evolutionary processes. Because land facets cannot serve as surrogates for all species (Beier and Brost 2010), such an approach should be used as a complement to existing systematic conservation planning processes that also focus on land cover and species as conservation features.

For conservation organizations, this approach to adaptation requires a shift from focusing on individual species and communities or ecosystems defined by dominant vegetation to geophysical settings. However, this shift is neither philosophically nor practically as large as it might seem. For example, Anderson and Ferree (2010) demonstrated that most rare species and community types that many conservation organizations in the northeastern U.S. strive to conserve already represent unique environmental settings, precisely because species and settings are so correlated. In addition, systematic planning efforts in the marine and freshwater realm already focus on physical habitats because of the lack of biodiversity information for many of these communities (Higgins et al. 2005).

In many ecosystems, climate change is already leading to rapid ecological change that can be construed as negative for biodiversity conservation (e.g., bleaching events for coral reefs—Berkelmans et al. 2004; drought-related mortality of Pinus edulis in the southwestern United States—Breshears et al. 2005). Because the probability, speed, type, and extent of these changes is unlikely to be uniform across a region, a relatively straight forward and intuitive approach to adaptation in regional conservation plans is to focus on identifying and protecting biodiversity in those areas least likely to undergo rapid climate-induced changes. Such places may serve as important climate refugia for species and habitats that become marginalized through ecological changes elsewhere. Climate refugia can exist both in places where changes in climate are attenuated (e.g., Saxon 2008), or where biodiversity is likely to be particularly robust to changes in climate, perhaps due to a broad climate tolerance (e.g., West and Salm 2003). For example, as part of a national conservation plan for Papua New Guinea (PNG), Game et al. (2011) identified climate refugia based on projected changes in seven climate dependent variables (potential evapotranspiration, precipitation/potential evapotranspiration, precipitation of the coldest quarter of the year, precipitation of the warmest quarter of the year, mean temperature of the coldest quarter of the year, mean temperature of the warmest quarter of the year, and average monthly temperature) (Fig. 2). The current value for these variables in 5-km pixels was compared with their projected value in the year 2100, and the expected change normalised with the value 1 being assigned to the pixel expected to experience the greatest climatic change across PNG.

There are multiple ways to define refugia from climate change, and different definitions require different methods of identification and data inputs. Ashcroft (2010) recommends that discussions of refugia explicitly distinguish between macrorefugia and microrefugia (i.e., the scale at which refugia are being identified, and therefore what resolution climate data are necessary or appropriate), in situ and ex situ refugia (whether refugia from future climate change are likely to be located within or outside of a species’ current distribution), and refugia based on climatic versus habitat stability. The issue of scale is particularly important as it has been shown to influence patterns of species richness and species turnover, particularly as they relate to changes along environmental gradients (Jetz and Rahbeck 2002). For example, species turnover in amphibians and birds is closely linked to environmental turnover, and this effect is more pronounced in tropical than temperate realms (Buckley and Jetz 2008), suggesting that tropical systems may be more susceptible to impacts from climate change. These results suggest that some areas identified as refugia and also containing high species richness and turnover may represent “win–win” situations for conservationists.

Conserving climate refugia represents only a partial solution to climate change adaptation. Many areas exposed to large climatic changes may become or remain important areas for biodiversity even if they contain a different suite of species. Similarly, identifying refugia relies largely on climate projections with all their associated uncertainties. While it is particularly hard to predict what species and communities are likely to colonize an area as a result of climate change, we have a better ability to predict what species and communities may be lost from an area. Conserving projected refugia will offer some ecosystems a better chance of adapting to climate change, but it certainly does not guarantee their viability. As such, the potential for an area to serve as a refugium should not be used as the sole basis for identifying important conservation areas.

A recent modification of the climate refugia approach involves identifying areas where high topographic diversity creates a wide array of microclimates in close proximity (Ashcroft et al. 2009; Fridley 2009). Because coarse-scale climate envelope models often fail to capture topographic or “microclimatic buffering” (Willis and Bhagwat 2009), they may overestimate or misrepresent the projected extinction rates for a given area. Thus, the climates experienced by individual organisms may differ dramatically from the regional norm and species are likely to shift their locations to take advantage of nearby microclimates.

Increasing landscape, watershed, and seascape connectivity is the most commonly cited climate change adaptation approach for biodiversity management (Heller and Zavaleta 2009). From an adaptation perspective, maintaining or improving the linkages between conservation areas serves at least two purposes. First, it provides the best opportunity for the natural adaptation of species and communities that will respond to climate change by shifting their distribution (Fig. 3). Second, improving connectivity can improve the ecological integrity of conservation areas, thereby enhancing the resilience of ecosystems to changes in disturbance regimes characteristic of climate change in many places. Even in the absence of climate change, connectivity is considered important to prevent isolation of populations and ecosystems, provide for species with large home ranges (e.g., wide-ranging carnivores), provide for access of species to different habitats to complete life cycles, to maintain ecological processes such as water flow (Khoury et al. 2010), and to alleviate problems deriving from multiple meta-populations that are below viability thresholds (Hilty et al. 2006). As a result, many regional assessments already consider the connectivity of conservation areas, albeit with varying degrees of sophistication.

The term connectivity has taken on many meanings in the context of biodiversity conservation. Crooks and Sanjayan (2006) identify two primary components of connectivity: “(1) the structural (or physical) component: the spatial arrangement of different types of habitat or other elements in the landscape, and (2) the functional (or behavioral) component: the behavioral response of individual, species or ecological processes to the physical structure of the landscape.” Connectivity has longitudinal, lateral (e.g., rivers to floodplains), vertical (e.g., recharge of subterranean ground water) and temporal (e.g., changing habitat distributions through time) dimensions. In regional conservation, connectivity has most commonly focused on developing corridors between areas to accommodate animal movement (e.g., Bruinderink et al. 2003; Fuller et al. 2006), and aquatic connectivity for fish migrations (e.g., Schick and Lindley 2007; Khoury et al. 2010). However, connectivity is also critical for the movement of water, sediment and nutrients, especially in marine and freshwater systems (Abrantes and Sheaves 2010; Beger et al. 2010; Khoury et al. 2010). Temporal connectivity has not received the same attention as spatial connectivity, but is likely critical in the creation of climatic refugia, such as during prolonged drought periods (Klein et al. 2009).

At regional scales, conservation planners can affect connectivity in four general ways: altering the size, placement and number of conservation areas; changing the shape and orientation of conservation areas; adding specific linkages between conservation areas; and improving management of the intervening land, water and sea matrix. Regional conservation plans can inform each of these decisions.

Although improving connectivity is a commonly recommended and widely applicable approach to adaptation (Heller and Zavaleta 2009; Krosby et al. 2010; Beier et al. 2011), implementing it can be difficult. First, we lack a complete understanding of exactly what types and locations of connectivity are needed to enable climate change-induced species movements, and whether they are similar to or different from connectivity needs under current climate conditions (Cross et al. 2012). Second, the optimal connectivity pattern will be different for nearly every species and community. Third, for most species we know very little about their connectivity needs and can answer the “how much is enough” question for only a few species—often large carnivores that are highly mobile and arguably the least challenged by movements needed for climate adaptation. A fourth challenge is determining how to measure and map connectivity patterns. There are many new metrics for doing such measurements, but each comes with its own set of assumptions and technical requirements (Beier et al. 2008; McRae et al. 2008). Fifth, most connectivity modeling of species or habitats is focused on their current distributions, which will likely prove inadequate for many species whose distributions will be changing. Finally, the suitability of corridor areas may change over time as climate changes (Williams et al. 2005).

In its early years, systematic conservation planning was largely focused on conserving the patterns of biodiversity with little attention given to ecological process and function (Groves et al. 2002). Conservation planners and scientists increasingly promote incorporation of ecological processes and function (e.g., Leroux et al. 2007; Manning et al. 2009). In the climate adaptation arena, Halpin (1997) was among the first to recommend the need to manage for the maintenance of natural disturbance regimes such as fire as an adaptation response to climate change. More recently, Millar et al. (2007) suggested that for forests that are far outside historical ranges of variability in terms of fire regime or forest structure, it may be necessary to manage for future expected conditions as well as implement restoration treatments. In freshwater ecosystems, ecologists are calling for large-scale reconnection of floodplains through levee setbacks that will reduce anticipated flooding risks while allowing more natural flow regimes (Opperman et al. 2009). In marine ecosystems, shellfish restoration efforts can restore important ecosystem functions including nutrient removal, shoreline stabilization and coastal defense against rising sea level and storm surges (Beck et al. 2011).

Sustaining current and future ecosystem process and function may be at the challenging end of the adaptation spectrum, but it is not a new idea in conservation planning (Baker 1992). The Nature Conservancy, for example, has incorporated the conservation of ecological process in its ecoregional conservation plans for over a decade (Groves et al. 2002). Cowling et al. (1999) and Pressey et al. (2003) were among the first to test methods for incorporating ecological process in specific systematic planning efforts. Despite over 20 years of recommendations to place more emphasis on ecological process and function in conservation plans, challenges remain. Establishing explicit conservation goals and objectives for these processes and functions in the face of climate change is among the most significant of these. For example, in the Murray-Darling Basin in Australia, the Victorian Environmental Assessment Council (VEAC) compiled and synthesized data on natural flood requirements for all flood-dependent vegetation classes and rare and threatened species in major parts of the Basin (Aldous et al. 2011). Although VEAC had already recommended setting aside 4000 giga-liters every 5 years for environmental flows, new estimates of runoff that had taken climate change into account suggested that the amount of water available for environmental flows could be reduced as much as 32% over earlier projections. Even modest climate change scenarios implied that water necessary for natural overbank flows that sustain the ecosystem would not be available in many parts of the system and that new infrastructure would be required in the future to deliver those environmental flows (Aldous et al. 2011).

Opportunities for conservation planning that may emerge as climate changes will range from ecological to social. Climate change may improve conditions for some species, ecosystems, and processes of conservation concern, allowing conservation resources currently directed at these elements to be redirected elsewhere. Societal responses to climate change can provide novel opportunities to increase both the success and cost effectiveness of conservation. For example, strategies for REDD—Reduced Emissions from Deforestation and Forest Degradation (Angelsen 2008) use payments from developed countries to developing countries to reduce greenhouse gas emissions from deforestation and forest degradation. This approach provides a potentially powerful and well-funded mechanism to maintain ecologically intact forests that are also likely to have substantial biodiversity benefits, such as conserving greater numbers of species (Venter et al. 2009; Busch et al. 2010). In addition to these biodiversity benefits, increasing the representation and extent of ecosystem types under conservation management have been identified as two key principles for climate adaptation (Kareiva et al. 2008). While REDD itself is a climate change mitigation activity, using REDD to help conserve biodiversity at a regional scale is an adaptation strategy taking advantage of an emerging opportunity. In addition to REDD, opportunities might also emerge from carbon/biodiversity off-sets (Kiesecker et al. 2010), renewable energy developments (Wiens et al. 2011), human responses to climate change (Hale and Meliane 2009), and perhaps other ecosystem service opportunities (Tallis et al. 2008). These opportunities could influence the priorities for conservation areas that emerge from systematic conservation planning processes, and plans may need to explicitly consider how such opportunities might best intersect with conservation priorities. For example, initial efforts to incorporate ecosystem services into systematic conservation planning are promising (Chan et al. 2006; Egoh et al. 2010) but may involve trade-offs with biodiversity conservation.

The climate change policy arena presents a special opportunity to focus on conservation actions that promote the ability of ecosystems, and the societies that depend on them, to deal with climate-induced changes. This approach is referred to as Ecosystem-Based Adaptation (EBA), a term favored by the International Union for the Conservation of Nature (IUCN; www.​iucn.​org/​) and the Climate Action Network (www.​climatenetwork.​org/​). These groups define EBA as: “a range of local and landscape scale strategies for managing ecosystems to increase resilience and maintain essential ecosystem services and reduce the vulnerability of people, their livelihoods and nature in the face of climate change” (CAN 2009). Strategies commonly proposed under the banner of EBA include maintaining or restoring wetlands and estuaries that help protect against flooding; maintaining coral reef systems that protect islands and coastlines from wave erosion; and protecting or restoring forests that can reduce flood damage and erosion from more frequent and severe storms while preserving access to clean water and food (Hale and Meliane 2009). In some cases, implementing these strategies is straightforward and involves actions similar to those necessary to establish most new conservation areas, except that in this case the focus is on conserving natural ecosystems that also provide a direct benefit to human communities.

EBA opportunities may represent the greatest departure from traditional systematic planning methods. For example, rather than planning to conserve a representative set of coral reef habitats in a region, we might choose to prioritize those reefs systems most critical for the protection of coastal human communities. To do this, we would need additional data not traditionally included in regional assessments such as the vulnerability of coastal communities to storm surges (e.g., USAID 2009) or the volume of carbon and rates of deforestation associated with implementing a REDD strategy (Venter et al. 2009). We will also likely need alternative decision support tools to communicate future climate scenarios and potential EBA solutions, such as interactive Web-based mapping applications (e.g., Ferdaña et al. 2010) (Fig. 4). Regional conservation plans can be used to identify the best places to implement EBA strategies. Early results are promising. For example, we increasingly recognize that we can re-operate dams to both improve their benefits to people and their natural flow regimes and connectivity for nature (Richter et al. 2010). In terrestrial systems, we now understand that the intensity and frequency of fire regimes are being amplified by climate change which may require larger areas to accommodate these disturbances and pro-active steps to “fireproof” local communities (Brown et al. 2004).

Each of the approaches to climate change adaptation in systematic conservation planning may require the collection and inclusion of additional data sets (Table 1). These data sets are additional to, not in place of, data on the distribution of biodiversity, as well as on the opportunities and constraints on conservation action, which are required for all regional assessments. Future climate change projections can be readily explored and obtained from various sources, such as the Climate Wizard tool (Girvetz et al. 2009), but additional data, information and analyses are needed to conduct climate change impact or vulnerability analyses (IPCC 2007b; Ferdaña et al. 2010; Game et al. 2010; Glick and Stein 2010).

To a large degree, incorporating adaptation in regional conservation plans involves acknowledging that we undertake conservation in a world where many species distributions, disturbance regimes, and ecological processes are changing at much faster rates than in the past and in ways we often have little certainty about. This recognition necessitates a shift in traditional planning along four lines:

The strength of the approaches identified in this paper is that they are largely robust to these uncertainties. By delivering conservation solutions that would be good for biodiversity regardless of future climates, they represent “no-regrets” approaches. Although other approaches and strategies may be employed depending on the specific ways climate change occurs on-the-ground, these five general approaches provide a good foundation for regional biodiversity conservation.