Ecology Letters, (2014) 17: 637–649
doi: 10.1111/ele.12262
REVIEW AND SYNTHESIS
Emily V. Moran,* and Jake M. Alexander ETH Zurich, Universitatstrasse 16, 8092, Zurich,Switzerland *Correspondence: E-mail: [email protected]
Evolutionary responses to global change: lessons from invasive species Abstract Biologists have recently devoted increasing attention to the role of rapid evolution in species’ responses to environmental change. However, it is still unclear what evolutionary responses should be expected, at what rates, and whether evolution will save populations at risk of extinction. The potential of biological invasions to provide useful insights has barely been realised, despite the close analogies to species responding to global change, particularly climate change; in both cases, populations encounter novel climatic and biotic selection pressures, with expected evolutionary responses occurring over similar timescales. However, the analogy is not perfect, and invasive species are perhaps best used as an upper bound on expected change. In this article, we review what invasive species can and cannot teach us about likely evolutionary responses to global change and the constraints on those responses. We also discuss the limitations of invasive species as a model and outline directions for future research. Keywords Biotic interactions, climate change, cline, genetic constraints, invasive species, local adaptation, niche, range expansion, rapid evolution, selection. Ecology Letters (2014) 17: 637–649
INTRODUCTION
Over the past decade, biologists have devoted increasing attention to the role of evolution in species’ responses to environmental change. Much of this interest has arisen from concerns about the ability of species to respond to rapid anthropogenic global change (Davis et al. 2005; Jump & Penuelas 2005; Bradshaw & Holzapfel 2006; Parmesan 2006; Reusch & Wood 2007; Hendry et al. 2008). Indeed, human influences have given rise to many classic examples of rapid evolution, including industrial melanism, tolerance of heavy metals or air pollutants and insecticide and herbivore resistance (reviewed by Reznick & Ghalambor 2001). Other studies have documented the rapid evolution of invasive species in response to novel biotic and abiotic conditions, and of native species in response to the invaders (Callaway & Maron 2006). Furthermore, theoretical studies and evolutionary experiments suggest that local adaptation can affect the ability of populations to grow, persist and colonise new habitat (Lavergne et al. 2010). These and other lines of evidence show that evolution may be an important component of species’ responses to diverse agents of global change (Visser 2008), especially given that the pace of change is likely to limit species’ ecological responses, such as migration to areas that are becoming climatically suitable following rapid climate change (Jump & Penuelas 2005). Nevertheless, until recently both models and empirical studies of global change responses largely ignored evolution (Lavergne et al. 2010), and it remains a challenge to predict what evolutionary responses can be expected and at what rates they will occur. This is partly because of the difficulty of
performing experiments over relevant timescales and partly because some of the most drastic challenges to species, such as global climate change, are ongoing and are themselves difficult to predict. In this article, we consider the extent to which the natural experiments offered by the spread of invasive species can help overcome these challenges and inform us about likely evolutionary responses to environmental change in native taxa. We focus particularly on climate change because of the close parallels between the novel conditions experienced by species introduced to new geographic areas and those resulting from climate change, but many insights are applicable to other drivers of global changes. Evolutionary changes consistent with responses to current climate change have already been detected in some species, including modest shifts towards earlier breeding phenology, in critical photoperiod for diapause timing, in clines of climaterelated chromosomal inversions and in the proportion of high-dispersal phenotypes in species undergoing range shifts or expansions (Balanya et al. 2006; Bradshaw & Holzapfel 2006; Hanski 2012). These examples are as yet limited, in part because distinguishing evolutionary change from plastic or ecological responses is challenging (Skelly & Freidenburg 2010). Moreover, the degree of climate change experienced in most areas of the world is still small compared to future projections (IPCC 2007), so selection pressures may still be too moderate to provoke strong evolutionary responses in many species. Indeed, studies of several mammal and bird species, most with generation times longer than a year, have confirmed plastic phenotypic responses but not genetic changes (Merila 2012); to date, genetic responses have mostly been observed in species with generation times of 1 year or © 2014 John Wiley & Sons Ltd/CNRS
638 Emily V. Moran and Jake M. Alexander
less (Reusch & Wood 2007; Merila 2012). Virtually all of these studies have been based on longitudinal field observations – only rarely, as in the case of pitcher-plant mosquito photoperiod responses, have the genetic change and its fitness consequences been confirmed in the lab (Reusch & Wood 2007). Experimentation is inherently difficult, because evolutionary change is slow relative to the length of most experiments, and chronosequences (e.g. of seeds or spores) from populations that have experienced climate change are not available for most species. Thus, much controversy remains regarding the extent to which evolutionary processes must be considered when attempting to predict species and community responses over the next 50–200 years (Davis et al. 2005; Parmesan 2006; Lavergne et al. 2010). Given these limitations, complementary approaches must be considered to study evolutionary responses of species to climate change. One possibility is to exploit the large-scale experiments provided by the transcontinental transport of non-native species (Callaway & Maron 2006; Gilchrist & Lee 2007; Sax et al. 2007). A recent review examined what invasive species – i.e. species that are actively spreading in a new region (Daehler 2001) – may teach us about the possibility for, and consequences of, range shifts in response to climate change (Caplat et al. 2013). As yet, however, limited attention has been given to what invasive species may teach us about evolutionary adaptation to climate change. The parallels between the invasion of new environments and the environmental changes associated with climate change are strong, though imperfect. Many of the novel conditions encountered by invasive species, including new temperature or precipitation regimes, altered mutualistic or antagonistic biotic interactions and the availability of new suitable habitat unoccupied by conspecifics, are similar to those that species will confront following climate change (Table 1). Some of these conditions apply to other global changes as well. Invasive species and native species exposed to rapid environmental change can also experience similar demographic perturbations. For example, both newly founded non-native populations and native populations in fragmented landscapes can experience severe population bottlenecks, and range expansions of nonnative populations are analogous to expansions of native populations following climatic or land-use changes. In both cases, comparative studies using invasive species could provide insights into the role of gene flow, genetic variation and life history traits for population recovery and adaptive evolution following these demographic events. Although these processes are also studied in the native range, one major advantage of invasions is that they provide the opportunity to directly compare populations that have experienced changing environmental conditions during spread with those in the native range that have not experienced those changes. The ‘experiments’ offered by biotic invasions have other advantages. Rapid evolution and eco-evolutionary interactions frequently occur both in the invader and in the native community (Lambrinos 2004; Callaway & Maron 2006; Lankau 2012) and they do so in the context of multiple interacting factors, rather than the simple single-factor conditions used in most selection experiments (Reusch & Wood 2007). Moreover, biological invasions span the same timescales, from dec© 2014 John Wiley & Sons Ltd/CNRS
Review and Synthesis
ades to centuries, over which we attempt to predict global change responses. Before attempting to extrapolate from such data sets, however, one must keep in mind that successful invasive species are, by definition, good at spreading and exploiting new environments. Not all introduced species become invasive, and there are potentially important differences in the genetic structure of native and invasive populations and in the selective pressures that they face; thus, many native species may exhibit more constrained ecological or evolutionary responses than do successful invaders. Nevertheless, invasive species may provide a useful upper bound to the amount of evolutionary change we should expect to see in response to climate change. We do not intend here to present an exhaustive review of all studies of evolutionary change in invasive species, nor of the theoretical background behind eco-evolutionary interactions in the context of climate change. These topics have been reviewed previously, at least in part (Holt 1990; Lambrinos 2004; Huey et al. 2005; Callaway & Maron 2006; Reusch & Wood 2007; Sax et al. 2007; Prentis et al. 2008; Visser 2008; Lavergne et al. 2010; Matesanz et al. 2010). Rather, our aim is to focus on how knowledge derived from species introductions can shed light on two major questions about evolutionary responses to climate change. First, which traits are likely to exhibit evolutionary responses to the novel environments encountered during invasion and rapid climate change, and how rapidly? Second, what limits evolutionary responses to novel environments, and is evolution likely to rescue populations at threat of extinction from climate change? We also discuss the potential limitations of using invasive species to derive lessons applicable to native species experiencing environmental change and outline promising avenues for future research. WHAT KIND OF EVOLUTIONARY RESPONSES TO CLIMATE CHANGE CAN WE EXPECT, AND HOW QUICKLY?
Climate change is likely to alter selection pressures in several ways. Many species exhibit local adaptation to climate, and as a result frequently exhibit clinal distributions of climaterelated traits, such as size and reproductive timing, along climatic gradients. Changes in temperature and precipitation regimes will select for shifts in these clines – for example, with warm-climate phenotypes being favoured in historically colder regions (Visser 2008). Even populations that are able to migrate to areas with climates similar to their current habitat will likely experience novel selection pressures due to altered biotic interactions (Parmesan 2006; Williams & Jackson 2007), as well as novel combinations of photoperiod cues and climate, which means that plastic responses alone are unlikely to be sufficient in the long term (Visser 2008). In addition, while the availability of newly suitable habitat beyond the current range edge could select for increased dispersal ability (Phillips et al. 2008; Hanski 2012), habitat fragmentation leading to high dispersal costs may create an opposing selective force for reduced dispersal in some populations (Lambrinos 2004; Cheptou et al. 2008). Changes in average dispersal ability, whether positive or negative, would in turn affect migration
Review and Synthesis
Invasion and evolution to global change 639
Table 1 Biological invasions as a model system for evolutionary responses to selection pressures imposed by climate change. Examples can be found in the text
Selection pressure
Predicted response in native species
Observations during biological invasions
Novel climatic conditions
Shifts in climate-related traits to match new local optima
Evolution of clinal trait variation along climate gradients (Table 2)
Availability of habitat unoccupied by conspecifics
Evolution of increased dispersal ability to exploit habitat made suitable by climate change
Evolution of increased dispersal ability at invasion front
Novel biotic interactions
Adaptation to novel antagonists/mutualists encountered due to range shifts
Evolution of native species in response to invasive species acting as a novel host/ predator/prey/competitor (Table 3)
ability, meta-population dynamics and gene flow, and thus feed back to affect ecological and evolutionary responses to climate. Invasive species have provided classic examples of rapid evolution in traits related to climate, biotic interactions and dispersal, and how quickly this evolution can occur. Abiotic selection pressures: climate
One of the most commonly documented forms of rapid evolution in invasive species is the evolution of clines in climaterelated traits (Table 2). Invasive species can face strong selection pressures to match the local climate, especially when the founders come from an area with a very different climate (Alexander 2013). Because species differ in their life history, the most useful measure of time is generations rather than years. Examples from the invasion literature show that geographic clines generally develop 50–150 generations after introduction, and occasionally in < 25 generations (Table 2). Extensive genetic change can sometimes occur despite both limited environmental variation in the native range or reduced genetic diversity in the non-native range. For example, the perennial shrub Hypericum canariense exhibits half as much allelic diversity and a third of the heterozygosity in California
and Hawaii compared to native Canary Islands populations. Despite this, invasive populations evolved a latitudinal cline in flowering time that exceeds the variation in flowering time seen in native populations and also exhibit an increase in growth rate (Dlugosch & Parker 2008b). The observed phenotypic changes documented for the examples in Table 2 were confirmed to be due to genetic (and/or epigenetic) rather than plastic responses to different environments through common garden experiments. However, unless multiple generations are studied (e.g. Huey et al. 2000), common garden experiments alone cannot distinguish between genetic and epigenetic effects. More multi-generational common garden studies would therefore be desirable. Although all of the genetic changes documented in Table 2 are consistent with local adaptation (Weber & Schmid 1998; Lounibos et al. 2003; Dlugosch & Parker 2008b), random drift or founder effects could also contribute to apparent clines. Two methods used to test for local adaptation are reciprocal transplants to examine survival or reproduction in different environments (Quinn et al. 2001), or the comparison of phenotypic clines to neutral expectations based on colonisation history (Keller et al. 2009). For Echinochloa crus-galli, Hakam & Simon (2000) confirmed that the greater cold tolerance of northern populations was due to the higher activity of a protective enzyme, and Roy et al. (2000) showed that these populations had descended from southern cold-sensitive North American populations. Either the direct measurement of fitness or the use of null models should be applied more widely when assessing evolutionary responses to invasion or global change. Nevertheless, it should be possible to confirm, in some cases from existing data, whether the parallel clines frequently observed in both native and introduced ranges are indeed adaptive. If so, further experiments, as in the Echinochloa example, could shed light on the mechanisms of adaptations. This in turn may help us identify what traits to focus on when measuring genetic diversity in native populations experiencing global change. Abiotic selection pressures: habitat availability
For organisms encountering a large area of suitable but unoccupied habitat, high dispersal rates can be strongly advantageous (Fig. 1). If this is a widespread response in native species, it could enhance species’ ability to track climate
Table 2 Examples of rapid evolution in climate-related traits in invasive species
Species
Common name
Trait
Years
Generations
Reference
Drosophila subobscura Aedes albopictus (Fig. 1) Drosophila subobscura Hypericum canariense Oncorhynchus tshawytscha Solidago altissima Eschscholzia californica Lythrum salicaria (Fig. 1) Silene vulgaris, S. latifolia Solidago gigantea, S. canadensis Echinochloa crus-galli
Fruit fly Tiger mosquito Fruit fly Canary Isl. St. John’s wort Chinook salmon Late goldenrod California poppy Purple loosestrife Campion Goldenrod Barnyard grass
Chromosomal inversions Photoperiodic diapause Wing size Flowering phenology Growth and reproductive traits Growth traits Flowering and growth traits Time of and size at flowering Various growth, flowering traits Flowering time, growth traits Photosynthetic enzyme activity
10–15 15 c. 20
doi: 10.1111/ele.12262
REVIEW AND SYNTHESIS
Emily V. Moran,* and Jake M. Alexander ETH Zurich, Universitatstrasse 16, 8092, Zurich,Switzerland *Correspondence: E-mail: [email protected]
Evolutionary responses to global change: lessons from invasive species Abstract Biologists have recently devoted increasing attention to the role of rapid evolution in species’ responses to environmental change. However, it is still unclear what evolutionary responses should be expected, at what rates, and whether evolution will save populations at risk of extinction. The potential of biological invasions to provide useful insights has barely been realised, despite the close analogies to species responding to global change, particularly climate change; in both cases, populations encounter novel climatic and biotic selection pressures, with expected evolutionary responses occurring over similar timescales. However, the analogy is not perfect, and invasive species are perhaps best used as an upper bound on expected change. In this article, we review what invasive species can and cannot teach us about likely evolutionary responses to global change and the constraints on those responses. We also discuss the limitations of invasive species as a model and outline directions for future research. Keywords Biotic interactions, climate change, cline, genetic constraints, invasive species, local adaptation, niche, range expansion, rapid evolution, selection. Ecology Letters (2014) 17: 637–649
INTRODUCTION
Over the past decade, biologists have devoted increasing attention to the role of evolution in species’ responses to environmental change. Much of this interest has arisen from concerns about the ability of species to respond to rapid anthropogenic global change (Davis et al. 2005; Jump & Penuelas 2005; Bradshaw & Holzapfel 2006; Parmesan 2006; Reusch & Wood 2007; Hendry et al. 2008). Indeed, human influences have given rise to many classic examples of rapid evolution, including industrial melanism, tolerance of heavy metals or air pollutants and insecticide and herbivore resistance (reviewed by Reznick & Ghalambor 2001). Other studies have documented the rapid evolution of invasive species in response to novel biotic and abiotic conditions, and of native species in response to the invaders (Callaway & Maron 2006). Furthermore, theoretical studies and evolutionary experiments suggest that local adaptation can affect the ability of populations to grow, persist and colonise new habitat (Lavergne et al. 2010). These and other lines of evidence show that evolution may be an important component of species’ responses to diverse agents of global change (Visser 2008), especially given that the pace of change is likely to limit species’ ecological responses, such as migration to areas that are becoming climatically suitable following rapid climate change (Jump & Penuelas 2005). Nevertheless, until recently both models and empirical studies of global change responses largely ignored evolution (Lavergne et al. 2010), and it remains a challenge to predict what evolutionary responses can be expected and at what rates they will occur. This is partly because of the difficulty of
performing experiments over relevant timescales and partly because some of the most drastic challenges to species, such as global climate change, are ongoing and are themselves difficult to predict. In this article, we consider the extent to which the natural experiments offered by the spread of invasive species can help overcome these challenges and inform us about likely evolutionary responses to environmental change in native taxa. We focus particularly on climate change because of the close parallels between the novel conditions experienced by species introduced to new geographic areas and those resulting from climate change, but many insights are applicable to other drivers of global changes. Evolutionary changes consistent with responses to current climate change have already been detected in some species, including modest shifts towards earlier breeding phenology, in critical photoperiod for diapause timing, in clines of climaterelated chromosomal inversions and in the proportion of high-dispersal phenotypes in species undergoing range shifts or expansions (Balanya et al. 2006; Bradshaw & Holzapfel 2006; Hanski 2012). These examples are as yet limited, in part because distinguishing evolutionary change from plastic or ecological responses is challenging (Skelly & Freidenburg 2010). Moreover, the degree of climate change experienced in most areas of the world is still small compared to future projections (IPCC 2007), so selection pressures may still be too moderate to provoke strong evolutionary responses in many species. Indeed, studies of several mammal and bird species, most with generation times longer than a year, have confirmed plastic phenotypic responses but not genetic changes (Merila 2012); to date, genetic responses have mostly been observed in species with generation times of 1 year or © 2014 John Wiley & Sons Ltd/CNRS
638 Emily V. Moran and Jake M. Alexander
less (Reusch & Wood 2007; Merila 2012). Virtually all of these studies have been based on longitudinal field observations – only rarely, as in the case of pitcher-plant mosquito photoperiod responses, have the genetic change and its fitness consequences been confirmed in the lab (Reusch & Wood 2007). Experimentation is inherently difficult, because evolutionary change is slow relative to the length of most experiments, and chronosequences (e.g. of seeds or spores) from populations that have experienced climate change are not available for most species. Thus, much controversy remains regarding the extent to which evolutionary processes must be considered when attempting to predict species and community responses over the next 50–200 years (Davis et al. 2005; Parmesan 2006; Lavergne et al. 2010). Given these limitations, complementary approaches must be considered to study evolutionary responses of species to climate change. One possibility is to exploit the large-scale experiments provided by the transcontinental transport of non-native species (Callaway & Maron 2006; Gilchrist & Lee 2007; Sax et al. 2007). A recent review examined what invasive species – i.e. species that are actively spreading in a new region (Daehler 2001) – may teach us about the possibility for, and consequences of, range shifts in response to climate change (Caplat et al. 2013). As yet, however, limited attention has been given to what invasive species may teach us about evolutionary adaptation to climate change. The parallels between the invasion of new environments and the environmental changes associated with climate change are strong, though imperfect. Many of the novel conditions encountered by invasive species, including new temperature or precipitation regimes, altered mutualistic or antagonistic biotic interactions and the availability of new suitable habitat unoccupied by conspecifics, are similar to those that species will confront following climate change (Table 1). Some of these conditions apply to other global changes as well. Invasive species and native species exposed to rapid environmental change can also experience similar demographic perturbations. For example, both newly founded non-native populations and native populations in fragmented landscapes can experience severe population bottlenecks, and range expansions of nonnative populations are analogous to expansions of native populations following climatic or land-use changes. In both cases, comparative studies using invasive species could provide insights into the role of gene flow, genetic variation and life history traits for population recovery and adaptive evolution following these demographic events. Although these processes are also studied in the native range, one major advantage of invasions is that they provide the opportunity to directly compare populations that have experienced changing environmental conditions during spread with those in the native range that have not experienced those changes. The ‘experiments’ offered by biotic invasions have other advantages. Rapid evolution and eco-evolutionary interactions frequently occur both in the invader and in the native community (Lambrinos 2004; Callaway & Maron 2006; Lankau 2012) and they do so in the context of multiple interacting factors, rather than the simple single-factor conditions used in most selection experiments (Reusch & Wood 2007). Moreover, biological invasions span the same timescales, from dec© 2014 John Wiley & Sons Ltd/CNRS
Review and Synthesis
ades to centuries, over which we attempt to predict global change responses. Before attempting to extrapolate from such data sets, however, one must keep in mind that successful invasive species are, by definition, good at spreading and exploiting new environments. Not all introduced species become invasive, and there are potentially important differences in the genetic structure of native and invasive populations and in the selective pressures that they face; thus, many native species may exhibit more constrained ecological or evolutionary responses than do successful invaders. Nevertheless, invasive species may provide a useful upper bound to the amount of evolutionary change we should expect to see in response to climate change. We do not intend here to present an exhaustive review of all studies of evolutionary change in invasive species, nor of the theoretical background behind eco-evolutionary interactions in the context of climate change. These topics have been reviewed previously, at least in part (Holt 1990; Lambrinos 2004; Huey et al. 2005; Callaway & Maron 2006; Reusch & Wood 2007; Sax et al. 2007; Prentis et al. 2008; Visser 2008; Lavergne et al. 2010; Matesanz et al. 2010). Rather, our aim is to focus on how knowledge derived from species introductions can shed light on two major questions about evolutionary responses to climate change. First, which traits are likely to exhibit evolutionary responses to the novel environments encountered during invasion and rapid climate change, and how rapidly? Second, what limits evolutionary responses to novel environments, and is evolution likely to rescue populations at threat of extinction from climate change? We also discuss the potential limitations of using invasive species to derive lessons applicable to native species experiencing environmental change and outline promising avenues for future research. WHAT KIND OF EVOLUTIONARY RESPONSES TO CLIMATE CHANGE CAN WE EXPECT, AND HOW QUICKLY?
Climate change is likely to alter selection pressures in several ways. Many species exhibit local adaptation to climate, and as a result frequently exhibit clinal distributions of climaterelated traits, such as size and reproductive timing, along climatic gradients. Changes in temperature and precipitation regimes will select for shifts in these clines – for example, with warm-climate phenotypes being favoured in historically colder regions (Visser 2008). Even populations that are able to migrate to areas with climates similar to their current habitat will likely experience novel selection pressures due to altered biotic interactions (Parmesan 2006; Williams & Jackson 2007), as well as novel combinations of photoperiod cues and climate, which means that plastic responses alone are unlikely to be sufficient in the long term (Visser 2008). In addition, while the availability of newly suitable habitat beyond the current range edge could select for increased dispersal ability (Phillips et al. 2008; Hanski 2012), habitat fragmentation leading to high dispersal costs may create an opposing selective force for reduced dispersal in some populations (Lambrinos 2004; Cheptou et al. 2008). Changes in average dispersal ability, whether positive or negative, would in turn affect migration
Review and Synthesis
Invasion and evolution to global change 639
Table 1 Biological invasions as a model system for evolutionary responses to selection pressures imposed by climate change. Examples can be found in the text
Selection pressure
Predicted response in native species
Observations during biological invasions
Novel climatic conditions
Shifts in climate-related traits to match new local optima
Evolution of clinal trait variation along climate gradients (Table 2)
Availability of habitat unoccupied by conspecifics
Evolution of increased dispersal ability to exploit habitat made suitable by climate change
Evolution of increased dispersal ability at invasion front
Novel biotic interactions
Adaptation to novel antagonists/mutualists encountered due to range shifts
Evolution of native species in response to invasive species acting as a novel host/ predator/prey/competitor (Table 3)
ability, meta-population dynamics and gene flow, and thus feed back to affect ecological and evolutionary responses to climate. Invasive species have provided classic examples of rapid evolution in traits related to climate, biotic interactions and dispersal, and how quickly this evolution can occur. Abiotic selection pressures: climate
One of the most commonly documented forms of rapid evolution in invasive species is the evolution of clines in climaterelated traits (Table 2). Invasive species can face strong selection pressures to match the local climate, especially when the founders come from an area with a very different climate (Alexander 2013). Because species differ in their life history, the most useful measure of time is generations rather than years. Examples from the invasion literature show that geographic clines generally develop 50–150 generations after introduction, and occasionally in < 25 generations (Table 2). Extensive genetic change can sometimes occur despite both limited environmental variation in the native range or reduced genetic diversity in the non-native range. For example, the perennial shrub Hypericum canariense exhibits half as much allelic diversity and a third of the heterozygosity in California
and Hawaii compared to native Canary Islands populations. Despite this, invasive populations evolved a latitudinal cline in flowering time that exceeds the variation in flowering time seen in native populations and also exhibit an increase in growth rate (Dlugosch & Parker 2008b). The observed phenotypic changes documented for the examples in Table 2 were confirmed to be due to genetic (and/or epigenetic) rather than plastic responses to different environments through common garden experiments. However, unless multiple generations are studied (e.g. Huey et al. 2000), common garden experiments alone cannot distinguish between genetic and epigenetic effects. More multi-generational common garden studies would therefore be desirable. Although all of the genetic changes documented in Table 2 are consistent with local adaptation (Weber & Schmid 1998; Lounibos et al. 2003; Dlugosch & Parker 2008b), random drift or founder effects could also contribute to apparent clines. Two methods used to test for local adaptation are reciprocal transplants to examine survival or reproduction in different environments (Quinn et al. 2001), or the comparison of phenotypic clines to neutral expectations based on colonisation history (Keller et al. 2009). For Echinochloa crus-galli, Hakam & Simon (2000) confirmed that the greater cold tolerance of northern populations was due to the higher activity of a protective enzyme, and Roy et al. (2000) showed that these populations had descended from southern cold-sensitive North American populations. Either the direct measurement of fitness or the use of null models should be applied more widely when assessing evolutionary responses to invasion or global change. Nevertheless, it should be possible to confirm, in some cases from existing data, whether the parallel clines frequently observed in both native and introduced ranges are indeed adaptive. If so, further experiments, as in the Echinochloa example, could shed light on the mechanisms of adaptations. This in turn may help us identify what traits to focus on when measuring genetic diversity in native populations experiencing global change. Abiotic selection pressures: habitat availability
For organisms encountering a large area of suitable but unoccupied habitat, high dispersal rates can be strongly advantageous (Fig. 1). If this is a widespread response in native species, it could enhance species’ ability to track climate
Table 2 Examples of rapid evolution in climate-related traits in invasive species
Species
Common name
Trait
Years
Generations
Reference
Drosophila subobscura Aedes albopictus (Fig. 1) Drosophila subobscura Hypericum canariense Oncorhynchus tshawytscha Solidago altissima Eschscholzia californica Lythrum salicaria (Fig. 1) Silene vulgaris, S. latifolia Solidago gigantea, S. canadensis Echinochloa crus-galli
Fruit fly Tiger mosquito Fruit fly Canary Isl. St. John’s wort Chinook salmon Late goldenrod California poppy Purple loosestrife Campion Goldenrod Barnyard grass
Chromosomal inversions Photoperiodic diapause Wing size Flowering phenology Growth and reproductive traits Growth traits Flowering and growth traits Time of and size at flowering Various growth, flowering traits Flowering time, growth traits Photosynthetic enzyme activity
10–15 15 c. 20
