Abrupt Climate Changes During the Holocene Across North America From Pollen and Paleolimnological Records - Abrupt Climate Change: Mechanisms, Patterns, and Impacts - page 171

for example, the modern analogue technique (MAT) [Over-
peck et al., 1985]. This work has been done in the context of
an extensive knowledge of the ecology and biogeography of
the principal plant species of the region, which lends theo-
retical justification to the conclusions.
Records of vegetation change preserved in pollen time
series have been used to document climate variability on
several time and space scales [Wright et al., 1993]. Studies
have shown that vegetation responds rapidly to climate
change [e.g., Webb,1986;Gajewski, 1987], and there is
evidence of synchronous vegetation response to abrupt cli-
mate changes during the late glacial and glacial period
[Grimm and Jacobson, 1991; Shuman et al., 2002a, 2002b;
Viau et al., 2002; Williams et al., 2002, 2004; Peros et al.,
2008] (see also collection of papers in special section

also show coherence with marine and ice core records [Viau
et al., 2002, 2006].
Increasingly, studies are using several different proxy-
climate indicators in the same core to provide more informa-
tion about past environments. In cases when different fossil
organisms from the same core are used for quantitative
paleoclimate reconstructions, the time series based on the
different indicators can be compared. Other indicators besides
pollen are being used, although these are far from being as
well studied or understood. For example, chironomids, the
larval form of nonbiting midges (insects) are fossilized in
lake sediments and can be identi ed. Initial work suggested

that these organisms provide useful paleoclimate estimates
[Batarbee, 2000], and although subsequent work has sug-
gested interpretation may be more subtle [Walker and Cwynar,
2006], several local studies have been accomplished. Other
fossils, such as diatoms, record changes in aquatic produc-
tion and factors such as pH. To the extent that the lake
chemical environment is affected by climate, diatoms may
therefore be an indirect recorder of past climates, although in
practice, it has proven dif

Veg-
etation Response to Millennial-scale Variability during the
Last Glacial ” in Quaternary Science Reviews, 29(21 - 22),
2010) as well as during the Little Ice Age (LIA) and Medie-
val Warm Period (MWP) of the past 1000 years [Gajewski,
1987]. The pollen-based paleoclimate reconstructions are
beginning to be supplemented by results using other proxy-
climate records found as fossils in lake sediments.
In this chapter, we will discuss Holocene climate variabil-
ity in North America using a spatially extensive pollen
database [Grimm, 2010] supplemented with other terrestrial
proxy-climate data. Quantitative reconstructions of Holo-
cene climates are made possible due to a large publically
available modern pollen database [Whitmore et al., 2005] for
quantitative calibration of the fossil paleoecological time
series. This research effort provides information of how
climate variability affects terrestrial ecosystem structure and
function. Although radiocarbon dating of these series is
currently a limitation to the resolution that can be obtained,
this can be overcome, and high-resolution series of past
vegetation and ecosystem dynamics can be obtained.
Using these pollen databases and the modern analogue
method, maps have been prepared of the climates of North
America at 100 year intervals for the past 12 kyr, and when
compared to climate model simulations of 6 kyr, both simu-
lated and reconstructed patterns show important similarities
[e.g., Wright et al., 1993; Gajewski et al., 2000; Sawada et
al., 2004]. Regional averaging of several pollen sequences is
one method to enable high-resolution time series to be ob-
tained [Viau et al., 2006] and is a methodology comparable
to the development of tree ring chronologies. Time series of
the climate of the Holocene have been developed for North
America, Europe, and several regions. Although this chapter
will focus on North America, studies have shown that the
timing of millennial-scale transitions is coherent in North
American and European pollen diagrams, as might be expected
[Gajewski et al., 2006]. The pollen-based reconstructions

“

cult to quantify past changes due
to the huge diversity of species and large differences even in
neighboring lakes [Bouchard et al., 2004]. Many other fossil
groups have been studied, but extensive enough data sets and
knowledge are not yet available for their use in other than
local studies. Chemical and physical components of the
sediment are also widely used as indices of past carbon
dynamics and temperature [Willemse and Törnqvist, 1999;
Kaufman, 2009; Fortin and Gajewski,2009].Theseare
especially useful as they are easy to obtain, and thus, high-
temporal-resolution records can be developed.

2. DATA AND METHODS

Pollen data are collected from sedimentary deposits and,
over the past few decades, mostly from lake sediments.
Cores from the sediments are sampled and pollen extracted
from a series of levels. The cores are dated using radiocarbon
and other methods such as 210 Pb. A pollen diagram from a
sediment core describes the vegetation history of the region,
and since plant growth is limited in part by climate, changes
in climate can be interpreted. However, pollen data are
multivariate time series, and to be useful as paleoclimate
indicators, they must be transformed, using some kind of
transfer function, to climate series in the units of degrees
Celsius or millimeters precipitation.

2.1. The North American Pollen Database

Continental-scale maps and time series of Holocene cli-
mate are made possible due to the availability of databases of