Amazonian biodiversity and protected areas:: Do they meet?

Protected areas are crucial for Amazonian nature conservation. Many Amazonian reserves have been selected systematically to achieve biodiversity representativeness. We review the role natural-scientific understanding has played in reserve selection, and evaluate the theoretical potential of the existing reserves to cover a complete sample of the species diversity of the Amazonian rainforest biome. In total, 108 reserves ( 604,832 km(2)) are treated as strictly protected and Amazonian; 87 of these can be seen as systematically selected to sample species diversity ( 75.3% of total area). Because direct knowledge on all species distributions is unavailable, surrogates have been used to select reserves: direct information on some species distributions ( 15 reserves, 14.8% of total area); species distribution patterns predicted on the basis of conceptual models, mainly the Pleistocene refuge hypothesis, ( 5/ 10.3%); environmental units ( 46/ 27.3%); or a combination of distribution patterns and environmental units ( 21/ 22.9%). None of these surrogates are reliable: direct information on species distributions is inadequate; the Pleistocene refuge hypothesis is highly controversial; and environmental classiWcations do not capture all relevant ecological variation, and their relevance for species distribution patterns is undocumented. Hence, Amazonian reserves cannot be safely assumed to capture all Amazonian species. To improve the situation, transparency and an active dialogue with the scientific community should be integral to conservation planning. We suggest that the best currently available approach for sampling Amazonian species diversity in reserve selection is to simultaneously inventory indicator plant species and climatic and geological conditions, and to combine Weld studies with remote sensing.