Neogene and Quaternary development of the neotropical rain forest: the forest refugia hypothesis, and a literature overview

The upheaval of the northern Andes in Miocene and Pliocene time changed the drainage system in northern South America significantly and caused the present-day rain forest areas of Choco and the Lower Magdalena Valley became separated from Amazonas. Plant diversity may have reached the highest level in the Miocene or Pliocene, and excessive present-day phytodiversity may be regarded as a legacy of the Tertiary, rather than an evolutionary product of the Quaternary. In the Quaternary strong temperature oscillations, related to the series of ice-ages, were superposed on the Late Tertiary forest dynamics, which included river displacement and latitudinal migrations of the equatorial rain belt (caloric equator) with the rhythm of the precession cycle of orbital climate forcing. The hypothesis that claims a permanent rain forest cover all over the Amazon basin during the last glacial is in contrast with the 'forest refugia hypothesis', which accepts replacement of rain forest by savanna, or savanna forest, during dry climatic intervals. Both scenarios have been evidenced by pollen records. In this paper, it is suggested that both hypotheses are not necessarily conflicting and apparently did occur in different parts of the Amazon basin, and in different periods, depending on the climatological constraints. A compilation of the most important literature concerning the vegetational, climatic, and environmental history of the rain forest areas of Amazonas and Choco, and surrounding dry ecosystems has been included. (C) 1998 Elsevier Science B.V. All rights reserved.