Landscape Ecol (2008) 23:757–769 DOI 10.1007/s10980-008-9243-6 REVIEW PAPER Holocene palaeo-invasions: the link between pattern, process and scale in invasion ecology? Lindsey Gillson Æ Anneli Ekblom Æ Katherine J. Willis Æ Cynthia Froyd Received: 28 March 2008 / Accepted: 10 June 2008 / Published online: 10 July 2008 Ó Springer Science+Business Media B.V. 2008 Abstract Invasion ecology has made rapid progress in recent years through synergies with landscape ecology, niche theory, evolutionary ecology and the ecology of climate change. The palaeo-record of Holocene invasions provides a rich but presently underexploited resource in exploring the pattern and process of invasions through time. In this paper, examples from the palaeo-literature are used to illustrate the spread of species through time and space, also revealing how interactions between invader and invaded communities change over the course of an invasion. The main issues addressed are L. Gillson (&) Plant Conservation Unit, Botany Department, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa e-mail: [email protected] A. Ekblom
K. J. Willis
C. Froyd Oxford Long-Term Ecology Laboratory, Biodiversity Research Group, Oxford University Centre for the Environment, South Parks Rd, Oxford OX1 3QY, United Kingdom e-mail: [email protected] K. J. Willis Department of Biology, University of Bergen, N-5007, Bergen, Norway K. J. Willis e-mail: [email protected] C. Froyd e-mail: [email protected] adaptation and plant migration, ecological and evolutionary interactions through time, disturbance history and the landscape ecology of invasive spread. We consider invasions as a continuous variable, which may be influenced by different environmental or ecological variables at different stages of the invasion process, and we use palaeoecological examples to describe how ecological interactions change over the course of an invasion. Finally, the use of palaeoecological information to inform the management of invasions for biodiversity conservation is discussed. Keywords Climate change
Disturbance
Landscape connectivity
Multi-factor hypothesis
Homogenisation Introduction It is well known that alien species can negatively affect biodiversity and ecosystem function, and they are one of the main causes of extinction, as well as huge economic cost (Perrings et al. 2005). Even without extinctions, local extirpations and invasions of widespread species lead to the breakdown of biotic realms and biological homogenization (Olden et al. 2004). It is also generally accepted that the scale and rate of today’s biological invasions is unprecedented, and that present-day anthropogenic introductions differ from natural invasions by the increased spatial and temporal scale of the dispersal of organisms, as well as in the 123 758 Landscape Ecol (2008) 23:757–769 frequency and magnitude of previously rare, longdistance dispersal events (Ricciardi 2007). While the means and frequency of introductions have increased, however, invasions—the process of spread through a landscape of self-perpetuating populations—remains comparable to the past, and it is in understanding the long-term dynamics of this process that palaeoecology has the most to offer invasion biology (Rejmánek 1999; Richardson et al. 2000). The palaeoecological records described here include fossil pollen, plant macrofossils, tree-rings and charcoal, dating from the Holocene (ca. the past 11,000 years) at temporal resolutions from annual to decadal. As will be shown in this review, such records can also elucidate processes of introductions and invasions over more recent time period. In many cases, fossil pollen records can be calibrated using present-day pollen rain, and this enhances the interpretation of pollen assemblages in terms of changes in community structure or species abundance and distribution. In utilising the palaeo-record in this way, it is important to distinguish how a species arrives and establishes at a new site (introduction) from the process of spread through a landscape of self-perpetuating populations (invasion) (Richardson et al. 2000). Both native and alien species can undergo the latter process, which may be studied in some detail using palaeoecological records that provide both the spatial and temporal scales that are relevant to the process of invasion. A long-term perspective on invasions raises opportunities for utilising palaeoecological data in five main ways, which we will exemplify here with selected paleoecological studies from the literature. (1) Range shift and climate change: Palaeo-data can elucidate patterns and rates of spread through landscapes in response to climate variability, including the ongoing Holocene (i.e. the last 11,000 years, the time since the end of the last Ice Age) warming and pulsed climatic anomalies nested within it, like droughts. (Björkman and Bradshaw 1996; Parshall 2002; Lyford et al. 2003; Bradshaw and Lindbladh 2005). Understanding the mechanisms behind Holocene plant invasions or migrations provides information on the development of modern ecosystems, patterns of alien species spread, lags and inertia in the rate of response, and the potential impacts of future climate and land-use change. 123 (2) (3) (4) (5) Eco-evolution and climatic niche shift: An understanding of past changes in distribution in response to climate change can help to identify species that are able to adapt to climate change through niche shifts, from those whose dispersal and rates of migration will not match changes in climate (Mooney and Cleland 2001; Kinnison and Hairston 2007). This can be used to inform climate change integrated conservation strategies by pinpointing those species that may require assisted migration (McLachlan et al. 2007). Disturbance history, landscape connectivity and invasive spread: Palaeoecological data can help determine how the pattern of spread is affected by landscape connectivity and how current patterns of landscape fragmentation may affect the landscape ecology of invasive spread (Ronce 2001; With 2002, 2004; Pearson 2006). Invasions as a multi-factor process: Palaeo-data can indicate how ecological interactions change over the course of an invasion and how the application of a multifactor hypotheses can explain long-term patterns in invasive spread (Mitchell et al. 2006). Implications of palaeo-invasions for conservation and ecosystem management: Palaeo-data on the time since introduction can help to establish the status of ‘‘doubtful natives’’, and raises interesting philosophical and scientific questions, for conservation management (van Leeuwen et al. 2005; Willis and Birks 2006). All of the case studies presented in this review illustrate how integration of palaeoecology, ecological theory, evolutionary ecology, and landscape ecology can act synergistically, providing the link between pattern, process and scale in invasion ecology. They also demonstrate the potential of palaeo-data to interface with theoretical advances in invasion ecology. We argue that the additional information provided through addressing these long-term ecological records is essential to the conservation and management of invasive species. Range shift and climate change: climatic change as a driver of palaeo-invasions A key research question associated with current and future climate change is where will biota move to and Landscape Ecol (2008) 23:757–769 how quickly? Long-lived tree species are unlikely to be at equilibrium with the prevailing climate, particularly in times of rapid climate change, when dispersal abilities are poor, or when mature trees have a broader climatic tolerance to germination and recruitment phases (Bennett 1998; Millar and Woolfenden 1999; Davis and Shaw 2001; Von Holle et al. 2003; Svenning and Skov 2004, 2005). The present distribution of long-lived trees is therefore a result of the interplay between these three factors (rapid climate change, poor dispersal, and differences in climatic tolerance of different life-stages), as well as disturbance history and geographical barriers. Studies of Holocene invasions provide important information on time-lags between environmental change and distributional response, and can help distinguish the effects of biological inertia and dispersal limitation from climatic tolerance (Von Holle et al. 2003; Hierro et al. 2005). These findings have potential for improving the accuracy of bioclimate envelope models and species distribution models, which currently assume that species distributions are in equilibrium with climate (Guisan and Thuiller 2005; Guisan et al. 2006). Knowledge of climatic conditions at the time of germination and recruitment is also essential in predicting how forests might respond to future climate conditions (Millar and Woolfenden 1999; Svenning and Skov 2004, 2005). Many of the old-growth forests that are seen today would have recruited in the cooler conditions of the Little Ice Age, between the 14th and 18th centuries (Millar and Woolfenden 1999). As a result, matching the current distribution of long-lived trees to present climatic parameters may lead to a serious under-estimate of the impact of climate change on tree distribution, because ecosystem inertia and persistence of mature trees may mask for centuries the fact that climate sensitive stages like germination recruitment are no longer occurring (Gillson and Willis 2004). Since climate change of the next 100 years is likely to be greater than the warming experienced at the little ice age, the mis-match between present-day and future temperature sensitive plant growth stages may also be correspondingly greater. While rare, long-distance dispersal events often dramatically increase the rate at which tree species can migrate (Clark 1998), there is also evidence that some species are still responding to ongoing postglacial climatic warming (Svenning and Skol 2007), 759 or are responding to pulses in climate, such as alternating drier and wetter periods. (Swetnam et al. 1999; Allen et al. 2003). The packrat midden record along a latitudinal gradient in western USA, for example, showed local extinction of pinyon pines (Pinus remota and Pinus edulis) in more southerly lowlands during the last deglaciation, and sequential migration to higher altitudes and latitudes. At the most northerly point in the record, Owl Canyon in Colorado, pinyon pines arrived just a few hundred years ago, suggesting that migration in response to Holocene warming is still ongoing (Betancourt et al. 1991; Swetnam et al. 1999). A further study of pinyon pine expansion at Dutch John Mountain, northeastern Utah, utilising evidence from woodrat middens and tree rings, suggests that expansion of an isolated population may have been prevented by episodic drought. Initial colonization in the thirteenth century was prevented by a catastrophic drought that occurred during the Medieval Warm Period. The drought probably also caused extensive mortality of the previously dominant Utah Juniper (Juniperus osteosperma), allowing pinyon pine to expand rapidly in the aftermath, during a cool wet period at the start of the Little Ice Age (fourteenth century) (Gray et al. 2006). Since these climatic anomalies are likely to have also affected the Owl Canyon site, at the other side of the Rocky Mountains, it seems that pulses of drought and wetter periods may also be part of the explanation for the lack of expansion from this outlier population, and more generally the lack of spread of Pinyon at landscape to regional scales. While ecosystem inertia can buffer ecosystems against climate change within certain limits, dramatic reorganization can take place if critical ecological thresholds are crossed. In south-central Spain, for example, analysis of fossil pollen, microfossils and charcoal revealed rapid transitions between pine, deciduous oak, and evergreen oak. The authors concluded that ecosystem inertia and stability was punctuated by periods of change, which occurred when ecological thresholds were crossed (Carrión et al. 2001). Rapid transitions also occurred during the cold conditions of the Younger Dryas (ca. 12,900–11,500 B.P.), when many European and North American palaeo-records suggest fast dieoffs and rapid community reorganization (Williams et al. 2002; Birks and Ammann 2000). A further 123 760 consideration is that climatic stress can increase the vulnerability of mature trees to pathogens, causing rapid and widespread regional die-offs (Breshears et al. 2005). Understanding the link between present distribution, past climates, dispersal abilities, ecosystem inertia and ecological thresholds will be critical to accurate predictions of future suitable climate space. Palaeoecological records have much to offer in this respect. Eco-evolution and climatic niche shift Species can adapt surprisingly rapidly to their new ranges (Davis and Shaw 2001). While phenotypic variation and ecological effects like enemy release account for some of this adaptation (see below), in some cases this may be due to rapid evolution on ecological times-scales (eco-evolution) (Kinnison and Hairston 2007). Rapid evolution of introduced species may occur because persistent, strongly selected individuals might represent extremes of competitive or dispersive ability (Davis and Shaw 2001; Kinnison and Hairston 2007). Equally, host communities may adapt rapidly in response to new opportunities for predation, herbivory and mutualism (Mooney and Cleland 2001; Kinnison and Hairston 2007). Palaeo studies can help to elucidate the interplay between these factors; invasions facilitated by pathogen release have been observed in the palaeo-record, and can be used to identify genetic adaptation over time (Kerfoot and Weider 2004). Adaptation of introduced species and their host communities may be by phenotypic plasticity or by genetic selection of individuals better adapted to the invaded community and its abiotic environment. Contemporary, rapid evolution (eco-evolution), might be favoured in introduced species because founder populations are physically separated from their source populations, and/or because successful founder populations may represent the extremes of a populations dispersive or competitive ability (Davis and Shaw 2001; Kinnison and Hairston 2007). Equally, invasive species might drive evolution in the host community as new predators, pathogens, herbivores and mutualists adapt to the new opportunities presented by exotics (Mooney and Cleland 2001; Kinnison and Hairston 2007). These adaptations contravene a major assumption of bioclimate 123 Landscape Ecol (2008) 23:757–769 envelope models—currently the best available tool for predicting species response to climate change—in that future ecological and environmental tolerances need not remain the same over time, and thereby further complicate the interpretations of model outputs. Recent demonstrations of eco-evolution and climatic niche shift during invasion of new areas (Broennimann et al. 2007; Kinnison and Hairston 2007), contrasts with previous work suggesting that ecological niches may serve as ‘‘stable distributional constraints’’ (Martinez-Meyer et al. 2004). An ingenious use of experimental palaeoecology to distinguish these possibilities in the case of Daphnia retrocurva, a Cladoceran that colonized Lake Superior in 1889 (Kerfoot and Weider 2004). D. retrocurva replaced D. dentiphera during a eutrophication period in the 1950s. Diapausing D. retrocurva eggs of different ages were extracted from dated sediment cores, and comparison of the hatchlings revealed morphological adaptation over a 60-year period. Furthermore, genetic analyses revealed changing allele frequencies during the eutrophication event, suggesting natural selection and/or founder effects. The authors interpreted these results in terms of the ‘‘Red Queen Hypothesis’’; having colonized the lake, D. retrocurva continuously evolved relative to changing predator pressure, (in turn precipitated by environmental changes in nutrient status), in order to maintain its foothold in the lake community (Kerfoot and Weider 2004). Disturbance history, landscape connectivity and invasive spread In understanding invasive spread, the palaeo-record provides opportunities to study how interactions between rare long-distance dispersal events, landscape connectivity, disturbance and ecosystem inertia determine the pattern and rate of spread of species through landscapes (Davis 1963; Björkman and Bradshaw 1996; Parshall 2002; Lyford et al. 2003; Von Holle et al. 2003; Bradshaw and Lindbladh 2005). The palaeo approach to understanding patterns of invasion is especially effective when pollen and plant macrofossils are used together, because macrofossils provide greater taxonomic and spatial precision, which allows detection of small, isolated Landscape Ecol (2008) 23:757–769 populations that may be missed in the pollen record (Lyford et al. 2003; Petit et al. 2004). A study of naturalized populations of European trees showed that 36 out of 55 species occupy less than 50% of their potential, based on climatic parameters relating to physiological threshold responses (Svenning and Skov 2004; Svenning and Skov 2005). The reasons for this include disturbance history, dispersal limitations in the face of ongoing postglacial climatic warming, patchiness in environmental resources, biological inertia and the various competitive and facilitative interactions encountered by migrating species (Von Holle et al. 2003; Petit et al. 2004; Parmesan et al. 2005; Brooker 2006). The interplay between climatic variability, dispersal and landscape structure was explored using palaeo methods in a study of the invasion of Utah juniper (Juniperus osteosperma) in the United States (Lyford et al. 2003). Macrofossils from woodrat middens were used to trace the spread of Utah Juniper from small pioneer colonies established by multiple long-distance dispersal events that occurred from 7,500 and 5,400 years ago. Following a wetter period of c. 2,600 years, subsequent back-filling began 2,800 years ago and proceeded in episodic expansions that coincided with periods of drought and warmth, until ca. 1,000 years ago. The authors concluded that climate was the driving factor in range expansion, but that its effect may have been partly mediated though increased connectivity of suitable habitat during the warmer drier periods. Isolated populations were able to persist even in unfavourable conditions, suggesting that regeneration rather than persistence was the critical life history stage, with the narrowest environmental requirements (Lyford et al. 2003). In terms of invasion theory, then, this study illustrates how rare, long-distance dispersal events, ecosystem inertia, climatic variability and landscape connectivity interacted in the invasive spread of Utah juniper. On a management level, knowledge of disturbance history can help to distinguish invasions from recovery of species from previous anthropogenic or natural disturbance events. The encroachment of woody plant species onto open grass-dominated habitats, for example, is of increasing concern, and may be linked to global drivers like increasing CO2 (Bond and Midgley 2000). Such invasions threaten species and assemblages that are unique to grassland (Fig. 1), 761 but in some cases may represent a return to previously forested conditions. In the Montane Grassland Restoration Project, New Mexico, a long-term perspective was used to determine the causes of tree encroachment and as a guide for restoration plans (Swetnam et al. 1999). Palaeobotanical evidence was used to establish that montane grasslands were a persistent feature of the landscape for millennia, and then evidence from tree rings, aerial photographs and historical sources was used to determine the extent and cause of the encroachment. The area of open montane grasslands had been reduced by 55% between 1935 and 1981, and a study of tree demography, soils, fire, climate and land-use history showed that tree invasion in montane grasslands was caused by changes in livestock grazing and fire exclusion. The cause of the invasion was therefore anthropogenic, and tree cutting and prescribed burning were used to restore the grasslands (Swetnam et al. 1999; Allen et al. 2003). In contrast to the grassland restoration project described above, a long-term perspective can reveal how some tree ‘‘invasions’’ are actually recoveries from past disturbance. In the Chaco Canyon, New Mexico, for example, palaeo work indicated that pinyon pine (Pinus edulis) forests are apparently still recovering in some places from overexploitation that occurred 800–1,000 years ago, and in other areas are yet to recover (Swetnam et al. 1999; Allen et al. 2003). It is only with a long-term perspective that this and other areas where forests are still recovering from past anthropogenic disturbance, can be distinguished from unprecedented tree invasions that could threaten ancient grassland ecosystems. The palaeo records presents an interesting opportunity to apply and test the theories of invasive spread, by comparing patterns and rates of spread at regional, landscape and local scales. The effects of landscape structure and connectivity in invasion ecology is a hot topic (Ronce 2001; With 2002, 2004; Pearson 2006). At the landscape level, whether a species can spread through a landscape to new areas depends to a large extent on the balance between ecosystem inertia, which can allow long-lived species to persist even when the environment has become unfavourable (Von Holle et al. 2003), and disturbance patterns, which can facilitate percolation through a landscape (With 2002, 2004). Disturbance facilitates invasions by reducing competition between native 123 762 Landscape Ecol (2008) 23:757–769 Fig. 1 Expansion of pinyon pine and juniper at the expense of grassland and sagebrush, in Utah, USA, as a result of decreased anthropogenic burning. Source: http://extension/usa.edu/rra and invasive species, creating or freeing up niche space (Petit et al. 2004) and sometimes by physically removing the barriers to dispersal. In order for a species to spread through a landscape, disturbance must reach a critical threshold, the position of which depends on the dispersal characteristics of the species. Percolation theory suggests that good dispersers prefer small localized disturbance diffused through the landscape while poor dispersers prefer large, concentrated disturbances (With 2002, 2004). For example, outputs from a natural landscape model suggest that a poor disperser will spread through the landscape at a percolation threshold of 70% for small, localized disturbances, but at 30% for large, concentrated disturbances. 123 The rate of spread of species through landscapes has been extensively studied in the palaeo record (Hunter et al. 1988; Davis and Shaw 2001). Detailed stand-scale analysis of fossil pollen provides the possibility to test the validity of percolation theory. The study of isolated stands of European beech (Fagus sylvatica) in mixed deciduous woodlands, southern Sweden showed that the establishment was facilitated by small ground fires that created suitable seed beds (Björkman and Bradshaw 1996). This initial establishment at the regional scale was climatically driven, but the widespread expansion of beech lagged by 1,000 years, and again was triggered by fire (Björkman and Bradshaw 1996; Bradshaw and Lindbladh 2005). The severity of disturbance Landscape Ecol (2008) 23:757–769 determined whether beech was able to completely replace lime (Tilia), suggesting that below a critical disturbance threshold, Tilia was resistant to invasion. This pattern of invasion is consistent with the concept of biological inertia as well as percolation theory, in that beech expansion lagged climatic change (inertia), and expansion throughout the landscape by infilling between pioneer populations could only occur when disturbance reached a critical threshold. A similar pattern of expansion was observed in the USA, where eastern hemlock (Tsuga canadensis) showed a gradual increase and westerly migration at the regional scale over the past 6,000 years, but a more recent, abrupt rapid increase at the stand scale from 5,000 to 100 years ago (Parshall 2002). In this case, the author’s interpretation was that the initial colonization reflected the expansion of small, outlier populations of T. canadensis, and that subsequent infilling only occurred later, when a critical threshold of cooler, wetter conditions was crossed, and the occurrence of fire was reduced (Parshall 2002). It may be that outlier populations persisted in favourable microhabitats, but an alternative interpretation of these data is in terms of source-sink dynamics; small populations could persist outside of their fundamental climatic niche because of repeated recolonization from a source population (the ‘‘rescue effect’’) (Pulliam 2000). In these examples from the palaeo literature, patchily distributed populations were a precursor to invasive spread, as predicted by percolation theory (With 2002, 2004). An understanding of the climatic requirements of the species concerned allowed contrasting interpretations for the mechanism of invasive spread, illustrating how palaeo-data can interface with current theoretical debates in invasion ecology. Ecological interactions: invasions as a multi-factor process The success (or otherwise) of introduced species is predicted by various hypotheses concerning release from enemies (competitors and pathogens), the presence of mutualist facilitators, the availability of ‘‘empty niches’’, the presence of facilitators, the possession of ‘‘novel weapons’’ and the suitability of the abiotic environments (see Table 1 for a summary). Mitchell et al. (2006) suggest that invasions 763 can be considered as a process in which different ecological factors dominate at different stages (Mitchell et al. 2006; Moorcroft et al. 2006). Their ‘‘multi-factor’’ hypotheses suggests that invader success depends on the interaction between biotic resistance, habitat filtering, and niche availability. The strength of these effects varies over time and also depends on whether close relatives are present in the resident community. Furthermore, these authors discuss the role of phylogenetic relatedness in modifying these effects, suggesting that unrelated introduced species are more likely to benefit from competitive release, enemy release, novel weapons and empty niches, whereas abiotic suitability and the presence of mutualist facilitators is more likely when introduced species are closely related to resident species (Mitchell et al. 2006). The palaeo-record provides a possibility to study these changing interactions over time-scales of decades to centuries, providing critical information on the process of invasion and ecological effects of long-lived tree species (Moorcroft et al. 2006). Mitchell et al.’s hypothesis does not concern the means of introduction of a species, but rather how and why that species spreads, and therefore, as explained in the introduction, can be applied to invasions of native as well as introduced species, and it is in this context that the palaeo-record is discussed here. Mitchell et al. (2006, p. 737) suggest that ‘‘Enemies [including pathogens], mutualists and competitors may all influence different stages of the invasion process, such as colonization, growth and spread, and long-term adaptation’’. They therefore suggest that invasion should be treated as a continuous rather than a categorical variable and that the study of chronosequence sites of different invasion stages could capture the shifts in dominant ecological interactions over time. Release from enemies has been suggested to facilitate spread in contemporary invasions (Keane and Crawley 2002; Mitchell and Power 2003). The effect of enemy release can also be observed in the palaeo-record. During the Holocene, the pollen record indicates that beech (Fagus grandifolia) spread rapidly into landscapes already dominated by eastern hemlock (Tsuga canadensis), a close competitor with very similar ecological and climatic requirements (Woods and Davis 1989). The results of modelling experiments indicate that this rapid rate of 123 764 Landscape Ecol (2008) 23:757–769 Table 1 Summary of contemporary ecological mechanisms explaining the success of introduced species (based on (Callaway and Aschehoug 2000; Siemann and Rogers 2001; Hierro et al. 2005; Mitchell et al. 2006) Hypothesis Summary Enemy release Introduced species benefit from reduced herbivory, predation and pathogens Competitive release Introduced species benefit from absence/reduction in competition Empty niche Introduced species are able to expand rapidly by utilizing available resources differently from native species Mutualist facilitation The acquisition of new mutualists is essential for invasive success Novel weapons Exotic plants exude allelopathic chemicals that are more effective on naı̈ve plant communities Evolution of increased competitive ability Multifactor hypothesis Resources allocated to defence against enemies in the native range can instead be used to increase vigour in the introduced range Invader success depends on the interaction between biotic resistance, habitat filtering, and niche availability. The strength of these effects varies over time and also depends on whether close relatives are present in the resident community spread was possible because beech was able to temporarily outstrip its species-specific pathogens at the leading edge of colonization, increasing vigour and conferring a transient competitive advantage in the pathogen free individuals, an example of enemy release. The modelling experiments showed that the lag between uninfected and infected individuals could occur even when host and pathogen had equal dispersal abilities, because pathogens depend on their hosts for colonization and there is a critical threshold in host population density that must be reached before a pathogen can successfully colonize (Moorcroft et al. 2006). A lagged wave of infected beech then arrived, returning the competitive ability of beech back to normal, and enabling a dynamic, patchy co-existence of both beech and hemlock to develop. The modelling experiments based on community assembly rules suggest that it would take approximately 250 years for this dynamic stability to develop, assuming no further environmental change and a homogeneous environment. This example demonstrates the interaction between enemy release (escape from host-specific pathogens) and community assembly; it was the transient competitive advantage conferred by enemy release that allowed beech to temporarily outcompete hemlock, its close ecological neighbour. It is not clear from the model outputs and palaeoevidence whether today’s highly fragmented landscapes would facilitate or hinder the preliminary invasion of non-infected trees. Further evidence for the impact of pathogens on invasions of eastern hemlock and beech can be found in the fossil pollen 123 record from Quebec. Here, hemlock showed a pathogen-linked decline at about 5,400 B.P., which was followed by a rapid expansion of beech 3,000– 2,500 years B.P. (Allison et al. 1986; Woods and Davis 1989; Moorcroft et al. 2006). The subsequent recovery of hemlock at ca. 200 years B.P. may have been linked to the evolution of pathogen resistance (Allison et al. 1986; Moorcroft et al. 2006). In this case, there appears to be an interplay between pathogen induced hemlock decline as a cause of beech invasion. Although it has been argued that the decline in hemlock, beech, as well as oak, was driven by an arid event (Foster et al. 2006), as explained above, climatic stress can increase the susceptibility of trees to pathogens (Breshears et al. 2005), and it may be that the two explanations are not mutually exclusive. This supports the argument of Gurevitch and Padilla (2004), that some invasions are a symptom rather than a cause of species decline. In keeping with Mitchell et al.’s (2006) idea of multifactor hypotheses, in which different factors dominate at different stages of invasion, it is interesting to consider how the patterns of invasion described above will interact with disturbance and demographic factors over time (Fig. 2). Colonization through a dynamic, heterogeneous vegetation landscape will generate even-aged, patchily distributed stands, which in turn will senesce at similar rates, creating temporally and spatially discrete opportunities for colonization. This pattern will interact with localized disturbances to create a heterogeneous and dynamic pattern of colonization. Such pulses of temporally and spatially discrete colonization events Landscape Ecol (2008) 23:757–769 T0 Ecosystem Inertia 765 T1 Disturbance Propagule Dispersal T3 Dispersal, establishment and enemy release T4: Dynamic co-existence Fig. 2 A multifactor hypotheses describing the dynamics of an invasion of over time. At time 0, species 2 (shaded trees) cannot establish in a forest of species 1 (white trees), even though climate space is suitable (ecosystem inertia). At T1, disturbance increases the permeability of the landscape. At T3, a long-distance dispersal event allows species 2 to establish in the disturbed area, and out-compete species 1, because of enemy release. At T4, species 2’s pathogens arrive, and reduce its competitive ability, allowing co-existence of both species in a patchy landscape (after Mitchell et al. 2006; Moorcroft et al. 2006; Björkman and Bradshaw 1996; Davis et al. 1998) have been observed in the palaeo record, and have been variously attributed to disturbance by fire and wind-throw (Woods 2000; Moorcroft et al. 2006). In addition, pollen records can show which landscape elements are most susceptible to invasion, even in the absence of disturbance. High resolution pollen analysis from forest hollows in a hardwood forest mosaic in northern Michigan, showed that hemlock preferentially invaded stands dominated by white pine (Pinus strobus) rather than red oak (Quercus ruber) or maples (Acer saccharum and A. rubrum), probably because pine provides a better seedbed for hemlock, with more light and a more penetrable surface litter (Davis et al. 1998). In this case, resident species composition rather than disturbance determined the invasiblity of the landscape at the stand scale (Davis et al. 1998). This finding might be linked to Allee effects and invasion pinning, in that only in white pine stands could hemlock reach a critical population threshold and eventually become dominant in the stand (Keitt et al. 2004). Understanding this dynamic forest pattern required the integration of palaeo-data to understand long-term, stand scale dynamics, combined with contemporary ecological knowledge 123 766 of the germination, recruitment, and demographics of both resident and invading species (Rejmánek 1999). Implications of palaeo-invasions for conservation and ecosystem management Understanding the time since introduction and the history and causes of invasive spread is essential in distinguishing ecosystem adaptation from deleterious biological invasions (Willis and Birks 2006). In both cases, there may be species turnover, and it is only by adopting a long-term perspective that net gains and losses can be assessed. If newly introduced species tend to be perceived as a threat to biodiversity, does the length of time since introduction change this perspective? Understanding how long a species has been present can help to inform management of introduced species, in that recent introductions are often perceived as the greatest immediate threat to biodiversity (Willis and Birks 2006; Willis et al. 2007). Palaeo-data clarifies whether species invasions are driven by anthropogenic disturbance or are actually part of migratory responses to ongoing, or pulsed climatic change—an essential adaptation for species survival in today’s rapidly changing climate (Hannah et al. 2002; Araújo et al. 2004). The palaeorecord may also help to identify those species which do not keep pace with climatic change, thereby helping to identify those which may require ‘‘assisted migration’’ in order to reach areas of suitable climate space (McLachlan et al. 2007). Isolated islands are particularly vulnerable to invasions, but control of invasive species is often complicated because of uncertainty over the exotic or native status of some species. Some of these ‘‘doubtful natives’’ can be effectively categorised using the fossil pollen record, to show whether they were present before anthropogenic settlement. In an example from the Azores, van Leeuwen et al. (2005) used fossil pollen records to determine the status of Selaganiella krassiana, previously classified as native, introduced (invasive), and uncertain in the literature. The pollen record confirmed that S. krassiana was present in the Azores from at least 6,000 years ago, confirming that the plant had not been introduced by Portuguese or Flemish settlers in the fifteenth century. The authors thus confirmed the species as native to the Azores, and inferred that other 123 Landscape Ecol (2008) 23:757–769 populations on the Canary Islands were also native, linking between the Azores and the distant mainland populations in Africa (van Leeuwen et al. 2005). In terms of the management of individual sites, a historical perspective can show whether species that are increasing in abundance pose a potential threat to the existing ecosystem or merely represent a recovery from previous disturbances. Knowledge of the demography and biology of invasions can help managers to time interventions effectively and to manage landscapes to facilitate or suppress invasions, depending on conservation goals. Many wetlands along the mid-Atlantic coasts and Mississipi Delta (USA), for example, have been invaded by an aggressive, European strain of giant reed grass (Phragmites australis), which causes dramatic changes in community composition and biodiversity loss (Lynch and Saltonstall 2002). Management of invasive P. australis colonies is complicated, however, because a native, non-invasive variety is also present in the region. An ingenious combination of pollen, macrofossils, radiocarbon dating and genetic analysis was used to resolve the management dilemma in a Lake Superior wetland known as Barks Bay Slough. Lynch and Saltonstall (2002) found that large grass pollen percentages and Phragmites stems occurred only in the top 12 cm of the peat core, suggesting that the reedbeds were established in the past few decades. However, genetic analyses revealed that the Phragmites beds at Bark Bay Slough, Lake Superior, belonged to the native chloroplast DNA haplotype, and microsatellite analysis showed no evidence of gene flow between the native and introduced populations. The authors therefore concluded that the colonization by Phragmites in Barks Bay Slough represented the recent expansion of the native North American lineage. The management dilemma was not completely resolved, however, because although native, the expansion may have been facilitated by human disturbance (Lynch and Saltonstall 2002; Willis and Birks 2006). On broader spatial scales, a long-term knowledge of the rate and pattern of species invasions in response to climatic and environmental changes can help in planning reserve networks that accommodate both present and future climate space, and the effects of landscape connectivity on the rate of migration (Jackson and Booth 2002; Lyford et al. 2003; Araújo et al. 2004). Comparing differences in species range Landscape Ecol (2008) 23:757–769 in native and introduced habitat over time can help in understanding the effects of enemy release, and therefore more accurately parameterizing the fundamental niche, with implications for improving the accuracy of Bioclimatic Envelope Models (Hierro et al. 2005). Discussion and conclusions Using selected examples from the palaeo-literature, this review explores synergies between palaeo-data, evolutionary ecology, landscape ecology and invasion ecology. The examples presented here demonstrate the potential of using the palaeo-literature to look for patterns and processes predicted by contemporary theory on invasion ecology. The palaeo-record reveals the continuous behaviour of invasion as opposed to a short time perspective that represents invasion as categorical variable (Mitchell et al. 2006). The process of invasion can be investigated using palaeoecological methods, both spatially in terms of landscape connectivity and invasive spread, as well temporally, in terms of the interplay between environmental variability and shifting ecological interactions (With 2002). A long-term perspective reveals past variability, information which helps to distinguish apparent invasions from cyclical changes, phase shifts and recovery from past disturbance. A long-term perspective on invasions raises several interesting issues in terms of conservation philosophy and approach. If most species naturally experienced rare, long-distance dispersal events in the past (Petit et al. 2004), then are today’s long distance dispersals also acceptable? Can the boundary between exotic and native be clearly made, if longdistance dispersal has happened throughout history, and all species distributions are a result of past invasions? If invasions are associated with increased eco-evolution in both invaders and host communities (Kinnison and Hairston 2007), and if today’s species richness has been enhanced by invasions (Pascal and Lorvelec 2005), then what criteria should be used to decide which invasions are good and which bad? How does this relate to current concerns over biotic homogenization (Olden et al. 2004; Rooney et al. 2007)? Can potentially harmful invasions be distinguished from those that reflect recovery from past 767 disturbance or migrational responses to climate change? Can species that have persisted and naturalized in a new environment over centuries or millennia ever achieve ‘‘native’’ status? Finally, given emerging debates over whether invasions are an effect rather than a cause of species loss (Gurevitch and Padilla 2004; Didham et al. 2005), can conservation strategies be developed that differentiate and respond appropriately to these two possibilities? Acknowledgements The authors thank two anonymous referees for their comments on the manuscript. References Allen CD, Betancourt JL, Swetnam TW (2003) Landscape changes in the southwestern United States: techniques, long-term data sets, and trends. In: Land use history of North America. http://biology.usgs.gov/luhna/chap9.html Allison TD, Moeller RE, Davis MB (1986) Pollen in laminated sediments provides evidence for a mid-holocene forest pathogen outbreak. Ecology 67:1101–1105. doi:10.2307/ 1939835 Araújo MB, Cabeza M, Thuiller W, Hannah L, Williams PH (2004) Would climate change drive species out of reserves? An assessment of existing reserve-selection methods. Glob Change Biol 10:1618–1626. doi:10.1111/j.1365-2486. 2004.00828.x Bennett KD (1998) The power of movement in plants. Trends Ecol Evol 13:339–340. doi:10.1016/S0169-5347(97) 01324-4 Betancourt JL, Schuster WS, Mitton JB, Anderson RS (1991) Fossil and genetic history of a pinyon pine (Pinus edulis) isolate. Ecology 72:1685–1697. doi:10.2307/1940968 Birks HH, Ammann B (2000) Two terrestrial records of rapid climatic change during the glacial-Holocene transition (14,000–9,000 calendar years B.P.) from Europe. Proceedings of the National Academy of Sciences 97:1390– 1394 Björkman L, Bradshaw R (1996) The immigration of Fagus sylvatica L., Picea abies (L.) Karst. into a natural forest stand in southern Sweden during the last 2000 years. J Biogeogr 23:235–244. doi:10.1046/j.1365-2699.1996. 00972.x Bond W, Midgley G (2000) A proposed CO2-controlled mechanism of woody plant invasion in grasslands and savannas. Glob Change Biol 6:865–869. doi:10.1046/j. 1365-2486.2000.00365.x Bradshaw RHW, Lindbladh M (2005) Regional spread and stand-scale establishment of Fagus sylvatica and Picea abies in scandinavia: paleoperspective in ecology. Ecology 86:1679–1686. doi:10.1890/03-0785 Breshears DD, Cobb NS, Rich PM, Price KP, Allen CD, Balice RG et al (2005) Regional vegetation die-off in response to global-change-type drought. Proc Natl Acad Sci USA 102:15144–15148. doi:10.1073/pnas.0505734102 123 768 Broennimann O, Treier UA, Müller-Schärer H, Thuiller W, Peterson AT, Guisan A (2007) Evidence of climatic niche shifts during biological invasion. Ecol Lett 10:701–709. doi:10.1111/j.1461-0248.2007.01060.x Brooker RW (2006) Plant-plant interactions and environmental change. New Phytol 171:271–284. doi:10.1111/j.14698137.2006.01752.x Callaway RM, Aschehoug ET (2000) Invasive plants versus their neighbours: a mechanism for exotic invasion. Science 290:521–522. doi:10.1126/science.290.5491.521 Carrión JS, Andrade A, Bennett KD, Munuera M, Navarro C (2001) Crossing forest thresholds. Inertia and collapse in a Holocene sequence from south-central Spain. The Holocene 11:635–653 Clark JS (1998) Why trees migrate so fast: confronting theory with dispersal biology and the paleorecord. Am Nat 152:204–224. doi:10.1086/286162 Davis MB (1963) On the theory of pollen analysis. Am J Sci 261:897–912 Davis MB, Calcote RR, Sugita S, Takahara H (1998) Patchy invasion and the origin of a hemlock-hardwoods forest mosaic. Ecology 79:2641–2659 Davis MB, Shaw RG (2001) Range shifts and adaptive responses to quaternary climate change. Science 292:673– 679. doi:10.1126/science.292.5517.673 Didham RK, Tylianakis JM, Hutchinson MA, Ewers RM, Gemmell NJ (2005) Are invasive species the drivers of ecological change? Trends Ecol Evol 20:470–474 Foster D, Oswald W, Faison E, Doughty E, Hansen B (2006) A climatic driver for abrupt mid-Holocene vegetation dynamics and the hemlock decline in New England. Ecology 87:2959–2966 Gillson L, Willis KJ (2004) As Earth’s testimonies tell’ wilderness conservation in a changing world. Ecol Lett 7(10):990–998 Gray S, Betancourt J, Jackson S, Eddy R (2006) Role of multidecadal climate variability in a range extension of pinyon pine. Ecology 87:1124–1130 Guisan A, Thuiller W (2005) Predicting species distribution: offering more than simple habitat models. Ecol Lett 8:993–1009. doi:10.1111/j.1461-0248.2005.00792.x Guisan A, Lehmann A, Ferrier S, Austin M, Overton JMC, Aspinall R et al (2006) Making better biogeographical predictions of species’ distributions. J Appl Ecol 43:386– 392. doi:10.1111/j.1365-2664.2006.01164.x Gurevitch J, Padilla DK (2004) Are invasive species a major cause of extinctions? Trends Ecol Evol 19:470–474. doi: 10.1016/j.tree.2004.07.005 Hannah L, Midgley GF, Millar D (2002) Climate changeintegrated conservation strategies. Glob Ecol Biogeogr 11:485–495. doi:10.1046/j.1466-822X.2002.00306.x Hierro JL, Maron JL, Callaway RM (2005) A biogeographical approach to plant invasions: the importance of studying exotics in their introduced and native range. J Ecol 93:5– 15. doi:10.1111/j.0022-0477.2004.00953.x Hunter ML, Jacobson GL Jr, Webb TIII (1988) Paleoecology and the coarse-filter approach to maintaining biological diversity. Conserv Biol 2:375–385 Jackson ST, Booth RK (2002) The role of Late Holocene climate variability in the expansion of yellow birch in the western Great Lakes region. Divers Distrib 8:275–284 123 Landscape Ecol (2008) 23:757–769 Keane RM, Crawley MJ (2002) Exotic plant invasions and the enemy release hypothesis. Trends Ecol Evol 17:164–170. doi:10.1016/S0169-5347(02)02499-0 Keitt TH, Lewis MA, Holt RD (2004) Allee effects, invasion pinning and species’ borders. Am Nat 157:203–216. doi: 10.1086/318633 Kerfoot WC, Weider LJ (2004) Experimental paleoecology (resurrection ecology): chasing Van Valen’s Red Queen. Limnol Oceanogr 49:1300–1316 Kinnison MT, Hairston NG Jr (2007) Eco-evolutionary conservation biology: contemporary evolution and the dynamics of persistence. Funct Ecol 21:444–454. doi:10.1111/j.13652435.2007.01278.x Lyford ME, Jackson ST, Betancourt JL, Gray ST (2003) Influence of landscape structure and climate variability on a late Holocene plant migration. Ecol Monogr 77:567– 583. doi:10.1890/03-4011 Lynch EA, Saltonstall K (2002) Palaeoecological and genetic analyses provide evidence for recent colonization of native Phragmites Australis populations in a Lake Superior Wetland. Wetlands 22:637–646. doi:10.1672/02775212(2002)022[0637:PAGAPE]2.0.CO;2 Martinez-Meyer E, Peterson AT, Hargrove WW (2004) Ecological niches as stable distributional constraints on mammals species. Glob Ecol Biogeogr 13:305–314. doi: 10.1111/j.1466-822X.2004.00107.x McLachlan JS, Hellmann JJ, Schwartz MW (2007) A framework for debate of assisted migration in an era of climate change. Conserv Biol 21:297–302. doi:10.1111/j.15231739.2007.00676.x Millar CI, Woolfenden WB (1999) The role of climate change in interpreting historical variability. Ecol Appl 9:1207– 1216. doi:10.1890/1051-0761(1999)009[1207:TROCCI] 2.0.CO;2 Mitchell CE, Power AG (2003) Release of invasive plants from fungal and viral pathogens. Nature 421:625–627. doi: 10.1038/nature01317 Mitchell CE, Agrawal AA, Bever JD, Gilbert GS, Hufbauer RA, Klironomos JN et al (2006) Biotic interactions and plant invasions. Ecol Lett 9(6):726–740 Mooney HA, Cleland EE (2001) The evolutionary impact of invasive species. Proc Natl Acad Sci USA 98:5446–5451. doi:10.1073/pnas.091093398 Moorcroft PR, Pacala SW, Lewis MA (2006) Potential role of natural enemies during tree range expansions following climate change. J Theor Biol 24:601–616. doi:10.1016/ j.jtbi.2005.12.019 Olden JD, Poff NL, Douglas MR, Douglas ME, Fausch KD (2004) Ecological and evolutionary consequences of biotic homogenization. Trends Ecol Evol 19:18–24. doi: 10.1016/j.tree.2003.09.010 Parmesan C, Gains S, Gonzalez L, Kaufman DM, Kingsolver J, Peterson AT et al (2005) Empirical perspectives on species borders: from traditional biogeography to global change. Oikos 108:58–75. doi:10.1111/j.0030-1299.2005.13150.x Parshall T (2002) Late-Holocene stand-scale invasion by hemlock (Tsuga canadensis) at its western range limit. Ecology 83:1386–1398 Pascal M, Lorvelec O (2005) Holocene turnover of the French vertebrate fauna. Biol Invasions 7:99–106. doi:10.1007/ s10530-004-9639-4 Landscape Ecol (2008) 23:757–769 Pearson RG (2006) Climate change and the migration capacity of species. Trends Ecol Evol 21:111–113. doi:10.1016/j. tree.2005.11.022 Perrings C, Dehnen-Schmutz K, Touza J, Williamson M (2005) How to manage biological invasions under globalization. Trends Ecol Evol 20:212–215. doi:10.1016/j.tree.2005. 02.011 Petit RJ, Bialozyt R, Garnier-Gére P, Hampe A (2004) Ecology and genetics of tree invasions: from recent introductions to Quaternary migrations. For Ecol Manag 197:117–137. doi:10.1016/j.foreco.2004.05.009 Pulliam HR (2000) On the relationship between niche and distribution. Ecol Lett 3:349–361. doi:10.1046/j.14610248.2000.00143.x Rejmánek M (1999) Holocene invasions: finally the resolution ecologists were waiting for!. Trends Ecol Evol 14:8–10. doi:10.1016/S0169-5347(98)01517-1 Ricciardi A (2007) Are modern biological invasions an unprecedented form of global change? Conserv Biol 21:329–336. doi:10.1111/j.1523-1739.2006.00615.x Richardson DM, Pyšek P, Rejmánek M, Barbour MG, Panetta FD, West CJ (2000) Naturalization and invasion of alien plants: concepts and definitions. Divers Distrib 6:93–107. doi:10.1046/j.1472-4642.2000.00083.x Ronce O (2001) Understanding plant dispersal and migration. Trends Ecol Evol 16:663–664. doi:10.1016/S0169-5347(01) 02348-5 Rooney TP, Olden JD, Leach MK, Rogers DA (2007) Biotic homogenization and conservation prioritization. Biol Conserv 134:447–450. doi:10.1016/j.biocon.2006.07.008 Siemann E, Rogers WE (2001) Genetic differences in growth of an invasive tree species. Ecol Lett 4:514–518. doi: 10.1046/j.1461-0248.2001.00274.x Svenning J-C, Skov F (2004) Limited filling of the potential range in European tree species. Ecol Lett 7:565–573. doi: 10.1111/j.1461-0248.2004.00614.x Svenning J-C, Skov F (2005) The relative roles of environment and history as controls of tree species composition and 769 richness in Europe. J Biogeogr 32:1019–1033. doi: 10.1111/j.1365-2699.2005.01219.x Swetnam TW, Allen CD, Betancourt JL (1999) Applied historical ecology: using the past to manage for the future. Ecol Appl 9:1189–1206. doi:10.1890/1051-0761(1999) 009[1189:AHEUTP]2.0.CO;2 van Leeuwen JFN, Schäfer H, van der Knapp WO, Rittenour T, Björck S, Ammann B (2005) Native or introduced? Fossil pollen and spores may say. An example form the Azores islands. NEOBIOTA 6:27–34 Von Holle B, Delcourt HR, Simberloff D (2003) The importance of biological inertia in plant community resistence to invasion. J Veg Sci 14:425–432. doi:10.1658/1100-9233 (2003)014[0425:TIOBII]2.0.CO;2 Williams JW, Post DM, Cwynar LC, Lotter AF, Levesque AJ (2002) Rapid and widespread vegetation responses to past climate change in the North Atlantic region. Geology 30:971–974 Willis KJ, Birks HJB (2006) What is natural? The need for a long-term perspective in biodiversity conservation science. Science 314:1261–1265. doi:10.1126/science. 1122667 Willis KJ, Gillson L, Knapp S (2007) Biodiversity hotspots through time: using the past to manage the future: an introduction. Philos Trans R Soc Lond B Biol Sci 362:169–174. doi:10.1098/rstb.2006.1976 With KA (2002) The landscape ecology of invasive spread. Conserv Biol 16:1192–1203. doi:10.1046/j.1523-1739. 2002.01064.x With KA (2004) Assessing the risk of invasive spread in fragmented landscapes. Risk Anal 24:803–815. doi:10.1111/ j.0272-4332.2004.00480.x Woods KD (2000) Long-term change and spatial pattern in a late-successional hemlock-northern hardwood forest. J Ecol 88:267–282. doi:10.1046/j.1365-2745.2000.00448.x Woods KD, Davis MB (1989) Paleoecology of range limits: beech in the upper peninsula of Michigan. Ecology 70:681–696. doi:10.2307/1940219 123