Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Contents lists available at ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology
journal homepage: www.elsevier.com/locate/palaeo
The Coexistence Approach—Theoretical background and practical
considerations of using plant fossils for climate quantification
T. Utescher a,j,⁎, A.A. Bruch b, B. Erdei c, L. François d, D. Ivanov e, F.M.B. Jacques f, A.K. Kern g, Y.-S.(C.) Liu h,
V. Mosbrugger a, R.A. Spicer i
a
Senckenberg Research Institute/BiK F (Loewe), Senckenberganlage 25, 60325 Frankfurt am Main, Germany
Heidelberg Academy of Sciences and Humanities, Research Center ‘The Role of Culture in Early Expansions of Humans’ at Senckenberg Research Institute, Senckenberganlage 25,
60325 Frankfurt am Main, Germany
c
Hungarian Natural History Museum, Botanical Department, Könyves K. krt. 40, Budapest 1087, Hungary
d
Unité de Modélisation du Climat et des Cycles Biogéochimiques (UMCCB), Université de Liège, Bât. B5c, Allée du Six Août 17, B-4000 Liège, Belgium
e
Institute of Biodiversity and Ecosystem Research, Bulgarian Academy of Sciences, Acad. G. Bonchev Str., bl. 23, BG-1113 Sofia, Bulgaria
f
Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Menglun, Mengla, PR China
g
State Museum of Natural History Stuttgart, Rosenstein 1, 70191 Stuttgart, Germany
h
Department of Biological Sciences, East Tennessee State University, Johnson City, TN 37614, USA
i
Environment, Earth and Ecosystems, Centre for Earth, Planetary, Space and Astronomical Research, The Open University, MK7 6AA, UK
j
Steinmann Institute, University of Bonn, Nussallee 8, 53115 Bonn, Germany
b
a r t i c l e
i n f o
Article history:
Received 18 November 2013
Received in revised form 13 May 2014
Accepted 27 May 2014
Available online 2 June 2014
Keywords:
Coexistence approach
Guidelines
Palaeoclimate reconstruction
Palaeobotany
Cenozoic
a b s t r a c t
The Coexistence Approach was established by Mosbrugger and Utescher (1997) as a plant-based method to reconstruct palaeoclimate by considering recent climatic distribution ranges of the nearest living relatives of each
fossil taxon. During its existence for over more than 15 years, its basics have been tested and reviewed in
comparison with other terrestrial and marine climate reconstruction techniques and climate modelling data.
However, some controversies remain about its underlying data or its applicability in general.
In view of these controversies this paper discusses the power and limitations of the Coexistence Approach by
summarising past results and new developments. We give insights into the details and problems of each step
of the application from the assignment of the fossil plant to the most suitable nearest living relative, the crucial
consideration of the usefulness of specific taxa towards their climatic values and the correct interpretation of
the software-based suggested palaeoclimatic intervals. Furthermore, we reflect on the fundamental data integrated in the Coexistence Approach by explaining different concepts and usages of plant distribution information
and the advantages and disadvantages of modern climatic maps. Additionally, we elaborate on the importance of
continually updating the information incorporated in the database due to new findings in e.g., (palaeo-)botany,
meteorology and computer technology.
Finally, for a transparent and appropriate use, we give certain guidelines for future applications and emphasize to
users how to carefully consider and discuss their results. We show the Coexistence Approach to be an adaptive
method capable of yielding palaeoclimatic and palaeoenvironmental information through time and space.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Plant-based palaeoenvironmental reconstruction is a widely accepted method used for studying palaeoclimate. This potential was already
recognised by Shen Kuo in the 11th century in China (Needham, 1986),
as well as later in the 19th century of Europe by Heer (1855, 1856,
1859), who quantitatively estimated Neogene palaeoclimate based on
plant fossils for the first time. Subsequently, biological palaeoclimate
proxies have mostly relied on the principle of physiological actualism,
⁎ Corresponding author at. Steinmann Institute, Nussallee 8, 53115 Bonn, Germany.
E-mail addresses: [email protected] (A.A. Bruch), [email protected]
(A.K. Kern).
http://dx.doi.org/10.1016/j.palaeo.2014.05.031
0031-0182/© 2014 Elsevier B.V. All rights reserved.
which assigns identical dependencies of modern morphological or taxonomical units on environmental constraints to their comparative fossil
equivalents. Without this principle, there would not be any correlation
between the dependencies of modern and fossil phytocoenoses on
changing environments (e.g., Mai, 1995; Tiffney, 2008). Therefore, several methods were developed in an attempt to reconstruct and quantify
palaeoenvironmental and palaeoclimatic parameters.
Among various more specific approaches such as the analysis of stomatal data, plant–insect interaction, or biomarkers, there are two classical complementary procedures to trace past climate from plant fossils
(Chaloner and Creber, 1990). One is based on physiognomy, which
takes advantage of empirical correlations between climate parameters
and specific plant traits like leaf anatomy (e.g., Bailey and Sinnott,
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
1915, 1916; MacGinitie, 1969; Parkhurst and Loucks, 1972; Wolfe,
1979; Wing and Greenwood, 1993; Wolfe, 1993, 1995; Wilf, 1997;
Wiemann et al., 1998; Spicer et al., 2009; Teodoridis et al., 2011) or
wood anatomy (e.g., Wheeler and Baas, 1993; Wiemann et al., 1998,
1999; Terral and Mengüal, 1999). As a consequence, to a large extent
these are independent of taxonomic determinations. These methods
are still being actively developed through more comprehensive calibration sets (e.g. Jacques et al., 2011a) and improved analytical techniques
(e.g. Teodoridis et al., 2011) to overcome discrepancies between estimated and observed modern data (Wiemann et al., 2001). Thus, precision and universal applicability in time and space are continuously
improved.
The second method to reconstruct palaeoclimate is based on the
nearest living relative (NLR) approach. This method relies on the close
relationship between modern and fossil plants and assumes that the
fossil taxon had the same climatic requirements as its modern representatives. Hence, it is applied mainly in the Quaternary and Neogene,
where evolutionary change of environmental requirements of each
plant is regarded as minimal (e.g., MacGinitie, 1941; Hickey, 1977;
Chaloner and Creber, 1990; Mosbrugger, 1999). Based on this assumption, the environmental tolerance of a modern plant with a known climatic distribution can be used to estimate the past climate. Likewise,
by using the taxa of a plant fossil assemblage, the climatic information
of each modern representative can be compared. If no evolutionary
changes in environmental tolerance have taken place, overlapping climatic ranges between modern representatives should be the result
and represent the palaeoclimate. Initial approaches in this direction
were made using selected key taxa (Iversen, 1944; Hintikka, 1963;
Grichuk, 1969; Zagwijn, 1996). Later, due to increasing computational
facilities, all available information could be analysed and climatic ranges
incorporating all relevant taxa were included in analyses. Various techniques were developed (Kershaw and Nix, 1988; Kershaw, 1997;
Mosbrugger and Utescher, 1997; Fauquette et al., 1998; Klotz, 1999;
Kühl et al., 2002; Greenwood et al., 2003, 2005; Klotz et al., 2006; Kou
et al., 2006; Utescher et al., 2009a) differing in the method of compilation of climatic requirements, specialisation regarding organ type of fossils including the use of abundance, continental or global adaptability, as
well as the application of calibration procedures, and statistical treatment of data and outliers, each having its specific strengths. The
59
Bioclimatic Analysis approach (Kershaw, 1997; Greenwood et al.,
2003, 2005) uses climatic envelopes of NLRs obtained from bioclimatic
modelling of distributions of modern plant taxa (BIOCLIM). It includes
a statistical treatment of outliers, and is mainly applied to micro- and
megafloras of North America, Oceania, Arctic, and Antarctica. The Climate Amplitude Method (CAM) introduced by Fauquette et al. (1998)
employs, secondary to the coexistence aspect, modern European pollen
profiles for calibration, and therefore has its focus of application in
younger Neogene European microfloras, while the Probability Coexistence Spheres (PCS) developed by Klotz (1999) is based on European
plant distribution and the specific structure of the modern climate
space of Europe serving for a statistical calibration procedure. Therefore,
both latter methods are well best suited for the analysis of later Neogene
and Holocene European palynomorph records. In the Probability Density Functions (pdf) method introduced by Kühl et al. (2002), pdfs estimated for monthly mean January and July temperatures of preselected taxa are used in order to define the most likely climate conditions. The Calibrated Coexistence Approach (Utescher et al., 2009a) employs global modern climate space in order to calibrate fossil climate
conditions and at the same time specifies present-day regions that correspond to the fossil climate.
Among those technique based on the NLR concept the Coexistence
Approach (CA) by Mosbrugger and Utescher (1997) was one of the
first to propose a standardised procedure. The straightforward use of
the climatic requirements of modern plants in a global context with respect to single parameters, and the absence of further calibration using
modern pollen profiles or chorological aspect brings about robustness
and universal applicability. Therefore, it has become a widely used
tool to reconstruct palaeoenvironment based on a fossil plant assemblage and has yielded results in various applications from late Cretaceous to Pleistocene, well interpretable in the context of data obtained
from other proxies (more than 150 publications, see ‘publications’
at www.neclime.de and published CA data from Eurasia on Fig. 1;
Appendix 2). For comparative studies on regional and continental
scale (e.g., Bruch et al., 2004; Mosbrugger et al., 2005; Bruch et al.,
2006; Akgün et al., 2007; Bruch et al., 2007; Utescher et al., 2007;
Akkiraz et al., 2011; Bruch et al., 2011; Ivanov et al., 2011; Liu et al.,
2011; Quan et al., 2011, Utescher et al., 2011; Erdei et al., 2012; Miao
et al., 2012; Popova et al., 2012; Quan et al., 2012; Yao et al., 2012) or
Fig. 1. Cenozoic sites of Eurasia with CA climate data sets published in Pangaea (www.pangaea.de). For coordinates and corresponding doi codes cf. Appendix 2.
60
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
model proxy comparisons (e.g., Utescher et al., 1997; Gebka et al., 1999;
Micheels et al., 2007; Steppuhn et al., 2007; Micheels et al., 2009, 2011;
Tang et al., 2013), CA results have provided valuable information for the
understanding of large-scale climate patterns and processes over time
and space. Moreover, the application of the CA evidenced the close correlation of marine and continental time series (Mosbrugger et al., 2005;
Utescher et al., 2009a; Larsson et al., 2011; Utescher et al., 2012;
Rasmussen et al., 2013), and revealed orbital forcing (Utescher et al.,
2009b; Ivanov et al., 2011; Kern et al., 2011; Utescher et al., 2012) and
short-term climate variability at suborbital scales (Kern et al., 2012).
Methodological comparisons and validations of the CA have been
performed by applying the CA on different plant organ groups (leaves,
seeds, wood, pollen) of one fossil flora (e.g., Mosbrugger and Utescher,
1997; Pross et al., 1998; Utescher et al., 2000; Pross et al., 2001; Bruch
and Kovar-Eder, 2003; Liang et al., 2003; Utescher et al., 2012;
Bondarenko et al., 2013), on the application of different methods on
the same flora (e.g., Liang et al., 2003; Uhl et al., 2003; Roth-Nebelsick
et al., 2004; Poole et al., 2005; Uhl et al., 2006, 2007a,b; Pross et al.,
2012; Bondarenko et al., 2013), or on the comparison of CA results
with other proxies (e.g., Uhl et al., 2006; Eronen et al., 2011;
Harzhauser et al., 2011, 2012; Pross et al., 2012; van Dam and
Utescher, submitted for publication). In particular, the methodological
differences between physiognomic and NLR approaches have been
widely discussed in the literature, where a general agreement of results
has been stated by various authors (e.g., Wing and Greenwood, 1993;
Herman and Spicer, 1997; Mosbrugger and Utescher, 1997; Wilf,
2000). Especially, applications of the two most common leaf-based physiognomic methods, the LMA (Leave Margin Analysis) as well as CLAMP
(Climate Leaf Analysis Multivariate Program), and CA on an identical
flora show consistent data in general (e.g., Liang et al., 2003; Uhl et al.,
2003; Roth-Nebelsick et al., 2004; Uhl et al., 2006, 2007a,b; Xing et al.,
2012; Bondarenko et al., 2013). Thus, the application of the CA has proven its consistency and reliability in palaeoclimatic reconstructions and its
potential to provide important and reproducible results.
The collection of extensive CA-based palaeoclimate data derived
from the palaeobotanical record has been made possible due to the international research network NECLIME (Neogene Climate Evolution in
Eurasia) which was established in 1999. The main objectives of
NECLIME are: (1) the quantitative reconstruction of the Neogene climate evolution in Eurasia and its patterns in time and space based on
proxy data and their quantitative climatic interpretation by means of
standardised techniques, (2) the reconstruction of Neogene regional
and global atmospheric circulation patterns via climate modelling,
(3) the analysis of the interaction between palaeogeography, vegetation
and climate. More details about the concept, structure and members of
NECLIME can be found on the NECLIME website (www.neclime.de).
More than 15 years have passed since the original publication of the
Coexistence Approach by Mosbrugger and Utescher (1997); therefore
now is an appropriate time to summarise experience, discuss the methodology accompanying its use, and to revisit the method in the light of
new technological developments. Finally, we draw conclusions about
the future of the Coexistence Approach as a method to reconstruct
palaeoclimate.
its nearest living relative.
(3) The climatic requirements or tolerances of a nearest living relative can be derived from its area of distribution.
(4) The modern climatic data used are reliable and of good quality.
Accepting those assumptions, quantitative climatic results for a fossil
flora can easily be reconstructed by the CA according to the following
steps:
(i) For each fossil taxon, the NLR is determined.
(ii) For each NLR the modern distribution area is compiled.
(iii) From each distribution area the range of single climate parameters is determined separately.
(iv) For each climate parameter analysed, the climatic ranges, in
which a maximum number of NLRs of a given fossil flora can coexist (i.e. the coexistence interval), is determined independently
and considered the best description of the palaeoclimatic situation in which the given fossil flora once lived (Fig. 2A).
The application of the classical CA is facilitated by the Palaeoflora
database (www.palaeoflora.de) which compiles information for steps
(i) to (iii) as derived through ongoing analyses and contains climatic
information about more than 1600 mostly European and Asian plant
taxa. Their climatic requirements are derived from meteorological stations within the respective distribution area (see Section 4.2). Furthermore, the computer program ClimStat (CLIMate STATistic tool; noncommercial java application by A. Heinemann) provides an easy tool
for generating the coexistence intervals in step (iv). The database provides climate data for seven different climatic parameters, including
the mean annual temperature (MAT), coldest month mean temperature
2. The Coexistence Approach: the basics
The Coexistence Approach (CA) by Mosbrugger and Utescher (1997)
is a nearest living relative method, which relies only on the presence/
absence of a plant taxon within a fossil assemblage and the climatic requirements of its modern relatives. It avoids any statistical processing or
further assumptions, except those given in Mosbrugger and Utescher
(1997), which read as follows:
(1) For fossil taxa systematically closely related nearest living relatives (NLRs) can be identified.
(2) The climatic requirements of a fossil taxon are similar to those of
Fig. 2. Case studies in the reconstruction of the coexistence intervals (dashed lines) using
hypothetic MAT ranges for NLR taxa A through E″. A: all taxa coexist; B: the majority of
taxa coexist, taxon E′ is identified as outlier; C: an ambiguous solution is obtained.
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
(CMT), warmest month mean temperature (WMT), mean annual precipitation (MAP), mean precipitation of the wettest month (MPwet),
mean precipitation of the driest month (MPdry) and mean precipitation
of the warmest month (MPwarm).
Because the Coexistence Approach relies only on the presence or absences of taxa and not on their abundances it is largely independent of
sampling size or sampling intensity and can be applied on all plant fossil
organs, and where the assemblage is diverse, the CA is robust to taphonomic filtering. Each taxon (even each specimen) provides its climatic
information for the analysis, independent of the occurrence of other
taxa or other specimens. In extremis, a single (reliably determined) fossil can provide valuable palaeoclimatic information. However, climatic
intervals of a single taxon usually are wide and therefore do not offer
useful precision. A higher number of taxa – the consideration of all
taxa of a given flora – can usually yield narrower intervals (higher resolution) and more precise results. Although the CA can work with at least
two known NLRs with climate data, only the analysis of a high number
of taxa has the potential to detect errors, or evolutionary drift within the
dataset, or evolutionary drift of a taxon regarding its environmental tolerances, and thus increases the reliability of reconstruction. Generally, a
fossil flora should at least have ten climatically significant taxa, i.e. with
NLRs with available climate data (Mosbrugger and Utescher, 1997). In
single cases, a lower number of taxa contributing with climate data
can be admitted, depending on the problem and resolution required.
There is no definite way to identify the accuracy of the method.
Accuracy is likely to vary according to various factors such as regional
and stratigraphical representativeness of the considered flora, diversity,
floristic composition, taxonomic level, and quality of climate data.
Accuracy also varies with respect to the parameter examined (see
Section 4.2). A high diversity of the analysed record generally enhances
the climatic resolution of CA reconstructions and, as a function of the
grade of overlapping, accounts for a higher statistical significance of
the results (Mosbrugger and Utescher, 1997). The quality of the primary
climate data defining the boundaries of resulting coexistence intervals
likewise plays an important role for estimating the accuracy of the calculated climate data. Therefore, these data should be verified, especially
in the case of non-overlapping and if ambiguous solutions are obtained
(see Sections 3.1 and 3.3). Short of inventing a time machine, accuracy
of the CA can only be assessed using consilience between multiple
palaeoclimate proxies, and this applies to all palaeoclimate proxies.
Even if the CA potentially offered perfect accuracy, each of the analytical steps includes the potential for error. Moreover the basic assumptions needed for the CA, such as the identification of proper NLRs (see
Section 3.3), the reliability of the climate data set used (see Section 4),
or the applicability of the method in general (see Section 6) needs examining. These issues are discussed in the following sections.
3. Determination of nearest living relatives
Being a taxonomic approach, the CA is closely connected to the quality
of determination of the fossils and the reliability of recognising their NLRs.
Those are all general tasks and problems in palaeobotany and are therefore not confined to the application of the CA. However, the level of determination of each taxon as well as the assignment of one (or several) NLRs
to a fossil taxon considerably impacts the CA results. The identification of
palaeobotanical remains and their NLRs depends on organ type and preservation status. Depending on morphological traits displayed by the fossil
taxon, and closeness of phylogenetic relations to modern species the
taxonomic level of NLR identification varies between species, genera or
family levels.
Considering fossil pollen grains, it has been notoriously difficult for a
long time, to trace their botanical affinities. Nowadays, however, many
problems can be solved due to the combined use of the light and scanning electron microscopy (TLM/SEM) (e.g., Ferguson et al., 2007). In addition, the introduction of a standardised terminology of recent and
fossil pollen micromorphology by Punt et al. (2007) and Hesse et al.
61
(2009) has improved the possibility to compare and correctly identify
the NLRs. Moreover, several important pollen atlases were published,
e.g. the Atlas of Pollen and spores of the Polish Neogene (Stuchlik,
2001, 2002, 2009) for the Cenozoic of Central Europe, which provide
not only highly valuable light and scanning electronic microscopic pictures for comparison, but also important remarks towards the pollen
taxon's NLRs. Other examples can be found for lower vascular plants
(Grimsson et al., 2011), gymnosperms (Grimsson and Zetter, 2011;
Grimsson et al., 2011), and angiosperms (e.g., Quercus, Liu et al., 2007;
Lagerstroemia, Liu et al., 2008; Fagaceae, Denk et al., 2012).
With megafloral remains, major progress has been made in the identification of suitable NLRs within the past two decades of palaeobotanical
research by using classical techniques such as comparative morphology
and leaf anatomy (e.g., Mai and Walther, 2000; Kunzmann and Mai,
2005; Kovar-Eder et al., 2006; Erdei et al., 2007; Teodoridis et al., 2009;
Collinson et al., 2012; Kvaček and Walther, 2012). Moreover, the NLR
concept of taxonomic units has benefited from recent advances in the
study of morphotypes. In particular, the “whole plant approach” and
the introduction of complex taxa may prove beneficial for developing
sound NLR concepts and is, in addition, essential to unravel true
palaeobiodiversity. Examples for recently established complexes are
Craigia bronnii Ung. complex (Worobiec et al., 2010), and Reevesia
hurnikii Kvaček complex (Worobiec et al., 2012), allowing the attribution
of leaves, fruits and pollen to one single fossil species. Moreover, phylogenetic studies became increasingly important over the last decade
(e.g., Fagaceae, Manos et al., 2001; Denk et al., 2012; Cupressaceae,
Yang et al., 2012; Meliaceae, Müllner et al., 2010). The improved understanding of phylogenetic relations within plant families and genera may
facilitate the better identification of modern reference taxa or clades thus
rendering NLR concepts more reliable (cf. also reports of the NECLIME
working group on taxonomy of the Neogene macrobotanical records at
http://www.neclime.de).
Although these innovations are important, data are constantly being
improved and therefore the related data sets needed for the CA have to
be checked by the user before each application because a precise taxonomy is the first step in improving CA results and precision. As this
represents a crucial step in the CA, NECLIME members have set up
working groups organising data input from current taxonomical research. These are separated into a working group on ‘taxonomy of the
Neogene macrobotanical record of Eurasia’, ‘taxonomy of Neogene
palynomorphs’, and the ‘working group on fossil wood’. The main
focus of the working groups is to continuously update the Palaeoflora
database. This work encompasses (1) the taxonomical and climatic
(re-)evaluation of (critical) taxa, (2) systematic studies on families
and form taxa (e.g., due to new advances in molecular phylogenetics),
(3) changes in taxonomy and NLR concepts (e.g., due to new advances
in molecular phylogenetics), (4) the search for data on missing taxa,
and (5) considerations and re-evaluation of Neogene palynomorph records using TLM/SEM in combination. In this context the role of reference collections is essential. Collection of new materials and exchange
between existing collections underpins the sound identification of fossil
plant remains and related NLRs. Reports on the results of each of the
working group meetings and updates concerning the fossil taxa and
their NLR concept can be accessed at the NECLIME website (www.
neclime.de ‘groups’).
3.1. The concept and power of outliers
The Coexistence Approach determines the climatic interval in which
all fossil taxa could coexist (Fig. 2A). The result of the application is a coexistence interval, analysed separately for each climate parameter.
However, in some cases, not all intervals of individual taxa overlap the
resulting coexistence interval and this can be caused by various factors.
Plant taxa whose climatic requirements do not overlap with those
of the majority of taxa in a fossil flora are denoted as outliers
(cf. Mosbrugger and Utescher, 1997, p. 66, and Fig. 2B). The concept of
62
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
outliers, their presence and proportion constitutes a crucial aspect in CA
analyses because it provides an estimate regarding the significance and
reliability of the determination of the fossil taxon and hence its
palaeoclimatic reconstruction (cf. relative percentage of overlapping
taxa, climatic homogeneity of a fossil flora; Mosbrugger and Utescher,
1997). Moreover, it represents a particular strength of the CA due to
unravelling various sources of uncertainty.
Uncertainties in CA analysis can be referred to (1) an unclear/wrong
NLR concept for the fossil taxon, (2) differing climatic requirements of
fossil and NLRs (evolution), (3) lack of climatic representativeness of
the present-day plant distribution area (relict, recent evolution of competitor taxa and/or pathogens, unclear natural distribution due to early
cultivation or anthropogenic restriction), and (4) incomplete database
entries concerning climate data of NLRs (Mosbrugger and Utescher,
1997; cf. Section 4.2). Having excluded errors regarding point (4),
problematic palaeobotanical taxa regarding points (1) to (3) become
obvious when applying the CA to numerous fossil floras, because these
taxa frequently stand out in the analyses. These are denoted as permanent outliers (Mosbrugger and Utescher, 1997).
According to the above, the reasons for the permanent outlier status
of a plant taxon are its unclear NLR concept (1), differing climatic requirements of a fossil and its NLRs (2), and problematic climatic representativeness of the extant reference taxon (3) (cf. also ‘relics’ in
Section 3.2). While reason (1) might be avoided by new taxonomical/
phylogenetic evidence, those of (2) and (3) represent sources of possible error that cannot be overcome.
Our experience using the CA, has led to several new conclusions.
Once identified, permanent outliers should be excluded from CA analyses because they do not contribute reliable data. In theory, outliers
should not influence the CA results at all (cf. Mosbrugger and
Utescher, 1997, p. 73). Nevertheless, practice has shown that while outliers may not affect the whole palaeoclimatic reconstruction, they sometimes form consistent coexistence intervals for one climate parameter
only (e.g., permanent outlier regarding temperature variables/overlapping for precipitation). As a consequence the program ClimStat suggests
an erroneous climatic interval for the whole fossil flora and so biases
the reconstruction. To avoid this, the CA reconstruction of each climatic
parameter has to be determined in detail before the outlier is removed
from the analysis manually. The characteristics defining a “permanent climatic outlier” are not universal but related to the specific palaeoclimatic
situation existing in a given region at a given time; e.g., tropical outlier
such as Cedrela in a temperate climate (Mosbrugger et al., 2005) or
the dry adapted Tetraclinis in a wet climate (van Dam and Utescher,
submitted for publication). More examples of permanent outliers are
summarised in Table 1.
Despite the above, every study must consider each possible outlier
separately, and even then it cannot be guaranteed that an outlier is successfully identified in a CA application. However, the outlier concept
employed in the CA constitutes an interesting alternative to other concepts (cf. relict, false friends; Kvaček, 2007) because of the empirical
way in which ‘soft’ data are treated (climate requirements of extant
plants in a fossil world). Furthermore, although the identification of outliers related to the evolution of a single taxon appears to be highly problematic, in a few cases such as in the late Cenozoic adaption of Tetraclinis
and Sequoia to seasonally dry climates it visualizes the potential of
plants to adapt towards changing environmental factors. As regards
modern Sequoia, it grows under more humid conditions than it appears
from ‘classic’ precipitation values, owing to its adaption to absorb water
directly from clouds (Burgess and Dawson, 2004). Thus, the relevance of
‘unconventional’ environmental parameters in habitat preference of
taxa, such as fog occurrence, and their identification may also contribute
to the outlier status of a taxon. Likewise, the identification of outliers
represents a positive side-effect in the application of the CA to multiple
palaeofloras and may help to understand changes in the climatic adaptations of taxa throughout Earth history.
3.2. Relict taxa in CA analyses
One possible cause of low precision in the CA is the lack of regional
and/or climatic representativeness of the present day distribution of
NRLs. This is often linked to the concept of ‘relict taxa’. From the perspective of chorology of modern plants, the term is used for plants which
were abundant in a large area in geological times, but now only occur
at one or a few small areas. Relictual plant populations are remnants
‘left behind’ from larger populations of plants, which have managed to
survive despite changes in the environment. Usually it is considered
that the reduction of the area of distribution of relicts is a climatically
driven (e.g., the extinction of plant taxa in the course of the Pleistocene
glaciation events). However, other abiotic and biotic factors influence
plant distribution, e.g., palaeogeographic changes (mountain uplifting,
geographic isolation of populations, appearance/closure of land bridges,
edaphic changes, etc.), fast spread of herbivorous animals, viruses and
plant disease, competition between plants in an ecosystem, etc. In modern plant geography, many of these so-called relict taxa are also termed
Table 1
Examples for permanent outliers (observation based on CA analyses of ca. 600 records). The outlier status of the taxa refers to palaeofloras of the mid-latitudes of Europe only, and to the
stratigraphical time frame specified in the second column.
Taxon/NLR
Time frame
Type
Temperature
Sequoia, diverse morphospecies/S. sempervirens
Palaeogene
Monotypic, relic
Comptonia, diverse morphospecies/Comptonia peregrina
Palaeogene
Monotypic
Scindapsites, Urospathites, diverse morphospecies/
Scindapsus, Urospatha
Cassia, diverse morphospecies/Cassia
Neogene
Neogene
Unclear botanical affinity: NLR
Monsteroideae? (cf. Mai, 1995)
Unclear botanical affinity
Cedrela, diverse morphospecies/Cedrela
Neogene
In many cases uncertain botanical affinity
Cathaya, few morphospecies/Cathaya
late Neogene
Relic
Glyptostrobus, diverse morphospecies belonging to 2
ecotypes (Kvaček, 2007)/Glyptostrobus lineatus
Matudaea, few morphospecies/Matudaea trinervia
late Neogene
Later Palaeogene
Relic, natural distribution problematic
(cf. Kvaček, 2007)
Relic
MAT ca. 9–15 °C
Too cool
MAT ca. 3–15 °C
Too cool
MAT ca. 25–28 °C; 22–27 °C
Too warm
MAT ca. 17–28 °C
Too warm
MAT ca. 20–26 °C
Too warm
MAT ca. 17–22 °C
Too warm
MAT ca. 17–22 °C?
Too warm
MAT ca. 19–26 °C
Too warm
Tetraclinis, diverse morphospecies/Tetraclinis articulata
Cenozoic
Relic (cf. also Kvaček, 1986; Mai, 1995)
Sciadopitys tertiaria/Sciadopitys verticillata
Cenozoic, some regions
Relic
Precipitation
MAP ca. 250–500 mm
Too dry
MAP ca. 1300–6000 mm
Too wet
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
palaeoendemic (Ferguson et al., 1997). Their use in the CA will depend
on their modern diversity. For example, the genus Engelhardia is commonly found in the Neogene throughout the whole Northern Hemisphere. It disappeared from Europe during the earlier Pleistocene
glacial cycles (Bertini and Sadori, 2010) while it persisted in East and
South-East Asia. The seven extant species of Engelhardia widely cover
several climatic zones (tropical to warm temperate), and are competitive in a variety of phytocoenoses and environments. Therefore, we consider the group a reliable proxy in CA analysis. Another example is
Cathaya (argyrophylla), which is common in the fossil record (Liu and
Basinger, 2000), but today is represented by a single species, restricted
to the subtropical mountains of southern China (Mao, 1989; Ge et al.,
1998). In this case, the present distribution area does not necessarily reflect the natural (potential) distribution of this taxon, and thus does not
represent its natural (potential) climatic range. The climatic interval of
recent Cathaya would most likely obscure the CA palaeoclimatic reconstructions and should therefore be excluded from the analysis. On this
account, it is listed among the ‘difficult taxa’ in Table 2.
From the palaeobotanical perspective it has been stated that ‘most
plants with Neogene near-related ancestors are in fact relicts in the
present day flora’ (Kvaček, 2007). Cenozoic vegetation history involves
substantial floristic evolution and turn-overs, leading from a rather
63
simple pattern in the Palaeocene to more differentiated situation with
zonal biomes in the Neogene (e.g., Wolfe, 1985; Akhmetiev, 1987).
This more complex biome heterogeneity reflects the diversification of
the climate space caused by the steepening of spatial gradients along
the Cenozoic Cooling (e.g., Bruch et al., 2011; Utescher et al., 2011),
leading to greater potential for evolutionary novelties. In terms of evolution of plant families and genera, from the extinction of Cretaceous elements in the earlier Palaeogene to the radiation of modern genera such
as grasses in the later Neogene, the Cenozoic has a long history of development centres, but also of relicts and refuge areas. An impressive outline of this topic and the main theories behind it (static vs. dynamic
geoflora concept) are given in Mai (1995; Section 4). One major driving
factor for these substantial changes was certainly climate evolution during the Cenozoic. So-called “palaeotropical” vegetation existed under
the equable climate conditions of the early Palaeogene greenhouse
world and subsequently broke up along with the evolving differentiation of climate and the steepening of global gradients (i.e. the equator
to pole temperature gradient) as well as more local gradients (e.g.
breakup of the Turgay flora due to continental drying). Later on, this
vegetation type was partly replaced by deciduous forest biomes mainly
composed of ‘arctotertiary’ taxa. The deciduous forest biomes had originated in the Arctic (e.g. partly extinct genera belonging to the
Table 2
Relic taxa considered to bias CA reconstruction (Kvaček, 2007). Climate data estimated from vegetation (Kvaček, 2007) and Palaeoflora data used in CA reconstructions. Highlighted data
show no overlapping.
Fossil taxon
NLRs
Climate data estimated from vegetation (Kvaček, 2007)
Palaeoflora climate data used in CA application
Glyptostrobus europaeus,
brevisiliquata
Glyptostrobus lineatus
MAT 9.1–25 °C CMT −2.7–19.8 °C
(use climate data for Taxodioideae)
Engelhardia, diverse
morphospecies
Engelhardia
Craigia (complex taxon)
Craigia
Tetraclinis, diverse
morphospecies
Tetraclinis articulata
Platanus, diverse
morphospecies
Platanus
Trigonobalanopsis
E.g. Trigonobalanus/Castanopsis
(T. rhamnoides)
Early Miocene
Zittau Basin — MAT 20 °C, CMT 10 °C; Köflach — MAT 14–17 °C;
Kymi/Aliveri—MAT 11–13 °C, CMT 6–10 °C;
Late Miocene
Vegora — MAT 10–15 °C CMT 0–4.5 °C
Moravská Nová Ves — MAT 12–16 °C
Pliocene
Gerstungen — MAT 13 °C, CMT 0–1 °C
Engelhardia sp. Middle Eocene
Messel — MAT 25–30 °C, CMT 10 °C
Engelhardia orsbergensis Oligocene
Haselbach — MAT 10–16 °C, CMT 2−4 OC
Early–Middle Miocene
Wackersdorf — MAT 14–18 °C, CMT 2–8 °C
Mydlovary — MAT 14–18 °C, CMT 0–5 °C
Early Oligocene:
Kundratice — MAT 15 °C, CMT 1 °C
Bechlejovice — MAT 15.3–16.6 °C; CMT 10.0–10.2 °C
Early Miocene
Cypris Shale — MAT 15 °C, CMT N1 °C
Pliocene:
Platana — MAT 13–15 °C, CMT 1–10 °C
Willershausen — CMT 0−5 OC
Late Eocene (see Mai, 1995):
Zeitz — MAT 15–20 °C, CMT 6–13 °C
Early Oligocene
Haselbach — MAT 10–15.5 °C, CMT 2–4 °C
Kundratice — MAT 15 °C, CMT N1 °C
Early to Middle Miocene
Wackersdorf — MAT 14–18 °C, CMT 2–8 °C
Kreuzau — MAT 11–16 °C, CMT 3 °C
Pliocene
Willershausen—CMMT 0–5 °C,; WMT 13–17 °C
Late Eocene
Zeitz — MAT 15–20 °C, CMT 6–13 °C
Early Oligocene
Haselbach — MAT 10–15 °C, CMT 2–4 °C
Early Miocene
Ipolytarnóc — MAT 20 °C, CMT 1 °C
Late Miocene
Delureni — MAT 15 °C, CMT 2–4 °C
Late Eocene
Staré Sedlo — MAT 15–20 °C, CMT 6–13 °C
Late Oligocene
Early Miocene
Wackersdorf — MAT 14–18 °C, CMT 2–8 °C
MAT 13.8–27 °C
CMT 3.1–25 °C
MAT 15.6–22 °C
CMT 3.8–13.6 °C
MAT 15.6–19.9 °C
CMT 7.1–12.3 °C
MAT 6.6–27.4 °C
CMT −10.9–25.6 °C
MAT 9.3–27.7 °C
CMT −1.5–27.2 °C
64
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Cercidiphyllaceae, Hamamelidaceae, Platanaceae, Betulaceae,
Juglandaceae, Tiliaceae, Ulmaceae, Fagaceae, and deciduous conifers
such as Metasequoia and Glyptostrobus; Mai, 1995). A comparable evolutionary change is also evident in the southern hemisphere (e.g., continental drying, fragmentation/dislocation of continental area due to
plate tectonics). Taking these facts into account, it is obvious that
palaeotropical taxa progressively gain relict status throughout the Cenozoic, leading to 70 palaeotropic relict taxa in the Plio-/Pleistocene in
western Eurasia (Mai, 1995). The growing and shrinking of plant distribution areas over time contribute to increasing uncertainties and reduced preciseness of the CA further back in time.
In order to address the above aspects, a specific study about the use
of relicts in quantitative palaeoclimate reconstruction using taxonomical approaches (in this case the CA) was carried out by Kvaček (2007).
In that work the author discusses a selection of seven relict taxa
(cf. Table 2) which he refers to as ‘false friends’ because they appear to
bias the actual palaeoclimate reconstructions (Table 2). Next to their
relict character (e.g., Glyptostrobus), these partly have a complex or
unclear NLR concept as well (e.g., Trigonobalanoids). Moreover, the
author points out problems when attempting the reconstruction of
palaeorequirements for these selected taxa by employing climatic tolerances of extant reference phytocoenoses for the fossil communities in
which these plants thrived. The considerations presented in Kvaček
(2007) are very helpful for identifying fossil plant taxa potentially affecting CA reconstructions although a clear identification of ‘false friend’
taxa based solely on their relict status remains doubtful. On the other
hand, climate reconstructions (CA and physiognomic approaches such
as LMA and CLAMP) for fossil sites where ‘false friends’ are recorded
can reveal the climatic tolerances these plants had in the past.
In regard to practical consequences for CA applications, several taxa
considered as problematic by Kvaček (2007) also appear as permanent
outliers in CA analyses and are therefore listed in Table 2. Other taxa in
question overlap with the majority of NLRs identified for the fossil flora
or may even delimit the coexistence interval. In the latter case, the climatic range identified by the CA might not be very well constrained,
but is also not severely biased. As the application of the CA either identifies ‘false friends’ as outliers or as overall unproblematic, an accurate
and careful analysis should not lead to erroneous palaeoclimatic results.
distribution. Before beginning considerations described in the above
paragraph, the NLR concept should be evaluated critically (see also
Section 3.2). In fossil megafloras, identifications of NLRs at species
level may be problematic because the number of diagnostic characters
identifiable from the fossil materials usually is limited and only available
for a single organ. Also, many NRL concepts at the species level are often
debated vigorously. In palynology, the ‘type’ status of identification accommodates these uncertainties. Even with extant taxa, taxonomy
may no longer be valid (e.g., Taxodiaceae, Tiliaceae) or/and some of
the modern taxa cited as NLRs in palaeobotanical literature have an unclear taxonomical status and may be treated in a different way depending on the flora used (e.g., treated as synonym, subspecies, variety). All
these points may introduce uncertainties and bias distribution information from analogue or digital resources. In the application of the CA on
modern floras, multiple coexistence intervals should not occur.
When ambiguous solutions are obtained in CA analysis, all fossil taxa
involved, their NLRs and their associated climate data sets should be
carefully checked for errors or incompleteness before interpreting the
result in terms of possible taphonomic effects. However, a prerequisite
for a successful survey is the availability of sufficient precise information. This leads directly to the problem area that we term “climatic requirements of modern taxa”. This includes retrieval techniques and
types of available resources.
3.3. Ambiguous solutions
The first step to obtain data on climate space for NLRs is to define the
spatial distribution of each taxon of interest. The distribution of a plant
(as well as any organism) is influenced by numerous factors, such as
abiotic components (e.g., climate, soil chemistry and geographical barriers), as well as biotic factors (e.g., competition, symbioses, availability
of specific pollinators and anthropogenic disturbances, the distribution
(environmental tolerance) of pathogens). For some taxa, these may be
the limiting requirements, while the actual distribution of others is
mainly caused by their climatic preferences. By focusing on the climatic
distribution, various sources of uncertainty are evident due to the different quality and resolution of information.
Basically, primary data on the plant distribution mainly derives from
herbaria. Several platforms allow inspecting specimens and all accompanying information (e.g., Chinese Virtual Herbarium, Muséum National d'Histoire Naturelle in France), although the precision of the
geographic information available in each case is highly inconsistent.
These vary from GPS coordinates for single trees to locations given at
county level, which fail to incorporate boundary changes that have
taken place over time (e.g., for most of the older Chinese specimen records). Because the primary information is based on samples, it corresponds to plots, which then are extrapolated into distribution areas
with specific problems of their own, particularly in mountain areas
and regions with steep relief.
Considering these issues, information is limited at best to an estimate
of the real areal distribution including its uncertainties. Nevertheless,
mapping plant distributions is a classical method in botany and the
basis for a variety of modern applications, especially in climate–
Ambiguous solutions in the CA analysis are obtained when the algorithm programmed in the ClimStat program yields multiple coexistence
intervals. This situation is shown for a hypothetic flora in Fig. 2C. Unlike
in the outlier concept, where the climatic range of the outlier taxon does
not overlap with those of the majority of taxa in a fossil flora (see
Section 3.1), several comparable ranges of a given climate variable are
generated, each showing overlapping taxa, but no coexistence interval
includes a significant majority.
Multiple coexistence intervals detected in the CA analysis of a fossil
flora may be evidence of the presence of various phytocoenoses in a
taphoflora, each reflecting differing climate conditions. Since the source
of the components plays a crucial role this case, it is mainly confined to
carpo-, xylo-, and palynofloras. The occurrence of more than one interval in the CA analysis can point to long-distance transport, large size of
catchment area and/or considerable relief in the surroundings of the
studied site (e.g., Ivanov et al., 2002; Hoorn et al., 2012). Especially distinct geographical relief close to large water bodies forces the formation
of local microclimates (lee effects) and steep climatic gradients at a
small regional scale. Hence, multiple intervals in CA analyses may provide important clues concerning palaeogeography and taphonomy of a
site and should be carefully traced. Moreover, reworking processes,
occurring quite common in continental to shallow marine facies, have
to be considered as possible reason for multiple coexistence intervals.
Another common reason leading to ambiguous results are problems
with the NLR concept or incomplete data about the NRL's climatic
4. Determination of climatic requirements of modern plant taxa
One basic assumption (3) of the Coexistence Approach is that ‘The
climatic requirements or tolerances of a nearest living relative, and
thus of the fossil taxon, can be derived from its area of distribution’
(Mosbrugger and Utescher, 1997). This strongly depends on two
sources of information, on plant distribution and on modern climate
data. Both issues have their own inherent problems and potential
sources of errors related to their methodologies. In order to address
this specific problem, a working group on ‘Digital data on plant distribution’ has been initiated in NECLIME discussing new techniques and updates to improve the datasets underlying the CA.
4.1. Distribution information and maps
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
vegetation modelling (Thompson et al., 1999), as well as species distribution modelling (Peterson et al., 2011). Several sources of distribution
maps are available online, e.g., at The Global Biodiversity Information
Facility (GBIF) (http://www.gbif.org/), The Atlas of United States Trees
(http://esp.cr.usgs.gov/data/atlas/little/), and Flora Europaea (http://
www.luomus.fi/english/botany/afe/index.htm). Additionally, various
books give printed maps for plant distributions in specific regions like
China (Fang et al., 2011), SW Asia (e.g., Browicz and Zielinski, 1982,
1984), and the former Soviet Union (Sokolov et al., 1977, 1980, 1986).
NECLIME is compiling such information for relevant taxa at Chorotree
(www.chorotree.de). Still more difficult are taxa, whose distribution information is given only as descriptive data only (e.g., Flora of China
(http://www.efloras.org), Seed Plants of China (Wu and Ding, 1999),
Atlas of Woody Plants in China (Fang et al., 2009, 2011)). These unfortunately provide insufficient data to determine reliably plant spatial distributions and climatic requirements. Understanding this variability in
data quality and resolution is crucial for the successful application of
the CA.
precision of distribution information. Currently new databases are
under construction to compile requirements of modern taxa based on
gridded climate data for areas with low coverage of meteorological stations (e.g., for Africa: Bruch et al., 2012). However, in some regions it is
difficult to obtain station data at all, but in those regions gridded data
also offer the lowest accuracy for exactly the same reason.
Experience has shown that the accuracy of gridded climate data extracted for a given plant distribution area can be problematic, especially
if chorological data are incomplete or of low resolution. Contrary to the
deliberate choice of extreme meteorological stations, avoiding subareas
with extreme orographic structuring or highly patchy occurrence of a
plant taxon, the extraction of all gridded data within a distribution
area with ill-defined boundaries bears high uncertainties. Those can
be considered during the analysis by buffering the distribution data
40
30
4.2. Climate data: meteorological station data and gridded data
MAT
20
10
0
-10
-20
-40
-20
0
20
40
20
40
20
40
MAT_30s
40
30
MAT
20
10
0
-10
-20
-40
-20
0
MAT_25m
40
30
20
MAT
After the distribution area of a NLR is reliably defined, or defined
with known uncertainties, the climatic data within this spatial range
can be assessed. In the first publication of the Coexistence Approach
(Mosbrugger and Utescher, 1997), the climate observation data were
retrieved for six ‘extreme’ meteorological stations situated within the
distribution range of each NLR. Actually, the availability of digital station
data, e.g. the Global Climate Data Atlas (Müller and Hennings, 2000) facilitates the identification of station providing extremes for the respective variables. The development of geographic information systems
(GIS) in recent years allows for alternative methods, such as using
gridded global datasets (New et al., 2002; Hijmans et al., 2005). Climatological information for points or areas can thus be easily extracted
from such datasets using GIS software.
Issues with modern climate data are not specific to the CA but are
common to methods and applications that require calibration based
on observations of the modern world. Generally, point data from meteorological stations are similar to those obtained for those locations from
gridded datasets, as these are constrained by the actual data. They are,
however, rarely identical. Calculating the differences between mean annual temperature values of meteorological data from Müller and
Hennings (2000) with WORLDCLIM datasets (Hijmans et al., 2005) at
different resolutions shows that these can still differ by as much as
5 °C for the same location with the best results being at 2.5 arcmin resolution (Fig. 3; Table 3). Similar results have been obtained across three
different gridded data sets for New Zealand (Kennedy et al., 2014, unpublished data). The different WORLDCLIM data sets also show increased disagreement with increasing altitudes emphasizing the
problem of correctly representing climate in mountain regions as
discussed by Hijmans et al. (2005) (Fig. 4). Moreover, gridded datasets
from different sources can also vary considerably in detail due to different interpolation methods, especially for regions with low coverage
with meteorological stations like islands and high latitudes (Hijmans
et al., 2005). It is worth mentioning that in topographically complex
areas a major problem is that aspect (slope angle and orientation) is
not incorporated into gridded data sets if the grid size, or the average
distance between the meteorological stations used in the interpolation,
are larger than the characteristic length of topographic granularity. This
can result in large temperature and humidity differences for the same
elevation. Station data, on the other hand, have inherent problems as
well. They invariably run for different time intervals, have gaps in the records, and have different instrument calibrations. This means that in
general, the uncertainties are difficult to quantify, unlike gridded data
which can be filtered for poor data more easily.
The decision as to which climate data source is chosen for the CA depends on various factors, for example the availability of data and the
65
10
0
-10
-20
-40
-20
0
MAT_10m
Fig. 3. Comparison of meteorological data from Müller and Hennings (2000) (covering
1961–1990) with gridded WORLDCLIM data (interpolations of observed data, representative of 1950–2000; Hijmans et al., 2005) in different resolution, i.e., 10 arcmin, 2.5 arcmin,
and 30 arcsec.
66
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Table 3
Statistical characteristics of the differences between MAT from different sources. Preciseness is highest and scatter lowest for 30 arcsec (mean of differences).
ΔMAT [°C]
n
Correlation coefficient
Arithmetic mean
Standard deviation
Standard error
Minimum
Maximum
Meteo minus 10 arcmin
Meteo minus 2.5 arcmin
Meteo minus 30 arcsec
1088
1088
1088
0.9925
0.9962
0.9966
0.1797
−0.0108
−0.0288
1.1564
0.8186
0.7703
0.0351
0.0248
0.0234
−8.7
−5.9
−7.9
7.8
5.2
5.6
(according to its resolution) and by reducing the data set to the altitudinal range of the given taxon if known in enough detail (Bruch et al.,
2012). We recognise that selection, educated or otherwise, of specific
station data to define a climatic envelope for a taxon introduces an unwelcome element of subjectivity, but until adequate digital chorological
information for the required NLRs is available in a global context, and a
standardised algorithmic expression, corrected for altitude and aspect,
can be appointed for gridded data, the original approach of
Mosbrugger and Utescher (1997) retains considerable merit.
The figures given as coexistence interval boundaries stem directly
from the primary climate data sets without further processing. Thus, it
is obvious that these boundaries are of the same type and quality as
the modern climate data used to identify the requirements of a plant.
Based on the application on 357 Cenozoic floras with a mean diversity
of 29 taxa having climate data (Fig. 7; see also Appendix 1), the precision of the CA results, i.e. the width of the coexistence interval is determined as 2.1 °C at a mean (std. 1.35 °C) for MAT. Precipitation-related
parameters are in general more difficult to assess because of the high
local variability in the modern data sets; here, the achievable precision
is even more dependent on the quality of the available chorological information and primary climate data. Coexistence intervals b 1 °C for
MAT and b 100 mm for MAP, respectively, represent valid solutions in
the sense of the CA algorithm, but are usually considered to be beyond
the resolution of the primary data, even if high resolving chorological
and climatological sources are employed (see also Section 5).
In general, palaeoclimatic reconstructions using the CA should be
based on a single source of data consistently, i.e. a single clearly specified meteorological data set. It also has to be appreciated that the character of data resulting from CA analysis is the same as the character of
the input data. If meteorological station data are chosen the results represent a best estimate of the palaeoclimatic conditions as they would
have been measured by a ‘palaeo-station’. If global gridded data in
10 arcmin resolution are considered, the results represent values of
10 arcmin ‘paleo-grid cells’. Those values cannot be interpreted as
being equivalent to recent meteorological station data as they are interpolated and therefore generally smoothed.
Palaeoflora climate data are mainly based on meteorological stations,
which are carefully chosen in order to represent the climatic space of
a modern taxon with respect to the variables. As regards the selection
of suitable stations, we follow the procedure described in Mosbrugger
and Utescher (1997). Furthermore, climate datasets in Palaeoflora provide additional parameter ranges such as potential evaporation and relative humidity, although these entries are currently incomplete. Data
5. The database — Palaeoflora
Palaeoflora (Utescher and Mosbruggger, 2013) has been designed to
provide essential information about fossil taxa, their related NLRs, and
their climatic requirements required for the CA. The database was
established in 1990 and has been continually growing, as well as
being updated and corrected on a regular basis. Palaeoflora currently includes ca. 5800 macrobotanical and 2500 microbotanical taxa, and for
ca. 1630 modern taxonomical units (species, genera, families) for
which climate data are available. Palaeoflora is maintained in the context of NECLIME and hence profits immensely from co-operations within the network and the activities of the NECLIME working groups on
taxonomy of the macro- and microbotanical record (see Section 3 and
materials and reports available on the NECLIME website). Details
on data collected, literature resources and main contributors can
be accessed at the Palaeoflora website (http://www.palaeoflora.de).
The Palaeoflora website provides free access to fossil macro- and
microbotanical taxa stored in the database, corresponding NLRs, as
well as MAT requirements. Additional climatic parameters with respect
to two temperature variables (CMT, WMT), and four precipitation parameters (MAP, MPdry, MPwet, MPwarm) are provided on request.
Fig. 4. Divergence of meteorological data from Müller and Hennings (2000) (covering
1961–1990) and gridded WORLDCLIM data (interpolations of observed data, representative of 1950–2000; Hijmans et al., 2005) in different resolutions, i.e., 10 arcmin,
2.5 arcmin, and 30 arcsec, related to altitude (resolution 30 arcsec).
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
67
Table 4
Comparison of MAT ranges of 25 North American arboreal taxa according to the Palaeoflora data base and Thompson et al. (1999).
Modern taxon
Distribution area represented
by Palaeoflora MAT data;
(% of gridded cells)
MATmin
(Thompson
et al., 1999)
MATmax
(Thompson
et al., 1999)
MAT range
(Thompson
et al., 1999)
MATmin
(Palaeoflora)
MATmax
(Palaeoflora)
MAT range
(Palaeoflora)
Climate range represented
by Palaeoflora database;
% of range (Thompson et al.,
1999)
Acer pennsylvanicum
Acer rubrum
Acer saccharinum
Acer saccharum
Aesculus glabra
Aesculus octandra
Alnus rubra
Alnus rugosa
Alnus serrulata
Betula lenta
Betula nigra
Betula papyrifera
Carpnius caroliniana
Carya aquatic
Carya cordiformis
Carya ovata
Carya tomentosa
Castanea pumila
Celtis occidentalis
Cornus florida
Fraxinus americana
Juglans cinerea
Juglans nigra
Picea glauca
Picea mariana
91.55
88.93
99.15
84.23
83.77
96.43
96.80
98.70
99.57
98.81
99.94
99.99
96.75
92.55
98.47
99.43
98.61
92.88
99.91
89.64
96.28
98.83
99.51
96.40
91.78
−0.1
−1.1
1.3
−1.1
8.0
8.0
−3.7
−12.5
3.4
4.4
5.9
−12.2
2.5
13.7
3.6
3.6
5.1
8.0
1.8
5.5
0.1
1.4
5.9
−12.2
−12.4
15.7
23.8
20.0
15.8
20.3
15.5
15.1
12.1
20.9
16.3
20.3
11.4
28.2
23.0
20.1
22.4
21.9
21.5
18.7
22.0
21.7
17.0
21.4
8.6
10.7
18.4
24.9
18.7
16.9
12.3
7.5
18.8
24.6
17.5
11.9
14.4
23.6
25.7
9.3
16.5
18.8
16.8
13.5
16.3
16.5
21.6
15.6
15.5
20.8
23.1
2.8
3.4
3.4
3.4
9.9
9.9
−0.3
−8.9
5.4
5.7
6.6
−11.9
3.4
13.3
5.7
5.7
7.9
11.8
2.5
9.3
4.4
5.6
6.6
−8.9
−8.9
15.6
23.9
20.8
15.6
17.9
15.6
13.8
11.8
20.8
15.6
21.3
12.5
21.1
21.1
21.3
21.1
20.8
21.3
17.1
20.8
21.1
16.7
18.8
6.9
9.3
12.8
20.5
17.4
12.2
8.0
5.7
14.1
20.7
15.4
9.9
14.7
24.4
17.7
7.8
15.6
15.4
12.9
9.5
14.6
11.5
16.7
11.1
12.2
15.8
18.2
86.49
82.33
93.05
72.19
65.05
76.00
75.00
84.15
88.00
83.19
102.08
103.39
68.87
83.87
94.55
81.91
76.79
70.37
89.57
69.70
77.31
71.15
78.71
75.96
78.79
not freely accessible are usually provided in exchange for information
on fossil evidence and NLR assignments. This is useful to keep the NLR
concept of the database up to date and for detecting erroneous entries.
Unlike data retrieval in the early years of the CA, where a total of six
stations were used to describe the climatic range of a taxon, today a considerable number of stations can be effectively checked for extremes by
using GIS compatible files of the Müller and Hennings (2000) selection
of global station data, regional compilations (e.g., for China: State
Bureau of Meteorology, 2002), as well as the digital collection of various
chorological resources compiled in the Chorotree project (http://www.
chorotree.de). The use of station data accommodates the lack of digital
data on plant distribution, coarse raster data (e.g., Flora Europaea) and/
or incomplete point data (e.g., GBIF). Moreover, station data can be retrieved easily also for analogue plant distribution maps (e.g., Meusel
et al., 1965) and to some extent mere descriptions of plant distribution.
Being based on chorological information with heterogeneous data
sources that vary in quality and resolution it is difficult to quantify predictive uncertainty. A few degrees difference regarding temperature,
and some tens of millimetres with respect to annual precipitation can
be cited as a rough assessment (cf. also Mosbrugger and Utescher,
1997), and similar values can also be empirically obtained when tracing
and comparing grid or station data along the boundary of a distribution
area following a well-defined climatic gradient. Even with high resolution chorological data, greater precision can hardly ever be achieved
due to uncertainties in the meteorological observations, be they station
data or derived from gridded datasets. Nevertheless, temperature data
in the Palaeoflora data base are given with first decimals because they
represent the original values from the meteorological stations selected.
Thus, this makes it possible to relate the climatic limit of one plant directly to a specific meteorological station.
Recently, the accuracy of NLR climate data provided in the Palaeoflora
database was challenged (Grimm and Denk, 2012). Considering the
highly heterogeneous quality of chorological data alone, it seems almost
impossible to adduce evidence for a guaranteed standard. However,
probability considerations for a high number of taxa overlapping
(Mosbrugger and Utescher, 1997), successful tests with modern floras
(Mosbrugger and Utescher, 1997; Grimm and Denk, 2012, Fig. 7), and
numerous comparisons with other, complementary, proxy-data-based
climate reconstruction methods such as LMA and CLAMP (e.g., Uhl
et al., 2003; Roth-Nebelsick et al., 2004; Yang et al., 2007; Xia et al.,
2009; Jacques et al., 2011a,b; Xing et al., 2012; Bondarenko et al., 2013)
underline the suitability of Palaeoflora climate data for palaeoclimate reconstructions. Nevertheless, it has never been claimed that Palaeoflora
entries are perfect; the improvement of the climate data stored there is
an ongoing process supported by valuable comments of users, and
newly available chorological resources data are continually being added.
To demonstrate the nature of Palaeoflora climate data, a comparison
between data obtained from overlapping digital plant distribution and a
gridded climatology (Thompson et al., 1999) is shown for 25 important
arboreal North American plant taxa, frequently encountered in modern
surface samples (Whitmore et al., 2005) (Table 4; Fig. 5). As evident
from the plot, MAT data cited in the Palaeoflora comply at a mean for
only ca. 80% of the entire MAT ranges reported for the selected taxa by
Thompson et al. (1999), but reflect mean MAT conditions of 95% of
their distribution area according to Thompson et al. (1999) (measured
in number of grid cells). From a different perspective, plant occurrences
in only 5% of the distribution area grid cells account for the extension of
the climatic range of a plant by a mean of 15%. In extreme cases, 0.5% of
recorded grid cells cause widening of the MAT range by ca. 28% (Juglans
cinirea, cf. Fig. 5). Hence, Palaeoflora data may not include the extremes
of climatic ranges where the taxa exist close to their limit. On
the one hand, too narrow ranges may lead to non-overlapping in CA
analysis and thus to ambiguous solutions (cf. wrong climate data in
Section 3.3) making revision of the primary data necessary. On the
other hand, the preclusion of the extreme ends minimizes errors introduced by a rare climatic situation, such as specific local microclimatic
conditions, unusually high snow depths, stands that are not in equilibrium with climate, or the impact of poor climatic data in high altitudinal
areas. For the majority of NLR taxa required for Cenozoic palaeoclimate
reconstruction, available chorological information does not precisely resolve the extremes (cf. Section 4.1). Hence, the tendency of the
Palaeoflora database to reflect core ranges where taxa potentially are
encountered with a high probability seems to us as overall suitable solution for reconstructing regional climate of the past.
Apart from the above considerations about the climatic space of
plants represented by Palaeoflora data, it is quite clear that Palaeoflora
68
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
does contain erroneous entries in view of the heterogeneity of available
chorological information for extant plants alone (see Section 4.1). This
fact has been already emphasized in the original description of the CA
method and Palaeoflora database (Mosbrugger and Utescher, 1997;
p. 65). To make this transparent, users of the online version of the
Palaeoflora database are asked to report on data entries ‘needing
updating or correction’ (http://www.palaeoflora.de/home). Only with
such information from a group effort is it possible to continuously improve the quality of the database. Thus, input from Palaeoflora users
and our own revisions allow regular updates of Palaeoflora data entries.
Furthermore, the CA is not confined to the use of Palaeoflora data.
Any source for the climatic requirements for extant plants can be used
as shown in several studies applying the CA based on other data sets,
e.g., restricted to China (Zhao et al., 2004; Kou et al., 2006; Xia et al.,
2009) or gridded data (Bruch et al., 2012).
6. The principal applicability of the CA
It is largely accepted that, apart from other factors outlined in
Section 3.2, climate intensely affects the distribution of plants. Examples
for exceptionally sensitive variables are e.g., the length and temperatures of the growing season, water availability during germination, or
the number of days with ground frost or extreme cold. These are commonly denoted as bioclimatic variables (Hijmans et al., 2005), as they
directly influence the distribution of many plants and therefore are
used to model vegetation distribution (e.g., François et al., 2011). Moreover, climate variables can be tuned to address specific question such as
past monsoon intensity (Jacques et al., 2014). Also, plant species may be
more sensitive to the variability of climate over time and to climatic extremes than to climatological means (i.e., averages over 30 or 50 years).
A change in climate variability is expected to significantly impact vegetation (Reyer et al., 2013). For instance, Dury et al. (2011) demonstrated
using a dynamic vegetation model that the increase in climate interannual variability anticipated under global warming will have large impacts on the net primary productivity of European vegetation in the
future, especially in the Mediterranean area. To involve this effect in
palaeoclimate reconstruction time series instead of climatological
means could be used to identify the requirements of taxa. In principle,
all these variables can directly be assessed from the distribution area
of a modern plant taxon and thus can be used in CA climate reconstructions, provided that the information is available for all NLRs of the fossil
flora analysed.
Annual and monthly means of temperature and rainfall, as commonly reconstructed in CA applications, reflect the sensitivity of plants less
directly, but are important key variables as they can easily be compared
with respective climatic data obtained from other proxies such as stable
isotopes or with data obtained from palaeoclimate modelling. It is
beyond the scope of this paper to discuss dependence of plants on e.g.
MAT in more detail, but the high dependence becomes obvious when
plant distribution maps are overlain with MAT isotherms. Fig. 6 might
be suggestive for a rough impression of signals that plants can resolve
with respect to MAT. This plot shows MAT requirements for ca. 250
North American woody plants (data from Thompson et al., 1999) sorted
by their MATmin tolerances. It has to be pointed out, however, that not
all climatically compatible taxa shown in Fig. 6 overlap in natural vegetation (e.g. East/West Coast taxa). Of course, the relation depicted in
Fig. 6 shows an empiric pattern, and hence the effectiveness of other factors such as precipitation, soil, autoecology, and competition cannot be
deduced from this pattern alone. The numerous examples of CA applications outlined in Section 1 and CA data compiled in Fig. 7 document the
ability of plants to reflect different climatic states in general. CA studies
show that these states are not only consistent with reconstructions from
other proxies (Zachos et al., 2001, 2008) but also converge to modern
conditions when taking into account the results obtained from the
youngest floras (e.g., Mosbrugger et al., 2005; Bondarenko et al., 2013).
When discussing the principal applicability of the CA in palaeoclimate
reconstruction one should consider that climatic situations might have
existed in geological history that are not present on Earth nowadays,
so-called extinct or fossil climates, defined by a variable combination
plotting outside modern continental climate space. Extinct climates
Fig. 5. Representativity of Palaeoflora MAT data for 25 North American arboreal taxa. Dark blue bars show the percentage of Palaeoflora MAT range with respect to the spans cited in
Thompson et al. (1999). Light bars indicate the percentage of the plant distribution area corresponding with the Palaeoflora MAT range (in % of grid cells of the total area according to
Thompson et al., 1999).
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
may have occurred e.g. under differing continent configuration or higher
global mean temperature under a raised contents in atmospheric greenhouses gases. In general, such conditions may still be reconstructed with
methods calibrated on modern climate (Jacques et al., 2014). This is possible because today's ‘interglacial’ world has a wider climatic range compared to the equable conditions mainly prevailing during the Cenozoic
(Wing and Greenwood, 1993). However the possibility still exists that
past climates may include conditions not found in today's world, which
may account for the continued failure of climate models to reproduce
palaeoclimates consistent with a range of proxies despite decades of
model development (Valdes, 2000; Spicer et al., 2013).
When making CA reconstructions for climate variables close to their
extant limit, real values are expected to be close or equal to the upper
limit of the suggested coexistence interval. However, the fact that in
the CA each climate variable is considered separately allows for the reconstruction of a climatic situation characterised by a parameter combination not existing today. Examples for extinct climates proposed are
climates of Northern Germany during several Neogene time spans,
using a 6-dimensional climatic space for reconstruction (Utescher
et al., 2009a). These obviously are periods with very mild winters and
generally low seasonality as compared to today. This supports the assumption of the existence of Cenozoic, non-analogue climates as assumed before by various other authors (e.g., Mosbrugger et al., 2005;
Williams and Jackson, 2007; Böhme et al., 2008).
Other than these moderate parameter combinations unfamiliar in
the present day the existence of extinct climates substantially differing
from modern are as yet a purely theoretical concept and it is difficult
to see how they could be detected. Climate modelling would be one approach but models tend to be inherently conservative (Spicer et al.,
2008, 2013) and like the real world are constrained by the laws of energy conservation and motion. Any extinct climate would have been similarly constrained. However, being adjusted to optimally reproduce
present-day conditions models still fail to simulate characteristic patterns in past climate such as continental interior equability (Wing and
Greenwood, 1993; Spicer et al., 2008, 2013), or high latitude warmth
(e.g. Uhl et al., 2007a,b; Spicer and Herman, 2010) as suggested by a diversity of proxies, including those based on plants using both NLR and
physiognomic approaches.
Other than extinct climates, there is clear evidence for past time periods with substantially raised atmospheric CO2 (e.g., Pagani et al.,
2011). Affecting photosynthesis, the atmospheric level of this greenhouse gas potentially impacts the temperature and humidity demands
of plant taxa. It is known from CO2 enrichment experiments that
stomatal conductance decreases with increasing atmospheric CO2
(e.g., Leuning, 1995, and references therein) resulting in transpiration
loss. Hence, water demand of a plant is expected to be reduced at elevated CO2 (Bader et al., 2013). At the same time it was shown that some
40
30
20
10
69
Fig. 7. CA intervals for mean annual temperature as reconstructed for Cenozoic floras
using the CA (published and unpublished data, cf. Appendix 1), sorted by CA interval
means. Data cover the range from cool temperate to tropical. Pre-selection for Ntaxa contributing with climate data N10 and width of CA interval b= 5 °C.
plants respond to elevated CO2 levels with a lower freeze-tolerance
(Royer et al., 2002). According to this, the atmospheric CO2 level may
modify the relationship between climate (especially precipitation) and
the presence/absence of plant species. These effects are demonstrated
by modelling studies at the level of plant functional types (CARAIB dynamic vegetation model) by François et al. (2006), who obtained very
different biome reconstructions for the Late Miocene under two different CO2 forcings using the same climatic conditions, and by model projections of species distribution shifts in dependence of prescribed CO2
(Cheaib et al., 2012). Hence, it could be argued that the climatic requirements defined for the NLR taxa are only valid under atmospheric CO2
close to modern. However, the above observations regarding the effects
of CO2 are the result of short term experiments using Free-Air CO2 Enrichment (FACE) or carried out on juvenile trees in growth cabinets or
glass houses. Because plants have the ability to moderate growth and
architecture under different environmental constraints (e.g. stomatal
characteristics and leaf size are not only CO2 dependent but are also
a function of water availability, light environment and wind stress)
(e.g., Hikosaka et al., 2005; Taub, 2010), cell wall versus lumen size
in wood (e.g., Wiemann et al., 1998), and even differentially allocate
resources in the face of potential pathogen threat (e.g., Freeman
and Beattie, 2008) it is possible that CO2 responses observed in the
short term (b102 years) can only be extrapolated to longer timescales
(N 103 years) with great caution. If short term observations regarding
CO2 effects on temperature requirements of plants species do reflect
long term responses it is difficult to see why both taxon-based
methods like the CA and physiognomic methods such as CLAMP yield
palaeotemperature estimates similar to those obtained by isotopic
methods and do not show any systematic offset (Mosbrugger et al.,
2005). Thus it is assumed that errors in CA reconstructions introduced
by CO2, if any, are fairly minor and presumably impact only the reconstruction of precipitation, essentially in areas where water is the main
limiting factor, such as in sub-desertic areas or savannahs, i.e., where
the change in stomatal conductance induced by CO2 increase is expected to have the largest effect on the plant–water relationship.
7. Conclusions and outlook
0
0
50
100
150
200
-10
-20
Fig. 6. MAT ranges of ca. 250 North American woody plant taxa (data from Thompson
et al., 1999), sorted by minimum temperatures of occurrence.
With more than 15 years of experience of using the Coexistence Approach on Cenozoic micro- and macropalaeobotanical records, it is clear
that the method provides a robust palaeoclimatic proxy. Results are
consistent with respect to global climate and environment, and predominantly consistent with data obtained from a large variety of other
proxies for example isotope geochemistry, climate data based on the interpretation of small mammal communities or other independent
plant-based palaeoclimatic approaches. New technologies have allowed
70
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
the methodology to be tested and refined and have given new insights
in the value of palaeoclimatic reconstructions achieved with the CA.
Constant updates of the primary data applied in the CA, including taxonomy, knowledge of the most appropriate nearest living relative for each
fossil taxon, and collection of climate datasets outlining the climatic requirements of extant plant taxa are necessary and will always be work
in progress. Note, however, that like all proxies that require calibration
based on modern climate data the precision can only be as good as that
of the calibration. In this respect until gridded data sets can be improved
in respect of altitudinal and aspect variability it is unrealistic to make reconstructions using such proxies to a precision better than 1–2 °C.
We hope that this paper elucidates the complexity and variety of single factors that potentially affect CA reconstructions, as well as some
basic concepts that underlie the use of this method. We have discussed
common difficulties, such as the existence of outliers and ambiguous
resolution, which despite confounding the palaeoclimate estimates
can reveal information concerning taphonomic and taxonomic issues.
Additionally, we discussed the specific potential of the CA to better assess palaeoenvironmental conditions.
What do we want to pass on to colleagues who want to reconstruct
palaeoclimate using plant fossils? When reconstructing from organ
types with complementary methods, all available procedures should
be applied and critically discussed (CA, CLAMP, Wiemann et al., 1999).
In many cases, overlapping data are obtained (Uhl et al., 2003;
Roth-Nebelsick et al., 2004; Yang et al., 2007; Xia et al., 2009; Jacques
et al., 2011a,b; Xing et al., 2012; Bondarenko et al., 2013), in others a
systematic difference (Mosbrugger and Utescher, 1997) may be revealed that is well worth investigating further. At least for pollen and
carpofloras, the CA is the most widely used, significant and versatile
tool for pre-Holocene/Pleistocene climate reconstruction. The limited
take up of the CA in Holocene studies is because other related methods
such as direct transfer functions are available. Where possible proxy records from non-biological proxies should also be compared to those derived from plant fossils.
To increase the transparency of the CA use, we suggest including detailed information about the analysis in each publication (e.g. as Supplementary material, figures, etc.). This will not only give more details for
readers who are familiar with the method (e.g. information about climate data used, list of taxa included in the analysis, etc.), but also allows
others to understand the data and mechanisms behind the climatic intervals given (e.g. why are specific taxa excluded from the analysis).
This all contributes to reproducibility and allows more meaningful comparisons to be made with other proxies. In this way, new information
can quickly be integrated into the Palaeoflora database (e.g. Liu et al.,
2011; Quan et al., 2011, 2012; Bondarenko et al., 2013).
Finally, it has to be emphasized that the whole CA procedure, especially quality improvement of the primary data needed for a successful
reconstruction benefits immensely from open discussion and exchange
of ideas fostered by the NECLIME community. Within this framework,
open workshops are routinely organised, as well as large training field
courses for the methods are offered, in the use of palaeobotanical data
kindly provided by its members. Currently, a glossary for the NECLIME
website is prepared where all issues discussed here will be summarised
and freely available at the homepage. This glossary is meant as a tool for
future development and improvement of the CA methodology for
palaeoclimate quantification.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.palaeo.2014.05.031.
Acknowledgements
We would like to express our thanks to all colleagues who advanced
the Coexistence Approach with their research and by sharing their data.
We are grateful to Teresa Spicer for the workshop facilities kindly provided. We thank both reviewers for their most constructive comments
and helpful suggestions. This work is a contribution to NECLIME
(Neogene Climate Evolution of Eurasia).
References
Akgün, F., Kayseri, M.S., Akkiraz, M.S., 2007. Palaeoclimatic evolution and vegetational
changes during the Late Oligocene-Miocene period in the Western and Central
Anatolia (Turkey). Palaeogeogr. Palaeoclimatol. Palaeoecol. 253, 56–106.
Akhmetiev, M.A., 1987. Cenozoic floras. Salient features of the Cenozoic palaeofloristics.
Major phytochoria. In: Meyen, S.V. (Ed.), Fundamentals of Palaeobotany. LondonNew York, Chapman and Hall, London-New York, pp. 329–344.
Akkiraz, M.S., Akgün, F., Utescher, T., Bruch, A.A., Mosbrugger, V., 2011. Precipitation gradients during the Miocene in Western and Central Turkey as quantified from pollen
data. Palaeogeogr. Palaeoclimatol. Palaeoecol. 304, 276–290.
Bader, M.F.K., Leuzinger, S., Keel, S.G., Siegwolf, R.T.W., Hagedorn, F., Schleppi, P., Körner,
C., 2013. Central European hardwood trees in a high‐CO2 future: synthesis of an 8year forest canopy CO2 enrichment project. J. Ecol. 101, 1509–1519.
Bailey, I.W., Sinnott, E.W., 1915. A botanical index of Cretaceous and Tertiary climates.
Science 41, 831–834.
Bailey, I.W., Sinnott, E.W., 1916. The climatic distribution of certain types of angiosperm
leaves. Am. J. Bot. 3, 24–39.
Bertini, A., Sadori, L., 2010. Pliocene to Pleistocene palynoflora and vegetation in Italy:
state of the art. Quat. Int. 225, 1–4.
Böhme, M., Ilg, A., Winkelhofer, M., 2008. Late Miocene “washhouse” climate in Europe.
Earth Planet. Sci. Lett. 275, 393–401.
Bondarenko, O.V., Blochina, N.I., Utescher, T., 2013. Quantification of Calabrian climate in
southern Primory'e, Far East of Russia — an integrative case study using multiple
proxies. Palaeogeogr. Palaeoclimatol. Palaeoecol. 386, 445–458.
Browicz, K., Zielinski, J., 1982. Chorology of Trees and Shrubs in South-West Asia and Adjacent Regions, vol. I. Polish Academy of Sciences, Institute of Dendrology, Kórnik.
Browicz, K., Zielinski, J., 1984. Chorology of Trees and Shrubs in South-West Asia and Adjacent Regions, vol. IV. Polish Academy of Sciences, Institute of Dendrology, Kórnik.
Bruch, A.A., Kovar-Eder, J., 2003. Climatic evaluation of the flora from Oberdorf (Styria,
Austria, Early Miocene) based on the Coexistence Approach. Phytologia Balc. 9,
175–185.
Bruch, A.A., Utescher, T., Alcalde Olivares, C., Doláková, N., Ivanov, D.A., Mosbrugger, V.,
2004. Middle and Late Miocene spatial temperature patterns and gradients in
Europe — preliminary results based on palaeobotanical climate reconstructions.
Cour. Forschungsinstitut Senckenb. 249, 15–27.
Bruch, A.A., Utescher, T., Mosbrugger, V., Gabrielyan, I., Ivanov, D.A., 2006. Late Miocene
climate in the circum-Alpine realm — a quantitative analysis of terrestrial
palaeofloras. Palaeogeogr. Palaeoclimatol. Palaeoecol. 238, 270–280.
Bruch, A.A., Uhl, D., Mosbrugger, V., 2007. Miocene Climate in Europe — patterns and Evolution. A first synthesis of NECLIME. Palaeogeogr. Palaeoclimatol. Palaeoecol. 253,
1–7.
Bruch, A.A., Utescher, T., Mosbrugger, V., 2011. Precipitation patterns in the Miocene of
Central Europe and the development of continentality. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 304, 202–211.
Bruch, A.A., Sievers, C., Wadley, L., 2012. Quantification of climate and vegetation from
Southern African Middle Stone Age sites — an application using Late Pleistocene
plant material from Sibudu, South Africa. Quat. Sci. Rev. 47, 7–17.
Burgess, S.S.O., Dawson, T.E., 2004. The contribution of fog to the water relations of
Sequoia sempervirens (D. Don): foliar uptake and prevention of dehydration. Plant
Cell Environ. 27, 1023–1034.
Chaloner, W.G., Creber, G.T., 1990. Do fossil plants give a climatic signal? J. Geol. Soc. 147,
343–350.
Cheaib, A., Badeau, V., Boe, J., Chuine, I., Delire, C., Dufrêne, E., François, C., Gritti, E.S.,
Legay, M., Pagé, C., Thuiller, W., Viovy, N., Leadley, P., 2012. Climate change impacts
on tree ranges: model intercomparison facilitates understanding and quantification
of uncertainty. Ecol. Lett. 15, 533–644.
Collinson, M.E., Manchester, S.R., Wilde, V., 2012. Fossil fruits and seeds of the Middle
Eocene Messel biota. Abh. Senckenb. Ges. Naturforsch. 570, 1–251.
Denk, T., Grímsson, F., Zetter, R., 2012. Fagaceae from the Early Oligocene of Central
Europe: persisting New World and emerging Old World biogeographic links. Rev.
Palaeobot. Palynol. 169, 7–20.
Dury, M., Hambuckers, A., Warnant, P., Henrot, A., Favre, E., Ouberdous, M., François, L.,
2011. Responses of European forest ecosystems to 21st century climate: assessing
changes in interannual variability and fire intensity. Biogeosci. For. 4, 82–89.
Erdei, B., Hably, L., Kázmér, M., Utescher, T., Bruch, A.A., 2007. Neogene flora and vegetation development of the Pannonian domain in relation to palaeoclimate and
palaeogeography. Palaeogeogr. Palaeoclimatol. Palaeoecol. 253, 115–140.
Erdei, B., Utescher, T., Hably, L., Tamás, J., Roth-Nebelsick, A., Grein, M., 2012. Early Oligocene continental climate of the Palaeogene Basin (Hungary and Slovenia) and the
surrounding area. Turk. J. Earth Sci. 21, 153–186.
Eronen, J.T., Micheels, A., Utescher, T., 2011. A comparison of estimates of mean annual
precipitation from different proxies: a pilot study for the European Neogene. Evol.
Ecol. Res. 13, 851–867.
Fang, J., Wang, Z., Tang, Z., 2009. Atlas of Woody Plants in China. Vol. 1 and Index. Higher
Education Press, Beijing.
Fang, J., Wang, Z., Tang, Z., 2011. Atlas of Woody Plants in China. Vol. 2 and Index. Higher
Education Press, Beijing.
Fauquette, S., Guiot, J., Suc, J.-P., 1998. A method for climatic reconstruction of the Mediterranean Pliocene using pollen data. Palaeogeogr. Palaeoclimatol. Palaeoecol. 144,
183–201.
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Ferguson, D.K., Liu, Y.-S., Zetter, R., 1997. The paleoendemic plants of East Asia: evidence
from the fossil record for changing distribution patterns. In: Jablonski, N.G. (Ed.), The
Changing Face of East Asia during the Tertiary and Quaternary. Centre of Asian
Studies, The University of Hong Kong, Hong Kong, pp. 359–371.
Ferguson, D.K., Zetter, R., Paudayal, K.N., 2007. The need for the SEM in palaeopalynology.
C. R. Pale 6, 423–430.
François, L., Ghislain, M., Otto, D., Micheels, A., 2006. Late Miocene vegetation reconstruction with the CARAIB model. Palaeogeogr. Palaeoclimatol. Palaeoecol. 238, 302–320.
François, L., Utescher, T., Favre, E., Henrot, A.-J., Warnant, P., Micheels, A., Erdei, B., Suc, J.-P.,
Cheddadi, R., Mosbrugger, V., 2011. Modelling Late Miocene vegetation in Europe: results of the CARAIB model and comparison with palaeovegetation data. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 304, 359–378.
Freeman, B.C., Beattie, G.A., 2008. An overview of plant defenses against pathogens and
herbivores. The Plant Health Instructor,. http://dx.doi.org/10.1094/PHI-I-20080226-01.
Ge, S., Hong, D.-Y., Wang, H.-Q., Liu, Z.-Y., Zhang, C.-,.M., 1998. Population genetic structure and conservation of an endangered conifer, Cathaya argyrophylla (Pinaceae).
Int. J. Plant Sci. 159, 351–357.
Gebka, M., Mosbrugger, V., Schilling, H.-D., Utescher, T., 1999. Regional scale palaeoclimate
modelling on soft proxy-data basis — an example from the Upper Miocene of the
Lower Rhine Embayment. Palaeogeogr. Palaeoclimatol. Palaeoecol. 152, 225–258.
Greenwood, D., Moss, P., Rowett, A., Vadala, A., Keefe, R., 2003. Plant communities and climate change in southeastern Australia during the early Paleogene. Geol. Soc. Am.
Spec. Pap. 369, 365–380.
Greenwood, D.R., Archibald, S.B., Mathewes, R.W., Moss, P.T., 2005. Fossil biotas from the
Okanagan Highlands, southern British Columbia and northeastern Washington State:
climates and ecosystems across an Eocene landscape. Can. J. Earth Sci. 42, 167–185.
Grichuk, V.P., 1969. An attempt to reconstruct certain elements of the climate of the northern hemisphere in the Atlantic period of the Holocene. In: Neishtadt, M.I. (Ed.),
Golotsen. Izd-vo Nauka, Moscow, pp. 41–57.
Grimm, G.W., Denk, T., 2012. Reliability and resolution of the coexistence approach — a
revalidation using modern-day data. Rev. Palaeobot. Palynol. 172, 33–47.
Grimsson, F., Zetter, R., 2011. Combined LM and SEM study of the Middle Miocene
(Sarmatian) palynoflora from the Lavanttal Basin, Austria: part II. Pinophyta
(Cupressaceae, Pinaceae and Sciadopityaceae). Grana 50, 262–310.
Grimsson, F., Zetter, R., Baal, C., 2011. Combined LM and SEM study of the Middle Miocene
(Sarmatian) palynoflora from the Lavanttal Basin, Austria: part I. Bryophyta,
Lycopodiophyta, Pteridophyta, Ginkgophyta, and Gnetophyta. Grana 50, 102–128.
Harzhauser, M., Piller, W.E., Müllegger, S., Grunert, P., Micheels, A., 2011. Changing seasonality patterns in Central Europe from Miocene Climate Optimum to Miocene Climate
Transition deduced from the Crassostrea isotope archive. Glob. Planet. Chang. 76, 77–84.
Harzhauser, M., Mandic, O., Latal, C., Kern, A., 2012. Stable isotope composition of the Miocene Dinaride Lake System deduced from its endemic mollusc fauna. Hydrobiologia
682, 27–46.
Heer, O., 1855. Flora Tertiara Helvetica I. Wurster and Comp, Winterthur.
Heer, O., 1856. Flora Tertiara Helvetica II. Wurster and Comp, Winterthur.
Heer, O., 1859. Flora Tertiara Helvetica III. Wurster and Comp, Winterthur.
Herman, A.B., Spicer, R.A., 1997. New quantitative palaeoclimate data for the Late Cretaceous Arctic: evidence for a warm polar ocean. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 128, 227–251.
Hesse, M., Halbritter, H., Zetter, R., Weber, M., Buchner, R., Frosch-Radivo, A., Ulrich, S.,
2009. Pollen Terminology — An Illustrated Handbook. Springer, Wien.
Hickey, L.J., 1977. Stratigraphy and paleobotany of the Golden Valley Formation (early
Tertiary) of western North Dakota. Mem. Geol. Soc. Am 150, 1–181.
Hijmans, R.J., Cameron, S.E., Parra, J.L., Jones, P.G., Jarvis, A., 2005. Very high resolution interpolated climate surface for global land areas. Int. J. Climatol. 25, 1965–1978.
Hikosaka, K., Onoda, Y., Kinugasa, T., Nagashima, H., Anten, N.P.R., Hirose, T., 2005. Plant
responses to elevated CO2 concentration at different scales: leaf, whole plant, canopy,
and population. Ecol. Res. 20, 243–253.
Hintikka, V., 1963. Über das Großklima einiger Pflanzenareale in zwei
Klimakoordinatensystemen dargestellt. Ann. Bot. Soc. Zool. Bot. Fenn. 34, 1–64
(Vanamo).
Hoorn, C., Straathof, J., Abels, H.A., Xu, Y., Utescher, T., Dupont-Nivet, G., 2012. A late
Eocene palynological record of climate change and Tibetan Plateau uplift (Xining
Basin, China). Palaeogeogr. Palaeoclimatol. Palaeoecol. 344–345, 16–38.
Ivanov, D.A., Ashraf, A.R., Mosbrugger, V., Palamarev, E., 2002. Palynological evidence for
Miocene climate change in the Forecarpathian Basin (Central Paratethys, NW
Bulgaria). Palaeogeogr. Palaeoclimatol. Palaeoecol. 178, 19–37.
Ivanov, D., Utescher, T., Mosbrugger, V., Syabryaj, S., Djordjević-Milutinović, D.,
Molchanoff, S., 2011. Miocene vegetation and climate dynamics in Eastern and Central Paratethys (Southeastern Europe). Palaeogeogr. Palaeoclimatol. Palaeoecol. 304,
262–275.
Iversen, J., 1944. Viscum, Hedera and Ilex as climate indicators. Geol. Fören. Förh. Stockh.
66, 463–483.
Jacques, F.M.B., Su, T., Spicer, R.A., Xing, Y., Huang, Y.-J., Wang, W.-M., Zhou, Z.-K., 2011a.
Leaf physiognomy and climate: are monsoon systems different? Glob. Planet. Chang.
76, 56–62.
Jacques, F.M.B., Guo, S.-X., Su, T., Xing, Y.-W., Huang, Y.-J., Liu, Y.-S., Ferguson, D.K., Zhou,
Z.-K., 2011b. Quantitative reconstruction of the Late Miocene monsoon climates of
southwest China: a case study of the Lincang flora from Yunnan Province.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 304, 318–327.
Jacques, F.M.B., Shi, G.L., Li, H.M., Wang, W.M., 2014. An early-middle Eocene Antarctic
summer monsoon: evidence of ‘fossil climates’. Gondwana Res. 25, 1422–1428.
Kern, A., Harzhauser, M., Mandic, O., Roetzel, R., Ćorić, S., Bruch, A.A., Zuschin, M., 2011.
Millennial-scale vegetation dynamics in an estuary at the onset of the Miocene Climate Optimum. Palaeogeogr. Palaeoclimatol. Palaeoecol. 304, 247–261.
71
Kern, A.K., Harzhauser, M., Soliman, A., Piller, W.E., Gross, M., 2012. Precipitation driven
decadal scale decline and recovery of wetlands of Lake Pannon during the Tortonian.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 317–318, 1–12.
Kershaw, A.P., 1997. A bioclimatic analysis of Early to Middle Miocene brown coal floras,
Latrobe Valley, South-eastern Australia. Aust. J. Bot. 45, 373–387.
Kershaw, A.P., Nix, H.A., 1988. Quantitative palaeoclimatic estimates from pollen data
using bioclimatic profiles of extant taxa. J. Biogeogr. 15, 589–602.
Klotz, S., 1999. Neue Methoden der Klimarekonstruktion - angewendet auf quartäre
Pollensequenzen der französischen Alpen. Tübinger Mikropaläontologische Mitt. 21,
1–169.
Klotz, S., Fauquette, S., Combourieu-Nebout, N., Uhl, D., Suc, J.-P., Mosbrugger, V., 2006.
Seasonality intensification and long-term winter cooling as a part of the Late Pliocene
climate development. Earth Planet. Sci. Lett. 241, 174–187.
Kou, X.Y., Ferguson, D.K., Xu, J.X., Wang, Y.F., Li, C.S., 2006. The reconstruction of
paleovegetation and paleoclimate in the Late Pliocene of West Yunnan, China. Clim.
Chang. 77, 431–448.
Kovar-Eder, J., Kvaček, Z., Martinetto, E., Roiron, P., 2006. Late Miocene to Early Pliocene
vegetation of southern Europe (7–4 Ma) as reflected in the megafossil plant record.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 238, 321–339.
Kühl, N., Gebhardt, C., Litt, T., Hense, A., 2002. Probability density functions as botanical climatological transfer functions for climate reconstruction. Quat. Res. 58,
381–392.
Kunzmann, L., Mai, D.H., 2005. Die Koniferen der Mastixioideen-Flora von Wiesa bei
Kamenz (Sachsen, Miozän) unter besonderer Berücksichtigung der Nadelblätter.
Palaeontogr. Abt. B 272, 67–135.
Kvaček, Z., 1986. The fossil Tetraclinis Mast. (Cupressaceae). Časopis Národního Muzea
Praze 155, 45–54.
Kvaček, Z., 2007. Do extant nearest relatives of thermophile European Tertiary elements reliably reflect climatic signal? Palaeogeogr. Palaeoclimatol. Palaeoecol.
253, 32–40.
Kvaček, Z., Walther, H., 2012. European Tertiary Fagaceae with chinquapin-like foliage
and leaf epidermal characteristics. Feddes Repertorium 121, 248–267.
Larsson, L.M., Dybkjær, K., Rasmussen, E.S., Piasecki, S., Utescher, T., Vajda, V., 2011. Miocene climate evolution of northern Europe: a palynological investigation from
Denmark. Palaeogeogr. Palaeoclimatol. Palaeoecol. 309, 161–175.
Leuning, R., 1995. A critical appraisal of a combined stomatal-photosynthesis model for
C3 plants. Plant Cell Environ. 18, 339–355.
Liang, M., Bruch, A.A., Collinson, M.E., Mosbrugger, V., Li, C., Sun, Q., Hilton, J., 2003. Testing
the climatic signals from different palaeobotanical methods: an example from the
Middle Miocene Shanwang flora of China. Palaeogeogr. Palaeoclimatol. Palaeoecol.
198, 279–301.
Liu, Y.-S., Basinger, J.F., 2000. Fossil Cathaya (Pinaceae) pollen grains from the Canadian
High Arctic. Int. J. Plant Sci. 161 (5), 829–847.
Liu, Y.-S., Zetter, R., Ferguson, D.K., Mohr, B.A.R., 2007. Discriminating fossil evergreen and
deciduous Quercus pollen: a case study from the Miocene of eastern China. Rev.
Palaeobot. Palynol. 145, 289–303.
Liu, Y.-S., Zetter, R., Ferguson, D.K., Zou, C., 2008. Lagerstroemia (Lythraceae) pollen from
the Miocene of eastern China. Grana 47, 262–271.
Liu, Y.-S., Utescher, T., Zhou, Z.-K., Sun, B., 2011. The evolution of Miocene climates in
North China: preliminary results of quantitative reconstructions from plant fossil records. Palaeogeogr. Palaeoclimatol. Palaeoecol. 304, 308–317.
MacGinitie, H.D., 1941. A Middle Eocene Flora from the Central Sierra Nevada, 534.
Carnegie Institution of Washington Publication, pp. 1–94.
MacGinitie, H.D., 1969. The Eocene Green River flora of northwestern Colorado and northeastern Utah. Univ. Calif. Publ. Geol. Sci. 83, 1–203.
Mai, D.H., 1995. Tertiäre Vegetationsgeschichte Europas. Gustav Fischer, Jena.
Mai, D.H., Walther, H., 2000. Die Fundstellen eozäner Floren des Weißelster-Beckens und
seiner RandgebieteAltenburger Naturwiss. Forsch. 13, 1–59.
Manos, P.S., Zhou, Z.-K., Cannon, C.H., 2001. Systematics of Fagaceae: phylogenetic tests of
reproductive trait evolution. Int. J. Plant Sci. 162, 1361–1379.
Mao, Z.-Z., 1989. Cathaya argyrophylla, and endemic tree of China — its resources,
distribution and environment. Guihaia 9, 1–11 (in Chinese, English summary).
Meusel, H., Jäger, E., Weinert, E., 1965. Vergleichende Chorologie der zentraleuropäischen
Flora. Band 1 (Text und Karten)Fischer-Verlag, Jena.
Miao, Y., Herrmann, M., Wu, F., Yan, X., Yang, S., 2012. What controlled Mid-Late Miocene
long-term aridification in Central Asia? — global cooling or Tibetan Plateau uplift: a
reviewEarth Sci. Rev. 112, 155–172.
Micheels, A., Bruch, A.A., Uhl, D., Utescher, T., Mosbrugger, V., 2007. A Late Miocene
climate model simulation with ECHAM4/ML and its quantitative validation
with terrestrial proxy data. Palaeogeogr. Palaeoclimatol. Palaeoecol. 253,
251–270.
Micheels, A., Bruch, A.A., Mosbrugger, V., 2009. Miocene climate modeling sensitivity experiments for different CO2 concentrations. Palaeontol. Electron. 12 (2), 6A (http://
palaeo-electronica.org/2009_2/172/index.html).
Micheels, A., Bruch, A.A., Eronen, J., Fortelius, M., Harzhauser, M., Utescher, T., Mosbrugger,
V., 2011. Analysis of heat transport mechanisms from a Late Miocene model experiment with a fully-coupled atmosphere–ocean general circulation model.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 304, 337–350.
Mosbrugger, V., 1999. The nearest living relative method. In: Jones, T.P., Rowe, N.P. (Eds.),
Fossil Plants and Spores Modern Techniques. The Geological Society, London, pp.
261–265.
Mosbrugger, V., Utescher, T., 1997. The coexistence approach — a method for quantitative
reconstructions of Tertiary terrestrial palaeoclimate data using plant fossils.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 134, 61–86.
Mosbrugger, V., Utescher, T., Dilcher, D.L., 2005. Cenozoic continental climatic evolution of
Central Europe. Proc. Natl. Acad. Sci. 102 (42), 14964–14969.
72
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Müller, M.J., Hennings, D., 2000. The Global Climate Data Atlas on CD Rom. University
Flensburg, Institute für Geografie, Flensburg.
Müllner, A.N., Pennington, T.D., Koecke, A.V., Renner, S.S., 2010. Biogeography of Cedrela
(Meliaceae, Sapindales) in Central and South AmericaAm. J. Bot. 97, 511–518.
Needham, J., 1986. Science and Civilization in China: Volume 3, Mathematics and the
Sciences of the Heavens and the Earth. Caves Books Ltd., Taipei.
New, M., Lister, D., Hulme, M., Makin, I., 2002. A high-resolution data set of surface climate
over global land areasClim. Res. 21, 1–25.
Pagani, M., Huber, M., Liu, Z., Bohaty, S.M., Henderiks, J., Sijp, W., Krishnan, S., DeConto, R.M.,
2011. The role of carbon dioxide during the onset of Antarctic glaciation. Science 334,
1261–1264.
Parkhurst, D., Loucks, O., 1972. Optimal leaf size in relation to environment. J. Ecol. 60,
505–537.
Peterson, A.T., Soberón, J., Pearson, R.G., Anderson, R.P., Martínez-Meyer, E., Nakamura, M.,
Araújo, M.B., 2011. Ecological Niches and Geographic Distributions. Princeton University Press, Princeton, Princeton.
Poole, I., Cantrill, D., Utescher, T., 2005. A multi-proxy approach to determine Antarctic
terrestrial palaeoclimate during the Late Cretaceous and Early TertiaryPalaeogeogr.
Palaeoclimatol. Palaeoecol. 222, 95–121.
Popova, S., Utescher, T., Gromyko, D.V., Bruch, A.A., Mosbrugger, V., 2012. Palaeoclimate
evolution in the Cenozoic of Siberia — evidence from fruit and seed floras. Turk. J.
Earth Sci. 21, 315–334.
Pross, J., Bruch, A.A., Kvaček, Z., 1998. Paläoklima-Rekonstruktionen für den Mittleren
Rupelton (Unter-Oligozän) des Mainzer Beckens auf der Basis mikro-und
makrobotanischer Befunde. Mainz. Geowiss. Mitt. 27, 79–92.
Pross, J., Bruch, A.A., Mosbrugger, V., Kvaček, Z., 2001. Tertiary pollen and spores as a tool
for quantitative paleoclimate reconstructions: the Rupelian (Oligocene) of Central
Europe. Proceedings IX International Palynological Congress, Houston.
Pross, J., Contreras, L., Bijl, P.K., Greenwood, D.R., Bohaty, S.M., Schouten, S., Bendle, J.A.,
Röhl, U., Tauxe, L., Raine, J.I., Huck, C.E., van de Flierdt, T., Jamieson, S.S.R., Stickley,
C.E., van de Schootbrugge, B., Escutia, C., Brinkhuis, H., 2012. Integrated Ocean Drilling
Program Expedition 318 Scientists, 2012. Persistent near-tropical warmth on the Antarctic continent during the early Eocene epoch. Nature 488, 73–77.
Punt, W., Hoen, P.P., Blackmore, S., Nilsson, S., Le Thomas, A., 2007. Glossary of pollen and
spore terminology. Rev. Palaeobot. Palynol. 143, 1–81.
Quan, C., Liu, Y.C., Utescher, T., 2011. Paleogene evolution of precipitation in Northeastern
China supporting the Middle Eocene intensification of the East Asian Monsoon.
Palaios 26, 743–753.
Quan, C., Liu, Y.-S., Utescher, T., 2012. Paleogene temperature gradient, seasonal variation
and climate evolution of northeast China. Palaeogeogr. Palaeoclimatol. Palaeoecol.
313–314, 150–161.
Rasmussen, E.S., Utescher, T., Dybkjaer, K., 2013. Drowning of the Miocene Billund delta,
Jylland: land–sea fluctuations during a global warming event. Geol. Surv. Den. Greenl.
Bull. 28, 9–12.
Reyer, C., Leuzinger, S., Rammig, A., Wolf, A., Bartholomeus, R.P., Bonfante, A., De
Lorenzi, F., Dury, M., Gloning, P., Jaoudé, R.A., Klein, T., Kuster, T.M., Martins,
M., Niedrist, G., Riccardi, M., Wohlfahrt, G., de Angelis, P., de Dato, G.,
François, L., Menzel, A., Pereira, M., 2013. A plant's perspective of extremes:
terrestrial plant and ecosystem responses to changing climatic variability.
Glob. Chang. Biol. 19, 75–89.
Roth-Nebelsick, A., Utescher, T., Mosbrugger, V., Diester-Haass, L., Walther, H., 2004.
Changes in atmospheric CO2 concentrations and climate from the late Eocene to
early Miocene; palaeobotanical reconstruction based on fossil floras from Saxony,
Germany. Palaeogeogr. Palaeoclimatol. Palaeoecol. 205, 43–67.
Royer, D.L., Osborne, C.P., Beerling, D.J., 2002. High CO2 increases the freezing sensitivity of
plants; Implications for paleoclimatic reconstructions from fossil floras. Geology 30,
963–966.
Sokolov, S., Svjseva, O., Kubli, V., 1977. Ranges of Trees and Shrubs of the USSR, Volume 1.
Nauka, Leningrad.
Sokolov, S., Svjseva, O., Kubli, V., 1980. Ranges of Trees and Shrubs of the USSR, Volume
2Nauka, Leningrad.
Sokolov, S., Svjseva, O., Kubli, V., 1986. Ranges of Trees and Shrubs of the USSR, Volume
3Nauka, Leningrad.
Spicer, R.A., A.B., Herman, A.B., 2010. The Late Cretaceous environment of the Arctic: a
quantitative reassessment using plant fossilsPalaeogeogr. Palaeoclimatol. Palaeoecol.
295, 423–442.
Spicer, R.A., Ahlberg, A., Herman, A.B., Hofmann, C.-C., Raikevich, M., Valdes, P., Markwick,
P.J., 2008. The Late Cretaceous continental interior of Siberia: a challenge for climate
models. Earth Planet. Sci. Lett. 267, 228–235.
Spicer, R.A., Valdes, P.J., Spicer, T.E.V., Craggs, H.J., Srivastava, G., Mehrotra, R.C., Yang, J.,
2009. New development is CLAMP: calibration using global gridded meteorological
data. Palaeogeogr. Palaeoclimatol. Palaeoecol. 283, 91–98.
Spicer, R.A., Herman, A.B., Yang, J., Spicer, T.E.V., 2013. Why future climate change is
likely to be underestimated: evidence from Palaeobotany. J. Bot. Soc. Bengal 67,
75–88.
State Bureau of Meteorology, 2002. Climate Observation of China (1971–2000). Chinese
Meteorological Publishing House, Beijing.
Steppuhn, A., Micheels, A., Bruch, A.A., Uhl, D., Mosbrugger, V., 2007. The sensitivity of
ECHAM4/ML to a double CO2 scenario for the Late Miocene and the comparison to
terrestrial proxy dataGlob. Planet. Chang. 57, 189–212.
Stuchlik, L. (Ed.), 2001. Atlas of pollen and spores of the Polish Neogene, Spores, vol. 1.
Polish Academy of Sciences, W. Szafer Institute of Botany, Kraków.
Stuchlik, L. (Ed.), 2002. Atlas of pollen and spores of the Polish Neogene, Gymnosperms,
vol. 2. Polish Academy of Sciences, W. Szafer Institute of Botany, Kraków.
Stuchlik, L. (Ed.), 2009. Atlas of pollen and spores of the Polish Neogene, Angiosperms,
vol. 3. Polish Academy of Sciences, W. Szafer Institute of Botany, Kraków, p. 1.
Tang, H., Micheels, A., Eronen, J.T., Ahrens, B., Fortelius, M., 2013. Asynchronous responses
of East Asian and Indian summer monsoons to mountain uplift shown by regional climate modelling experiments. Clim. Dyn. 40, 1531–1549.
Taub, D., 2010. Effects of rising atmospheric concentrations of carbon dioxide on plants.
Nat. Educ. Knowl. 3, 21.
Teodoridis, V., Kvaček, Z., Uhl, D., 2009. Pliocene palaeoenvironment and correlation of
the Sessenheim–Auenheim floristic complex (Alsace, France). Palaeodiversity 2,
1–17.
Teodoridis, V., Mazouch, P., Spicer, R.A., Uhl, D., 2011. Refining CLAMP — investigations towards improving the Climate Leaf Analysis Multivariate Program. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 299, 39–48.
Terral, J.-F., Mengüal, X., 1999. Reconstruction of Holocene climate in southern France and
eastern Spain using quantitative anatomy of olive wood and archaeological charcoal.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 153, 71–92.
Thompson, R.S., Anderson, K.H., Bartlein, P.J., 1999. Atlas of relations between climatic
parameters and distributions of important trees and shrubs in North America. U.S.
Geological Survey Professional Paper 1650 A–B, (available online b http://pubs.
usgs.gov/pp/p1650-a/).
Tiffney, B.H., 2008. Phylogeography, fossils, and northern hemisphere biogeography: the
role of physiological uniformitarianism. Ann. Mo. Bot. Gard. 95, 135–143.
Uhl, D., Mosbrugger, V., Bruch, A.A., Utescher, T., 2003. Reconstructing palaeotemperatures
using leaf floras — case studies for a comparison of Leaf Margin Analysis and the Coexistence Approach. Rev. Palaeobot. Palynol. 126, 49–64.
Uhl, D., Bruch, A.A., Traiser, C., Klotz, S., 2006. Palaeoclimate estimates for the Middle Miocene Schrotzburg flora (S-Germany) — a multi-method approach. Int. J. Earth Sci. 95,
1071–1085.
Uhl, D., Klotz, S., Traiser, C., Thiel, C., Utescher, T., Kowalski, E., Dilcher, D.L., 2007a. Cenozoic paleotemperatures and leaf physiognomy — a European perspective.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 248, 24–31.
Uhl, D., Traiser, C., Griesser, U., Denk, T., 2007b. Fossil leaves as palaeoclimate proxies in
the Palaeogene of Spitsbergen (Svalbard). Acta Palaeobotanica 47, 89–107.
Utescher, T., Mosbruggger, V., 2013. The Palaeoflora database. Available online at http://
www.palaeoflora.de/.
Utescher, T., Gebka, M., Mosbrugger, V., Schilling, H.-D., Ashraf, A.R., 1997. Regional
palaeontological–meteorological palaeoclimate reconstruction of the Neogene
Lower Rhine Embayment. Proceedings 4th EPPC, Mededelingen Nederlands Instituut
voor Toegepaste Geowetenschappen TNO, 58, pp. 263–271.
Utescher, T., Mosbrugger, V., Ashraf, A.R., 2000. Terrestrial climate evolution in Northwest
Germany over the last 25 million years. Palaios 15, 430–449.
Utescher, T., Mosbrugger, V., Bruch, A.A., Milutinovic, D., 2007. Climate and vegetation
changes in Serbia during the last 30 Ma. Palaeogeogr. Palaeoclimatol. Palaeoecol.
253, 141–152.
Utescher, T., Mosbrugger, V., Ivanov, D., Dilcher, D.L., 2009a. Present-day
climatic equivalents of European Cenozoic climates. Earth Planet. Sci. Lett.
284, 544–552.
Utescher, T., Ivanov, D., Harzhauser, M., Bozukov, V., Ashraf, A.R., Rolf, C., Urbat, M.,
Mosbrugger, V., 2009b. Cyclic climate and vegetation change in the late Miocene of
Western Bulgaria. Palaeogeogr. Palaeoclimatol. Palaeoecol. 272, 99–114.
Utescher, T., Bruch, A.A., Micheels, A., Mosbrugger, V., Popova, S., 2011. Cenozoic climate
gradients in Eurasia — a palaeo-perspective on future climate change? Palaeogeogr.
Palaeoclimatol. Palaeoecol. 304, 351–358.
Utescher, T., Ashraf, A.R., Dreist, A., Dybkjær, K., Mosbrugger, V., Pross, J., Wilde, V., 2012.
Variability of Neogene continental climates in Northwest Europe — a detailed study
based on microfloras. Turk. J. Earth Sci. 21, 289–314.
Valdes, P.J., 2000. Warm Climate Forcing Mechanisms. Cambridge University Press, Cambridge, pp. 3–20.
van Dam, J., Utescher, T., submitted. Inter-calibrating paleoprecipitation proxies: comparing plant- and micromammal-based reconstructions. Palaeogeogr. Palaeoclimatol.
Palaeoecol.
Wheeler, E.A., Baas, P., 1993. The potentials and limitations of dicotyledonous wood anatomy for climatic reconstructions. Paleobiology 19, 487–498.
Whitmore, J., Gajewski, K., Sawada, M., Williams, J.W., Minckley, T., Shuman, B., Bartlein,
P.J., Webb III, T., Viau, A.E., Shafer, S., Anderson, P., Brubaker, L.B., 2005. A North
American modern pollen database for multi-scale Paleoecological and Paleoclimatic
applications. Quat. Sci. Rev. 24, 1828–1848.
Wiemann, M.C., Manchester, S.R., Dilcher, D.L., Hinojosa, L.F., Wheeler, E.A., 1998. Estimation of temperature and precipitation from morphological characters of dicotyledonous leaves. Am. J. Bot. 85, 1796–1802.
Wiemann, M.C., Mancheter, S.R., Wheeler, E.A., 1999. Paleotemperature estimation from
dicotyledonous wood anatomical characters. Palaios 14, 459–474.
Wiemann, M.C., Dilcher, D.L., Manchester, S.R., 2001. Estimation of mean annual temperature from leaf and wood physiognomy. For. Sci. 47, 141–149.
Wilf, P., 1997. When are leaves good thermometers? A new case for Leaf Margin Analysis.
Paleobiology 23, 373–390.
Wilf, P., 2000. Late Paleocene-early Eocene climate changes in southwestern Wyoming:
Paleobotanical analysis. Geol. Soc. Am. Bull. 112, 292–307.
Williams, J.W., Jackson, S.T., 2007. Novel climates, no-analog communities, and ecological
surprises. Fron. Ecol. Environ. 5, 475–482.
Wing, S.L., Greenwood, D.R., 1993. Fossils and fossil climate: the case for equable continental interiors in the Eocene. Philos. Trans. R. Soc. Lond. B 341, 243–252.
Wolfe, J.A., 1979. Temperature parameters of humid to mesic forests of Eastern Asia and
relation to forests of other regions of the Northern Hemisphere and Australasia: analysis of temperature data from more than 400 stations in Eastern Asia. U.S. Geol. Surv.
Prof. Pap. 1106, 1–37.
Wolfe, J.A., 1985. Distribution of major vegetational types during the Tertiary. Geophys.
Monogr. 32, 357–375.
T. Utescher et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 410 (2014) 58–73
Wolfe, J.A., 1993. A method of obtaining climatic parameters from leaf assemblages. U.S.
Geol. Surv. Bull. 2040, 1–71.
Wolfe, J.A., 1995. Paleoclimatic estimates from Tertiary leaf assemblages. Annu. Rev. Earth
Planet. Sci. 23, 119–142.
Worobiec, G., Worobiec, E., Kvaček, Z., 2010. Neogene leaf morphotaxa of Malvaceae s.l. in
Europe. Int. J. Plant Sci. 171, 892–914.
Worobiec, G., Worobiec, E., Szynkiewicz, A., 2012. Plant assemblage from the Upper Miocene deposits of the Bełchatów Lignite Mine (Central Poland). Acta Palaeobotanica
52, 369–413.
Wu, Z.Y., Ding, T.Y., 1999. Seed Plants of China. Yunnan Science and Technology Press,
Kunming, China, (distributed as CD; in Chinese).
Xia, K., Su, T., Liu, Y.S., Xing, Y., Zhou, Z.-K., 2009. Quantitative climate reconstructions of a
late Miocene Xiaolongtan megaflora from Yunnan, southwest China. Palaeogeogr.
Palaeoclimatol. Palaeoecol. 276, 80–86.
Xing, Y.-W., Utescher, T., Jacques, F.M.B., Tao, S., Liu, Y.-S., Huang, Y.-J., Zhou, Z.-K., 2012.
Palaeoclimatic estimation reveals a weak winter monsoon in southwestern China during the late Miocene: evidence from plant macrofossils. Palaeogeogr. Palaeoclimatol.
Palaeoecol. 358–360, 19–26.
73
Yang, J., Wang, Y.-F., Spicer, R.A., Mosbrugger, V., Li, C.-S., Sun, Q.-G., 2007. Climatic
reconstruction at the Miocene Shanwang basin, China, using leaf margin analysis, CLAMP, coexistence approach, and overlapping distribution analysis. Am. J.
Bot. 94, 599–608.
Yang, Z.-Y., Ran, J.-H., Wang, X.-Q., 2012. Three genome-based phylogeny of Cupressaceae
s.l.: further evidence for the evolution of gymnosperms and Southern Hemisphere
biogeography. Mol. Phylogenet. Evol. 64, 452–470.
Yao, Y.-F., Bruch, A.A., Cheng, Y.-M., Mosbrugger, V., Wang, Y.-F., Li, C.-S., 2012. Monsoon
versus Uplift in Southwestern China — Late Pliocene Climate in Yuanmou Basin,
Yunnan. PLoS ONE 7 (5), e37760. http://dx.doi.org/10.1371/journal.pone.0037760.
Zachos, J.C., Pagani, M., Stone, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and
aberrations in global climates 65 Ma to present. Science 292, 686–693.
Zachos, J.C., Dickens, G.R., Zeebe, R.E., 2008. An early Cenozoic perspective on greenhouse
warming and carbon-cycle dynamics. Nature 451, 279–283.
Zagwijn, W.H., 1996. An analysis of Eemian climate in Western and Central Europe. Quat.
Sci. Rev. 15, 451–469.
Zhao, L.-C., Wang, Y.-F., Liu, C.-J., Li, C.-S., 2004. Climatic implications of fruits and seed
assemblage from Miocene of Yunnan, southwestern China. Quat. Int. 117, 81–89.