Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Definition of the Subject

This entry focuses on two ecological phenomena. The first is “keystone species” which is defined by Paine [1] as a species (mostly of high trophic status) whose activities exert a disproportionate influence on the patterns of species occurrence, distribution, and density in a community. The second is the concept of “ecosystem engineers” defined by Jones et al. [2] as organisms that directly or indirectly modulate the availability of resources (other than themselves) to other species by causing physical state changes in biotic or abiotic materials.

Introduction: Keystone Species and Ecosystem Engineers: Analysis of Concepts

Paine’s definition of keystone species was inspired from the large effects of the removal of the carnivorous starfish (Pisaster ochraceus) from intertidal habitat, which reduced prey species diversity due to intense competition from mussel prey [3], and represents now a classic textbook in ecology. The original keystone species concept of Paine [1, 4] thus identified a very specific mechanism: the top-down regulation of community structure and diversity by a top predator (Fig. 4.1). The concept of keystone species has been later extended to a broader definition and now includes any species whose effect on ecosystems is disproportionately large relative to its low biomass in the community as a whole [5]. Keystone species are thus species which have large effects on communities and ecosystems through many different processes such as trophic interactions, pollination, or habitat modification [6, 7]. Examples include rabbits that can increase abundance and diversity of lizards [8] and sea otters whose hunting in the late nineteenth century caused a population explosion of their sea urchin prey and consequent overgrazing of kelp which led to numerous extinctions of local species [6].

Fig. 4.1
figure 1

Two examples of keystone species impacts. (a) Effects of the removal of Pisaster on prey species diversity as a consequence of mussel population explosion. (b) Consequences of the removal of sea otters on species diversity due to overgrazing of kelp by sea urchins. Keystone species are represented in grey boxes. Small grey arrows indicate the direction of species abundance changes following the removal of the keystone species. The large grey arrows indicate the global consequences of keystone species loss on the ecosystem

Full size image

The concept of ecosystem engineering was proposed two decades later than the “keystone species” concept by Jones and colleagues [2]. They defined ecosystem engineers as “organisms that directly or indirectly modulate the availability of resources (other than themselves) to other species by causing physical state changes in biotic or abiotic materials. In so doing they modify, maintain and/or create habitats” [2]. They further distinguish between two types of ecosystem engineers (Fig. 4.2): autogenic engineers that change the environment via their own physical structures, i.e., their living and dead tissues, and allogenic engineers that change the environment by transforming living or nonliving materials from one physical state to another via mechanical or other means. The idea that organisms can have important effects on abiotic processes occurring in the environment had been recognized before; indeed, Darwin devoted a whole book to the impact of earthworms on soil formation [9]. However, since the development of the concept of ecosystem engineer, engineering effects have been described for many organisms, from classic examples such as beavers, termites, or earthworms [10–12] to mollusks [13], fish [14], caterpillars [15], polychaete worms [16], grasses [17, 18], burrowing shrimp [19], ants [20], and many other species (see Table 1 of [21]).

Fig. 4.2
figure 2

Two examples of ecosystem engineering with kelp (a) and earthworms (b) as ecosystem engineers. Ecosystem engineering corresponds to changes in physical state (state 1–2) of biotic (i.e., kelp for a) or abiotic (i.e., soil for b) materials. Ecosystem engineers are represented in grey boxes. (a) Case of autogenic engineering, the engineer is part of the new physical state (via growth here). (b) Case of allogenic engineering, the new physical state is caused by the engineer (, caused via feeding here), but the engineer is not part of the new physical state

Full size image

Both the keystone species and the ecosystem engineer concepts point out to species which have important effects in ecological communities and ecosystems. Although these concepts partly overlap – an ecosystem engineer can be a keystone species – they however insist on different aspects: the keystone concept focuses on species which have disproportionate effects on community structure and ecosystem functioning (“outcome focused” sensu [22]) whereas the ecosystem engineering concept considers organisms which influence the abiotic environment with consequences on other species and related ecosystem processes (“process focused” sensu [22]). These differences between the two concepts are reflected in the literature: these concepts generally appear in distinct studies as less than 5% of the studies on these topics refer to both ecosystem engineers and keystone species (source: ISI Web of Science).

The keystone species concept has been strongly related to food web theory since its first definition [1, 23]. In particular, the identification of keystone species in food webs is an important issue. Theoretical studies have tried to pin down the characteristics of keystone species through food web models [24, 25] and several indices based on food web topology have been developed to identify keystones [26]. Models have shown that the loss of species with a large number of trophic interactions can trigger high numbers of secondary extinctions with serious consequences for species persistence; a result which highlights the potential keystone role of highly connected species in food webs [24, 27, 28].

In contrast, the concept of ecosystem engineering has been rarely related to food web studies. Recent studies acknowledge that ecosystem engineers may also play an important role in the network of trophic interactions but separating the trophic effects from the engineering effects to determine their relative importance is difficult [20, 29–31].

The importance of keystone species can also be strongly linked with ecosystem engineering. For example, the large impact of sea otters in kelp forest ecosystems results from the coupling between engineering effects and a trophic cascade [32]. In these ecosystems, kelp provides habitat for many species and dampens wave action; the keystone effect of sea otters is thus mediated through their indirect trophic effect on kelp densities which is a main ecosystem engineer.

Issue-1: How to Find Keystone Species and Ecosystem Engineers in Communities?

Keystone species and ecosystem engineers may affect ecosystem processes, such as nutrient cycling, and thereby ecosystem functioning. In the face of rapid biodiversity loss, a considerable amount of studies were dedicated to investigate a possible link between species richness and ecosystem function [33] and the threat of diversity loss on the loss of ecosystem services to man. First indications show positive relationships between species richness and ecosystem productivity, stability, and sustainability, with more species being able to fully and complementarily run ecosystem functions due to niche differentiation and facilitative interactions (reviewed by [34]). However, there is now a growing consensus that functional diversity, rather than species numbers per se, strongly determines ecosystem functioning [35]. This means that the presence of a particular species with specific traits may play a larger role in determining ecosystem function than merely the number of species [36]. The apparent diversity-ecosystem function relationship can thus be partly caused by a greater chance of an influential species with particular traits being present in more diverse communities than in species-poor communities.

If it is possible to predict and identify a priori a set of species traits that determine keystone interactions in a system, this would greatly benefit management and conservation purposes. Species’ traits determine how species contribute to ecosystem processes, so the presence and distribution of such traits can be utilized to indicate aspects of ecosystem functioning [37]. To identify keystone species various methods have been used ranging from experimental removal or addition manipulations to comparative studies and natural history observations [5]. Partly because of these methodological issues, identifying keystone species has so far proved elusive [5, 38] although some progress has been made and its concept now widely investigated in the context of complex ecological networks [25, 39–41]. Some examples of specific traits are for instance trophic level, body size, connectance, or traits concerning tolerance and resilience to disturbances. Organisms that influence their environment strongly and contribute disproportionately to the functioning of ecosystems often seem to occupy higher trophic levels in food webs [5]. Top predators have been described as highly interactive keystone species [42], have been shown to play an important role in stabilizing food webs [43], and play important roles in marine ecosystems [44] and terrestrial ecosystems [45].

Also, the loss of top predators has been linked to secondary extinctions [46, 47]. This has been attributed to their ecological role as suppressors of medium-sized predators (mesopredators) (e.g., [48, 49]) and generalist herbivores [50, 51]. In terrestrial ecosystems, organisms that influence their environment strongly also often seem to be large bodied (e.g., [52]). Larger bodied organisms require a high resource and energy use per individual [53, 54] and have greater mobility, home ranges, and longevity [55, 56] and, thereby, control more resources over greater and coarser spatial scales [52, 57]. It is also proposed that well-linked and interacting species as key interactors are more important for the community [28, 58–62]. This approach characterizes the interaction structure of species placed in an ecological network. Among plants, on the other hand, some studies have shown that species within the same functional types but with different requirements and tolerances may provide insurance to the system in the form of long-term resilience against changes in environmental factors, such as global warming, grazing, drought or frost [35].

The latter example indicates that the keystone status of a species often appears to be context dependent, and may change with successional status, productivity, diversity, and other ecosystem traits [63]. It is therefore important to identify how the importance of traits that define keystone species change across a gradient of conditions, measuring environmental factors, community composition, trophic dynamics, and distribution of strong and weak links in the community (e.g., [24]). Without droughts, a specific plant species may not play an important role in maintaining community composition or ecosystem functioning. The Australian brushtail possum (Trichosurus vulpecula) may function as a keystone species in rata-kamahi forests by defoliating and killing canopy trees, but not in beech-dominated forests where floristic composition, but not forest structure, is typically affected [64–66]. A species in its native grounds may play no specific role in the system, whereas an invasive species may have devastating effects in the system it got introduced into, e.g., feral cats and rats on islands [67], Alewife (Alosa pseudoharengus) in nonnative freshwater lakes and ponds [68], and Cheatgrass (Bromus tectorum) in nonnative grasslands [69].

Identifying keystone species therefore is not without its problems. It is also important to notice that ecologically important species might not necessarily be the ones that are also considered important by traditional conservationists (i.e., rare species; [70]).

Issue-2: Usefulness for Management

Because of the limited resources available in comparison to conservation needs, it has been proposed to design protection of single species in the aim of indirectly protecting the regional biota. These “surrogate species” are roughly of three categories [71]: (1) flagships, charismatic species that attract public support; (2) umbrellas, species requiring such habitats that their protection might protect other species; and (3) biodiversity indicators, taxa whose presence may indicate high species richness. However the effectiveness of these policies has been questioned and [70] suggested that single-species management might be more effective when directed toward keystone species. Indeed, the importance of keystone species and ecosystem engineers in communities make these species particularly important conservation targets, since the loss of these species can affect entire communities and ecosystems [72]. However, the main difficulty for applying these concepts to conservation issues lays on both the identification of keystone species and ecosystem engineers in communities and on the context dependence of their impacts, as discussed in the previous section. Thus, although these concepts appear relevant for conservation policies, it is still a long way from providing general and practical recommendations for conservationists and managers [71].

The concepts of keystone species and of ecosystem engineers could also be useful for other management issues in natural and anthropized ecosystems, such as for ecosystem restoration or agriculture. For example, in agro-ecosystems, several well-known ecosystem engineers have been used to improve soil fertility and crop yield. In some countries, farmers make use of the soil fertilizing effect of termites by spreading termite mound soil in their field [73]. Similarly, earthworm inoculation has generally positive effects on crop yield [74].

Future Directions

The notions of ecosystem engineers and keystone species have been playing prominent roles in ecology for several decades, still many questions and uncertainties ask for further investigations. Three of them are briefly described here. The first is how keystone roles and engineering effects are related to body-size: Do larger organisms have larger effects than smaller organisms? The second regards the context dependence of keystone roles and engineering effects: As the composition and structure of ecological communities are dynamic both in terms of species composition and species abundances, what does that imply for the role species have in communities and ecosystem functioning? The third line of research might be the most relevant for our society: How can the concepts for nature conservation, biodiversity protection, and the enhancement of environmental quality be used?