ISSN 1028-334X, Doklady Earth Sciences, 2011, Vol. 437, Part 1, pp. 380–382. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © A.Yu. Puzachenko, A.K. Markova, 2011, published in Doklady Akademii Nauk, 2011, Vol. 437, No. 1, pp. 114–116.
380
The transition between the Pleistocene and the
Holocene is the key period in integrated research of
the reaction of biota to global climate variations. The
time–space variations for European natural com-
plexes are of particular interest, since the contrasts
caused by the direct influence of glaciation were espe-
cially manifest there. The history of the latest glacia-
tion is described in a comprehensive way in the scien-
tific literature. Models of spatial differentiation of
European natural complexes have been created with
varying degrees of detail for a series of climatic periods
during the Valdai (Vistula) glaciation and the Early
Holocene based on a vast body of paleontological and
paleoclimatic data [1–3]. In mid-2010, detailed data
on the climatic variations in the northern hemisphere
at the end of the Pleistocene and the beginning of the
Holocene became available; these data were generated
based on analysis of the isotope composition of ice in
the Greenland ice sheet [4–6]. The information accu-
mulated allows conducting general analysis of the
effect of global climatic variations on biodiversity in
Europe during the transition from the Pleistocene to
the Holocene (33–8 ka).
The investigation was based on the results obtained
over the last two decades [3]. The source data on diver-
sity of the vegetation cover are represented by spore–
pollen spectra (431 sections), and the data on diversity
of theriocomplexes, by lists of mammal species
(400 locations). The methodological approaches to
application of multidimensional analysis to paleobiolog-
ical data were taken from the studies of A.V. Kozharinov
and Yu.G. Puzachenko [7, 8]. We extended these
approaches to paleozoological data for the first time
[3]. Nonmetric multidimensional scaling was used as
the principal tool of multidimensional analysis [9]. As
a result, a limited set of linear independent macrovari-
ables was generated, which contained generalized
information on the spatial structure of the diversity.
The method used by us is similar to thermodynamic
description of physical systems using special macrova-
riables (temperature, pressure, volume, and entropy).
The objective of the study was to justify the use of
the information parameters determined from the mac-
rovariables for description of variations in spatial
biodiversity.
The descriptive models were created independently
for each of the six consecutive time periods: the Bryansk
interstadial (33–24 ka), the maximum cooling during the
latest glaciation (24–17 ka), the Late Glacial (17–12.4
ka), the Bolling–Allerod interstadial (12.4–10.9 ka), the
late Dryas stadial (10.9–10.2 ka), and the Early
Holocene (Preboreal–Boreal) (10.2–8 ka).
We demonstrated previously that the number of
macrovariables is different for different time periods,
but in each of them the principal macrovariables have
correlations (in the general case, nonlinear correla-
tions) with both geographic positions of the locations
and their taxonomic diversity [3].
At the second stage of the investigation, the Shan-
non–Weaver entropy, the Kullback–Leibler diver-
gence measure [10], and the sums of these parameters
for all variables were calculated for each of the macro-
variables.
The Kullback–Leibler divergence is a quantitative
measure of deviation from a selected model distribu-
tion. The normal distribution with the average and
variance being equal to the average and variance of the
macrovariable was used as the model. The model cor-
Using Multidimensional Analysis and Information Functions
for Macrodescription of European Natural Complexes
in the Second Part of the Late Pleistocene and the Holocene
A. Yu. Puzachenko and A. K. Markova
Presented by Academician V.M. Kotlyakov October 30, 2010
Received November 17, 2010
Abstract—This article analyzes the effect of geographic factors on variations in the species diversity of mam-
mals and plants for the European territory. Multidimensional analysis and a number of information parame-
ters have been applied for the first time, which allowed identifying objective indicators of mammal and plant
diversity for different time periods in the Late Pleistocene and the Holocene.
DOI: 10.1134/S1028334X1103007X
Institute of Geography, Russian Academy of Sciences,
Staromonetnyi per. 29, Moscow, 119017 Russia
GEOGRAPHY