Late
Quaternary Biomes of Canada and the Eastern United States
J. W. Williams, T. Webb III, P. J.
H. Richard, and P. Newby
2000 Journal of Biogeography 27: 585-607
Abstract
-
Pollen data have been used to
construct
biome maps for today, 6000 yr. B.P., and 18,000 yr. B.P. for Canada and
the eastern United States.
-
The inferred modern biome
distributions
agree well with independent reconstructions of North American
vegetation
prior to European settlement. Some discrepancies between the pollen
data
and the modern potential vegetation are caused by post-settlement
clearing
of the landscape and the consequent increase of herbaceous types in the
recent pollen record.
-
Biome distributions at 6000 yr.
B.P. reflected
the warmer and drier conditions then prevalent in the continental
interior,
but the overall position of biomes was similar to that today. Boreal
treeline
in North America was not significantly north of its present position,
in
contrast to the 100-300 km shift reported for Siberia. At the last
glacial
maximum (18,000 yr. B.P.), steppe and tundra were prevalent in the
Midwest
and north-west Canada, and coniferous forests and woodlands grew in
eastern
North America. The open vegetation at 18,000 yr. B.P. was likely due to
drier conditions and/or lower concentrations of atmospheric CO2.
-
The composition and physical
structure
of biomes is not constant over time. Mid-Holocene biomes were similar
in
structure to those today, but shifts in the relative importance of
individual
plant functional types are large enough that the physical properties of
biomes such as albedo, canopy conductance, and surface roughness likely
varied even during the Holocene. Last glacial maximum biomes were
structurally
different from their modern counterparts. The biome maps therefore may
obscure significant vegetational changes in space and time during the
late
Quaternary.
-
The difference between the
highest and
next highest affinity scores for each sample measures how strongly
affinity
scores discriminate among biomes. For many biomes, the difference is
not
large, and affinity score ties are not uncommon, highlighting the
importance
of tiebreak procedures when using the biomisation method.