Opening communities to colonization – the impacts of invaders on biodiversity and ecosystem functioning
Opening communities to colonization – the impacts of invaders on biodiversity and ecosystem functioning
Abstract and Keywords
Research on the relationship between biodiversity and ecosystem functioning typically varies biodiversity levels by establishing communities that are subsets of the species in the high diversity community. This chapter examines what happens when biodiversity change is not limited to these subsets but instead is open to colonization from a larger species pool. The chapter first examines species traits that are responsible for successful colonization, establishment, and impact on ecosystem processes. The chapter then addresses how novel species may produce cascading and irreversible effects, drawing on known processes (selection effect or complementarity effect) that drive relationships between biodiversity and ecosystem functioning. Finally, the chapter explores how information on species traits and processes driving the relationship between biodiversity and ecosystem functioning may be used to enhance the economic evaluation of invasion risks to society.
Biodiversity is becoming increasingly homogenized as dispersal barriers are broken down (McKinney 2004) and as species respond to global warming through range shifts (Moen et al. 2004, Thuiller et al. 2006b). Such homogenization is leading to a net decrease in global biodiversity (Sax and Gaines 2003) with alarming consequences for the world’s life support systems. Locally, however, addition of species through natural and anthropogenic causes can result in idiosyncratic, unpredictable, and sometimes latent changes in biodiversity. As found by studies reviewed throughout this volume and elsewhere (e.g. Hooper et al. 2005), changes in biodiversity may translate into measurable changes in ecosystem processes and the delivery of services to human societies. Thus, the addition of novel species can have impacts on the processing of energy and matter in ecosystems, either directly or indirectly, by affecting biodiversity.
Given the growing number of species additions to ecosystems worldwide, we would like to know whether it is possible to assess the risk of a large and potentially cascading and irreversible impact of a novel species on biodiversity and ecosystem functioning so that the riskiest species can be targeted for management. This requires linking two burgeoning fields in ecology: invasion ecology and biodiversity-ecosystem functioning research. Both fields are concerned with the loss of species, changes to ecosystem functioning, and measuring species traits to predict community and ecosystem-level impacts. However, whereas biodiversity-ecosystem functioning research has generally not allowed immigration from a regional species pool (but see Chapter 10), invasion ecology inherently focuses on open communities. Biodiversity-ecosystem functioning research focuses on the relationship and feedbacks between biodiversity and ecosystem functioning; invasion ecology generally focuses on one or the other. Both fields are therefore ripe for cross-fertilization.
In this chapter, we explore the consequences of opening communities to colonization and the new establishment of one or more species that strongly interact with resident species and the environment. We focus on the factors that determine the colonization, establishment, and impact of novel species on community properties and ecosystem functioning. We then use a risk analysis based on a bioeconomic framework to illustrate approaches to invasive species management that would be appropriate for different types of species. Most of our examples involve terrestrial plant communities, as these are the most commonly studied systems in the invasion ecology literature (Bruno et al. 2005), but we also draw from other trophic levels and systems. To avoid confusion in terminology regarding colonizers versus invaders, we follow the nomenclature of Davis and Thompson (2000; Box 16.1). Although our focus in this chapter is on the community and ecosystem-level impacts of the colonist, a large body of literature exists on the invasibility of the receiving community (Box 16.2). (p.218)
(p.219) 16.2 The relationship between biodiversity and ecosystem functioning
Biodiversity effects on ecosystem functioning are common, although the magnitude and direction of effects can vary across ecosystems, trophic levels, response variables, experimental designs (see Chapter 2), and, as we discuss in this chapter, communities that are relatively open or closed to immigration from regional or global species pools. Mechanisms that explain the relationship between biodiversity and ecosystem functioning can be placed into two main categories that differ fundamentally in how species interact. The sampling effect (or selection or dominance effect; see Chapter 7) occurs when greater diversity increases the probability that a highly competitive species is present and gains dominance in a community. Such species typically use more resources and produce more biomass than the average species and are therefore associated with higher levels of ecosystem functioning (e.g. higher productivity and higher resource use). An inverse sampling effect (Loreau 2000, Engelhardt and Ritchie 2002, Jiang et al. 2008) is a special case when the competitively dominant species does not have the greatest effects on ecosystem functioning, as can be the case when resource depletion is not the primary means for interspecific competition. In contrast, the complementarity effect (or niche differentiation effect; see Chapter 7) arises from resource partitioning among species. In this case, species-rich communities use more resources and are more productive than species-poor communities, up to a saturating point (Cardinale et al. 2006a), because species are using resources differently. Since both mechanisms are necessary for the maintenance of biodiversity (i.e. competitive asymmetries are common in communities and species partition resources through tradeoffs), they operate simultaneously in most systems. Quantifying the contribution of each mechanism to a biodiversity effect can be challenging and has received considerable attention in recent years (see Chapter 7).
Communities are not static, however, and the relative importance of these opposing community-structuring mechanisms may change as species immigrate, go locally extinct, and change in abundance. Hence the relationship between biodiversity and ecosystem functioning may shift between sampling and complementarity effects depending on the age of the community, the strength of interspecific interactions, species presence, and the per capita contribution of each individual within a community to ecosystem processes. Some biodiversity-ecosystem functioning (BEF) studies report that the effects of biodiversity on ecosystem functioning grow increasingly positive through time (Tilman et al. 2001, Jonsson 2006, Fargione et al. 2007). In nitrogen-limited grassland systems, for example, this effect is explained by a shift from the selection effect to the complementarity effect. In this case, complementarity in resource use increased the input and retention of nitrogen through time (Fargione et al. 2007). However, others have found that the effect of biodiversity on ecosystem functioning grows weaker over time (Bell et al. 2005b), especially in systems where facilitative interactions early in the development of the community are replaced by competitive interactions (Cardinale and Palmer 2002). Still others report that a positive BEF relationship is only transient (Hooper and Dukes 2004, Fox 2004a), possibly because interference competition allows monocultures to outperform polycultures in the long term (Fox 2004a).
(p.220) While instructive in understanding how the BEF relationship changes through time as constructed ecosystems mature, these long-term studies of closed communities do not address how the BEF relationship changes during succession and/or invasion, which are processes that are inherent to open communities that allow immigration. Specifically, what happens to the BEF relationship when new species are allowed to immigrate?
16.3 Impacts of colonizing species on the biodiversity-ecosystem functioning relationship
Immigration and extinction processes, which are a function of biogeographic, environmental, and biotic constraints (Naeem and Wright 2003), strongly determine the structure and composition of all biological communities. A species will be absent from a community if it cannot disperse to a site, survive in the new abiotic environment, and successfully reproduce in the presence of the resident biota. Understanding the potential impact of a colonizing species on local biodiversity and ecosystem functioning will therefore depend on four factors. These are (1) the species’ likelihood of colonizing a new area, (2) the species’ likelihood of establishing a viable population and increasing in abundance, (3) the response of resident communities to the species’ presence, and (4) the functional traits of the colonizing and resident species that determine biodiversity’s effect on ecosystem processes. Functional traits (see Chapter 4) are quantifiable biological properties of species that affect how species respond to the biotic and abiotic environment through changes in the distribution and abundance of organisms (‘response trait’) and that affect ecosystem processes (‘effect trait’; Lavorel and Garnier 2002, Naeem and Wright 2003, Engelhardt 2006).
16.3.1 The likelihood of colonization
The likelihood that a species will enter a new location is a function of the species’ current range with respect to that location, characteristics of the landscape surrounding that location, and the organism’s dispersal ability within that landscape. Among plants and invertebrates, wind, water, and vertebrate dispersal modes, as well as small seed or body size and high propagule output, are associated with dispersal ability (Fenner and Thompson 2005). For birds and mammals, large body size and carnivorous diet type allow organisms to disperse long distances (Sutherland et al. 2000, Jenkins et al. 2007).
If some species traits confer greater dispersal ability and allow a species to reach new locations, are they effective predictors of the invasiveness of a species? For plants, for which the greatest body of literature is available, traits associated with dispersal are indeed generally related to invasiveness. Broad, comparative studies associate high fecundity (including production of many offspring, short juvenile periods, and/or long flowering seasons), small propagule size, and long-distance dispersal capability with high abundance or broad distribution of invasive plants (Richardson and Rejmánek 2004, Hamilton et al. 2005). For example, alien woody species with fleshy fruits (which can be carried great distances by their avian dispersers) expanded their ranges to a greater extent in New York City during the 20th century than species with other fruit types (Aronson et al. 2007). Having two modes of dispersal (e.g. sinking seeds and floating vegetative fragments in Mimulus guttatus on rivers of northern Europe; Truscott et al. 2006) or high plasticity in seed mass (e.g. Ambrosia artemisiifolia along European rivers; Fumanal et al. 2007) are other dispersal-related traits attributed to invasive plant species. For vertebrates, in contrast, body size and diet type are less important in predicting an invasive species’ likelihood of reaching a new location than whether or not humans hunt it for sport – game species are brought across oceans and continents more frequently than are non-game species (Jeschke and Strayer 2006).
For our purposes, the question is whether the traits that increase the likelihood of a species reaching a new location are those that are likely to affect biodiversity, ecosystem functioning, and the strength of the relationship between and biodiversity ecosystem functioning. In the absence of human-mediated dispersal, the answer at this time appears to be ‘no’: in plants, studies (p.221) specifically addressing this question have generally shown that no consistent relationship emerges between the traits of the seed stage, which is responsible for most dispersal, and the traits of the mature stage (Westoby 1998, Lavorel and Garnier 2002), in which effects on ecosystem functioning are greatest. Many more empirical tests of this relationship are needed to determine the robustness of this answer, however (Suding et al. 2008).
In contrast, species intentionally introduced and cultivated by humans are sometimes chosen because they are robust in the new location and are useful to humans in some way; these characteristics are related to successful establishment, spread, and effects on the new ecosystem (Alpert 2006) and therefore warrant special attention from management and policy communities in determining the balance of their risks and benefits (Section 16.4). Other intentionally introduced species are remarkably uninvasive. For example, maize (Zea mays) survives in the Old World only because of intense care through the use of fertilizer, irrigation, and a variety of biocides. The case of unintentional dispersal by humans is more ambiguous. For example, freshwater zooplankton disperse only slowly by natural means, but can drastically change ecosystem processes in new unintended locations (Havel and Medley 2006). Unintentionally introduced species are not constrained by a potential tradeoff between dispersal ability and competitive ability. They may simply be in the right place at the right time, and whether they affect resident diversity, functioning, or both in their new location may depend on traits that allow them to become established and to spread. It is these species that represent the greatest uncertainty regarding their potential threats to ecosystems. Consequently, they should be the focus of investigations for traits that predict the impact of colonizing species on biodiversity and ecosystem functioning.
16.3.2 The likelihood of establishment
Some species traits inherently confer a high likelihood of establishment success (e.g. Rejmanek and Richardson 1996, Daehler 1998), with high propagule output (Kolar and Lodge 2001), fast growth rate (Newsome and Noble 1986), and high adaptability (genetic variation in fitness traits or phenotypic plasticity; Poulin et al. 2007) appearing consistently across a wide range of species. Novel weapons (Abhilasha et al. 2008, Bais et al. 2003, Callaway and Ridenour 2004, Callaway et al. 2008) and a release from enemies (Keane and Crawley 2002, Mitchell and Power 2003) can also increase the chances of establishment by a novel species by conferring a competitive advantage over native species. Most species that become established become part of the resident community without noticeable effects on diversity or ecosystem functioning. Therefore, we ask again, which, if any, of these traits are most likely to cause a novel species to impact biodiversity, ecosystem functioning, and/or the shape or strength of the relationship between them?
High propagule output and other traits associated with juvenile stages are unlikely to have significant effects for the same reasons that we discussed for dispersal traits above. Fast growth rate and high adaptability may be associated with biodiversity and/or ecosystem functioning changes under certain circumstances, such as disturbed conditions, which may result in an immediate or latent impact of the invader on biodiversity or ecosystem functioning depending on when the disturbance happens during establishment. Competitive superiority, whether caused by enemy release, novel weapons, or some other mechanism, is more likely to have significant effects. In this case, only a one-for-one substitution within a functional group of a resident species by the novel species has no effect on biodiversity. A more likely outcome is a shift in species’ abundances, which may include the complete elimination of one or more species or a change in abundances among functional groups. In any of these cases, ecosystem functioning may be impacted through a direct or indirect sampling effect. For example, Argentine ants (Linepithema humile) invading sub-tropical and temperate regions possess a different social structure that allows the formation of fast-growing, high-density ‘super colonies’ that deplete resources of an area faster than native ants can (Holway 1999). A direct effect of their resulting dominance may be changes in nutrient cycling due to differences in their nest construction from that of native ants (Holway (p.222) et al. 2002), whereas an indirect effect could act through their increased predation on other invertebrates that affect pollination or decomposition. While sampling effects are common in BEF experiments, their significance in nature is not known (Cardinale et al. 2006a). Therefore, predicting novel species’ effects on biodiversity, ecosystem functioning, or their relationship from traits that confer competitive advantage is tenuous.
On the other hand, some species become established not because of a specific trait that can be consistently traced across many communities, but because they use the environment in some novel way; i.e. they are initially complementary. For example, Clarke et al. (2005) showed that an invading grass can take advantage of both summer and winter rains, in contrast to natives, which used rain in only one of these seasons. Similarly, red brome (Bromus madritensis ssp. rubens) is an annual grass that can exploit water and other soil resources for 2–3 months before native perennials break dormancy in the Mojave Desert (DeFalco et al. 2007). We argue that traits that confer establishment success because of their complementarity (niche differences) will more consistently affect ecosystem functioning and the BEF relationship than will traits involved in a sampling effect by conferring competitive advantage. This is because complementarity implies a direct effect on resource use, which is a key part of ecosystem functioning (Vitousek 1990). Just how large these niche differences must be is a difficult question, however. The novel species needs to be similar enough to tolerate local environmental conditions, and disturbances can open space and release resources that allow species that do not differ from the residents to become established. The literature shows clearly that exotic diversity patterns mirror native diversity, suggesting that novel species are tracking similar conditions and resources (Stark et al. 2006). On the other hand, if the novel species is too similar to the resident species, its impact on biodiversity and ecosystem functioning will be negligible.
16.3.3 The impact of a colonizer on resident communities and ecosystem functioning
By definition (Box 16.1), an invader has a large impact on the native ecosystem. This usually involves a reduction of native species abundance or richness and/or a substantial change in ecosystem functioning. However, most species that colonize and become established in new environments have little or no impact (Williamson and Fitter 1996), and others even have a positive effect on the resident community (Hacker and Dethier 2006, Kondo and Tsuyuzaki 1999, Bruno et al. 2005). Determining the traits of the species and the characteristics of the corresponding receiving communities that lead to these different situations is a main goal of the science of invasion ecology, and it is equally important for determining how a new species will affect the relationship between biodiversity and ecosystem functioning.
A crucial, often overlooked, step in determining the traits that cause large impacts is a quantitative demonstration that a suspected invasive species actually does have an effect on the receiving community. Perceptions of species’ impacts are often not substantiated by quantitative studies. For example, of 196 exotic species in the Chesapeake Bay region, 20 percent were thought to have a negative impact on a resident population, community, or process, but only 6 percent were actually documented to have a negative impact (Ruiz et al. 1999). In addition, there are reports of species widely considered to be invasive actually having no impact on individual native species of concern (e.g. Menke and Muir 2004) or on native richness or abundance (e.g. Treberg and Husband 1999). On the other hand, a novel species may ameliorate limiting conditions and positively affect native species. For example, a non-native larch species intensively planted on the lower slopes of a Japanese volcano after its eruption spread into unplanted areas on the volcano, making some consider it invasive (Kondo and Tsuyuzaki 1999). However, diversity and richness of native species during primary succession on the volcano was greater under this non-native tree than under a native tree (Titus and Tsuyuzaki 2003). In the US Pacific Northwest, Spartina anglica, a marine grass from the UK, facilitates the growth of native species by quickly accreting sediment and creating more hospitable habitats for growth in unvegetated estuarine habitats, but decreases native species diversity in other habitats (Hacker and Dethier 2006). These examples suggest that colonizers with positive (p.223) impacts on residents often share traits with native species that play a similar role in succession.
Removal studies yield the strongest evidence for negative impacts of novel species on native abundance and richness (Levine et al. 2003). For example, growth rates and recruitment of two shrubs increased in response to the removal of three non-native grasses (D’Antonio et al. 1998). Hulme and Bremner (2006) observed a significant increase in α and γ diversity after the invasive riparian weed Impatiens glandulifera was removed. In these studies the role of competition is clear, but the traits that drive this competition vary among situations. Gould and Gorchov (2000) found that survival and fecundity of native annuals were greater when they were transplanted into forest plots without the invasive shrub Lonicera mackii than in plots where it was present. Although they did not investigate the specific traits responsible for interspecific competition, they suggested reduction of light availability early in the growing season due to the invader’s longer phenology as one possibility. Dyer and Rice (1999) found that vegetative growth and reproductive output of a native bunch grass were greater when grown with conspecifics compared to when grown with exotic annual grasses at a variety of densities. In this case, the invasive species’ early growth depleted shallow soil water resources and reduced light availability early in the growing season, thereby suppressing the root growth that the native perennial required for acquiring deep soil moisture later in the growing season. Other traits associated with competitive invasive species are those that confer high resource capture ability and utilization efficiency (Feng et al. 2007), the ability to forage on low-quality resources (Gido and Franssen 2007), and the ability to alter soil biotic communities (Callaway et al. 2004). A recurring theme in these traits is that they are somehow different from those of the species in the receiving community (D’Antonio and Hobbie 2005).
Once established, a species may also negatively impact resident species by disrupting their reproduction and dispersal. Animals may disrupt plant dispersal through displacement of more effective, native pollinators; predation of pollinators or dispersers; or destruction (eating or trampling) of flowers, pollen, or seeds (Traveset and Richardson 2006). A plant may impact dispersal of other plants by competing for pollinators or dispersers (Brown and Mitchell 2001), impeding dispersers (Traveset and Richardson 2006), or augmenting populations of generalist granivores (Ortega et al. 2004). For these interactions to cause substantial negative effects on the natives, however, these novel colonizers must somehow be more disruptive than the combined effect of all the other species already occurring in the community. Little information exists as to the traits that would yield this effect, but examples of such occurrences suggest that traits novel to the community, such as a carnivore in a previously carnivore-free system, are those that will have the largest impact.
Similarly, novel species that strongly affect ecosystem functioning also tend to do things differently than the residents (D’Antonio and Hobbie 2005). Vitousek (1990) described three ways that novel species alter ecosystem functioning: by altering resource supply rates, by changing trophic structure, and by modifying disturbance regimes. Plants that bring N2-fixing bacteria into naturally nitrogen-poor systems and thereby increase nutrient availability and cycling rates (Vitousek and Walker 1989) are one of the classic examples of a trait that causes transformation by modifying resource supply rates. Novel species can also accumulate a resource (e.g. salt) to the point that concentrations are toxic to other species (Vivrette and Muller 1977). Ponds created by North American beavers in Chile increase retention of fine particulate organic matter in streams, leading to increased food availability for, and therefore productivity of, stream macroinvertebrates (Anderson and Rosemond 2007). Novel fish that affect feeding behaviour of herbivores alter the trophic structure of stream systems so that primary productivity increases and nitrogen dynamics are altered (Simon et al. 2004). The fine, quickly drying or highly flammable leaves of grasses, combined with their nearly continuous ground cover, allow novel grasses to carry fire through woody ecosystems that previously burned infrequently (D’Antonio and Vitousek 1992). This greater fire frequency, combined with the different structure of the novel and resident species, can shift the ecosystem from being a carbon sink to a carbon source (Bradley et al. 2006).
(p.224) Currently, predicting the magnitude and direction of a species’ effects on biodiversity and ecosystem functioning is tenuous at best. Novel species with traits similar to those of the resident species are less likely to have cascading effects on the ecosystem than novel species with different traits. Many uncertainties remain, however. Myrica faya is a novel functional type in Hawaii and has extensive effects on biodiversity and ecosystem functioning (Vitousek and Walker 1989). In contrast, Acer platanoides has many similar traits compared to the resident Acer saccharum (Kloeppel and Abrams 1995); however, the former can substantially reduce forest biodiversity in the northeastern United States whereas the latter does not (Webb et al. 2000). Acer platanoides has a fertilizing effect of forest soils, which increases the growth of tree seedlings of four different species (Gomez-Aparicio et al. 2008). Thus the invasion of Acer platanoides may reduce forest biodiversity but increase forest productivity. These uncertainties highlight the need to know whether and how current and future invaders will affect the relationship between biodiversity and ecosystem functioning.
16.3.4 Tying it together: invaders’ effects on the BEF relationship
We know that the breakdown of dispersal barriers has allowed a degree of global biotic homogenization. In most cases, species novel to an ecosystem will not gain dominance and will simply blend into the saturating function of the complementarity effect. These species join the community because they have similar environmental tolerances to resident species; they displace few if any established individuals because they are complementary or equally competitive; and they increase functioning in a minor additive way, if at all. However, in some cases, novel species can have strong immediate or latent effects on biodiversity and ecosystem functioning. Here, the novel species may change the shape and trajectory of the BEF relationship by changing species richness, changing the point at which an ecosystem function saturates, and/or shifting the mechanism from complementarity to a sampling effect as the nature of species interactions shifts from resource partitioning to competition.
The traits of the species that substantially impact resident species or ecosystem processes vary considerably with the situation. Nonetheless, the overarching theme is that the traits must differ somewhat from those possessed by the species in the receiving community for novel species to impact both biodiversity and ecosystem functioning. Because functional diversity is a critical component of the biodiversity-ecosystem functioning relationship (see Chapter 4; Petchey and Gaston 2006; Wright et al. 2006), it is this difference between the colonizer and the resident species that determines whether a new species will impact biodiversity and ecosystem functioning. The challenge is learning how large of a difference between the colonizer and the resident species is required for the colonizer to have a significant effect on biodiversity and ecosystem functioning.
The type of effect that the novel species has on the BEF relationship will depend on the species interactions at play in the resident community and how the newcomer fits in. The diverse array of interactions witnessed in BEF studies (e.g. facilitation and competition) occur in experimental settings where extinction could occur but colonization of new species could not. Therefore it is to be expected that the arrival of a new species could have equally diverse ramifications. When a colonizer is common to the region (‘native’), this process is referred to as succession. When the colonizing species is new to the region and has not coevolved with the resident species (‘non-native’ or ‘exotic’), this process is generally called invasion if the newcomer achieves dominance (Box 16.1). The processes are fundamentally the same: a single species can have a small or large impact on the resident community or ecosystem functioning (Davis and Thompson 2000).
The lack of common evolutionary pressures between an exotic species and a receiving community makes it more likely that the new species will have a novel set of traits that allows it to be a ‘super-competitor’ that influences functioning through the sampling effect. The sampling effect is usually associated with a positive BEF relationship, but it can also create negative relationships when the competitive dominant does not have the greatest effect on functioning (the inverse sampling effect; (p.225) Loreau 2000, Engelhardt and Ritchie 2001, Weis et al. 2007, Jiang et al. 2008). We argue that the chance of an inverse sampling effect is higher when exotic species with novel sets of traits invade a system because the exotic species are less likely to follow the ‘rules’ that maintain biodiversity in the receiving community. Thus, an exotic invader may be a superior competitor but not be the most effective processor and transformer of nutrients and energy. Or, the exotic species might have a high growth rate but be a poor competitor. Indeed, in an experimental study that tested the effects of three native and one exotic species on wetland ecosystem functioning, Engelhardt and Ritchie (2001, 2002) found that the exotic species was the most productive species that retained the most nutrients, but it was a poor competitor. Hence ecosystem biomass production and nutrient retention in mixed communities was lower than in the monoculture of the exotic because interspecific competition led to the dominance of a poor biomass producer.
There is still much to be done in order to reliably predict the effects of individual species, no matter what their origin, on ecosystem functioning. It is clear, however, that understanding how the addition of a species impacts ecosystem functioning will require knowledge of species traits – those of the new species and those of the resident species. This focus on traits will occur in a BEF framework that so far has focused on the consequences of extinction on ecosystem functioning in closed systems. This framework is applicable to conservation biology and restoration science, which seek to understand how species extinctions or community assembly, respectively, may impact ecosystem processes (Figure 16.1). Opening BEF theory and experiments to unplanned species colonization poses five real challenges that must be overcome to better understand the effects of invasive species on biodiversity and ecosystem functioning. These are:
A. Biodiversity is both an independent and a dependent variable when species are added to the resident community through colonization, and species are potentially deleted through competitive interactions with the colonizer. We need to know how BEF studies, which typically manipulate diversity as the independent variable and only allow extinction, can be redesigned to incorporate the inevitable and non-random process of new colonizations, in the form of succession or invasion. This would probably entail long-term experimental studies that would establish resident communities at a desired richness and then intentionally release one or more potentially invasive species. Changes in biodiversity and ecosystem functioning would need to be monitored through time using a design that accounts for changes in covarying environmental factors.
B. The establishment of a new species is influenced by biotic factors including the diversity of the resident community and the presence of enemies (Box 16.2). It is also influenced by abiotic factors such as disturbances, which affect the resident community even in the absence of the new species (Box 16.2). Thus, changes in a community’s diversity in the face of a new addition may not be caused by the new species, but by abiotic conditions that drive changes in biodiversity and the successful establishment of the species. The challenge here is to test the relative impact of the biotic and abiotic factors that drive changes in biodiversity of the resident community. Again, a long-term experimental approach may be required as well as long-term monitoring of newly invaded systems.
C. Invaders can have a direct impact on ecosystem processes that can override or cancel the impacts of the resident community on ecosystem processes. However, invaders can also have an indirect impact on ecosystem processes by effecting change in the resident community. These direct and indirect effects of invaders need to be studied together to understand how the relationship between biodiversity and ecosystem functioning changes in communities that are open to colonization.
D. Many species that establish in a new location are not invasive and simply blend in to the saturating function of the BEF relationship. These species may be ‘time bombs’, however, that can have significant latent effects on biodiversity and ecosystem functioning when environmental conditions change. Similarly, some species may be invasive as soon as they establish at a new location, but their impact decreases through time. We need to know how frequent these latent effects are and whether they can be predicted, especially in the face of global climate change.
E. Finally, every species has a near endless number of traits. Thus it seems impossible to know whether a novel species will become invasive without knowing the traits of all species in a community and the traits of the colonizer. Little consensus exists about the functional traits of known invasive species and even less, if anything, is known about the traits of those yet to come, raising the question of whether the right traits have been measured. Instead of knowing all traits, which would be an unrealistic proposition, it might be useful to measure key functional traits that are reliable predictors of species’ impacts on biodiversity and ecosystem functioning, and estimate the plasticity of these traits. Species can vary broadly in their traits between their native and introduced range (Siemann and Rogers 2001), suggesting that understanding key traits and their plasticity under different abiotic and biotic conditions will be important to assess which species pose the greatest risk of changing the BEF relationship.
16.4 Invasions and ecosystem services: assessing risk for better management
The globalization of international commerce presents a policy challenge: sales and movement of live organisms create wealth, but measures to prevent unintended movement of organisms have costs and there is potential for non-native species to cause considerable economic and environmental harm. Bioeconomic theory and modelling incorporate current understanding of species’ impacts on biodiversity and ecosystem functioning to quantify the impact of non-native species on ecosystem services. This is an essential step in developing effective practices and policy for invasive species management. In this section, we take a brief look at how bioeconomic frameworks can be used to evaluate the economic costs and benefits of various actions regarding non-native species; then we illustrate how BEF principles are incorporated into a strategy for assessing the risk of economic impacts of non-native species.
16.4.1 The use of bioeconomic frameworks
Leung et al. (2002) presented a quantitative bioeconomic modeling framework to analyze risks from non-native species to economic activity and the environment. The model identifies the optimal allocation of resources to prevention versus control, acceptable invasion risks, and consequences of invasion to optimal investments (e.g. labour and capital). When applied to invasive zebra mussels (Dreissena polymorpha) in North America, the model indicated that society could benefit by spending up to US$324,000 per year to prevent invasions into a single lake. By contrast, the US Fish and Wildlife Service spent US$825,000 in 2001 alone to manage all aquatic invaders in all US lakes.
A bioeconomic approach was also used by Cook et al. (2007) to evaluate the economic benefit of biosecurity measures aimed at preventing the arrival and establishment of the parasitic bee mite, Varroa destructor, into Australia over the next 30 years. Specifically, this study evaluated the expected consequences of Varroa impact on feral bee populations and the flow-on effects in terms of loss of pollination services for the horticulture industry. The model estimated the benefits of exclusion to be between Aus$21.9 million and Aus$51.4 million per year, provided exclusion is maintained. The model further revealed that existing cost-sharing arrangements between government and industry do not (p.227) accurately reflect the spread of potential benefits, such as the substantial benefits derived by the horticulture industry from ‘free’ pollination services of feral bees when they are not impacted by the mite. These studies are significant in demonstrating the potential for ex-ante evaluations of the economic impact of invasive species. Unfortunately, our knowledge of key economic variables, such as the value of biodiversity and the societal discount rates for environmental goods, is extremely limited and currently makes evaluation of non-market effects challenging. Identifying these gaps in our economic understanding highlights the need for interdisciplinary approaches in the development of improved policy frameworks for biosecurity and invasive species management (Thomas and Reid 2007; Wilson et al. 2007; and see Chapter 17).
16.4.2 Risk assessment
The bioeconomic examples above add weight to the recent conclusion of Keller et al. (2007) that risk assessment and screening protocols to limit the introduction of damaging species can deliver positive net economic benefits. Determining how to conduct an effective risk assessment and prioritize investment in biosecurity measures is complex. Risk assessments generally combine some measure of hazard with a measure of likelihood to score risk. With an invasive species, the threat or hazard is essentially determined by the magnitude of its impact and the rate at which this impact occurs. The magnitude of impact is nil to little when it is restricted to a local scale. Impact magnitude increases as species, communities, ecosystems and ecosystem functioning are adversely affected, and/or as the spatial and temporal extent of the impact increase. Similarly, some species will spread slowly and/or take a long time to have an impact and, at the other extreme, the impact of some invaders can be almost instantaneous.
In Fig. 16.2 we combine impact and rate to create a matrix to inform biosecurity strategy and investment priorities. In general, species populating the bottom left of the matrix (labelled class 1) might be considered lowest threat since the rate and magnitude of impact is small. As such, they likely represent low priorities for biosecurity investment, as the costs might be expected to outweigh the benefits. Class 2' species have low impact but a more rapid rate of impact, as might result from higher rates of spread and/or lower impact–abundance thresholds. The higher a species is on the rate axis, the more important investment in biosecurity measures that prevent arrival and establishment (such as trade or movement barriers, quarantine and inspection) will be compared to investment in management measures aimed at mitigating the problem after arrival – if impact is rapid then prevention is better than cure. Nonetheless, with restricted impact magnitude class 2' species will still assume fairly low priority compared with species with greater, community- or ecosystem-level impacts (classes 3 and 4). Of these, species whose effects are rapid (class 4) represent the greatest priorities for preventive biosecurity measures, since the implications of an incursion are severe (e.g. a disease like foot and mouth, where a single case can impact a whole industry overnight). However, if the rate of impact is slow (class 3), this could create options for investment not just in preventive measures but also in mitigation measures aimed at longer-term control or eradication.
This broad framework rests on an understanding of the traits we discussed in the first section: those that determine dispersal and establishment influence the rate of impact, whereas those that determine (p.228) how a species affects the receiving community and its ecosystem processes govern the magnitude of impact. Equally critical for determining the benefit of various actions, however, is the extent to which invasive species are a direct cause of biodiversity decline or whether they are simply responding to other forms of ecosystem change (Gurevitch and Padilla 2004). Whether an invasive species is a ‘driver’ of biodiversity change or a ‘passenger’ (MacDougall and Turkington 2005) has important implications for whether control of an invasive species is expected to increase biodiversity (Thomas and Reid 2007; Fig. 16.3(a)). Moreover, even if an invasive species is the cause of initial biodiversity loss, it need not necessarily follow that management of the invader will result in biodiversity recovery (Thomas and Reid 2007) because of dispersal limitation or local extirpation (Laughlin 2003) or because of lasting biotic (e.g. altered soil fauna and flora) or abiotic (e.g. altered stream flow) effects from the invader. Finally, even if structurally similar communities do re-establish, the order of species assembly can have a marked impact on the pattern and rate of functional recovery (Wilby and Thomas 2002; Kremen 2005; Hooper et al. 2005). If functionally significant species respond quickly to removal of the invader, then ecosystem services can recover at a more rapid rate than biodiversity overall (Thomas and Reid 2007; Fig. 16.3(b)). On the other hand, if functional species respond slowly, then even substantial recovery in biodiversity will not necessarily result in restoration of function (Thomas and Reid 2007; Fig. 16.3(b)). Such factors identify a clear need for understanding the impacts of non-native and potentially invasive species (and their control) from a BEF perspective.
Like BEF science, invasion biology has become its own branch of ecology. As we have shown here, the two branches are linked through their search for traits that determine the mechanisms by which species impact communities and ecosystems. Much still needs to be learned about how the traits can be used to forecast how novel species introduced to a new location will affect the relationship between biodiversity and ecosystem functioning. We offer the following conclusions as hypotheses that we hope will stimulate further research into the linkages between biodiversity and ecosystem functioning in open communities.
1) Traits that allow species to be good colonizers are poor predictors of species impacts on ecosystem functioning because traits at the juvenile stage are poorly related to traits at the mature stage when impacts on ecosystem processes are most likely to occur. Good colonizers are often poor competitors. Thus, good colonizers would be species with limited effects on biodiversity and ecosystem functioning and blend into the saturating function of the biodiversity-ecosystem functioning relationship.
2) Intentionally or unintentionally introduced species do not need to be good dispersers to get to a new location. Thus, exotic species do not necessarily play by the rules, and therefore have the greatest potential to affect biodiversity and ecosystem functioning. Two types of species with potentially large impacts are (a) those that directly impact ecosystem processes by occupying a new niche and changing environmental conditions to precipitate cascading effects on biodiversity, and (b) those that establish dominance by gaining a competitive advantage and affecting ecosystem processes through effect traits that differ from resident species.
3) The magnitude and direction of a biodiversity effect on ecosystem functioning differs between open and closed systems. The relationship in closed systems becomes stronger with time and switches from a sampling to a complementarity effect. The relationship in open systems is less strong and will be dominated by a sampling effect as an invasive species, by definition, drives community abundance and/or ecosystem processes.
To forecast the impact of invasive species on BEF and to evaluate risk of invasion, we need to know which species are most likely to cause cascading effects on biodiversity and ecosystem functioning. Much of this will rely on understanding similarities and dissimilarities in species traits. This is a major task, but an important one if we want to preserve biodiversity and ecosystem functioning within a changing world, where the introduction of novel species is an inevitable consequence of global trade and human travel.
We thank the NSF-supported BioMERGE Research Coordination Network and the DIVERSITAS ecoSERVICES Core Project for supporting the workshop ‘The consequences of changing biodiversity – solutions and scenarios’ (30 November–5 December 2006, Ascona, Switzerland) and the workshop organizers for giving us the opportunity to participate in this volume. We also thank Shahid Naeem, Diane Larson, Qinfeng Guo, and two anonymous reviewers for their thoughtful feedback that substantially improved this chapter.