For more than a century it has been postulated that the Holocene vegetation of western Europe has changed in significant ways. A half-century ago a lively debate revolved on whether there were one or two dry intervals causing bogs to dry out and become forested, or whether instead the climate warmed to a maximum and then cooled. Today none of these climatic schemes is accepted without reservation, because two nonclimatic factors are recognized as significant: the differential immigration rates of dominant tree types (e.g., spruce in the north and beech in the south) brought unexpected changes in forest composition, and Neolithic man cleared the forest for agriculture and thereby disrupted the natural plant associations.In North America some of the same problems exist. In the hardwood forests of the Northeast, which are richer than but otherwise not unlike those of western Europe, the successive spread of white pine, hemlock, beech, hickory, and chestnut into oakdominated forests provides a pollen sequence that may yield no climatic message. On the other hand, on the ecotone between these hardwood forests and the conifer forests of the Great Lakes-St. Lawrence area, the southward expansion of spruce, fir, and tamarack in the late Holocene implies a climatic cooling of regional importance, although the progressive conversion of lakes to wetlands favored the expansion of wetland forms of these genera.In the southeastern states the late-Holocene expansion of southern pines has uncertain climatic significance. About all that can be said about the distribution and ecology of the 10 or so species is that some of them favor sandy soils and are adapted to frequent fires. In coastal areas the expansion of pines was accompanied by development of great swamps like Okefenokee and the Everglades—perhaps related to the stabilization of the water table after the early Holocene rise of sea level. The vegetation replaced by the pines in Florida consisted of oak scrub with prairie-like openings, indicating dry early Holocene conditions, which in fact had also prevailed during the time of Wisconsin glaciation.In the Midwest the vegetation history provides a clearer record of Holocene climatic change, at least along the prairie border in Minnesota. With the withdrawal of the boreal spruce forest soon after ice retreat, pine forest and hardwood forest succeeded rapidly, as in the eastern states. But prairie was not far behind. By 7000 years ago the prairie had advanced into east-central Minnesota, 75 miles east of its present limit. It then withdrew to the west, as hardwoods expanded again, followed by conifers from the north. The sequence easily fits the paleoclimatic concept of gradual warming and drying to a maximum, followed by cooling to the present day. It is supported by independent fossil evidence from lake sediments, showing that lakes were shallow or even intermittently dry during mid-Holocene time.Here we have a paleoclimatic pattern that is consistent with the record from glaciers in the western mountains—a record that involves a late-Holocene Neoglaciation after a mid-Holocene interval of distant glacial recession. Just as the Neoglaciation is time-transgressive, according to the review of its evidence by Porter and Denton, so also is the mid-Holocene episode of maximum warmth, and they are thus both geologicclimate units. The warm episode is commonly termed the Hypsithermal, which, however, was defined by Deevey and Flint as a time-stratigraphic unit that is supposed to have time-parallel rather than time-transgressive boundaries. It was defined on the basis of pollen-zone boundaries in western Europe and the northeastern United States that have a sound biogeographic but questionable paleoclimatic basis. Perhaps it should be redefined as Porter and Denton suggest, as a geologic-climate unit with recognizable time-transgressive boundaries that match the gradual geographic shifts in the general circulation of the atmosphere and the resulting location of storm tracks and weather patterns. Holocene glacial and vegetational progressions provide a good record of climatic change, if one can work out the lag effects related to the glacial economy and the geographic factors controlling tree migration. The terminology for the Holocene, where so much time control is available, should indicate the dynamic character not only of the climate but also of the geologic and biogeographic processes controlled by climate.
Effects of past climate variability on fire and vegetation in the cerrãdo savanna of the Huanchaca Mesetta, NE Bolivia
