==== Front Proc Biol Sci Proc Biol Sci RSPB royprsb Proceedings of the Royal Society B: Biological Sciences 0962-8452 1471-2954 The Royal Society 33234075 10.1098/rspb.2020.2061 rspb20202061 1001 60 Review Articles Review Article Racing against change: understanding dispersal and persistence to improve species' conservation prospects Climate change and species conservationhttp://orcid.org/0000-0002-3972-7560Kerr Jeremy T. Department of Biology, University of Ottawa, Ottawa, Ontario, Canada K1N 6N5 e-mail: jkerr@uottawa.caInvited review by the Canadian Society for Ecology and Evolution former President. Electronic supplementary material is available online at https://doi.org/10.6084/m9.figshare.c.5217361. 25 11 2020 25 11 2020 25 11 2020 287 1939 2020206121 8 2020 4 11 2020 © 2020 The Authors.2020http://creativecommons.org/licenses/by/4.0/http://creativecommons.org/licenses/by/4.0/Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited.Climate change is contributing to the widespread redistribution, and increasingly the loss, of species. Geographical range shifts among many species were detected rapidly after predictions of the potential importance of climate change were specified 35 years ago: species are shifting their ranges towards the poles and often to higher elevations in mountainous areas. Early tests of these predictions were largely qualitative, though extraordinarily rapid and broadly based, and statistical tests distinguishing between climate change and other global change drivers provided quantitative evidence that climate change had already begun to cause species’ geographical ranges to shift. I review two mechanisms enabling this process, namely development of approaches for accounting for dispersal that contributes to range expansion, and identification of factors that alter persistence and lead to range loss. Dispersal in the context of range expansion depends on an array of processes, like population growth rates in novel environments, rates of individual species movements to new locations, and how quickly areas of climatically tolerable habitat shift. These factors can be tied together in well-understood mathematical frameworks or modelled statistically, leading to better prediction of extinction risk as climate changes. Yet, species' increasing exposures to novel climate conditions can exceed their tolerances and raise the likelihood of local extinction and consequent range losses. Such losses are the consequence of processes acting on individuals, driven by factors, such as the growing frequency and severity of extreme weather, that contribute local extinction risks for populations and species. Many mechanisms can govern how species respond to climate change, and rapid progress in global change research creates many opportunities to inform policy and improve conservation outcomes in the early stages of the sixth mass extinction. range expansionrange lossclimate changedispersalpersistenceNatural Sciences and Engineering Research Council of Canadahttp://dx.doi.org/10.13039/501100000038Discovery Grant and Discovery Accelerator SupplemeMitacshttp://dx.doi.org/10.13039/501100004489University of Ottawahttp://dx.doi.org/10.13039/100008572cover-dateNovember 25, 2020 ==== Body 1. Introduction Human activities have caused extinction rates to rise sharply among populations and species in most regions of the world [1–3]. Land use change, land use intensification and overexploitation have decimated biological communities where such threats have been concentrated [4,5]. Climate has long been known to contribute vitally to the distribution of life on Earth [6,7]. Consequently, the rapidity of human-caused climate change has led to widespread biotic responses, which demonstrate the pervasive influences—and risks—of climate change on the life histories of species. Climate change is accelerating and its relative importance as a cause of present-day extinction is growing [8]. The link between where a species is found and prevailing climatic conditions in those areas is one of the oldest relationships that exists in ecology and evolutionary biology, predating the discovery of natural selection or even area effects [9,10]. Species’ distributions depend on the array of environmental conditions they tolerate after accounting for antagonistic or beneficial biotic interactions and dispersal limitation [11–13]. This understanding, which draws on classical approaches to describing species’ niches [14], led to early predictions that anthropogenic climate change would cause species' geographical ranges to shift [15]. There are many precedents for such shifts throughout the complex climatic history of the Earth, with past climate changes causing redistributions of species across broad areas [16], such as those detected through ancient DNA or fossil pollen records [16,17]. Although climate sensitivities to growth in atmospheric CO2 concentrations were predicted (and broadly acknowledged) in the nineteenth century [18], and confirmed and refined subsequently [19], the rapidity of both warming and the global torrent of biotic responses to it has been striking. Range shifts are one of the clearest signs of biotic responses to climate change. They require at least one of two processes to operate: dispersal from a leading range boundary that establishes a population in an area the species did not occupy historically, or the loss of populations from historically occupied areas, such as from a trailing range boundary [20]. Leaving aside assumptions about other vital processes, like local population dynamics or adaptation, climate change will benefit a species if the rate of expansion into new areas exceeds the rate of loss elsewhere (i.e. geographical ranges grow larger). Conversely, species’ extinction risks rise if range losses in some areas exceed range expansions in others (i.e. ranges get smaller), as geographical range size is strongly linked to species' extinction risk [21]. The establishment of populations in new areas is limited by, among other things, species’ dispersal and persistence capacities in particular environments over both short and longer time frames [22,23]. Both processes—dispersal and persistence—are vital to understanding range dynamics during climate change. Dispersal in this context is specifically the movement of individuals of a species into an area that was previously unoccupied (i.e. dispersal that directly relates to range shift), regardless of the life stage for their dispersal. Persistence is the degree to which individuals or populations of a species within its range remain present over time. Such species responses are not fixed and can respond to selection, accelerating eco-evolutionary dynamics [24] and changes in species' traits, such as dispersal capacity in different areas of their ranges [25,26]. Other mechanisms, such as the constellation of existing and potential biotic interactions, each represent an additional challenge in terms of understanding and predicting when species will maintain or grow their geographical ranges in response to climate change [22]. Here, I review recent developments around measurements and applications of two mechanisms necessary for understanding and predicting species’ range shifts during climate change. These mechanisms are (i) species' capacities to disperse to (and establish) in novel environments and (ii) how species’ exposure and sensitivity (or susceptibility; [27]) to emerging conditions can be detected and used to understand their persistence in areas undergoing climate change. Emerging computational and data gathering tools are reviewed here and have enabled measurement and prediction of dispersal and persistence using datasets that are unprecedented in their spatial and temporal extents. The result has been rapid progress. Early models measured success if they detected species' range shifts in directions that were qualitatively consistent with climate change effects, while new research links species’ extinction–colonization dynamics to highly resolved measurements of climate change and consequent short-term environmental variability. These developments lead to practical policy advice that could alter the trajectory of extinction rates. 2. Geographical range shifts and climate change Informed by extensive understanding of how changes in palaeoclimates affected species distributions and interactions, Peters & Darling [15] wrote presciently about the prospects that anthropogenic climate change would accelerate extinction rates and cause a widespread redistribution of species. This work anticipated the need to account for interactions between land uses and climate change, the potential requirements for managed relocation of species, the problems of protecting species in nature reserves whose boundaries were fixed in place while species' range boundaries became dynamic (figure 1a), potential interactions between land use change and climate change, and the myriad challenges of accounting for changes to biotic interactions. Moreover, many species’ dispersal capacities were thought to be far below rates required to track shifting climatic conditions, even in the absence of widespread habitat losses and fragmentation [15]. This work anticipated that extinction risks would rise as a consequence of climate change through impacts on species' range dynamics. Figure 1. Development of critical predictions around how species might respond to climate change was followed rapidly by detection of south–north trends in extinction risk in a western butterfly species. (a) A species' equatorward range boundary prior to human-caused habitat losses (top panel), its range after land use change, limited largely to protected areas (middle panel), and range losses consequent to climate change that caused this species to become extinct in one of the protected areas (adapted from [15]). (b) Latitudinal differences in observed population extinctions of the bay checkerspot butterfly (Euphydryas bayensis; inset photo by Walter Sigmund, licence CC BY-SA 3.0), with the likelihood of a population extinction shown by the length of bars, on the left. The latitudinal range of population resurveys is shown in black and corresponds to the edges of the bar graph. June 2018 temperatures vary from warm (in the south) to cool (in the north) based on brightness temperature measurements from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor on Terra. These temperature data were rendered slightly transparent and overlaid on a shaded relief map to depict topographic variation also. (Online version in colour.) Qualitative evidence that anthropogenic climate change was beginning to take a biotic toll emerged from work comparing the limits of species’ present-day geographical ranges against historical observations and asking a simple, but powerful, question: what changed [28,29]? In a series of resurveys of known populations of the bay checkerspot butterfly (Euphydryas editha), population extinctions were found to be most likely in the south [30] (figure 1b). Areas that had been strongly altered through land use changes were omitted from the study, leaving climate change prominent among potential causes for this pattern. Colonization of new areas was not measured because of risks that historical surveys might not have detected all populations. This work set a precedent for using butterfly populations as a kind of sentinel taxon for the biotic effects of climate change [31]. Its focus on patterns of population extinction also differed from later work, which often emphasized more strongly the detection of range expansions along poleward range margins. Inferences that climate change contributed to species' modern geographical range shifts grew rapidly stronger. Evidence of poleward range expansion came from a variety of taxa, though the geographical foci of early studies were most commonly European and often British. Poleward range shifts were detected among European butterfly species [28], and plant populations along elevational gradients [32,33] were found to be creeping upward altitudinally. Southern Finnish birds frequently expanded their ranges northward [34], and observations of British species across an array of taxa [35]—including several invertebrate and vertebrate groups—suggested similar trends. Such observations have expanded in both their geographical and taxonomic scopes [36]. Evidence from across many taxa, including birds, mammals, insects and plants, has led to upward adjustments in the rate at which shifts are occurring, to about 17 km decade−1 [37]. Early estimates were rarely spatially explicit except in the directional sense: range shifts assessed along latitudinal and altitudinal gradients indicated that shifts were broadly poleward and upward, respectively. The detection of such shifts and even their attribution specifically to climate change is still a vital step removed from establishing quantitative relationships with range dynamics or understanding the mechanisms underlying such relationships [38]. These mechanisms were largely and necessarily implicit in pioneering detection and attribution studies, drawing on realized niche concepts, long-term monitoring datasets, and extensive natural history expertise [12]. Measurements indicating that species were responding in various ways to anthropogenic climate change enabled correlative or other statistical tests of links between those responses and particular global change drivers [39]. That species are shifting poleward, for example, can as readily be a consequence of land use change as of warming conditions, and distinguishing between their relative roles is obviously necessary to understand whether their effects might benefit or harm species’ survival prospects [40]. For example, warming could facilitate poleward expansion for a species but so could land use changes that make new habitats suitable for potential range expansion, as has been observed among many generalist species [41]. Similarly, baseline forest cover and forest cover change, as well as climate change [42], have contributed to species redistributions along elevation gradients (based on 1464 separate observations of elevation shift from 46 locations). Among butterflies in the Sierra Nevada mountains, climate changes and habitat losses led to opposite responses, with range loss predominating in low elevation areas where habitat loss was extensive, and range expansion being more common at high elevations, where climate changes predominated [43]. Other kinds of biotic responses, such as phenological shifts, can also reflect signals of both climate and land use changes, respectively [44]. Statistical approaches provide, minimally, tests of the strength of the potential climate change signal in biological change data and can test alternative causes of those biotic responses [45]. They have the potential to be spatially explicit, testing or predicting specific trends that are particular to localities within the total study area, given a range of measured environmental changes in those areas. These models can specify mechanisms underlying particular hypotheses, creating a pseudo-experimental framework for testing the strengths of different global change drivers [46]. Distinguishing the relative roles of climate and land use change, respectively, is difficult without more sophisticated models incorporating mechanisms that logically govern range shifts, such as dispersal or tolerance to environmental changes. Nevertheless, quantitative links between warming and range shifts are now widely documented: species in terrestrial, freshwater and marine ecosystems are shifting their geographical distributions in relation to rapid warming [47]. 3. Dispersal and range dynamics In the context of geographical range shifts, dispersal results in the establishment of a species' population in an area that was previously unoccupied [48]. Successful dispersal consequently leads to local expansion of a species’ geographical range. Despite this vital role in global change responses, challenges associated with measuring dispersal—which varies contextually [23], potentially in response to selection [49] and interspecifically [50]—have slowed the development of models that assess its effects in specific terms. Dispersal has often been ignored entirely [51], for lack of data or because its inclusion complicates the modelling process, in predictions of species' movements in response to climate change. If the limits of individual movement are known, they can be included in models predicting shifts in species' geographical ranges over time, such as through the use of dispersal kernels (but see [52,53]). Models incorporating different dispersal scenarios are uncommon [53]. Dispersal that leads to range shifts can be modelled using species’ known movement limits. Across a range of dispersal capacities (including known dispersal limits, zero dispersal and unlimited dispersal, respectively) and climate scenarios, many bumblebee species in North America are confronted with steep declines in the habitable area [53]. For some species, the disappearance of existing combinations of climatic conditions [54], not just shifts in climate zones, makes range losses inevitable [55]. Yet, dispersal is inextricably linked with many species' conservation prospects during environmental change. Frameworks for assessing dispersal rates predate nearly all global change biology research but have been adapted to enable predictions for species’ range shifts during global change. Skellam [56] developed the reaction–diffusion equation to account for random dispersal of individuals from a point of origin, enabling the predictions of changes in population density spatially and temporally. Interestingly, a motivation for this work was to explain rates of climate change-driven recolonization of Britain by oaks, which appeared to have occurred too rapidly for oak dispersal rates to permit. The mechanisms in this model—including dispersal, population growth and spatial displacement of climate zones—create a constellation of subsequent, specific predictions that help inform conservation applications [57]. Applying this framework to poleward range expansion in response to climate change (reviewed in [58]), the rate of change of species' population density, u, depends on the diffusion rate, D, of individuals through space, x, from occupied areas: 3.1 ∂∂tu=D∂2∂x2u+ru, where r is the per capita growth rate of the dispersing species when its population size is small. The minimum critical patch size (Lc) for a species in a bounded space is 3.2 Lc= πDr. This model was adapted to test whether a species’ capacity to disperse into new areas is likely to track shifting climatic conditions (q, the rate at which climatically suitable areas shift in space) rapidly enough to enable it to persist, assuming habitat suitability is binary: 3.3 Lc(q)=πDr(1−q2Dr)−1. Finally, this model predicts species persistence if 3.4 q