Environmental diversity: on the best-possible use of surrogate data for assessing the relative biodiversity of sets of areas

• Published in: Biodiversity and conservation 1996, Vol.5 (4), p.399-415
• Main Authors: Faith, D. P., Walker, P. A.
• Format: Article
• Language: English
• ISSN: 0960-3115; 1572-9710
• DOI: 10.1007/BF00056387

The conservation goal of representation of biodiversity (in the broad sense of all species) in protected areas requires best-possible use of available surrogate information. One standard approach is based on 'indicator' groups of taxa. A minimum set of areas having at least one representation of each indicator species is taken to be representative of other organisms. This same minimum-set approach is adapted to other 'attributes' of biodiversity, for example, derived environmental clusters. A weakness of these approaches is that useful information is lost; for example, for environmental clusters, there is no distinction made either among or within clusters. A more powerful surrogate approach can use some expression of environmental and/or biotic pattern so that variation among areas is seen as part of a continuum rather than partitioned into arbitrary clusters/attributes. The challenge in using pattern effectively is to adopt a robust model for the relationship between pattern and the underlying units of biodiversity, i.e. species. An environmental space (a continuum or ordination pattern), combined with the standard ecological continuum model relating species to environmental space, has advantages over other patterns based on hierarchy or distance matrices. Because an environmental space can be estimated either directly (observed environmental data) or indirectly (data on indicator groups), the corresponding surrogate-measure of biodiversity, 'environmental diversity' (ED) makes best-possible use of either kind of data. We conclude that the arbitrariness of the 'attribute' approach can be replaced by a robust surrogate 'pattern' approach that is flexible and avoids unwarranted assumptions.