Determining Keystone Species

Larisa Khanina

De Leo and Levin (1997) provide a good description of the concepts "keystone species" and "functional groups." As a rule, however, only a very few (one or perhaps several) functional groups in any community are groups of keystone species.

Smirnova (1998) states, "The population dynamics of keystone species define the mosaic pattern of succession of vegetation. Turnover cycles of energy and matter are dominated by the life activities of keystone species, and these activities determine the major shifts at the spatial and temporal scales at which these species exist. Population mosaics of key species have largest spatial-temporal dimensions, and population mosaics of subordinate species are thereby determined by key-species." The last means that vegetation patterns produced by key species are larger and last longer than those produced by subordinate species. The unit of analysis is the "spatial extent" covered by the species' population.

It needs to be underlined that we propose to name as keystone species only those species whose populations (or flocks of animals, as rule) either support or essentially alter the main vegetation pattern of the ecosystem. Under such comprehension, for example, only trees can be considered as keystone species of forest communities (detritus ecosystems), and bison, for example, can be considered as keystone species of grassland communities (pasture ecosystems). Note also that, in this case, the ecosystem is considered in the frame of some type of vegetation community (forest, meadow, etc.), and keystone species "are responsible" for the existence of an ecosystem of a certain type. The type of ecosystem will alter when keystone species disappear for some reason, or when new "stronger" keystone species come. We assume that this restriction in applying the keystone species notion is useful in studying the structure and functioning of the Earth's living cover.

Successional mosaics in vegetation result from biotic and abiotic disturbance patterns at different spatiotemporal scales. This hierarchy of disturbances to vegetation results in a hierarchical pattern of plant population mosaics. Examples of relatively large-scale disturbances in temperate zones are (1) catastrophic events (such as fire, hurricane); (2) pathogens (such as fungi or insects); and (3) patterns of mammalian herbivory (such as bison or beaver). Thus, the key species of the temperate zone are trees, which create and support forest communities, as well as pathogens and large herbivores, which destroy forest communities and create possibilities for the existence of other types of communities (Zaugol'nova et al. 1998). The hierarchical pattern of their influences is reflected in the qualitative differences (several orders of magnitude) in the scales of population mosaics and turnover time. Tree species produce mosaics of 1-25 ha, insects and fungi produce mosaics of 1-50 ha, and large herbivores produce mosaics of 100-10000 ha. Mosaics produced by subordinate species are essentially smaller. For example, herb and shrub species produce mosaics of 0.25-1 m² and 2.5 x 1000 m, respectively (Smirnova 1994).

The asynchronous development and mutual replacement of key-species mosaics maintain high biodiversity at several scales. According to Smirnova (1998), "There is correlation between structural and taxonomic diversity. The maximum taxonomic diversity in a climax landscape develops due to the structural diversity of population mosaics produced by all key species of the biota and the spatial-temporal heterogeneity of these mosaics."

The application of this approach to ecosystem management permits reconstruction of climax landscapes in destroyed areas and the construction of elaborate models of biodiversity.

Responses to this article are invited. If accepted for publication, your response will be hyperlinked to the article. To submit a comment, follow this link. To read comments already accepted, follow this link.

De Leo, G. A., and S. Levin. 1997. The multifaceted aspects of ecosystem integrity. Conservation Ecology [online]1(1): 3. Available from Internet. URL: http://www.consecol.org/vol1/iss1/art3.

Smirnova, O.V., editor. 1994. East-European broad-leaved forests. Nauka, Moscow, Russia. (In Russian.)

_______ . 1998. Population organization of biocenosis design of forest landscapes. Uspehi sovremennoj biologii 118:148-165. (In Russian with English resume.)

Zaugol'nova, L. B., O. V. Smirnova, I. Istomina, and L. G. Khanina. 1998. Population mosaic cycles in forest ecosystems. Studies in Plant Ecology 20:31.