By David A. Neill
During the past 30 years, much scientific attention has focused on study of climatic fluctuations during the Pleistocene and their effects on plant distributions. Debate among scientists has centered on the question of whether lowland Amazonia was significantly drier during Pleistocene glacial maxima (Ab'Sáber, 1982) or colder (Colinvaux, 1987; Bush et al., 1990; Colinvaux et al., 1997) or both (Piperno & Pearsall, 1998), and whether or not the Amazon forest was reduced to a few Pleistocene refuges (Prance, 1982) with consequent effects on present-day diversity and endemism patterns of plants and animals in the Neotropical lowlands.
In the high Andes, temperatures were about 6–7°C lower than at present during the last Pleistocene glacial maximum (18,000–13,000 years ago), and the lower limits of glaciers in the northern Andes were at 3,100–3,800 m (van der Hammen, 1974, 1982). At present, the lower limits of glaciers on the Ecuadorian volcanoes are about 4,700–5,100 m (Jørgensen & Ulloa Ulloa, 1994). The upper limit of forest in the Andes during the Pleistocene glacial maxima was at about 2,000 m, i.e., about 1,500 m lower than the present-day upper limit of forest at about 3,500 m (van der Hammen, 1982). Therefore, most of the areas in the high Andes that are today covered with páramo, were beneath glacial ice during the Pleistocene maxima, and páramo vegetation then covered a much larger area than at present, on the slopes and the inter-Andean valleys above 2,000 m. Regional and world temperatures fluctuated up and down a number of times during the 2.5-million-year history of the Pleistocene, so the altitudinal belts of glacial ice, páramo, and montane forests also shifted up and down the slopes of the northern Andes a number of times (Hooghiemstra & Cleef, 1995).
Climatic shifts during the Pleistocene in the Neotropical lowlands certainly occurred, but the nature and magnitude of the changes, and their effect on vegetation patterns, are still being debated. The original Pleistocene refuge theory (Haffer, 1969)—which postulated that most of the Amazon basin was so dry during the Pleistocene glacial periods that the Amazon forest was replaced by savanna except for a few wetter refuges where forest remained—was popular during the 1970s and early 1980s (Prance, 1982), but has now been discredited because much more paleoecological evidence is available for the lowland Neotropics (summarized in Piperno & Pearsall, 1998). The Pleistocene climate was cooler in the Amazon lowlands than it is today, probably 5–7°C cooler, and also somewhat drier, with perhaps 25–40% less rainfall than at present. Central and western Amazonia, including the Ecuadorian portion of the basin, remained forested throughout the Pleistocene, but the floristic composition was different, with taxa that are now restricted to the montane forests of the Andes above 1,200 m, such as Podocarpus and Hedyosmum (Colinvaux et al., 1997; Athens, 1997), growing in the lowlands. Lowland Amazonian forests evidently had a different mix of species during the Pleistocene than is present there today, with montane species growing together with some of the lowland taxa that make up modern Amazonian forests.