Most biodiversity assessment methods tend to sample isolated areas of land cover such as closed forest or local land use mosaics. Contemporary methods of assessing biodiversity are briefly reviewed and focus on the relative roles of the Linnean species and plant functional types (PFTs). Recent case studies from central Sumatra and northern Thailand indicate how the range distributions of many plant and animal species and functional types frequently extend along regional gradients of light, water and nutrient availability and corresponding land use intensity. We show that extending the sampling context to include a broader array of environmental determinants of biodiversity results in a more interpretable pattern of biodiversity. Our results indicate sampling within a limited environmental context has the potential to generate highly truncated range distributions and thus misleading information for land managers and for conservation. In an intensive, multi-taxa survey in lowland Sumatra, vegetational data were collected along a land use intensity gradient using a proforma specifically designed for rapid survey. Each vegetation sample plot was a focal point for faunal survey. Whereas biodiversity pattern from samples within closed canopy rain forest was difficult to interpret, extending the sample base to include a wider variety of land cover and land use greatly improved interpretation of plant and animal distribution. Apart from providing an improved theoretical and practical basis for forecasting land use impact on biodiversity, results illustrate how specific combinations of plant-based variables might be used to predict impacts on specific animal taxa, functional types and above-ground carbon. Implications for regional assessment and monitoring of biodiversity and in improving understanding of the landscape dynamics are briefly discussed.
Année de publication
2004
Auteurs
Langue
English
Mots clés
biodiversity, assessment, indicators, vegetation, environmental factors, animals, plants
Géographique
Indonesia, Thailand
