FIGURE 5.14 Matching of a disturbance gradient with an index gradient for biotic integrity. (From Karr, J. R. 2000. Ecological Integrity, Integrating Environment, Conservation, and Health. D. Pimental, L. Westra, and R. F. Noss (eds.). Island Press, Washington, DC. With permission.)
ecosystem, such as ant abundance (Andersen and Sparling, 1997; Majer, 1983) or decomposition rate (Durall et al., 1985; Lawrey, 1977) for terrestrial systems and phosphorus concentration (Dillon and Rigler, 1975) for lakes. In these cases the individual parameters used as indicators are judged to have special importance in the particular ecosystem context under consideration. Different individual parameters of the ecosystem develop at different rates. Thus, as mentioned earlier, marsh vegetation was restored quickly at Kenilworth Marsh in Washington, DC, but the development of organic matter content in the soil has lagged far behind (Kassner, 2001).
In other cases, sets of parameters are combined together into composite indices, such as with the wetland evaluation technique (i.e., WET, see Adamus, 1988) or the lake trophic state index (Carlson, 1977; Lambou et al., 1983). One of the best known indices is the Index of Biotic Integrity (IBI) developed by James Karr (1981, 1991; Karr and Chu, 1999) which was initially used for stream fish communities but has been applied more widely over time. The IBI consists of a series of attributes, termed metrics, that reflect both structural and functional characteristics of an ecosystem. In general, metrics are summed and an overall index is calculated that has meaning relative to reference conditions. Karr's index is especially interesting because he developed a new concept with it, termed biotic integrity, which is meant to synthesize the qualities of an ecosystem. Karr and Dudley (1981) define biotic integrity as "a balanced, integrated, adaptive community of organisms having a species composition, diversity, and functional organization comparable to that of natural habitat of the region." The approach is to devise an index scale that matches a general scale of ecosystem structure and function (Figure 5.14). This and other related concepts, such as ecosystem health, represent new approaches for relating human impacts to the condition of the environment. However, it must be remembered that much subjectivity is involved in choosing metrics for the IBI. For example, exotic species are usually left out of the index because they are judged not to contribute to the biotic integrity of the system. But exotic species do add diversity and function to the actual ecosystem where they are found, which is thus unaccounted for by the convention of the IBI. Semantic issues with applying the idea of health to an ecosystem also are problematic (Slobodkin and Dykhuizen, 1991). For example, a polluted stream dominated by tolerant species (Figure 2.3) might be considered unhealthy by human standards but it is perfectly adapted to the conditions it is exposed to. The community of species in the polluted stream is appropriate (and healthy) for an energy signature that includes organic wastes. In fact, the clean water or intolerant species would be indicators of unhealthy conditions in a polluted stream because they are not adapted to the stream's energy signature! Efforts at assessing ecosystem function for wetlands with an indicator approach are perhaps the most developed in the field of restoration ecology, to the point of having an established American Society for Testing and Materials standard ASTM, 1998), but even here subjectivity still exists. This sense of subjectivity in restoration ecology sets it apart from other areas of ecological engineering and engineering in general where there is little subjectivity in what to measure in order to determine the success of a design. For example, standards in terms of BOD, total nitrogen, and suspended solids are well accepted as measures of success for treatment wetlands, and there is little disagreement between workers on the subject.
An important procedure that is often used in restoration ecology is to compare measures of a restored ecosystem to a reference system in order to evaluate success. In this situation the reference ecosystem is judged to represent ideal or at least appropriate conditions, which usually means as close as possible to the natural or undisturbed state. Thus, success is gauged by the degree to which the restored ecosystem matches with the reference ecosystem. The logic of this procedure is obvious, yet difficulties arise in establishing reference sites. First, a decision must be made on the type of ideal reference and then a search must be made to discover if any of these actually exist in the landscape. Here again, subjectivity is involved, which leaves open the possibility for critical debate. A good deal of literature concerns the reference ecosystem issue (Aronson et al., 1994; Brinson and Rheinhardt, 1996; Egan et al., 2001; Findlay et al., 2002; Hughes et al., 1986; White and Walker, 1997), and Hughes (1994) lists six approaches for establishing a reference. The most objective approach is to use historical data on the original ecosystem that was impacted and is now to be restored. Unfortunately, this kind of data is seldom available. The more typical approach is to match the ecosystem to be restored with nearby, similar ecosystems that have not been impacted (see, for examples, Confer and Niering, 1992; Galatowitsch and van der Valk, 1996). Because of the inherent variability of ecosystems, it is seldom possible to locate a single individual reference ecosystem, and typically, multiple reference ecosystems are used to account for natural variation. Inevitably, these studies deal almost as much with interpreting reference conditions as they do with comparing reference sites to restored sites. However, when high quality reference sites can be established, they become very valuable as examples of natural conditions, and they deserve study in their own right and preservation for their special environmental value.
A number of public policies relate to restoration ecology. These are parts of the complex hierarchies of laws that regulate environmental impacts and mandate mitigation. At the federal level there are examples such as the National Environmental Protection Act and the Clean Water Act, along with others that call for restoration after strip mining, oil spills, and other impacts. States also become involved in local regulation with various legislation. All of this makes understanding the "regulatory" environment a formidable task and one beyond the scope of this text. However, the important case of wetland regulation is discussed below to illustrate some facets of public policy in regards to restoration.
Wetlands regulations have evolved in the U.S. as society has become aware of the values of these ecosystems. At one time wetlands were viewed as wastelands with no value, and they were actively filled or drained. However, in the 1960s and 1970s their natural values in hydrology, water quality, wildlife habitat, and education began to be recognized, and by the 1980s and 1990s wetland values began to be quantified (see Chapter 8). One consequence of the recognition of these values was that laws were enacted to protect wetlands. A national policy emerged called "No Net Loss," in order to reverse the destruction of wetlands and to restore both their area and function (Davis, 1989; Deland, 1992). A mitigation process has been created as part of this overall policy to deal with cases where a land owner wishes to develop a wetland for commercial or residential purposes (Beck, 1994; Berry and Dennison, 1993). Mitigation refers to a set of actions or rules that seek to preserve and even increase wetland values while accommodating economic development. Three main categories of action are meant to be applied in a sequential fashion for each case of proposed wetland impact: first, attempt to avoid the impact; second, attempt to minimize the impact; and third, provide compensation where the impact is inevitable. Compensation usually requires the creation of new wetlands or the enhancement of existing wetlands, both of which require the technology of restoration ecology. In general, wetland regulations and restoration techniques have developed concurrently (Kruczynski, 1990; Wolf et al., 1986) and both are still evolving. A multimillion dollar industry of environmental consultants also has developed for the evaluation of natural wetlands and for the design and construction of mitigation wetlands. Thus, large areas of new wetlands are being created across the country to compensate for natural wetlands that are being destroyed by economic development. Usually, a larger amount of creation is required relative to wetland destruction, such as a 2:1 ratio of created vs. destroyed acreage. This is done in an effort to ensure that wetland functions are not lost and, hopefully, that their values actually will be increased. The entire topic has become controversial, primarily because evidence is scattered and incomplete on the question of whether restoration technology can create new wetlands that are equivalent to the natural ones that are lost to development (Harvey and Josselyn, 1986; Malakoff, 1998; Race, 1985, 1986; Race and Christie, 1982; Savage, 1986; Young, 1996).
Restoration clearly can create ecosystems, but do these new ecosystems provide the same services as the natural ones that are lost? A great deal of ecological research is being conducted on this topic (Kusler and Kentula, 1990; Zedler, 1996a,b, 2000), but the mitigation question remains unresolved and contentious. Essentially the situation is the ecological equivalent of the Turing test for determining artificial intelligence in computers. The mathematican Alan Turing (1950) invented this imitation game just as digital computer technology was being developed. In the most general form of the test a human is seated at a teletype console by which he or she can communicate with a teletype in another room. The second teletype is controlled either by another human or by a computer. The programmer at the first teletype asks questions through the console to determine whether he or she is in contact with a human or a computer in the other room. If the programmer cannot distinguish between responses of a human and a computer at the second teletype, then the computer is said to have passed the test and is considered to be intelligent. In the ecological equivalent of the Turing test, ecologists sample created and reference wetlands, like the programmer asking questions of the human and the computer (for examples, see Wilson and Mitsch, 1996; Zedler et al., 1997). The created wetland passes the test if the ecologist cannot distinguish it from the reference wetland. Unfortunately, at the current state of the art, created wetlands do not seem to be passing the ecological Turing test very often (Kaiser, 2001a; Turner et al., 2001). Despite this situation, though, the "No Net Loss" policy and the mitigation process are achieving at least some kind of balance between economic development and environmentalism. At the same time, these policies are creating a major source of employment for ecological engineers whose growing experience should lead to technologies for achieving functional equivalency between created and natural wetlands (Zentner, 1999).
Was this article helpful?