Refugia: keys to climate change management

Refugia are habitats that components of biodiversity retreat to, persist in and can potentially expand from under changing environmental conditions (Keppel et al., 2012), facilitating the survival of organisms during extreme climatic changes (Médail & Diadema, 2009). They may thus maintain favourable climatic conditions now absent in the surrounding landscape. This is enabled by higher local heterogeneity of microclimates and the decoupling of local from regional climates (Dobrowski, 2011; Keppel et al., 2012). Because refugia potentially safeguard the persistence of components of biodiversity over millennia, they may act as safe havens under projected anthropogenic climate change (Médail & Diadema, 2009; Keppel et al., 2012). This capacity makes them critical components in climate change management. As a result, the identification and protection of refugia has increased in priority in conservation planning (Noss, 2001; Game et al., 2011). However, refugia are currently loosely defined and their study is often based on ad hoc, descriptive data sources (Ashcroft, 2010). We therefore clarify the concept of refugia and related terms. We then argue for an increased focus on understanding how refugia will respond to climate change. In the context of rapid anthropogenic climate change, we suggest that this focus will lead to prioritization of conservation actions towards those habitats offering the greatest potential as safe havens for biodiversity. Despite an exponential increase in the use of the terms refugia and refuges in literature (Fig. 1), refugia are still confused with refuges, remnants and reserves. Most current uses of refugia are to places providing environmental conditions not available in the surrounding landscape over long periods of time (e.g., Ashcroft, 2010; Dobrowski, 2011), with an increasing interest in identifying future refugia under projected anthropogenic climate change. However, the term is also used to refer to a variety of other concepts. For example, in Global Change Biology refugia have recently been used to refer to temporary shelter from predation or disturbance (e.g., Ledger et al., 2011) and to remnant patches of suitable habitat (e.g., Rittenhouse et al., 2010). The term refugia has also been used in reference to reserves (see Fig. 1). While the term refugia has only recently been effectively defined (see Keppel et al., 2012), use of the other three R-terms has long been widely established. We therefore advocate the correct use of the relevant jargon. Refuges are places that through structures or processes provide shelter from disturbances, predation, herbivory or competition. They are therefore defined as locations providing spatial and/or temporal protection or advantages in biotic interactions over ecological time periods (i.e., within the life span of the relevant organism). Remnants are patches of suitable habitats for species intolerant of changed or modified landscapes that prevail in landscapes highly modified by human populations and resource exploitation. Reserves are areas legally or customarily protected from particular anthropogenic activities and may vary in size from remnants to portions of large ‘intact’ landscapes or so-called wilderness areas. While refuges, remnants and reserves continue to receive considerable research attention, refugia are an area of rapid recent growth (Fig. 1). The ability of refugia to locally mitigate the effects of regional climate change is increasingly considered important for climate change adaptation planning (Ashcroft, 2010; Game et al., 2011). This is because they are potential safe havens for biota and may offer the only hope for in-situ persistence of poorly dispersed species. Protecting such climate change refugia could allow some species to persist through anticipated anthropogenic climate change. They may subsequently recolonize the surrounding landscape, if environments become more favourable. Therefore, including refugia in protected area networks and in climate change management planning is a high priority for conserving biodiversity under anthropogenic climate change (Noss, 2001; Game et al., 2011). For refugia to be effective in climate change management they will need to be identified and their biological and environmental limits understood. Those with the greatest resilience and highest biodiversity values can then be targets for inclusion in protected area networks. Because of the relatively recent realization of the importance of future biodiversity safe havens under anthropogenic climate change, understanding of refugia is in its infancy. Therefore, targeted efforts towards elucidating the ecological and climatic functioning of refugia will be required. Several methods of identifying refugia have been proposed. Keppel et al. (2012) suggested a multi-disciplinary approach, utilizing topographical, species distribution, genetic, remote sensing and climatic data. While this approach is thorough, it may be time consuming in areas with little, imprecise or coarse-scale data. Considering the progression of climate change in some regions (Klausmeyer & Shaw, 2009) more rapid approaches may be needed. Ashcroft et al. (2012) recently proposed a time-efficient method that utilizes topoclimatic data. Remote-sensing could potentially identify refugia even more quickly, but may be affected by accuracy and resolution (Keppel et al., 2012). While it is increasingly feasible to identify refugia, the environmental limits of refugia remain poorly understood. There is urgent need to understand the buffering capacity or resilience of refugia. This will enable a determination of the ecological impacts of regional climate change and the likely responses of resident species. What impact will regional climate change have on refugia? For example, what would be the change in temperature within a refugium, if temperature increased by 3, 5 or 7 °C regionally? Are there tipping points for the buffering capacity of refugia? For example, would a refugium be able to maintain environmental conditions at a reduction of 10 mm in annual rainfall but not at a 20 mm decrease? Unless we can answer such questions regarding the buffering capacity of refugia, we will not be able to gauge their potential to act as safe havens for biodiversity. Simply identifying refugia may therefore not offer effective protection of biota from climate change. Probably the most efficient way of investigating the resilience of refugia is to compare them along climatic gradients that reflect predicted future changes in climate. For example, the gradient of reducing rainfall eastward across the muted landscape of south-western Australia provides an excellent natural experiment to test the likely effects of ongoing and predicted future reductions in annual precipitation. Results would facilitate management in the light of the likely effects of future reductions in rainfall. Not only may the capacity of refugia be compromised, but they are also likely to contract in size under climate change. Indeed refugia are best considered dynamic entities that may expand or contract in response to changes in regional climate (Dobrowski, 2011). Reductions in area generally affect population sizes and the number of species that can be supported. As a result, changes in the size of a refugium under anthropogenic climate change are important when evaluating the quality of refugia. Once we understand the ecological and environmental functioning and limits of refugia, we will be able to prioritize refugia on their capacity to mitigate the impacts of climate change on biodiversity. This would allow defining the limits of protection from anthropogenic climate change afforded by different refugia, and hence to prioritize available conservation funding. Incorporating this knowledge about refugia into climate change management would also facilitate the optimization of conservation efforts towards protection from anthropogenic climate change. This work is supported by an Australian Research Council (ARC) grant (LP 0990914). We would like to thank Hafeel Kalideen for his assistance with the literature review.