Combining landscape variables and species traits can improve the utility of climate change vulnerability assessments
Introduction
Conservation organizations worldwide are investing in climate change vulnerability assessments to help incorporate climate change into natural resource management plans. Vulnerability assessments have three primary objectives, to determine (1) which species will be most and least vulnerable, (2) where species will be most and least vulnerable, and (3) how to reduce climate change vulnerability (Williams et al., 2008, Watson et al., 2013). These objectives help ensure that conservation resources are devoted to the most vulnerable species and locations (Pacifici et al., 2015) and help identify areas where groups of species might be resilient to climate change (Klausmeyer et al., 2011, Nadeau et al., 2015). They also increase the likelihood that management actions designed to reduce climate change vulnerability will be effective (Mawdsley et al., 2009, Nadeau et al., 2015).
Most vulnerability assessments completed to date have estimated vulnerability by predicting the change in climatically suitable habitat for a suite of species (Urban, 2015). This approach has been heavily criticized for focusing too much on climate change exposure while ignoring important species traits and landscape features that can affect species sensitivity and adaptive capacity (Dormann, 2007, Beale et al., 2008, Randin et al., 2009, Sinclair et al., 2010, Willis et al., 2015). Many alternative vulnerability-assessment methods have been developed recently (Rowland et al., 2011); however, a large portion of these assessments focus on either species traits (e.g., dispersal ability; Galbraith and Price, 2009, Young et al., 2011, Moyle et al., 2013) or landscape features (e.g., landscape connectivity; Klausmeyer et al., 2011, Watson et al., 2013, Nadeau et al., 2015). Few assessments include both species traits and landscape features. This is surprising considering the likely influence of species traits, landscape features, and their interaction on all three primary objectives of vulnerability assessments.
Species traits strongly influence which species will be most vulnerable to climate change (Jiguet et al., 2007, Diamond et al., 2011). For example, fish species with broader diet breadths were less vulnerable to droughts similar to those predicted under future climate change (Chessman, 2013). Landscape features can also affect which species are most vulnerable if landscape features differ within and surrounding the distribution of different species. The amount of topographic diversity within and around a species distribution is a good example. Topographic diversity can decrease the distance between current and future suitable climates (Guralnick, 2007) and provide climate holdouts where species may persist for many years (Patsiou et al., 2014, Scheffers et al., 2014). Hence, species that have a large portion of their distribution in areas with high topographic diversity may be less vulnerable to climate change (Luoto and Heikkinen, 2008, Randin et al., 2009).
Landscape features within and surrounding a species distribution can also affect where species are likely to be most vulnerable (Klausmeyer et al., 2011, Nadeau et al., 2015). Species traits may interact with landscape features such that species with different traits are vulnerable in different locations. For example, many species with short dispersal distances will be vulnerable to climate change in flat areas because they will be unable to move fast enough to track changing climates (Loarie et al., 2009). However, species with long dispersal distances (e.g., many birds) may be less vulnerable in flat regions because they might be able to track suitable climates in these regions.
Species traits and landscape features can also affect the utility of an assessment for identifying management actions to reduce vulnerability. Many landscape features can be manipulated (e.g., landscape connectivity) or protected (e.g. topographic diversity) to reduce vulnerability. Including landscape variables in a vulnerability assessment can therefore increase the utility of the assessment for identifying potential management actions to reduce vulnerability. Including the interaction between species traits and landscape variables could further improve the utility of vulnerability assessments because species with different traits may require different management actions to reduce vulnerability. For example, many species may not be able to move fast enough to track changing local climates even in perfectly connected landscapes (Loarie et al., 2009, Schloss et al., 2012). Therefore, increasing landscape connectivity in fragmented landscapes may not reduce climate change vulnerability for all species.
Here, we evaluate how landscape features and species traits interact to affect all three primary objectives of climate change vulnerability assessments. We focus on 113 species of birds, herpetofauna, and mammals in the northeastern United States (Supplementary Fig. S1 and Table S1). We compare estimates of which species will be most vulnerable to climate change among a landscape-based vulnerability assessment, a trait-based assessment, and an assessment that combines landscape variables and species traits. We also compare predictions of where species are likely to be most vulnerable and the types of management actions recommended for each species between a landscape-based assessment and an assessment that combines landscape variables and species traits. Our aim is to better understand (1) which species traits and landscape variables have the largest influence on each of the three primary objectives and (2) which types of vulnerability assessments are most useful for each objective.
Section snippets
Focal species
We evaluated the vulnerability of terrestrial and semi-aquatic species on the list of New York State Species of Greatest Conservation Need (New York State Department of Environmental Conservation, 2005) that do not occur primarily in marine or coastal environments, are aquatic for the minority of their life, and have mapped distributions (Supplementary Table S1). This included 12 mammals, 72 birds, 10 amphibians, and 19 reptiles. Hereafter, reptiles and amphibians are grouped as herpetofauna.
Assessing which species will be most vulnerable to climate change
The landscape-based vulnerability scores were similar for most species (Fig. 1) because many species distributions include both highly vulnerable and less vulnerable areas. Species with exceptionally high landscape-based scores (i.e., upper outliers in Fig. 1) had higher scores for topoclimate homogeneity, landscape resistance, and climate change velocity than less vulnerable species (Fig. 2). Species with exceptionally high landscape-based scores had a large portion of their distribution in
Discussion
The importance of species traits and landscape variables depended on the objective of the vulnerability assessment. Species traits were much more important than landscape variables for determining which species will be most vulnerable to climate change. Landscape variables were less important in determining which species will be vulnerable because most species in our study have distributions that overlap areas with both high and low landscape-based vulnerability scores. This result may differ
Acknowledgements
The New York State Department of Environmental Conservation provided funding for this research through New York State Wildlife Grants program grant T-18 awarded to New York by the U.S. Fish and Wildlife Service, Wildlife and Sport Fish Restoration Program. Dan Rosenblatt, Patty Riexinger, and Gordon Batcheller provided useful guidance and feedback. Pat Sullivan at Cornell University also provided useful guidance. Kimberly Corwin provided lists of species experts. We obtained climate data from
References (44)
Identifying species at risk from climate change: traits predict the drought vulnerability of freshwater fishes
Biol. Conserv.
(2013)Promising the future? Global change projections of species distributions
Basic and Applied Ecology
(2007)- et al.
Biodiversity management in the face of climate change: a review of 22 years of recommendations
Biol. Conserv.
(2009) - et al.
Integrating climate change vulnerability assessments from species distribution models and trait-based approaches
Biol. Conserv.
(2015) - et al.
Conserving the stage: climate change and the geophysical underpinnings of species diversity
PLoS ONE
(2010) - et al.
Resilient Sites for Species Conservation in the Northeast and Mid-Atlantic Region
(2011) - et al.
Resilient Sites for Terrestrial Conservation in the Northeast and Mid-Atlantic Region
(2012) - et al.
Do species's traits predict recent shifts at expanding range edges?
Ecol. Lett.
(2011) - et al.
Opening the climate envelope reveals no macroscale associations with climate in European birds
Proc. Natl. Acad. Sci.
(2008) Conceptualizing and designing corridors for climate change
Ecol. Restor.
(2012)