Communities change through a number of processes that can operate over very short to very long time scales. "Succession" is the replacement of populations in a habitat through time due to ecological interactions. A "landscape," in ecology, is the particular spatial arrangement of components of the environment that are important in some way to population dynamics of a given species. Landscapes usually include patches of multiple habitats, as well as variability in conditions that affect habitat quality. Unlike some definitions of the term landscape, this definition does not link landscapes to a particular spatial scale. Instead, it recognizes that landscapes are different for different organisms, depending on the spatial scales over which the organisms interact with the environment (Wiens, 1997). Landscapes have an important impact on local and regional community structure. For example, the structure of a metapopulation (e.g., the number of and distance between populations) is embedded within a landscape.
The habitat that is present in the largest proportion in a landscape and that has the greatest connectivity is considered the habitat "matrix," within which other habitat patches are distributed. A habitat matrix can be occupied by a competitively dominant species or by a diversity of species that coexist through the various mechanisms previously discussed. Alternative nonmatrix habitat patches are created in many ways, and many species are adapted to exploit patchily distributed habitats. The dynamics of these habitats are obviously important to community structure. We have to ask, how are the habitats formed, and what proportion of the landscape do they cover?
One type of nonmatrix habitat is created where the competitively dominant species are absent. This habitat is characterized by an abundance of resources due to lack of competition. Fugitive species are adapted to exploit these patches. A process that causes the removal of an otherwise competitively dominant species or group of species is known as a "disturbance." Disturbances also alter distributions of resources or modulators. Many communities are dependent upon disturbances to maintain species diversity and ecosystem function. Light, for example, determines density and diversity of plants within a stand. In a closed-canopy forest, little light hits the ground. Density is generally high in these stands and diversity low. If these areas are subject to disturbances such as tree fall or fire, density of the stand is decreased, light will strike the forest floor, nutrients and water will not be captured as rapidly, and herbaceous layer species will be allowed to establish.
The initial species to appear after a disturbance are r-selected species with dispersal strategies (in space or time) designed to place them in such habitats first. These pioneer species are also capable of making opportunistic use of available resources or have mechanisms to increase rates of nutrient cycling such as N fixation. These species are replaced in time by more competitive species; for example, plant species more tolerant of shade or low soil nutrients. The "climax community" is a stable endpoint of succession, or at least an assemblage in which succession has slowed to the point at which other processes are more important. The initial model of a single climax community has been shown to be inaccurate. Instead, climax communities form a continuum that varies across environmental gradients largely characterized by local variations in climatic and edaphic conditions, local disturbance regimes, and biotic factors, particularly herbivores. Normally the climax community also dominates the landscape and is therefore the matrix community.
Secondary succession follows disturbances that leave soils largely unchanged and plant propagules in the seed bank. The progress of plant succession is often predictable based on climate, soil type, and the presence of seeds in the seed bank. Nutrient cycling is also often altered by disturbances. Enrichment phenomena increase available nutrients, such as by release from litter layers and humic materials through burning. Nutrient availability and rates of nutrient cycles may also be decreased such as when vegetation and humic layers are removed entirely from an area through hurricanes, floods, or intentional management such as plowing. Secondary succession cannot occur following a catastrophic disturbance that removes soil and all biota, such as glacial activity and volcanic eruptions. In this case primary succession, or succession without inputs from a dormant community at the site of the disturbance, occurs at the site. This type of succession often takes hundreds of years to return the community to the predisturbance state. The time required for soil development and recovery of soil populations such as decomposers and mycorrhizal symbionts can often delay recovery of plant communities (Allen et al., 1992).
Disturbances are an inherent part of community structure in a large number of systems. Most disturbances are caused by events that are repeated at some rate and spatial scale. The constant creation of disturbed patches and gradual return to a climax community creates a "shifting mosaic" of different habitats at different stages of succession (Wu and Loucks, 1995). The proportion of the landscape that is in the climax community should equal some steady-state value determined by the rate and spatial scale of the disturbance events and the rate of return to the climax community through succession. This allows fugitive species to depend entirely on the presence of minor habitats with relatively quick turnover rates. Some matrix communities are dependent on widespread repeated disturbance; in this case the matrix community is not the same as the climax community. Such species have evolved mechanisms to persist or regerminate following disturbance. Mangrove forests are dependent on hurricanes to remove colonizing species that without disturbance have the potential to outcompete mangroves. Many grasslands are maintained by fires, as they are responsible for removing tree seedlings that can lead to the establishment of deciduous forests. Unfortunately, catastrophic events such as fires in areas with high debris loads and in areas unaccustomed to fire, such as the tropical rain forest, result in a great deal of damage because species have not evolved mechanisms to tolerate such disturbances. These events have become more common due to human intervention, as have chronic disturbances such as acid deposition and excessive nutrient loading to which no communities are accustomed.
There are many habitats that are qualitatively different from the matrix habitat and are created through some process other than disturbance. Often an entirely different suite of organisms is adapted to exploit these habitats. Examples include the riparian zone near a river and the river itself. In soil, the rhizosphere, fecal matter, and decomposing plant tissue are important examples of this type of habitat (Blackwood and Paul, 2003). The latter two examples represent habitats defined by a limited pool of resources. Microbial succession in these habitats is driven by a constant change in environmental conditions as resources are used up and the environment is restructured. Some of these habitats, including all the soil habitats in the previous example, are created by events that, like disturbances, have a particular rate of occurrence and spatial scale, followed by community succession. Therefore, they also fit into our model of the landscape as a shifting mosaic of habitats.
The shifting mosaic picture of the landscape is based on a dynamic equilibrium model. However, the particular characteristics of a patch edge, the surrounding habitat patches, and ease of dispersal across other elements of the landscape may all be important determinants of metapopulation and local patch population dynamics. Climate change, human activity, and other novel events can result in nonequilibrium dynamics. Under these conditions, populations are kept from reaching carrying capacity or the stable equilibrium predicted by logistic models.
Was this article helpful?