FLAGSHIPS, UMBRELLAS, AND KEYSTONES - IS SINGLE-SPECIES MANAGEMENT PASSE IN THE LANDSCAPE ERA

Because it is so difficult to monitor and manage every aspect of biodi versity, several shortcuts have been proposed whereby we monitor and/o r protect single species. The indicator species concept is problematic because there is no consensus on what the indicator is supposed to in dicate and because it is difficult to know which is the best indicator species even when we agree on what it should indicate. The umbrella s pecies (a species that needs such large tracts of habitat that saving it will automatically save many other species) seems like a better app roach, although often whether many other species will really fall unde r the umbrella is a matter of faith rather than research. Intensive ma nagement of an indicator or an umbrella species (for example, by trans plant or supplemental feeding) is a contradiction in terms because the rest of the community to be indicated or protected does not receive s uch treatment. A flagship species, normally a charismatic large verteb rate, is one that can be used to anchor a conservation campaign becaus e it arouses public interest and sympathy, but a flagship need not be a good indicator or umbrella. And conservation of flagship species is often very expensive. Further, management regimes of two flagship spec ies can conflict. Ecosystem management, often on a landscape scale, is a proposed solution to problems of single-species management. Keep th e ecosystem healthy, according to this view, and component species wil l all thrive. However, conservationists have concerns about ecosystem management. First, if is variously defined, and many definitions empha size the commodities ecosystems provide for humans rather than how hum ans can protect ecosystems. Second, the term 'ecosystem health' is ill -defined and associated with an outmoded, superorganismic view of the ecosystem. Third, ecosystem management seems focused on processes and so would appear to permit losses of species so long as they did not gr eatly affect processes like nutrient-cycling. Fourth, ecosystem manage ment is often implemented by adaptive management. This may make it dif ficult to study the underlying mechanisms driving an ecosystem and to know when an entirely new management approach is needed. Thus, some co nservationists see ecosystem management as a Trojan horse that would a llow continued environmental destruction in the name of modern resourc e management. The recognition that some ecosystems have keystone speci es whose activities govern the well-being of many other species sugges ts an approach that may unite the best features of single-species and ecosystem management. If we can identify keystone species and the mech anisms that cause them to have such wide-ranging impacts, we would alm ost certainly derive information on the functioning of the entire ecos ystem that would be useful in its management. Some keystone species th emselves may be appropriate targets for management, but, even when the y are not, our understanding of the ecosystem will be greatly increase d. Keystone species may not be a panacea, however. We do not yet know how many ecosystems have keystone species, and the experiments that le ad to their identification are often very difficult. (C) 1998 Publishe d by Elsevier Science Ltd.