High-resolution palaeoclimatology of the
last millennium: a review of current
status and future prospects
P.D. Jones,1
* K.R. Briffa,1
T.J. Osborn,1
J.M. Lough,2
T.D. van
Ommen,3
B.M. Vinther,4
J. Luterbacher,5
E.R. Wahl,6
F.W. Zwiers,7
M.E. Mann,8
G.A. Schmidt,9
C.M. Ammann,10
B.M. Buckley,11
K.M. Cobb,12
J. Esper,13
H. Goosse,14
N. Graham,15
E. Jansen,16
T. Kiefer,17
C. Kull,18
M. Küttel,5
E. Mosley-Thompson,19
J.T. Overpeck,20
N. Riedwyl,5
M. Schulz,21
A.W. Tudhope,22
R. Villalba,23
H. Wanner,5
E. Wolff24
and E. Xoplaki5
(1
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4
7TJ, UK; 2
Australian Institute of Marine Science, Townsville MC, QLD 4810, Australia; 3
Australian
Antarctic Division & ACE CRC, Private Bag 80, Hobart Tasmania 7001, Australia; 4
Centre for Ice
and Climate, Niels Bohr Institute, University of Copenhagen, Juliane Maries Vej 30, DK-2100
Copenhagen Ø, Denmark; 5
Oeschger Centre for Climate Change Research (OCCR) and NCCR
Climate and Institute of Geography, Climatology and Meteorology, University of Bern, Hallerstrasse
12, CH-3012 Bern, Switzerland; 6
Division of Environmental Studies and Geology, Alfred University,
NOAA-Paleoclimatology, Boulder CO 80305, USA; 7
Climate Research Division, Environment Canada,
4905 Dufferin Street, Toronto Ont. M3H 5T4, Canada; 8
Earth System Science Center, Department of
Meteorology, Pennsylvania State University, 523 Walker Building, University Park PA 16802, USA;
9
NASA Goddard Institute for Space Studies, 2880 Broadway, New York NY 10025, USA; 10
Climate &
Global Dynamics Division, NCAR, Boulder CO 80307-3000, USA; 11
Tree-Ring Laboratory, LamontDoherty
Earth Observatory, Palisades, New York NY 10964, USA; 12
School of Earth and Atmospheric
Sciences, Georgia Institute of Technology, 311 Ferst Drive, Atlanta GA 30332-0340, USA; 13
Swiss
Federal Research Institute WSL, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland; 14
Institut
dAstronomie et de Géophysique G. Lemaître, Université Catholique de Louvain, Chemin du cyclotron
2, 1348 Louvain-la-Neuve, Belgium; 15
Hydrologic Research Center, 12780 High Bluff Drive, ú, La
Jolla CA 92130-3017, USA; 16
Department of Geology, University of Bergen, Bjerknes Centre for
Climate Research, Allegaten 55, NO-5007 Bergen, Norway; 17
PAGES International Project Office,
Sulgeneckstrasse 38, 3007 Bern, Switzerland; 18
Advisory Body on Climate Change (OcCC),
Schwarztorstrasse 9, CH-3007 Bern, Switzerland; 19
Department of Geography and Byrd Polar
Research Center, Ohio State University, 108 Scott Hall, 1090 Carmack Road, Columbus OH 43210,
USA; 20
Institute for the Study of Planet Earth, University of Arizona, 715 N. Park Avenue, 2nd Floor,
Tucson AZ 85721, USA; 21
MARUM – Center for Marine Environmental Sciences and Faculty of
The Holocene 19,1 (2009) pp. 3–49
© 2009 SAGE Publications 10.1177/0959683608098952at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
4 The Holocene 19,1 (2009)
Geosciences, Universität Bremen, Postfach 330 440, D-28334 Bremen, Germany; 22
School of
Geosciences, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, UK; 23
Argentinean
Institute for Snow, Ice and Environmental Sciences, IANIGLA-CRICYT, Casilla de Correo 330,
Mendoza 5500, Argentina; 24
Physical Sciences Division, British Antarctic Survey, High Cross,
Madingley Road, Cambridge CB3 OET, UK)
Received 20 March 2008; revised manuscript accepted 29 August 2008
*Author for correspondence (e-mail: p.jones@uea.ac.uk)
Abstract: This review of late-Holocene palaeoclimatology represents the results from a PAGES/CLIVAR
Intersection Panel meeting that took place in June 2006. The review is in three parts: the principal high-resolution
proxy disciplines (trees, corals, ice cores and documentary evidence), emphasizing current issues in their
use for climate reconstruction; the various approaches that have been adopted to combine multiple climate
proxy records to provide estimates of past annual-to-decadal timescale Northern Hemisphere surface temperatures
and other climate variables, such as large-scale circulation indices; and the forcing histories used in climate
model simulations of the past millennium. We discuss the need to develop a framework through which
current and new approaches to interpreting these proxy data may be rigorously assessed using pseudo-proxies
derived from climate model runs, where the ‘answer’ is known. The article concludes with a list of recommendations.
First, more raw proxy data are required from the diverse disciplines and from more locations, as well
as replication, for all proxy sources, of the basic raw measurements to improve absolute dating, and to better
distinguish the proxy climate signal from noise. Second, more effort is required to improve the understanding
of what individual proxies respond to, supported by more site measurements and process studies. These activities
should also be mindful of the correlation structure of instrumental data, indicating which adjacent proxy
records ought to be in agreement and which not. Third, large-scale climate reconstructions should be attempted
using a wide variety of techniques, emphasizing those for which quantified errors can be estimated at specified
timescales. Fourth, a greater use of climate model simulations is needed to guide the choice of reconstruction
techniques (the pseudo-proxy concept) and possibly help determine where, given limited resources, future sampling
should be concentrated.
Key words: Palaeoclimatology, high-resolution, last millennium, tree rings, dendroclimatology, chronology,
uncertainty, corals, ice-cores, speleothems, documentary evidence, instrumental records, varves,
borehole temperature, marine sediments, composite plus scaling, CPS, climate field reconstruction,
CFR, pseudo-proxy approach, time series, climate forcing.
Introduction and rationale
In its Fourth Assessment Report (AR4), Working Group 1 of the
Intergovernmental Panel on Climate Change (IPCC, 2007) concluded,
with respect to the palaeoclimate record of the last two millennia
(Jansen et al., 2007), that: ‘Average Northern Hemisphere
temperatures during the second half of the 20th century were very
likely (> 90% certainty according to IPCC’s definition) higher than
during any other 50-year period in the last 500 years and likely
(> 66% certainty) the highest in at least the past 1300 years’. A
similar conclusion was also reached by the US National Academy
of Sciences (National Research Council (NRC), 2006). Study of
palaeoclimate of the last 1–2 millennia (late Holocene) has undergone
dramatic developments in the last 15 years. Up to the early
1990s, there was little cross-disciplinary work and few studies
attempted to bring together reconstructions from diverse proxies
except at the subcontinental scale. Although there had been earlier
qualitative attempts to compare different reconstructions (eg,
Williams and Wigley, 1983), the first quantitative extension of the
hemispheric-scale instrumental record was produced by Bradley
and Jones (1993). This decadal-average curve showed that since
about 1930 onwards, summer (June to August) Northern
Hemisphere (NH) temperatures were warmer than they had been
for any time since 1400. This work was reported in the Second
Assessment Report (SAR, IPCC, 1995), replacing the schematic,
conceptual temperature history that had been presented in the First
Assessment Report (IPCC, 1990: figure 7.1c – see Appendix A for
a discussion of the validity and likely source of this figure).
Although the early results (Bradley and Jones, 1993) were
received with considerable scientific interest, they were not allotted
much prominence by the SAR, and did not receive much public
attention. The availability of several independent or partially independent
annually resolved NH average temperature reconstructions
(Jones et al., 1998; Mann et al., 1998, 1999 – hereafter MBH98,
MBH99; Briffa, 2000; Crowley and Lowery, 2000), along with the
explicit representation of quantitative uncertainty estimates led to
the reconstructions having greater prominence in the Third
Assessment Report (TAR, IPCC, 2001). The MBH98 reconstruction
was also prominently featured in the associated Summary for
Policymakers (SPM). The lack of awareness of the 1993 paper and
the SAR is evident in many studies which often only contrast the
IPCC position in 1990 with that in 2001 (see also Appendix A),
though this may not be for purely scientific reasons. Subsequent
analyses since the TAR have not substantially changed the interpretation
of recent palaeoclimatic data and, if anything, the earlier conclusions
have been strengthened (SPM of IPCC, 2007). While the
visibility of large-scale (average and spatially detailed) reconstructions
stems from their ability to contextualize ‘unprecedented’ climate
change in the twentieth century against a multicentury
backdrop, such multiproxy reconstructions are critical to a variety
of climate science studies. They provide a large-scale context with
which to compare regional climate variability as reconstructed by
single proxy records, which may ultimately help resolve the largescale
mechanisms of past low-frequency climate change. They also
provide much-needed tests of the response of large-scale climate to
a variety of climate forcings which occurred during the last millennium
(most notably solar and volcanic forcing).
The purpose of this review is to consider what direction underlying
scientific investigations might most profitably take in the
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
immediate future, to reduce existing uncertainties, eg, those
inherent in the basic proxy data and the reconstruction of past
temperature variability, and to provide greater insight into the
factors governing past climate change. More work is needed to
better understand and more comprehensively quantify the sources
of uncertainty in various climate proxy archives and errors in
reconstructions, and then to improve them. However, we first
need to understand the reasons for the differences among existing
climate reconstructions which make use of different types or
combinations of climate proxy data and different statistical methods
to combine these data within a climate reconstruction. There
is, as always, a clear need for more local and regional reconstructions
from diverse proxies in as many parts of the world as
possible. Climate model simulations can also play an important
role here, acting as a surrogate climate, whose variability through
time is perfectly known, and with which we may test the performance
of alternative reconstruction approaches. Such tests,
however, are tied to the underlying assumptions regarding the
error structure of the climate proxy data.
To make significant progress there is a parallel need to improve
our understanding of the nature of the processes, both climatic and
non-climatic, that influence climate proxy data, and the need to
recognize and account for these intrinsic limitations. It is also
widely recognized that future work ought to extend beyond the
reconstruction of simple hemispheric average temperature series
and important large-scale circulation indices. Instead, further
investigations should also attempt to resolve seasonally specific
variations and identify the associated patterns of temperature, precipitation
and circulation variability, perhaps examining specific
anomalous periods in the past (selected from analysis of climate
proxy observations, estimated forcing histories or model simulation
results). This paper represents the outcomes of discussions at
Wengen, Switzerland in June 2006 under the auspices of
PAGES/CLIVAR held in an attempt to define how progress could
be made on these issues.
Background
The purpose of the Wengen workshop was to synthesize the current
state of late-Holocene climate reconstruction efforts, to assess
the approaches to data–model comparisons and to elaborate on
what possible or likely advances are expected in the coming years.
The workshop discussions were organized into three principal
subject areas:
(1) Proxy data availability and reliability
(2) Large-scale/regional reconstruction approaches and their
uncertainties, and how these can be informed by climate modelling
(from simple Energy Balance Models (EBMs) to fully
coupled Atmosphere/Ocean General Circulation models
(A/OGCMs), even including atmospheric chemistry)
(3) Factors that influence/force the climate system (both natural
and anthropogenic external factors as well as internal variability)
and how these are treated within climate models.
This review is an extensive and critical examination based on
these discussions and is organized in a similar framework. A brief
summary from the meeting has already been published (Mann
et al., 2006). The section ‘Proxy data uncertainty’ below discusses
aspects of the use of various types of proxy data, with a specific
focus on their current availability and uncertainty. Although structured
by discipline, each subsection addresses the most important
issues: reducing uncertainties (improving both reconstruction reliability
and, for less-than-annually resolved proxies, improving
dating accuracy and resolution) and making best and full use of
instrumental records for calibration and verification (particularly
important for decadally to annually resolved proxies) of reconstructed
climatic parameters. The issues discussed in this section
differ in focus and detail, in part reflecting the different maturity
of each discipline. Some weight and a corresponding amount of
text is allotted to the dendroclimatic issues. The justification for
this is the generally large proportion of tree-ring based proxies
used in many of the large-scale reconstructions, but also because
it is felt that much of the discussion on interpretational limitations
of these data has relevance for the increasing use of other forms of
high-resolution proxy data, particularly as they become available
over increasingly longer periods of time.
The section ‘Combining proxies to reconstruct large-scale patterns,
continental and hemispheric averages’ discusses the various
approaches that have been developed to combine currently available
proxy series into large-scale (continental to hemispheric)
averages and to reconstruct internally consistent climate fields,
representative of seasonal or annual conditions during the past.
This section also describes the application of experiments using
pseudo-proxies (derived from GCM output) to provide benchmark
tests for assessing the performance of statistical reconstruction
techniques. The following section, ‘Climate forcing and histories’
reviews the development of past climate forcing histories, discussing
which are the most important for millennial-scale climate
integrations, how different models implement the forcings, and
discusses uncertainties in forcing histories. The final section concludes
with a summary of major findings and a comprehensive set
of recommendations for future work.
Proxy data uncertainty
Tree rings and the need for improved regional
and temporal coverage
Tree-ring-derived records have played a prominent role in
attempts to establish how climate has varied in the recent past.
Networks of climatically sensitive tree-ring chronologies have
long been used to reconstruct detailed spatial patterns of interannual
climate variability on regional and near-hemispheric scales,
typically extending observed climate records by several centuries
(Fritts, 1991; Schweingruber et al., 1991; Briffa et al., 1994,
2002b; Cook et al., 2004b, 2007). Several chronologies extending
over a longer time span, with variability displaying a strong
and direct association with changing local temperatures, have
been utilized in virtually all published studies aimed at reconstructing
Northern Hemisphere (NH) or global average surface
temperature changes during the millennium leading up to the
present (Jansen et al., 2007).
We do not attempt here to discuss in detail the well-known positive
attributes of dendroclimatology. Reviews of the scope and
general strengths of different types of tree-ring data (that they are
continuous, precisely defined with annual resolution or better,
accurately dated on a calendar timescale, widely distributed and
rigorously calibrated against observed climate data) are already
well described in a number of books and general articles (Fritts,
1976b; Cook and Kairiukstis, 1990; Briffa, 1995; Schweingruber,
1996; Treydte et al., 2001, 2006; Hughes, 2002; McCarroll and
Loader, 2004; Luckman, 2007) and the published proceedings of
international conferences (eg, Hughes et al., 1982; Fritts and
Swetnam, 1989; Bartholin et al., 1992; Dean et al., 1996).
However, the continuing advancement of dendroclimatology as a
discipline is not based solely on the exploitation of these strengths.
It also involves an explicit appreciation of limitations or weaknesses.
Hence, our purposes in this review article are best served
by drawing attention to some of the general lessons learned in
tree-ring research, with a focus on the shortcomings of this proxy:
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 5
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
those that are innate characteristics of tree-ring data themselves,
but also those that may be identified in current dendroclimatic
practise (Hughes, 2002; Esper et al., 2007b). The issues discussed
do not relate solely to the most commonly used tree-ring ‘width’
data per se, but apply to all forms of tree-ring-derived proxy data,
including densitometric (eg, Schweingruber, 1996), chemical or
stable isotope (McCarroll and Loader, 2004; Treydte et al., 2007)
data. The following discussion draws attention to some selected
aspects of tree-ring data and dendroclimatology that are relevant
to the continuing efforts to better understand late-Holocene
climate variability. We first discuss the aspects of temporal and
spatial coverage of tree-ring data, with a focus on the status
of dendroclimatic studies in the tropics and the Southern
Hemisphere. We then describe a number of issues relating to
potential improvements in statistical methods used to assemble
long tree-ring chronologies and how their statistical quality might
be better represented. Finally, we draw attention to problems in
the way chronologies are generally interpreted in terms of specific
climate parameters. The points we raise about the need for
improved ways to measure chronology confidence and possible
limitations in the reconstruction of longer-timescale climate variability
are described at some length here because we believe them
to be equally relevant to the analysis and interpretation of other
proxy records discussed in the following sections of this review.
Knowledge of the expanding geographic coverage and recent
regional developments in dendroclimatology may be gleaned
from various reviews (Hughes et al., 1982; Dean et al., 1996;
Hughes, 2002; Luckman, 2007). Here we focus on the most
recent developments in those regions highlighted in the AR4, as
virtually devoid of tree-ring data, specifically the tropics and the
Southern Hemisphere (SH), though we include discussion of the
state of ‘long-chronology’ development in these and other
regions of the world.
Prospects for tropical dendroclimatology
Tropical dendrochronology was long considered impractical because
the growth periodicity of most tropical tree species is seldom clearly
and unambiguously defined (eg, Jacoby, 1989; Gourlay, 1995;
Vetter and Wimmer, 1999; Worbes, 1995). However, this has turned
out not to be entirely true, as seasonally dry regions have produced
tree-ring records from tropical Asia (eg, Berlage, 1931; Buckley
et al., 1995, 2005, 2007a, b; D’Arrigo et al., 1994, 1997, 2006a;
Pumijumnong et al., 1995; Sano et al., 2008), from Africa (eg, Stahle
et al., 1999; Tarhule and Hughes, 2002; Therrell et al., 2006) and
from the American tropics (eg, Biondi, 2001; Schöngart et al.,
2004a, b; Therrel et al., 2004; Biondi et al., 2005; Brienen and
Zuidema, 2005, 2006; D’Arrigo and Smerdon, 2008). Furthermore,
the notion that tropical regions that lack clear seasonality pose an
insurmountable problem has turned out not to be the case. Evans and
Schrag (2004), Poussart et al. (2004) and Poussart and Schrag (2005)
are among the first to have applied improved methods of stable isotope
geochemistry that show the possibility of using many apparently
‘ringless’ species for dendroclimatic studies in tropical
environments. Evans and Schrag (2004) demonstrated the application
of these methods on species from Costa Rica, while Poussart et
al. (2004) and Poussart and Schrag (2005) applied them to Thai and
Indonesian trees. In all cases clear periodicity in δ18
O and δ13
C was
demonstrated and determined to be annual, and relationships to rainfall
were established. Poussart et al. (2006) also applied x-ray microprobe
synchrotron analysis to a ringless species from Thailand, in an
attempt to more easily define annual periodicity through chemical
cycles of calcium, and this too holds promise for wider application.
In spite of the successes, however, formidable obstacles still
restrict the development of tropical dendroclimatology, including
the scarcity of suitable tree species with easily identifiable and
measurable annual rings, and continued pressure on forest resources
due to logging and disturbance that limits the availability of oldgrowth
trees. A veritable absence of information regarding the ecophysiology
and phenology of tree species further exacerbates
problems of working in the tropics (Borchert, 1995). Even with the
use of isotopic time series the problem of crossdating is still severe,
given the nature of locally absent banding that is often severe in
many tropical trees. Furthermore, the time and costs associated with
isotopic geochemistry currently inhibit widespread application.
Temporal control remains an issue with tropical tree-ring analyses
under many situations, and radiocarbon (14
C) measurements
have been used to assess the annual nature of growth rings
through detection of the radiocarbon ‘bomb spike’ (eg, Biondi and
Fessenden, 1999; Hua et al., 1999) and to estimate the age and average
growth rates of some tropical trees (eg, Poussart and Schrag,
2005). However, species-specific effects can limit the application of
14
C dating because of uncertainties associated with soil respiration,
internal carbohydrate transfer and other ecophysiological factors
(Worbes and Junk, 1989). Dendrometer studies have been
employed (eg, Buckley et al., 2001; DaSilva et al., 2002), and these
provide a useful alternative, along with cambium-wounding or ‘pinning’
methods that give a reference for growth from a time of
known scarring of the cambium (eg, Mariaux, 1967; Nobuchi et al.,
1995). Analysis of tropical wood chemistry may also reveal seasonal
signatures of cambium activity, as exemplified by Poussart et
al. (2004) in Indonesia, and Poussart and Schrag (2005) in Thailand,
both of whom demonstrate that the generation of replicated subannual
δ18
O and δ13
C records from ringless tropical trees is possible
over several decades.
More traditional approaches to dendroclimatology in the tropics
still have their place. Buckley et al. (2007b) produced an inferred
reconstruction of Palmer Drought Severity Index (PDSI), a measure
of soil moisture availability (Palmer, 1965), based on ringwidths
of teak from northwestern Thailand that extended back
nearly 500 years. This record illuminated periods of decadal-scale
drought, the most significant of which persisted from 1690 to
1720 and 1735 to 1765, respectively, and coincided with periods
of extreme social unrest. Sano et al. (2008) produced a 535-yr
ring-width record from living Fokienia hodginsii (Dunn, A. Henry
and H.H. Thomas) of the family Cupressaceae. The authors used
this record to reconstruct PDSI for the pre-monsoon period of
March to May over the past 500 years for northern Vietnam.
Significantly this is the first record from the region that successfully
calibrated and verified the reconstruction statistically, using
the available instrumental records. When compared with the
Buckley et al. (2007b) teak record from Thailand, a similar drought
from 1750 to 1780 is revealed, suggestive of a ‘megadrought’ that
may have extended from Burma to Vietnam. This period coincides
with the outright collapse of all the major kingdoms in Southeast
Asia (Leiberman, 2003), pointing to a possible direct climatic
impact on the societies of the region.
In some instances it has been shown that the response of tropical
tree species to climate is not as straightforward as for temperate
regions. Rivera et al. (2002) showed that for a number of
species the influence of subtle changes in photoperiodicity in the
American and Asian tropics is more important than rainfall for
producing annual flushing of leaves in shade-sensitive species.
Buckley et al. (2007a) illustrated an apparent inverse response to
drought in Pinus merkusii across southeast Asia, possibly linked
to reduction of photosynthetically active radiation during times of
intense convective rainfall. Saleska et al. (2007) found that
Amazon forests significantly increased their biomass in response
to a prolonged drought in 2005, counter to model predictions of
forest collapse in response to drought (eg, Betts et al., 2004). At
the peak of the 2005 Amazon drought, the authors found a significant
increase in canopy ‘greenness’, indicating an ecological and
physiological vegetation response that is opposite to a presump-
6 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
tion that anomalously low rainfall would negatively influence
these forests. Clark et al. (2003) found a negative relationship
between minimum temperature and tropical tree growth at La
Selva in Costa Rica, but no relationship with interannual variations
in precipitation (Clark and Clark, 1994). These findings have
implications for the interpretation of isotopic time series from
tropical trees, illustrating the need for research into the physiological
response of tropical forest species to climate, and a greater
understanding of their anatomy.
Furthermore, the research of both Feely et al. (2007) and Clark
et al. (2003) suggest a ‘deceleration’ in growth rates of tropical
forests in Costa Rica, Panama and Malaysia, in opposition to some
predictions that pan-tropical forests could experience increased
growth induced by CO2 fertilization or increases in water use efficiency
(eg, Lewis et al., 2006). There continues to be much debate
about the response of tropical forests to climate change and fundamental
questions remain about whether or not tropical forests serve
as carbon sink or source, and the extent to which any changes in
biomass can be attributed to anthropogenic change (eg, Wright,
2006). Dendrochronology can help to answer many of these questions
if it can be more widely applied, and the developments so far
are encouraging. As more species are analysed for their eco-physiological
characteristics and their response to climate, more oldgrowth
species are discovered and improvements are made to
geochemical methodologies, the prospects for widespread application
of dendrochronology in the tropics look promising.
Recent developments in South America and
southern Africa
For the interval 1890–2000, tree-ring width chronologies (from
Swietenia macrophylla and Cedrela odorata) were developed by
Dünisch et al. (2003) in Mato Grosso, Brazil (10°09′S, 59°26′W).
Correlation analyses revealed a significant relationship between
seasonal precipitation and the growth of both species. Several
month-long inundation indices influence the formation of annual
rings in trees growing in the seasonally flooded Amazon plains (3–
4°S, 65°W). Ring widths are inversely related to duration of the
flood. A 200-yr long Euphorbiaceae chronology (Piranhea trifoliata)
has been used to estimate the length of the vegetation season,
which is, in turn, related to ENSO events (Schöngart et al., 2004a,
b). The results indicate that during the last two centuries, the severity
of ENSO events in the Amazon basin has significantly increased.
In a related study, Schöngart et al. (2004b) developed tree-ring
chronologies from Macrolobium acaciifolium in two different
floodplains in Central Amazonia. Maximum tree age in the nutrient-poor
‘black-water’ was more than 500 years, contrary to the
nutrient-rich ‘white-water’ floodplain, where ages are not older
than 200 years. Ring-width variations in both floodplain forests
were significantly correlated with the length of the vegetation
period derived from the daily recorded water level at the port of
Manaus since 1903. Both chronologies showed increased wood
growth during El Niño events associated with negative precipitation
anomalies and lower water discharge in Amazonian rivers.
Exploratory work has established the dendrochronological
potential of several tropical lowland species in the Bolivian sector
of the Amazon basin (11–15°S, 66–68°W). Brienen and Zuidema
(2005) crossdated six rainforest species and established the influence
of annual and seasonal rainfall on radial growth. Ongoing
work in Argentina and Brazil has also identified several species
typical of the dry tropical forest of southern Bolivia–northern
Argentina (16–24°S) and southeastern Brazil (22–25°S) that show
clear rings and potential for dendroclimatic studies.
A major advance in the effort to expand the spatial coverage of
tree-ring records across the Americas has been the recent development
of Polylepis tarapacana chronologies in the Bolivian
Altiplano (Argollo et al., 2004). These records located between 16
and 22°S and above 4500 m elevation, are the closest to the
Equator in the Andes and the highest-elevation chronologies in the
world. Most Polylepis records cover the past three to four centuries,
but some extend over seven centuries (Argollo et al., 2004;
Soliz et al., 2008). Examination of interannual variations in ring
width and climate in the Altiplano indicate that the growth of
Polylepis is associated with summer water balance (Morales et al.,
2004). In addition, P. tarapacana chronologies from the southcentral
tropical Andes provide high-resolution records that are
extremely sensitive to ENSO in the tropical Pacific, and represent
an important component to be considered in future multiproxy
ENSO reconstructions (Christie et al., 2008).
A major requirement in Southern African dendroclimatology is
the need to examine the many tropical and subtropical species to
assess their dendrochronological potential (see the pioneer work
of Lilly, 1977 and February, 1996). Stahle et al. (1997) have
begun surveying the diverse indigenous forests in tropical Africa
for species suitable for dendroclimatology. Two tropical chronologies
of Pterocarpus angolensis from Zimbabwe are strongly correlated
with total rainfall amounts during the wet season. Both
chronologies reach back only to 1870 at present, but P. angolensis
is the most important timber species in south tropical Africa, and
old samples survive in buildings and other diverse sources that
may permit the eventual development of 200–300 yr chronologies
in southeastern Africa (Stahle et al., 1997).
Developing more ‘long’ tree-ring-based chronologies
There are very few tree-ring chronologies from around the globe
that extend back 1000 years. This is very apparent in figure 6.11
of Jansen et al. (2007), which shows only 16 locations globally
from which tree-ring data have been used in large-scale temperature
reconstructions to date. Of these, three are in the SH. The
remainder are virtually all restricted to the western edge of North
America or the high latitudes of Eurasia, with the exceptions being
two sites adjacent to the Mediterranean and one in Mongolia. The
contribution of dendroclimatology to improved late-Holocene climate
reconstruction must involve a geographic expansion of work
developing long composite chronologies. Many areas, some with
a proven history of (admittedly short-timescale) tree-ring research
and demonstrably climatically sensitive trees, have yet to be systematically
investigated for their potential to produce subfossil
wood and hence much longer chronologies. More long records are
needed in the northern mid latitudes though work is ongoing in
Europe (Nicolussi and Schiessling, 2001; Grabner et al., 2001;
Helama et al., 2005; Büntgen et al., 2006, 2008; Popa and Kern,
2008), N Africa (Esper et al., 2007a), N America (Barclay et al.,
1999; Buckley et al., 2004; Luckman and Wilson, 2005) and Asia
(Sidorova et al., 2006; Esper et al., 2007c).
Many more long chronologies are needed in the SH. Significant
recent achievements in South America include the development of a
5666-yr-long composite chronology from Fitzroya cupressoides,
currently the longest continuous chronology in the SH (WolodarskyFranke,
2002). Chronologies of this species (from Argentina and
Chile) were used in the 1990s to develop the first millennium-long
temperature reconstructions for South America. However, recent
work suggests a lack of stability in the relationship between the lowfrequency
component of climate and F. cupressoides tree growth,
particularly during recent decades. Studies using Austrocedrus
chilensis have developed several new records. A 1864-yr-long
chronology from northern Argentinean Patagonia is a major component
of a preliminary SH ENSO reconstruction for the last 1300
years and collaborative studies between tree-ring laboratories in
Chile, Argentina and USA have developed 800-yr-long precipitation
reconstructions for central Chile (LeQuesne et al., 2006, 2008).
Development of millennia-long tree-ring chronologies from
Australia and New Zealand has been based on three tree species:
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 7
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Lagarostrobos franklinii from Tasmania, Lagarostrobos colensoi
from the South Island of New Zealand and Agathis australis from
the North Island of New Zealand. These are very long-lived
species and provide abundant quantities of well-preserved wood.
The L. franklinii record from a high-elevation site on Mount Read,
Tasmania presently covers 4136 years (Cook et al., 2006). It contains
a reliable warm-season temperature signal and its variability
is significantly correlated with SSTs in the southern Indian Ocean.
The L. colensoi record comes from low-elevation sites on the west
coast of the South Island and is now 2327 years long. It too represents
warm-season temperatures and correlates well with Tasman
Sea SSTs. Finally, the A. australis record comes from mostly lowelevation
sites on the North Island and now spans the past 3722
years. It contains a robust ENSO signal, but has not yet been used
for climate reconstruction purposes (Cook et al., 2006).
Improved tree-chronology-production methods,
measuring chronology confidence and statistical
limitations
Despite the development of wider networks of various dendroclimatic
proxies it is important to appreciate the essential requirement
for continuing effort to improve the number and quality of current
dendroclimatic resources, even in areas that currently possess
widely utilized long chronologies. There remain major issues concerning
statistical and interpretational confidence associated with
building and interpreting existing long tree-ring chronologies.
‘Standardization’ of tree-ring data is required to remove biases in
chronologies that arise during periods where the make-up of samples
is dominated by a concentration of measurements from either
relatively young-age trees (that typically have wider rings and
higher density) or old-age trees (with narrow rings and lower density).
Approaches used previously to overcome these potential
biases tended to operate as high-pass filters of the data and resulting
chronologies contained little, if any, evidence of climate variation
on timescales longer than the average length of the sample
trees, perhaps only on centennial timescales or even less (Cook
et al., 1995; Briffa et al., 1996). In recent years new methods, or
new applications of old methods, used in different contexts, have
been proposed specifically for the processing of tree-ring data
intended for long-period climate reconstruction work (Erlandsson,
1936; Becker, 1989; Briffa et al., 1992; Dupouey et al., 1992).
These methods are generally referred to under the generic title
‘Regional Curve Standardization’ or RCS. RCS methods have the
potential to preserve more low-frequency climate information. It is
important to recognize that the chronologies used in many recent
NH temperature reconstructions representing the last millennium
were produced, either using only tree-ring data processed with various
realizations of these RCS techniques (Briffa, 2000; Esper
et al., 2002; Cook et al., 2004a; D’Arrigo et al., 2006b) or, alternatively,
included at least some tree-ring data produced with them
(Mann et al., 1999; Mann and Jones, 2003; Hegerl et al., 2006).
Most long tree-ring chronologies must be constructed using a
combination of data extracted from living, recently dead and
somewhat older preserved subfossil wood. Processing of this
material, using the RCS techniques, includes assumptions about
the homogeneity of sample data through time (ie, that changes in
the apparent average growth rates of trees in one region result
solely from changes in local climate forcing and not because of a
lack of homogeneity in the origins of the sample trees, whose
average growth rates might reflect differences in elevation, aspect,
soil type or competition pressures). These methods also require
large replication of samples and that the samples together span a
long period of time. In practical situations where these methods
have been used, these requirements are unlikely to be entirely
satisfied. So while the chronologies do exhibit greater variability
on longer timescales than is apparent with earlier processing
techniques (Briffa et al., 1996; Cook et al., 2006; D’Arrigo et al.,
2006b), this long-period variance may be associated with wide
confidence limits. Hence, the possible improvement in preserved
low-frequency variance must be viewed in the context of realistically
assessed, and potentially large, uncertainty.
It is also known that various implementations of the basic RCS
approach (as for other standardization approaches), at least in the
way that they have been implemented until now, can impart systematic
biases near the beginnings and ends of tree-ring chronologies
(Cook and Peters, 1997; Briffa and Melvin, 2008; Melvin and
Briffa, 2008). Recent statistical processing techniques go some
way to correct for spurious variance inflation where sample replication
is low, or where ring-width measurements are extremely
small (Cook and Peters, 1997; Osborn et al., 1997; Frank et al.,
2007). Work is also continuing to explore other biases in the use
of the RCS technique, but this work must be expanded to take in
a wider range of geographical and ecological situations and to
widen the scope of RCS application to situations where sample
replication is limited, the data are known to be inhomogeneous or
where they cover only a relatively short (often recent) period
(Briffa and Melvin, 2008).
In the meantime, chronologies, even some based on similar
measurement data from the same locations, can exhibit very different
statistical properties because they were constructed using
different techniques, or different implementations of similar techniques
(eg, compare Hantemirov and Shiyatov, 2002 with Briffa,
2000; Naurzbaev et al., 2002 with Jacoby et al., 2000 or Esper
et al., 2002). This can cause confusion and may lead to a degree
of arbitrariness in the choice of seemingly similar published data
for inclusion in different hemispheric or global reconstruction
efforts. This situation will also be improved if future work is concentrated
on the provision of demonstrably robust regional
chronologies and associated regional climate reconstructions.
Dendroclimatology has always paid attention to the important
issue of quantifying chronology confidence, originally by calculating
generalized statistics such as overall chronology standard
error and through analysis of variance which was used to
provide a measure of the strength of the common forcing inherent
among the series of tree-ring indices that make up a chronology
(Fritts, 1976b; Cook and Kairiukstis, 1990). Subsequent
work demonstrated how a simple measure of the growth-forcing
signal could be conveniently estimated (as the average of all
correlation coefficients, r–, calculated from the multiple interseries
comparisons of standardized data) for different chronologies
(Wigley et al., 1984). More importantly, it was shown how
r– could be used to measure the time-dependent statistical quality
of a chronology, in terms of its similarity to a hypothetical
infinitely replicated series (ie, one with no expressed noise and
where the variance of the underlying growth-forcing process is
perfectly represented: the so-called expressed population signal,
EPS). This measure of chronology quality has been widely
adopted in dendroclimatic studies, to gauge the period over
which a chronology can be considered to be of ‘acceptable’ statistical
confidence, though a somewhat ad hoc EPS threshold
criterion is often used for defining acceptance (Wigley et al.,
1984). However, what is perhaps not widely appreciated is that
this use of EPS is often biased toward the measurement of shorttimescale
chronology confidence, because it is interannual variance
that dominates the routine calculation of r–, particularly
when this is done on the basis of a relatively short-period of
common overlap between constitute chronology index series.
So, while a chronology may be reasonably considered statistically
robust in terms of its representation of relative year-toyear,
or even decade-to-decade variability, assuming that high
EPS values, as they are generally calculated, are a reliable
measure of longer-timescale variance is questionable.
8 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
The common variability between low-pass-filtered index series
within a chronology (eg, representing timescales >100 years) can
be substantially less than that measured at medium-frequency
(~ multidecadal timescales), and the common variability at long
and medium timescales is invariably lower than that measured at
the interannual (< 10 years) timescale. Early work aimed at assessing
the level of sample replication required to achieve statistically
robust chronologies for dendroclimatic studies was carried out in
the southwest of the USA. Tree growth in this region is strongly
influenced by a lack of available moisture and inter-tree ring-width
variability displays strong common patterns, with large year-toyear
variations. Measurements of common variance within these
data led to the conclusion that ~20 trees were sufficient to provide
a robust chronology (DeWitt and Ames, 1978). This estimate is
unlikely to be generally valid where the focus is on relatively
longer timescales than the short-period variability that dominates
the growth patterns of drought sensitive conifers in the American
southwest. In many other regions of the world, where the climate
signal in tree-ring data is typically smaller, data from more trees are
required to represent, even high-frequency, variability reliably and
perhaps an order of magnitude more sample series will be required
to produce a robust representation of long-timescale variance (see
Cook and Kairiukstis, 1990). Many tree-ring chronologies, including
some that make up well-known spatially expansive networks in
parts of North America, Europe and Russia are not sufficiently
internally replicated to ensure reliable individual representation of
relatively low-frequency common forcing (ie, on timescales longer
than several decades). Typically these networks are made of
chronologies that span several centuries. However, for this time,
intraregional similarities in chronology variability can be used to
demonstrate the likely fidelity of inferred regional growth-forcing
signals (Schweingruber and Briffa, 1996; Esper et al., 2002;
D’Arrigo et al., 2006b; Esper et al., 2007c).
The value of large series replication in dendroclimatic reconstruction
of centennial and longer timescales cannot be overstated.
All proxy disciplines benefit from an appreciation of the need to
provide multiple measurement series both within and between
adjacent sampling locations. Reliable estimates of inherent timedependent
and timescale-dependent growth-forcing signals, and
subsequent quantification of proxy series reliability are only possible
where multiple sampling is routinely undertaken. Certainly,
within dendroclimatology, even greater attention needs to be paid to
the explicit quantification of chronology confidence for specific
timescales of variability. Empirical, Monte Carlo-based approaches
to assessing chronology confidence, such as by bootstrap techniques,
applied at individual site and wider inter-regional spatial
scales, could be usefully adopted much more widely (Cook, 1990;
Esper et al., 2002; Cook et al., 2004a; Guiot et al., 2005).
This brings us back to the particular issue of establishing the
realism of low-frequency, regionally coherent, climate signals in
existing high-resolution tree-ring series. In the same way that the
parallel behaviour of different sample series within a localized
chronology can be used explicitly to demonstrate the existence,
and measure the strength, of a common climate signal, it is important
to establish that different chronologies constructed within the
same region display common features over relatively long, multicentennial
periods (Esper et al., 2002).
Instrumental climate records have been used to estimate the distances
over which surface temperatures would be expected to covary
in different parts of the globe (Briffa and Jones, 1993; Jones
and Briffa, 1996). This covariance changes according to season
and timescale: it is greater over tropical oceans compared with
high-latitude land areas. It is greater in winter rather than summer
and appears to be greater on decadal compared with interannual
timescales. These results show that within ‘reasonable’ separation
distances, different chronologies would be expected to exhibit
common variability. At multidecadal, century and longer timescales,
evidence of mutual variability is a minimum prerequisite for establishing
whether or not chronologies provide genuine evidence of
regional climate signals. There is as yet not enough evidence to
demonstrate that the few relatively long chronologies that are generally
incorporated within numerous different reconstruction studies
robustly represent the low-frequency climate forcing of trees
within their source regions. In the light of this uncertainty, there
is, therefore, a requirement, not only for sampling ‘new’ areas, but
also for substantially greater effort to provide a much improved
density of samples within these critical regions, and over long
periods, where there is already a proven potential for tree-ring
data to contribute usefully to multimillennial-length, average
hemispheric and global reconstruction efforts.
The various methods by which proxy climate data are transformed
into estimates of past climate variability, on different time
and space scales are discussed in the following section. However,
several issues that arise in the context of interpreting tree-ring data
are worthy of separate mention here.
The first is, in part, associated with possibly limited expression
of long-timescale climate forcing discussed above: the potential
biasing of regression weights. When tree-ring data are scaled by
comparison with instrumental climate data there is generally no
explicit consideration of whether or not the scaling should be equal
for all timescales. Calculation of the optimum scaling factor takes
account of the match in total variance, with no separate assessment
made of the fit between growth and climate variability on short and
long timescales. Tree-ring records from different tree species have
very different variance spectra in terms of the ratio of short- to
long-timescale variance, irrespective of the standardization procedure.
For example, in northern Eurasia, pine species will display
relatively low high-frequency variance (a redder spectrum) while
larch series have a much greater proportion of year-to-year variability
(bluer spectrum). Similarly, ring-width series in conifers
have a redder spectrum than ring densities. Simple linear scaling or
least squares regression will be strongly influenced by the yearto-year
variability of the data, the more so where the comparison
(calibration) involves data with little or no trend. Where the comparison
period is short, the regression coefficients, besides being
subject to large uncertainty, are invariably biased towards representing
growth processes operating on year-to-year rather than century
or longer timescales (Osborn and Briffa, 2004).
Opportunities to explore whether a single scaling is ‘optimum’
for different timescales have been limited by a general lack of long
instrumental climate records, but what little work that has been
done to test this assumption suggests that it may not be appropriate
(Guiot, 1985; Osborn and Briffa, 2000). Whether this is a reflection
of a genuine difference in the temperature control of tree-growth
processes on different timescales, or merely a product of statistical
sampling error is not known. What is known is that regressing climate
data against the same tree-growth data, alternatively represented
at annual resolution or after common low-pass smoothing of
the climate and growth data, can result in very different scaling factors
and, hence, reconstructed climate series with notably different
long-term variability (Esper et al., 2005). This further complicates
the comparison of dendroclimatic reconstructions where different
scaling approaches have been used (Jansen et al., 2007).
In dendroclimatology, a long-recognized requirement to ‘verify’
the temporal stability of regression coefficients and to establish
that multiple regression equations have not been overfitted,
has led to the routine subdivision of climate series to allow the calculation
of a suite of goodness-of-fit statistics, comparing climate
estimates with independent observations (Cook et al., 1994).
Typically, the period of overlap between tree-ring and climate
data is split either in half or in the ratio two-thirds to one-third (eg,
Neter et al., 1990). The data in one period are used to derive the
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 9
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
optimum scaling of the tree-ring data and the regression estimates
are compared with the climate observations over the other period.
In the case of multiple regression, this comparison provides
more realistic estimates of the reconstruction uncertainty than provided
by the calibration-period measures alone (Fritts, 1976a).
Repeating the process with the periods reversed allows direct
comparison of the time-stability of regression coefficients and
the reconstruction calculated ranges. As this cross-calibration
approach requires that the climate data be subdivided into even
shorter series, the calculation of regression coefficients and the
assessment of reconstruction quality are unavoidably strongly
biased toward short-timescale variability. In an attempt to mitigate
this bias, it has been common practice to repeat the calibration of
the ultimate regression weights to be used for reconstruction, but
using all of the available climate data (eg, Briffa et al., 1992). In
such situations, however, no formal statistical verification of the
quality of the long-timescale variance estimates is possible.
Measurements of the similarity between estimated and observed
climate variability at different timescales made over the full overlap
period often reveal a higher association on interannual rather
than on decadal and longer timescales. In future work, it would be
desirable if reconstructions were more commonly presented with
timescale-specific confidence limits: ideally in a way that indicates
predictor (chronology) error and calibration error, separately
and together. Where climate estimates are extrapolations beyond
the range of the variance incorporated within the data used for calibration
(or more strictly, those used for independent reconstruction
verification) this should also be clearly indicated.
Recent divergence between tree-ring growth and
temperature
The final aspect of tree-ring studies that needs to be highlighted is
what has become known as the ‘divergence’ issue. This refers to
the apparent failure of some (established as temperature-responsive)
tree-ring data to follow the trend in instrumental temperatures
observed over the latter part of the twentieth century. Chronology
time series that vary largely in parallel with changing temperature
in earlier periods progressively fail to show the increasing trends
that would represent a continuing positive response to the strong
warming observed during recent decades. Originally this was noted
primarily in certain northern high-latitude areas for ring-width data
in Alaska (Jacoby and D’Arrigo, 1995) and in ring-width and particularly
ring-density data, in more extensive regions of northern
Europe and Russia (Briffa et al., 1998). In the earlier work, it was
suggested that the cause of the North American observations was a
shift from a direct dominant temperature control on tree growth to
one where lack of available moisture becomes increasingly influential,
possibly to an extent where the sign of the temperature influence
becomes negative rather than positive (Jacoby and D’Arrigo,
1995; D’Arrigo et al., 2004).
Subsequently, various studies focused mainly on recent treegrowth
in Alaska and Canada support the idea that current tree
growth may no longer be responding positively to increased
warming (Barber et al., 2000; Lloyd and Fastie, 2002; Davi et al.,
2003; Wilmking et al., 2004; Driscoll et al., 2005; Pisaric et al.,
2007). Other suggestions have been offered as the cause of the
widely observed loss of temperature response over northern
Eurasia. The increasing influence of drought has also been suggested
as the cause (Jacoby et al., 2000), though other suggestions
include possible reduced atmospheric clarity, localized persistence
of spring snow cover and seasonal changes in ozone-related
surface UV concentrations (Briffa et al., 1998, 2004; Vaganov et
al., 1999; D’Arrigo et al., 2008).
The IPCC recently laid particular stress on this issue, pointing out
that any significant shift in the recent growth response of trees would
invalidate the assumptions that underlie the simple regression-based
approach to reconstructing past temperature changes. This would
imply an inability to recognize potential underestimates of the degree
of warmth in earlier periods of reconstructions (Jansen et al., 2007).
It is important to stress that not all high-latitude regions display this
apparent decoupling between observed and dendroclimatically estimated
temperatures (Briffa et al., 2007; Wilson et al., 2007).
However, the issue remains a crucial one. Unfortunately, a comparative
scarcity of recent (ie, post-1980) tree-ring data remains a major
obstacle to further exploration of the extent and causes. Hence we
stress the vital requirement for widespread updating of major treering
networks, as well as for the acquisition of data for new regions.
Corals
Massive reef-building corals provide high-resolution, continuous
records of tropical ocean climate ranging from several decades to
several centuries in length. Several characteristics of corals make
them ideal tools for the reconstruction of tropical climate: (1) annual
skeletal density banding provides chronological control, (2) high
linear growth rates of 1–2 cm/yr combined with year-round growth
provide annual to subannual resolution, (3) growth of colonies to
several metres in height provides several hundred years of continuous
growth, (4) location in shallow-water tropical ocean regions
(~30°N to 30°S constrained by bathymetry and a variety of environmental
factors, Kleypas et al., 1999) provides information from
regions poorly represented by other sources of high-resolution
palaeoclimatic data, and (5) the calcium carbonate skeleton incorporates
a range of geochemical tracers that reflect local environmental
conditions (see reviews by Gagan et al., 2000; Cole, 2003;
Felis and Pätzold, 2003; Correge, 2006; Grottoli and Eakin, 2007).
Coral skeletons that are well preserved after death allow generation
of high-resolution proxy climate records during Uranium/Thorium
(U/Th)-dated windows of the more distant past (eg, Tudhope et al.,
2001; Cobb et al., 2003a, b; Felis et al., 2004). To date, proxy climate
records derived from corals have mostly been used to reconstruct
a specific feature of the tropical climate system, in particular
ENSO and, to a lesser extent, decadal–interdecadal variability over
the last several centuries (eg, Fleitmann et al., 2007). Here, after
summarizing the basis for extracting climate information from
corals, we focus on the broader objective of using coral records for
large-scale climate reconstruction.
Specific data
The construction of long, regionally representative palaeoclimate
records from massive corals requires careful site selection followed
by close inspection of coral skeletal morphology and lithology. The
process begins with the identification of a suitably large coral head
growing in an open ocean reef environment. The most commonly
used massive coral genera are Porites in the Indo-Pacific,
Montastraea in the Caribbean/Atlantic with a couple of records
obtained from Pavona and Platygyra. Recent studies have also
demonstrated the palaeoclimatic potential of the more slowly
growing species Diploria (Draschba et al., 2000), Diploastrea
(Watanabe et al., 2003; Bagnato et al., 2004) and Siderastrea
(Guzmán and Tudhope, 1998). Large several metre high coral
colonies, although not uncommon, are not found in appreciable
numbers on coral reefs (cf. dendroclimatology where there is often
a whole forest from which to select suitable samples). The palaeoclimatic
value of a particular core must be evaluated in the laboratory,
where cores are cross-sectioned into ~5–10 mm thick slices
for x-radiography and microscopic inspection. X-ray images reveal
density variations associated with annual banding (when present)
and growth direction, which guide the placement of transects for
subsequent geochemical sampling. Ideally, a good coral slice will
have a well-presented annual density banding pattern in which all
years are clearly and consistently identifiable along the length of
the coral core – this is rarely the case, especially in corals from sites
with a small seasonal cycle in temperature. In these cases, other
forms of annual banding, including luminescent banding (visible
10 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
under illumination with longwave ultraviolet light in corals from
some locations) and geochemical banding (in particular seasonality
in skeletal δ18
O and δ13
C) may be used to help establish the
chronology. Coral growth characteristics (linear extension rate,
skeletal density, calcification rate) can be measured from the x-rays
and from gamma or x-ray densitometry (eg, Chalker and Barnes,
1990; Helmle et al., 2002), and may provide information related to
environmental gradients in sea surface temperatures (SSTs) and
water quality (eg, Lough and Barnes, 2000; Carricart-Ganivet and
Merino, 2001). Microscopic analysis of thin coral sections is critical
to detecting subtle diagenetic alteration (eg, pore infilling by
secondary aragonite), which can compromise primary geochemical
signals recorded in basal sections of long coral cores (Enmar et al.,
2000; Muller et al., 2002; Quinn and Taylor, 2006). The occurrence
and intensity of luminescent lines (observed under ultraviolet
light) in near-shore corals from selected locations can provide
robust reconstructions of past river flow and rainfall (eg, Isdale,
1984; Nyberg, 2002; Hendy et al., 2003; Lough, 2007).
Reconstructions of tropical ocean palaeoclimatic variability for
the past several centuries are primarily based on geochemical
analyses of the coral skeleton. This requires drilling of carbonate
powder at regular intervals along a major coral growth axis.
Sampling resolution is most commonly in the range of 4–12 samples
per year (eg, Quinn et al., 1996). Homogenizing samples
milled across entire growth bands yields records with annual to
5-yr resolution (eg, Druffel and Griffin, 1993; Cole et al., 2000;
Hendy et al., 2002; Calvo et al., 2007). Commonly employed
analysis techniques include using stable isotope ratio mass spectrometry
for measuring oxygen and carbon isotopic composition
(δ18
O and δ13
C, respectively), and inductively coupled plasma
mass spectrometry (ICP-MS), inductively coupled plasma atomic
emission spectrometry (ICP-AES) and thermal ionisation mass
spectrometry (TIMS) for measuring trace and minor element such
as Sr/Ca, Mg/Ca, U/Ca and Ba/Ca (for examples, see Beck et al.,
1992; Shen et al., 1992; Min et al., 1995; Mitsuguchi et al., 1996;
Schrag, 1999; Gagan et al., 2000; Quinn and Sampson, 2002;
McCulloch et al., 2003; Fleitmann et al., 2007).
Most coral palaeoclimate proxy records are based on time series
of δ18
O and/or Sr/Ca. While skeletal δ18
O is generally interpreted
as a measure of SST, sea water salinity, or a combination of the
two, coral Sr/Ca variability is considered a more direct measure of
SST variations.
Coral skeletal: δ18
O
The well established temperature-dependent fractionation of oxygen
isotopes between sea water bicarbonate and calcium carbonate
means that changes in water temperature are recorded in the
coral skeleton, and there are now many empirical studies that
show that the temperature dependent slope of coral aragonite δ18
O
is close to that predicted from inorganic carbonate studies (eg,
Craig, 1965) at 25°C, ie, about −0.2‰ δ18
O per 1°C increase in
temperature. It is this temperature sensitivity of carbonate δ18
O
that is most often used in the interpretation of coral δ18
O records.
However, it is important to remember that changes in the isotopic
composition of seawater in which the corals grow also directly
impact the isotopic composition of the coral skeleton being precipitated.
On seasonal and interannual timescales, the water isotopic
composition (δ18
OW) is closely related to salinity, with
isotopically enriched, or more positive, δ18
OW values accompanying
high salinity (eg, see Fairbanks et al., 1997). In detail, however,
δ18
OW is controlled by the combined effects of local
precipitation/evaporation balances, regional sources and sinks of
water vapour, ocean mixing and advection, and whole ocean δ18
O
composition. Therefore, while the relationship between coral δ18
O
and temperature does not vary, the relationship between salinity
and δ18
O varies appreciably from site to site, and through time.
Nonetheless, many of the published records of coral δ18
O have
been used to reconstruct changes in rainfall associated with climatic
phenomena such as ENSO (eg, Cole et al., 1993; Le Bec
et al., 2000; Urban et al., 2000; Tudhope et al., 2001; Cobb et al.,
2003b), though poor constraints on the relationship with seawater
δ18
O on decadal-to-centennial timescales introduces significant
uncertainty to climatic interpretations on long timescales.
Coral skeletal: Sr/Ca
The coral skeletal Sr/Ca ratio is considered to be a more direct
measure of water temperature (Smith et al., 1979; Beck et al.,
1992), an interpretation based on temperature-dependent changes
in the Sr/Ca of inorganic aragonite (Kinsman and Holland, 1969)
and an ever-growing body of empirical coral-based calibrations
(eg, Alibert and McCulloch, 1997). Although this relationship
appears to be strong and robust in some cases, the mechanisms
responsible for the temperature relationship remain relatively
poorly understood, and some non-temperature-related artefacts
have been noted (eg, de Villiers et al., 1995; Cohen et al., 2002;
Gaetani and Cohen, 2006). Poor Sr/Ca-temperature calibrations are
observed in corals characterized by slow (less than 6 mm/yr) and/or
variable growth rates. Subtle diagenesis has a large effect on coral
Sr/Ca-based temperature reconstructions, owing to the shallow
slope of the Sr/Ca-temperature relationship (~ 0.7%/K). To date
there are relatively few published century-to-multicentury Sr/Ca
coral records (Linsley et al., 2000; Hendy et al., 2002). Further
study and more long Sr/Ca records, specifically multiple long
records from the same site (eg, Hendy et al., 2002), are required to
establish the strengths and weaknesses of this tracer for large-scale
temperature reconstruction over centuries to millennia.
Description of specific measured proxies and
analysed chronologies
There are now ~90 published coral geochemical records but only
~30 (predominantly δ18
O with its potentially mixed response to
both SST and salinity) that provide proxy tropical climate records
that extend earlier than 1900 (NOAA Paleoclimatology data base
http://www.ncdc.noaa.gov/paleo/index.html). The majority (21) of
these long coral records are from reefs in the tropical Pacific, six
from the Indian Ocean, three from the Atlantic/Caribbean and two
from the Red Sea. Several studies have demonstrated that even the
current limited spatial network of coral records can capture useful
amounts of large-scale climate variance (eg, Evans et al., 1998,
2000, 2002; D’Arrigo et al., 2005; Wilson et al., 2006; Gong and
Luterbacher, 2008). The main limitation of these large-scale reconstructions
is their reduced fidelity prior to ~1850 as the number of
available coral records declines, with only four records by 1700,
eight by 1750, 14 by 1800, 21 by 1850 and 32 by 1900. The longest
records date back to 1560 for the central Great Barrier Reef (Hendy
et al., 2002) and 1600 for the eastern equatorial Pacific (Dunbar
et al., 1994).
Uncertainties in high-resolution coral proxy
climate records
There are, as yet, no routine techniques (as used in dendroclimatology
see the section ‘Proxy data uncertainty’) for determining the
strength and reliability of climatic signals contained in these coral
geochemical records (Lough, 2004). One common problem with
calibration and verification of coral geochemical records is the lack
of suitable long-term instrumental SST and environmental measurements
at remote coral reef locations. Another major problem
stems from the use of ‘calibrations’ based on annual cycles of coral
δ18
O and SSTs to quantify interannual to multidecadal palaeo-SST
variability. When sufficiently long SST records are available, bandspecific
calibrations of coral geochemical variations can be performed
to check for non-linear coral proxy behaviours.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 11
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
The largest source of uncertainty in the climatic interpretation
of coral δ18
O records lies in the mixed temperature and sea water
δ18
O signal contained within these records, and the possible
changes in relative contributions of these factors over time or as a
function of timescale. For example, δ18
O records can overestimate
twentieth-century SST warming if the coral record is considered
only as a temperature record and the standard temperature dependence
relationship of about −0.2‰ δ18
O/°C is applied, as regional
freshening at some coral sites has likely contributed to coral δ18
O
trends. Indeed, observed tropical SST warming of +0.2°C at 16
Indo-Pacific coral reef sites from 1951–1990 compares with a
warming of +0.7°C estimated from coral δ18
O records at the same
sites, when attributing δ18
O to temperature only (Lough, 2004).
However, in some parts of the tropical oceans, such as the western
and central equatorial Pacific warm pool region, temperature and
salinity are tightly coupled on seasonal-to-interannual timescales,
with high SST driving increased atmospheric convection and
accompanying heavy rainfall. Therefore, it is possible to use calibration
and verification of the coral δ18
O signal over the recent
instrumental period to derive an empirical relationship with temperature,
using methods developed for dendroclimatology (eg,
Wilson et al., 2006). However, it is not clear how well these relationships,
which are often biased to interannual variance, will hold
for low-frequency changes on multidecadal–centennial timescales
for which the SST versus sea-surface salinity relationships may be
different. In these cases, an independent measure of SST variations
(eg, from skeletal Sr/Ca from the same core) may provide a
valuable check on the low-frequency SST variations and trends, as
well as allowing the water composition (salinity) signal to be separately
resolved. Another approach is to attempt to calibrate skeletal
δ18
O to temperature independently for different frequency
bands. The difficulty with the latter approach is the shortness of
the instrumental record in most locations.
Additional uncertainty results from our limited understanding
of the biological mechanisms leading to coral skeleton formation
(calcification) and how the incorporation of geochemical tracers
used for palaeoclimate reconstruction into the skeleton may be
modified by biological processes. This biological overprint has
been observed in studies of coral δ18
O and Sr/Ca, and can lead to
significant uncertainties in coral palaeoclimate reconstructions in
worst-case scenarios. For example, coral skeletal δ18
O values are
significantly offset from those predicted by inorganic equilibrium
calculations based on seawater δ18
O compositions and temperature.
This isotopic disequilibrium or ‘vital effect’ has been attributed
to kinetic isotopic effects (McConnaughey, 1989a, b, 2003).
The magnitude of this offset from equilibrium can be substantially
different in corals growing in the same location – amounting to
mean coral δ18
O offsets of up to 0.4‰ (giving an uncertainty
equivalent to ~2°C) in both modern and fossil corals (Linsley
et al., 1994; Cobb et al., 2003b). This coral-to-coral offset poses
no problem for interpreting relative changes in a single long coral
δ18
O record, as the offset is relatively constant through time if the
coral’s major growth axis is sampled (Felis et al., 2003). Sampling
transects that fall off the major growth axis are associated with
larger kinetic effects (McConnaughey, 1989a; Hart and Cohen,
1996). However, it is a serious hindrance for interpretation of isolated
fossil coral δ18
O sequences, as it is impossible to differentiate
a change in mean climate conditions from an offset associated
with coral-specific vital effects. One promising strategy for minimizing
such errors involves averaging the mean coral δ18
O values
across numerous fossil coral heads growing during the same time
period (Cobb et al., 2003b). Such coral-to-coral offsets likely exist
for coral Sr/Ca records and other coral geochemical proxies, but
have not been quantified as of yet. Another kinetic-related ‘vital
effect’ concerns the observation that coral δ18
O and Sr/Ca vary
with extension rate in slow-growing portions of corals (usually <6
mm/yr) (McConnaughey, 1989a, b; deVilliers et al., 1995; Felis
and Patzold, 2003). By focusing on faster-growing coral skeletons,
coral palaeoclimate researchers can minimize the contribution
from growth-related artefacts to coral palaeoclimate records.
Improving coral-based palaeoclimate records
We now know considerably more about the wide range of potential
palaeoclimatic and palaeoenvironmental records locked in the
calcium carbonate skeletons of massive corals than when Knutson
et al. (1972) suggested that they could be applied to determining
‘water temperature variations’. Indeed, corals have made a significant
contribution to understanding the nature and causes of climate
variability in the tropical oceans, in particular interannual
ENSO-related variability, but more recently also multidecadal to
centennial timescale variability (eg, Linsley et al., 2004; D’Arrigo
et al., 2005; Wilson et al., 2006; Gong and Luterbacher, 2008).
This contribution to the global, high-resolution palaeoclimatic
data base can be substantially extended by (1) improving the reliability
of coral geochemical time series through increased replication,
(2) increasing the temporal depth of coral records and (3)
distinguishing between temperature and hydrological sources of
variability in coral δ18
O records.
The fact that most published coral time series are based on
records extracted from single coral cores prevents the quantitative
separation of the desired climate signals from noise at a given site.
Replication is essential to identify non-climatic artefacts in individual
coral records, to ensure dating accuracy and to identify
large-scale environmental signals that are shared among different
coral records. Ideally, at least three cores from a reef location are
necessary. Given the rarity of long, multicentury cores, it may be
more practical to undertake the replication work on more common
shorter cores that span the last several decades.
A revised globally coordinated sampling strategy may be necessary
to increase the temporal depth of coral records to span the
last millennium. To date, much of the sampling focus has been on
obtaining proxy records from key locations influenced by ENSO
events. However, future sampling should also target cooler reef
sites, as average linear extension rates of massive corals vary
directly with average annual SST so that corals in cooler waters
grow more slowly than those in warmer locations (Lough and
Barnes, 2000). This also means for the same-sized coral colony,
one living in relatively cooler tropical waters will tend to contain
more years of record than one living in warmer waters.
Realistically, however, coral reconstructions that pre-date AD 1500
must be based on long-dead corals that are scattered on reef tops
and beaches throughout the tropics. Such corals can be U/Th
dated, analysed for δ18
O and Sr/Ca, and then spliced together to
yield multicentury windows on tropical marine climate (Cobb
et al., 2003a, b). Given appropriate material, such ‘fossil’ coral
reconstructions can be spliced to the oldest portions of corresponding
‘modern’ coral reconstructions
Reducing the uncertainties surrounding the interpretation of coral
δ18
O time series requires the implementation of multiple strategies.
First, better characterization of the environments in which sampled
coral colonies live would help reduce uncertainty in the climatic
interpretation of records derived from them. For example, more routine
use of in situ loggers (temperature and, ideally, conductivity)
would allow the response of the coral to be better understood with
respect to local and regional conditions. Even a few years of in situ
logger data can help identify the extent to which the coral’s local
environment is representative of the more relevant regional-scale
conditions. Such local monitoring should also include the generation
of rainfall and seawater δ18
O time series, where practical, as the
IAEA’s Global Network of Isotopes in Precipitation contains very
few ocean island sites, obscuring the relationship between largescale
climate and seawater δ18
O changes. A second, complementary
12 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
approach involves increasing the number of coral Sr/Ca records
from sites where coral δ18
O records already exist. This approach
could be used to isolate the seawater δ18
O component of coral δ18
O
records, which remains poorly constrained, especially on decadalto-centennial
timescales. Finally, as with any potential source of
proxy climate information, there is still a need for detailed process
studies of how to extract the most reliable palaeoclimate records
from corals, both modern and fossil. This is particularly the case for
the development of relatively new proxies such as Sr/Ca.
The coral palaeoclimatic community would also benefit from
the development of a set of standard sampling, screening and
reporting protocols that allow the quality and reliability of different
coral records to be assessed. A meaningful step towards this
goal would be to encourage the storage of more complete sets of
metadata (including water depth and GPS locations of drilling, xradiograph
images of the cores and chronological methodologies)
on national data servers such as NOAA/NCDC (see section ‘Tree
rings and the need for improved regional and temporal coverage’).
Such information would also facilitate the informed use of coral
records by researchers outside the coral community, particularly
those interested in multiproxy climate field reconstruction.
High-resolution ice core climate records for
the last millennium
Ice cores provide a rich record of environmental tracers trapped as
snow progressively accumulates on ice sheets and glaciers.
Climate processes leave an imprint on the delivery of moisture and
contaminants that provides a basis for reconstructing climate
parameters from the ever-increasing suite of measurable species.
The principal variable from ice cores that has been used in multiproxy
climate reconstructions (eg, Mann and Jones, 2003;
Pauling et al., 2003, Jones and Mann, 2004; Luterbacher et al.,
2004; Schneider et al., 2006) is the isotopic composition of the
water, which is generally treated as a proxy for temperature. The
main focus of this section is on these water isotope records: their
interpretation, associated uncertainties and availability. However,
the optimal use of the isotope records relies on other measurements,
particularly high-resolution trace chemical measurements
(eg, McConnell et al., 2002; Kaspari et al., 2004; Dixon et al.,
2005) which provide better resolved annual layers at lower accumulation
sites where diffusion of water vapour extinguishes
annual variability from the isotope record (see later discussion). In
the future, it is likely that climate reconstructions will be enhanced
by incorporating other ice core parameters. A growing range of
such additional climatic indicators is available, from the local net
annual snow accumulation to proxies for regional- and globalscale
circulation modes (eg, ENSO, Meyerson et al., 2002; Bertler
et al., 2004; NAO, Vinther et al., 2003; Mosley-Thompson et al.,
2005), mid- and high-latitude sea-level pressures, the Southern
Annular Mode and circulation indices (eg, Souney et al., 2002;
Goodwin et al., 2004; Shulmeister et al., 2004; Kaspari et al.,
2005), and sea-ice extent (Curran et al., 2003). At certain sites,
records of summer temperatures may be derived from the thickness
and frequency of melt layers in the ice (Fisher et al., 1995).
Ice cores have also provided significant information about
global-scale climate forcing over recent millennia. They provide the
main source of data on past concentrations of all the major longlived
greenhouse gases (MacFarling Meure et al., 2006). They also
provide annually resolved records of volcanic fallout that can be
used to estimate past forcing by volcanic aerosols (eg, Robock and
Free, 1995; Palmer et al., 2001) and they offer one means to estimate
the past strength of solar activity (eg, Bard et al., 2000).
Clearly, ice core records are limited to ice-covered sites, but frequently
such locations are precisely where other sources of proxy
climatic data are not available. Many ice cores provide sufficient
detail for ‘high resolution’ records, taken here to mean those with
resolvable annual variations that enable a countable chronology
with near-absolute precision (say a few percent). This precision is
aided by reference to markers of known age from volcanic events
and nuclear bomb testing (Mosley-Thompson et al., 2001). The
ability to resolve annual variations is determined by the amount of
snow deposited, the stratigraphic disturbance by meltwater percolation
(if any), postdepositional modification of the seasonal signal
by processes such as vapour diffusion, and the removal,
redeposition and/or mixing of near-surface snow by the wind.
The goal of obtaining records that extend over multiple millennia
imposes a further constraint of sufficient ice sheet thickness.
Sites which meet these combined requirements are found in
Greenland, across temperate and tropical alpine regions and in
higher accumulation zones in Antarctica.
The full geographic extent of potential sites has not yet been
exploited. Significant initiatives within the ice core community,
under the banner of IPICS (International Partnerships in Ice Core
Sciences, Brook and Wolff, 2006), are expected to expand the network
of ice cores, especially those that provide records for the last
two millennia. This work extends other programmes, notably
PARCA (Program for Arctic Regional Climate Assessment,
Mosley-Thompson et al., 2001; Thomas et al., 2001) and SCARITASE
(Scientific Committee on Antarctic Research–International
Trans Antarctic Scientific Expedition, Mayewski and Goodwin,
1997; Mayewski et al., 2005), which have collectively extracted
many ice core records from Greenland and Antarctica, respectively.
In the context of dwindling alpine glaciers, the impetus for
recovering records from these temperate locations is clear and
urgent. These are briefly discussed in a later part of this section.
Ice core isotopic records: interpretation and
uncertainties
The strong correlation between the isotopic composition of precipitation
and local temperature is a well-known property at midto-high
latitudes (Craig, 1961; Dansgaard, 1964). This provides an
essential tool for reconstructing palaeotemperatures from ice core
measurements of δ18
O or δD (hereafter abbreviated together as δ).
In fact, the processes controlling fractionation are complex and
reflect conditions at the moisture source, during atmospheric
transport and at the points of condensation and final precipitation.
However modelling studies confirm observations that at polar
sites it is the final stages of moisture transport, and site temperature
in particular, that dominates the δ signal (Helsen et al., 2006).
For alpine sites, depletion during transport and ‘washout’ (a
dependence of isotopic fractionation on precipitation amount),
together with seasonality in these effects, are often more important
than local temperature in controlling fractionation. A significant
body of literature is devoted to understanding and calibrating the
‘isotope thermometer’ (eg, Johnsen et al., 1972; Jouzel et al.,
1997), key aspects of which are considered here.
Before continuing the discussion of water isotopes in ice cores,
some general properties and constraints arising from the nature of
ice cores need to be addressed. The most fundamental among these
is the property that ice cores only record a δ signal during precipitation
events, which individually do not represent the mean climate.
Furthermore, seasonal variations in the timing and magnitude of
precipitation events can lead to additional bias, and this may also be
variable and subject to climatic modulation (Charles et al., 1995).
In situ studies, comparisons with meteorological data, model simulations
and direct comparisons among different ice core species can
help detect such bias and place limits on its influence.
In addition, non-climatic variability, or noise, results from surface
roughness (sastrugi) and from postdepositional reworking,
which lead to uneven stratification of the snow (Steffensen et al.,
1997). This depositional noise increases as the accumulation rate
diminishes relative to surface relief, so that the highest quality
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 13
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
records are expected from sites with higher accumulation rates.
The signal-to-noise ratio (SNR) in highly resolved Greenland
δ18
O data differs greatly across the ice sheet. SNR for annual δ18
O
at the south Greenland Dye-3 drill site (annual accumulation 0.56
m ice) is ~2.5 (Fisher et al., 1985, 1996), while the SNR is ~0.9
for the Greenland Ice Core Project (GRIP) drill site (annual accumulation
0.23 m ice) at the summit of the Greenland ice sheet
(Johnsen et al., 1997). Likewise, in Antarctica, the lowest noise
is found for higher accumulation rate coastal areas, where SNR is
comparable with that for sites with similar accumulation rates in
Greenland (Jouzel et al., 1997). Hence, while single ice core δ
records are useful from sites with high accumulation rates such as
Law Dome, Antarctica (0.7 m ice for the DSS site) and southern
Greenland, for lower accumulation sites (eg, Greenland summit)
replication is preferable and averaging of several annual δ records
should be undertaken before a regional climatic interpretation is
pursued (White et al., 1997).
A further influence on the fidelity of high-resolution reconstructions
from ice cores arises from the ‘timing noise’ in both
environmental markers and precipitation events. Annual markers
in the ice for defining years (typically mid-summer or mid-winter
extrema in a measurable parameter) will not correspond to a fixed
calendar date, and so computed annual averages will partition
some portion of the signal into the adjacent year. While timing
noise reduces the fidelity of annual and shorter-term reconstructions,
the impact diminishes rapidly for averages over a few years.
Timing noise can also be reduced by using multiple species to
define annual horizons (eg, Palmer et al., 2001).
Rapid ice flow can cause non-climatic trends in δ records. Such
trends arise because ice cores drilled on a sloping surface will contain
upstream ice originating from progressively greater elevations
at progressively greater depths in the core. The Dye-3 δ18
O record
is known to be affected by such upstream effects (Reeh et al.,
1985). It is essential to identify and correct ice core δ data for
flow-related trends, if the cores are not drilled at a site with modest
ice flow (ie, on an ice divide or on the summit of a dome).
Even where flow is minimal, evolution of the ice sheet itself (eg,
elevation changes) may introduce non-climatic temperature or isotope
fractionation trends. This possibility needs to be considered
before interpretation of the data.
Temporal resolution of ice records is limited by the accumulation
rate, postdepositional noise, diffusive smoothing (see below),
analytical sample size requirements and the finite size and number
of precipitation events. Modern analytical techniques permit some
species to be sampled at many tens of samples per year; however
the actual number of precipitation events that can be distinguished
above the surface reworking noise is much smaller. Glacier thickness
can impose a further limit on high-resolution records because
the progressive thinning of annual layers with depth, as a result of
ice flow, ultimately reduces the layer thickness to the point where
layers are not analytically resolvable. For alpine glaciers and thinner
marginal regions of Antarctica and Greenland this rapid thinning
can restrict annual records to the century-scale or less.
Turning specifically to issues affecting δ records, we first consider
the limiting effect of diffusive smoothing on resolution. In
the upper several tens of metres of the ice sheet where the ice
matrix is permeable (the firn zone) water vapour is free to move
diffusively. By the time the ice attains a state of impermeability,
where vapour movement ceases, the integrated effect is sufficient
to extinguish isotopic variations on vertical scales less than 20–25
cm of ice. This places a fundamental limit on the resolution of any
reconstruction. It also limits the ability to use annual cycles in δ
for layer counting, although signal processing techniques can be
used to enhance cycles and improve countability (Johnsen and
Vinther, 2007). Even at high accumulation sites, where seasonal
variations survive diffusive smoothing, the amplitude of variations
is progressively attenuated with depth in the firn. This introduces
non-climatic trends in winter and summer season δ values which
must be removed by prior correction for diffusion, if seasonally
separated records are to be reconstructed (Vinther et al., 2008).
Dating of most layer-counted cores uses a combination of
annual cycles in isotopes, where these can be resolved, and annual
cycles in trace chemicals. Most trace chemicals are typically
(though not exclusively) free from movement in the firn, and with
modern continuous analysis systems can be routinely sampled at
the centimetre scale. This facilitates identification of wellresolved
annual layers (sampled on the order of around ten times
per year) for annual accumulation rates above 10 cm ice-equivalent.
For such accumulation rates, vapour diffusion smoothes the
water isotope signal over 2–3 years.
The observed strong correlation between δ and site temperature
(T) suggests a straightforward approach for temperature reconstruction,
by simply applying the observed δ–Τ regression slope
(eg, Dansgaard, 1964). However, this spatially derived calibration
need not apply to the temporal variations at a given site, and a
more appropriate temporal calibration of the isotope thermometer
has been determined either from direct calibrations using colocated
data on water isotopes and site temperature (either instrumental
or derived from a different proxy method), or from a
modelling approach using atmospheric GCMs that have the capability
to explicitly simulate water isotopes (eg, Cuffey et al., 1995;
Johnsen et al., 1995; Werner et al., 2000). Temporal regression
slopes are generally difficult to compute directly from data since
there are few records of site temperature of sufficient length at ice
core sites to derive a calibration, and even where limited data are
available, the temperature excursion in the record is typically
small, limiting the calibration. At highly resolved sites, the larger
subannual temperature variability can be used for calibration (van
Ommen and Morgan, 1997), although the result will reflect seasonal
variations in transport in addition to temperature.
One major issue that emerges from both modelling and data
studies is that both the calibration slope and the quality of the δ–T
relationship varies according to the time period covered. Thus a
different value emerges from data studies comparing monthly values,
annual values and decadal trends (eg, Peel et al., 1988;
Rozanski et al., 1992; Kohn and Welker, 2005). It is, therefore,
necessary to use a calibration in a manner appropriate to the study
being conducted.
For coastal, high-accumulation Antarctic records, temporal calibration
over recent years gives low δ–T slopes (Peel, 1992;
McMorrow et al., 2004). This result is also found in isotopic GCM
studies (Werner and Heimann, 2002). Calibrations for high-elevation
areas of inland Antarctica are more difficult, since few recent
climatic data exist for these inland sites. An alternative approach
is to investigate the δ–T relationship over very large past climate
changes such as glacial–interglacial transitions. Even here, direct
estimates are difficult in central Antarctica since the low-accumulation
regime limits the utility of borehole temperature reconstructions,
while the absence of abrupt temperature changes in
Antarctica renders the thermal diffusion effect that has been used
successfully in Greenland (see next paragraph) inapplicable.
However, indirect evidence using palaeoprecipitation estimates to
infer temperature and comparing with δ gives calibration slopes
consistent with the spatially derived value (Jouzel et al., 2003) and
GCM modelling studies agree with this conclusion (Jouzel et al.,
2007). In any case, it is the lower elevation records which provide
the high-resolution chronologies considered here. So we are left
with a dilemma since the longest-term calibration data from cores
in the high interior of East Antarctica give a calibration slope similar
to the spatial value, while the shorter-term calibrations possible
from higher accumulation coastal sites give a much lower
slope: whether these calibrations apply at these sites on different
14 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
timescales, particularly for decadal and centennial changes is
unclear, and needs further data and model-based calibration work.
For Greenland, long-term calibrations have been made by comparing
borehole temperature data with δ values (eg, Cuffey et al.,
1995; Johnsen et al., 1995). Another method is based on estimating
the magnitude of past rapid temperature changes from combined
gravitation and thermal diffusion of the air in the firn
column (Severinghaus et al., 1998). These two approaches give
calibration slopes for δ–T at Greenland sites which are in good
agreement with slopes estimated from seasonal oscillations in
highly resolved central Greenland cores (Shuman et al. 1998), but
are considerably lower than the spatially derived values, and values
from simple Rayleigh-type distillation models. However, the
two techniques apply particularly to the major climate changes of
the past, and the reduced slope at least partly derives from a
change in the seasonality of snowfall (less winter snow in very
cold periods) (Werner et al., 2000; Krinner and Werner, 2003).
The different calibrations applicable at different sites and for
different timescales of change underscore the importance of
understanding the processes responsible. At higher accumulation
sites, precipitation is typically dominated by the large atmospheric
cyclonic systems that reach the Antarctic coast. Further inland,
more of the precipitation arrives in the form of ice crystals,
although infrequent penetration of large storms can still bring significant
snowfall. The net effect of these different depositional
processes on the isotope temperature calibration has yet to be
quantified. It is not clear that the fractionation relationships for the
two types of precipitation regime should be the same. It is clear,
for example, that at Law Dome transport, as well as temperature,
plays a considerable role on interannual timescales. On longer
timescales, it is likely that the transport variability diminishes in
importance and temperature correlation increases (van Ommen
and Morgan, personal communication, 2008).
Greenland ice cores
During the past five decades, numerous ice core records have
been retrieved from the Greenland ice sheet, many of them during
the Greenland Ice Sheet Program (GISP) in the 1970s
(Langway et al., 1985) and PARCA in the 1990s. Records covering
the past two millennia stem solely from ice cores drilled to
bedrock during deep drilling campaigns, ie, Camp Century, Dye-3,
GISP2, GRIP, Hans Tausen, NGRIP and Renland records
(Langway, 1967; Dansgaard et al., 1982, 1993; Johnsen et al.,
1992; Alley et al., 1995; Hammer et al., 2001; North Greenland
Ice Core Project (NGRIP) Members, 2004). Of these seven ice
cores, the Dye-3, GRIP and NGRIP cores have been measured at
sufficient resolution to provide annually resolved δ18
O data over
the whole of the past two millennia, and two shallow cores
drilled near the GRIP drill site also provide such data for the past
millennium (Johnsen et al., 1997). The GISP2 core has been
measured with annual resolution back ~1200 years (Grootes and
Stuiver, 1997).
Annually and seasonally resolved δ18
Ο data have been shown to
be closely related to Greenland coastal temperatures (Barlow
et al., 1993; White et al., 1997; Rogers et al., 1998; Vinther et al.,
2003) and a comparison with a new Greenland coastal temperature
record extending back to 1784, showed persistent and highly
significant correlations between winter season δ18
O from seven
Greenland ice cores and winter temperatures in SW Greenland
(Vinther et al., 2006b). A new study based on seasonal data from
multiple Greenland shallow ice cores indicates that winter season
δ18
O can also be used as a proxy for local and coastal annual average
temperatures (Vinther et al., 2008). Indeed winter season δ18
O
seem to be a significantly better proxy than annual average δ18
O
for both coastal Greenland annual average temperatures and local
temperatures reconstructed from borehole temperature profiles
(Vinther et al., 2008). Greenland winter season δ18
O has also been
found to contain a clear NAO signal (Vinther et al., 2003).
Recently the Dye-3, GRIP and NGRIP cores have been crossdated
throughout the Holocene using volcanic reference horizons
as tie-points (Rasmussen et al., 2006; Vinther et al., 2006a). It is
expected that the common dating for the three cores will facilitate
extraction of regional-scale Greenland climatic signals. Seasonal
variations can be resolved for the last two millennia in the Dye-3
and GRIP δ18
O records, while diffusional processes in the top 60
m of the ice sheet destroys the seasonal oscillations at the lower
accumulation NGRIP drill site (Johnsen, 1977; Johnsen et al.,
2000; Vinther et al., 2008). The upcoming North Greenland
Eemian (NEEM) deep drilling project in NE Greenland is
expected to produce at least one additional seasonally resolved δ
record spanning the last two millennia.
Antarctic ice cores
There are only a few long, annually resolved ice core records from
Antarctica dated with better than decadal precision for millennial
or longer timescales. Currently, the longest high-resolution continental
temperature reconstruction from ice cores (Schneider et al.,
2006) draws upon cores at five sites to construct a two-century
record. Of these five sites, the longer records come from Dronning
Maud Land (1000 years; Graf et al., 2002), Law Dome (700 years;
Morgan and van Ommen, 1997) and ITASE cores 2000-01 (~350
years) and 2000-02 (~300 years; Steig et al., 2005). Annually
resolved records are also available from the 302 m core (~550
years) drilled at Siple Station at the base of the Antarctic Peninsula
(Mosley-Thompson et al., 1990) and the upper 190 m (~480
years) of a core from the Dyer Plateau in the Antarctic Peninsula
(Thompson et al., 1994). When compared with the highest resolution
δ histories available at that time, the Siple and Dyer δ records
strongly suggest that the seesaw in temperature between the
Peninsula region and the East Antarctic Plateau evident in the
available meteorological data (1945 to 1985) likely persisted for
the last five centuries (Mosley-Thompson, 1992). Further highresolution
measurements of Law Dome core material (V. Morgan,
personal communication, 2008) will provide annually resolved δ
data back 2000 years.
New cores emerging from the West Antarctic Ice Sheet (WAIS)
Ice Divide deep ice coring site and Berkner Island may potentially
give interannual- to decadally resolved records. Earlier data from
Berkner Island provide annual cycles countable with subdecadal
precision back ~750 years. The accumulation rates at the WAIS and
Berkner sites are not sufficient for preservation of annually resolved
δ, because of diffusion, but trace chemical data may provide dating
(J. McConnell, personal communication, 2008) with high precision.
Lower latitude ice cores
Ice cores have also been recovered from high-elevation ice fields
at lower latitudes. The interpretation of δ histories is challenging
for lower latitude, high-elevation cores, particularly those in the
tropics and subtropics where large seasonal differences dominate
the precipitation regime. Calibration efforts have been limited by
the lack of contemporaneous instrumental temperature climatic
records as most drill sites are in very remote locations. There has
been vigorous discussion in the literature as to whether the δ
records from the highest tropical and subtropical ice caps and glaciers
reflect local temperature or precipitation (via the amount
effect) or large-scale patterns of tropical climate. Rozanski et al.
(1992, 1993, 1997) reviewed in situ data and suggested that to a
first approximation, temperature is the primary control on isotopic
fractionation in the high- to mid-latitudes but the amount of precipitation
is the more dominant process in the tropics. More recent
investigations of the processes controlling δ in snowfall at very
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 15
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
high elevations in low latitudes suggest a more complicated picture
(Thompson and Davis, 2005; Mosley-Thompson et al., 2006).
One of the first efforts to quantify the controls on δ in snowfall
over the Tibetan Plateau was conducted by Yao et al. (1996) who
measured δ in fresh precipitation collected at selected meteorological
stations where ambient temperature and precipitation data
were available. They found that on timescales ranging from a discrete
precipitation event to an average over several months the
trends in δ reflect trends in temperature more strongly than in precipitation
amount at sites in northern Tibet. In southern Tibet
where monsoonal precipitation dominates, they found trends in δ
more closely linked to precipitation amount than temperature. On
annual timescales the δ–T relationship reflected atmospheric
dynamical processes (more depletion in the warm, wet season);
however, on longer timescales atmospheric temperature appeared
to be the more dominant control on δ.
Another process affecting δ in precipitation on the high ice
fields in the Himalayas and the South American Andes is the
intensity of atmospheric convection. Increases in intense convection
result in condensation at higher elevations and hence at colder
temperatures (Thompson et al., 2000). The resulting precipitation
is more isotopically depleted. The interpretation of δ in Andean
precipitation on shorter timescales has been addressed by Grootes
et al. (1989), Henderson et al. (1999), Bradley et al. (2003) and
Vuille et al. (2003); however, no simple interpretation has
emerged. Bradley et al. (2003) reported a strong link between sea
surface temperatures across the equatorial Pacific and the δ histories
in ice cores from the tropical Andes and the Dasuopu Glacier
in the Himalayas. They also reported a strong link between trends
in δ and ENSO variability, which is not surprising as ENSO integrates
all the mechanisms mentioned above. On multicentury to
millennial timescales there is strong evidence that δ is more
strongly controlled by temperature than precipitation (Thompson
et al., 2000, 2003). Even on shorter timescales (multiannual to
centuries) there is abundant evidence that precipitation amount is
not the primary control on δ in precipitation over ice fields in the
tropics and subtropics (see Thompson et al., 2006: figure 5).
However, as with the polar ice cores, a better understanding of the
differential roles these various processes play in controlling δ on
high-elevation, low-latitude ice fields remains an area ripe for further
investigation (eg, Schmidt et al., 2007).
Ice cores offer a wealth of information that extends well beyond
that provided by stable isotopic ratios. They provide records of
regional changes in the climate and environment as well as histories
of larger, global-scale changes. A more complete picture of the
Earth system’s climate history requires the extraction and accurate
interpretation of high-resolution (ideally annual) ice core records
from the polar ice caps to the highest mountains in the tropics and
subtropics. These histories must extend back many centuries to
multiple millennia. Such a collection of ice-core-derived histories
is closer to becoming a reality with the IPICS initiative that calls
for a global array of 2000-yr cores (Brook and Wolff, 2006).
Documentary and early instrumental data
Documentary data include all forms of written historical information
about past climate and the weather. The sources are numerous
and varied ranging from direct human observations (of sea ice,
frozen lakes and rivers or snow lines), through the effects of the
weather on important aspects of crop growth, such as yields and
harvest dates, to paintings of glaciers in the Alps. The sources
encompass all possible aspects of written history that were routinely
noted in diaries and also because it was required for state,
legal and tax reasons. The use of documentary data in past climate
reconstructions is, therefore, restricted to locations where there are
long written histories (such as Europe, China, Japan and Korea,
where some extensive analyses have been undertaken, but also
potentially includes the Islamic World, where little has been
analysed and published). Brázdil et al. (2005) provide an extensive
summary of the type of proxy evidence used, European documentary
sources and available reconstructions. Luterbacher et al.
(2006) complement this with a review of the documentary evidence
from the Mediterranean area. Information from eastern Asia
and Japan is not summarized in a single recent paper, but is generally
split into those looking at China, Japan or Korea (Wang and
Zhao, 1981; Zhang and Crowley, 1989; Song, 1998, 2000;
Mikami, 1999, 2002; Wang et al., 2001; Qian et al., 2002; 2003;
Yang et al., 2002; Ge et al., 2003, 2005, 2008).
Although most analyses have been undertaken in these regions,
there is scope for some geographic expansion using early European
colonial archives (for Iceland and Greenland involving sea-ice
records in particular, Ogilvie, 1992, 1996; Ogilvie and Jónsdóttir,
1996; for the Americas, eg, Bradley and Jones, 1995; Overland and
Wood, 2003; Druckenbrod et al., 2003 for North America and
Quinn and Neal, 1992; Prieto and Herrera, 1999; Prieto et al.,
2000, 2004; Ortlieb, 2000; García-Herrera et al., 2008 for South
America) and in barely accessed archives of the Ottoman Empire
in Turkey (for the Middle East region). Further, records from
Spanish galleons crossing the Pacific Ocean during the sixteenth to
eighteenth centuries provide a description of secular changes in the
wind circulation (García et al., 2001). The systematic abstraction
of naval and merchant ship logbooks has made available a new data
base with daily wind observations over the Atlantic and Indian
Oceans for the period 1750–1850 (eg, Jones and Salmon, 2005;
Wheeler, 2005; García-Herrera et al., 2005a; Gallego et al., 2005)
and for the later decades will provide instrumental measurements
of SSTs and atmospheric pressure. The existence of possible secular
variations in the occurrence of Atlantic hurricanes in the sixteenth
to eighteenth centuries is also suggested from records
obtained from the same sources (García-Herrera et al., 2005b).
There are also a large number of as yet unexplored ship logbooks
(mainly British) that have great potential to improve and extend sea
level pressure reconstructions and SSTs further back in time (eg,
Jones and Salmon, 2005; García-Herrera et al., 2005a).
Arabic documentary sources have been widely used in astronomy
and geophysics to date phenomena such as eclipses and auroras.
However, they have not yet been explored from a climatic
perspective. J.M. Vaquero (University Extremadura, Spain, personal
communication, MedClivar Meeting, Carmona, 2006) provided
several examples of their use to construct series of extreme
events such as droughts and floods in southern Iberia. Significant
collections have been preserved in Spain, frequently translated
into Spanish that could provide the basis for the reconstruction of
extreme events in south-central Iberia for the period AD 750–1200.
The actual availability of these sources in the Mediterranean
Arabic countries is currently unknown.
Documentary sources, by their very nature, will tend to emphasize
extreme events, primarily in the winter and summer seasons
and to a lesser extent spring, as these are the types of records that
are more likely to have been recorded because of their socio-economic
relevance. Records indicating conditions in past autumns
tend to be more rare (Pfister, 1992; Xoplaki et al., 2005).
Luterbacher et al. (2007) endorse this latter conclusion but still
show that the warmth in autumn 2006 in Central Europe was very
likely exceptional within the last 500 years. Until the early 1970s
most documentary evidence was collected into compendiums
about past weather with little regard for whether the reporter was
local or contemporary in time (Brázdil et al., 2005 and references
therein). Since the mid-1970s, much greater rigour has been used,
with long series developed combining a whole array of documentary
sources into indices of past seasonal temperature and precipitation
conditions (eg, Pfister et al., 1998, 1999; Glaser, 2001;
Meier et al., 2007 for Switzerland and Central Europe; van
16 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Engelen et al., 2001 for the Low Countries and Le Roy Ladurie,
2004; Chuine et al., 2004 for France). Similar series have also been
developed for China, Japan and Korea (see earlier references).
Documentary data are the only kind of palaeoclimatic information
based on direct observations of meteorological variables in
terms of narrative descriptions. They are highly temporally
resolved with detailed descriptive precision. Important uses of
documentary data are the verification of extreme values in natural
proxies such as tree rings and for the detailed descriptions of past
weather conditions. Most importantly, they are the only evidence
that is directly related to the socio-economic impacts of rare but
significant disasters in the period prior to the organization of the
network of instrumental observations (eg, Brázdil et al., 2005). In
Europe, the most difficult problem is that the types of historical
sources available prior to the middle of the eighteenth century are
markedly less widely available since that time (Pfister, 1992;
Pfister et al., 1999; Jones et al., 2003; Luterbacher et al., 2004;
Xoplaki et al., 2005; Brázdil et al., 2005 and references therein).
The widespread use of new meteorological instruments at that
time meant the death knell for many types of documentary sources
(eg, diaries) as the observers (who were often scientists or medical
doctors) quickly moved to using the new instruments. The development
of long documentary-based records extending to the present
has led to a variety of approaches to fill the recent gap. The
most common approach is to use degraded instrumental records
for much of the last 200 years (Jones et al., 2003). In Europe, this
issue means that it is difficult to assess the accuracy of the true
documentary part of the long series for an accurate calibration, as
there are only a few areas with short periods with an overlap of
documentary and instrumental data (see Dobrovolný et al., 2008
and references therein). This problem is less of an issue in the Far
East, as here documentary sources continue well into the twentieth
century, particularly in China, and instrumental records only
really begin there in earnest in the late-nineteenth century.
Early instrumental data
Instrumental recording began in Europe in the late-seventeenth
century, but only a few long series exist from this time (see Jones
and Briffa, 2006). In most other parts of the world, the widespread
use of instruments began much later (see eg, Jones and Thompson,
2003). It is often believed that instrumental records only extend
back to the founding of National Meteorological Services
(NMSs), but for most countries there are generally much longer
records available. Sometimes they will take some tracking down,
but there are considerable amounts of early data for Africa, South
America and southern Asia in European colonial archives.
Recently early Japanese records have been found (in Dutch
archives) from the late-eighteenth century, taken by Dutch traders
at Nagasaki (Können et al., 2003) and data for other locations are
beginning to be found in Japanese archives (Zaiki et al., 2006).
Instrumental series are an essential part of palaeoclimatology, as
they provide the basis for assessing the usefulness and reliability of
all proxy records. For almost all sites, though, it is rare for the temperature
and precipitation records to be fully homogeneous over
their full length of record. Much work must generally be undertaken
to ensure that adequate adjustments (for changes in sites,
instruments, recording hours, etc.) are made to all instrumental
records to ensure their long-term homogeneity. Numerous methods
and approaches have been applied (see Jones and Thompson, 2003
and references therein). Because longer instrumental records
enable better proxy calibration and verification exercises, ensuring
the earliest parts of series are correct is vital. The most difficult
aspect of ensuring homogeneity relates to thermometer exposure
issues before the widespread use of variants of Stevenson screens
in the late-nineteenth century, affecting mostly European records.
The issue is particularly important for summer temperatures, which
might be too warm because of the possibility of direct solar influences
on the early thermometers (see the discussion in Moberg
et al., 2003; Böhm et al., 2009). Rebuilding pre-Stevenson screens
(from mid-nineteenth century diagrams) or using data loggers (at
the locations of the earliest sites) has recently enabled the issue to
be addressed in Spain and Sweden (Brunet et al., 2006 and
Klingbjer and Moberg, 2003, respectively), but much more work is
needed. The recent intensive digitization of many of the long daily
European records is providing new insights into long-term homogeneity
by additionally looking at air pressure and cloudiness
observations (Camuffo and Jones, 2002 and Moberg et al., 2003).
New approaches for homogeneity assessment of long daily series
are also important advances (Della-Marta and Wanner, 2006).
Regional climate reconstructions
Temporal and regional high-resolution temperature reconstructions
show important climatic features, such as regionally very hot
or cool summers or autumns and very mild or cold winters or
springs that may be masked in hemispheric or global reconstructions
(eg, Luterbacher et al., 2004, 2007; Brázdil et al., 2005;
Casty et al., 2005, Guiot et al., 2005; Xoplaki et al., 2005). Thus,
regional studies and reconstructions of climate are important when
climate impacts are evaluated. Seasonal extremes at regional-tocontinental
scales, such as the hot European summer of 2003
(Luterbacher et al., 2004), the mild winter of 2006/2007
(Luterbacher et al., 2007) and warm spring of 2007 (Rutishauser
et al., 2008) exhibit much larger amplitudes than extremes at the
global or NH scale, thus they may markedly affect the local-toregional
natural environment and impact societies and economies.
In their large-scale Northern Hemisphere (NH) temperature
reconstructions (MBH98) Mann et al. (2000) pointed out that
there is a decrease in the number of independent patterns of variability
(eg, principal components) that can be reliably reconstructed
prior to 1760. This leads to greater spatial smoothing and
thus to less high-frequency variance at continental spatial scales.
Consequently, to accurately reproduce the spatial pattern within
regions, an extensive and dense proxy network is required to capture
several of the principal components. By combining long
homogenized station temperature series and temperature-derived
indices from documentary evidence as well as a few seasonally
resolved natural proxies, Luterbacher et al. (2004) and Xoplaki
et al. (2005) were able to map seasonally resolved temperature
anomalies across European land areas for individual years back to
1500, along with area-specific error estimates. Multivariate principal
component (PC) regression (later in section ‘Combining
proxies to reconstruct large-scale patterns, continental and hemispheric
averages’ also termed truncated Empirical Orthogonal
Functions, or EOFs, regression) was used for the monthly (back to
1659) and seasonal (four values per year; 1500–1658) European
land surface air-temperature fields (at 0.5° by 0.5° resolution).
This method places a greater weight on specific proxy series that
exhibit a greater affinity with the twentieth-century large-scale
instrumental record (see Luterbacher et al., 2002b for a detailed
description of the methodology). The mapping permits a rigorous
assessment of the spatial coherency of past annual-to-decadal climatic
changes at a subcontinental scale, and also allowed Pauling
et al. (2003) to calculate the ‘best’ predictors of boreal winter and
summer temperatures from the available array of different proxy
climate data for different parts of Europe and the North Atlantic
Ocean (particularly involving documentary evidence from sea-ice
extents north of Iceland). It shows, for instance, as could be intuitively
expected, that tree rings have been a good predictor of past
summer temperatures across northern and central Europe, whereas
documentary sources are more reliable for reconstructing winter
temperatures. Brönnimann et al. (2007) recently investigated the
influence of the ENSO phenomenon on European climate using
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 17
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
these gridded continental-scale reconstructions. After removing
years following tropical volcanic eruptions, they find a consistent
and statistically significant ENSO signal in European climate in
late winter and spring.
Since some of the most important potential consequences
of past climate changes have been linked to changes in
regional/continental circulation patterns (eg, Shindell et al., 2001;
Luterbacher et al., 2002a; Casty et al., 2007), the frequency and
intensity of droughts and floods (eg, Jacobeit et al., 2003;
Mudelsee et al., 2003; Pfister et al., 2006) and hurricane activity
(eg, Gárcia-Herrera et al., 2005b), regional and large-scale reconstructions
of changes in other climatic variables, such as precipitation,
over the last centuries provide a valuable complement to
those for temperature (NRC, 2006). In recent work, Pauling et al.
(2006) used a multiproxy data set including precipitation measurements,
precipitation indices derived from documentary sources
as well as precipitation-sensitive natural proxies to derive
European-scale precipitation maps with associated uncertainties,
back to 1500. Casty et al. (2005) applied multivariate PC regression
to a similar data set as Luterbacher et al. (2004) to reconstruct
seasonal temperature and precipitation variability over the greaterAlpine
area. Both studies found that precipitation at regional and
continental scales is much more difficult to estimate back in time
than temperature, and explained variances are almost always
lower than for temperature.
Other potential high-resolution proxies
The most important feature of the high-resolution proxies already
discussed is that they involve as near to absolute dating as possible.
This is a fundamental aspect of the use of tree rings and documentary
information, where it is ensured through crossdating and
source assessment and replication, respectively. For both ice cores
and corals, there is a growing emphasis in recent years on the
analysis of multiple sample data sets and the use of tree-ring techniques
to improve dating accuracy. Better reconstructions of climate
variability have been achieved, for both these sources, when
data from more than one core have been integrated. Other potential
sources of absolutely dated records include those from varved
sediments (lacustrine and marine) and speleothems. As with ice
cores and corals, as well as other archives assumed to have
absolute time control, extra effort must always be made to confirm
the exact nature of the temporal resolution and accuracy before
using varved sediment and speleothem records for reconstructing
the high-resolution palaeoclimatology of the last millennium.
Varved sediment archives
Over the last couple of decades, quantitative palaeoclimatic reconstructions
from terrestrial (eg, lake and wetland) and marine sediments
have been widely generated and used. An important, and
growing, subset of these reconstructions includes records where
time control is provided by annually laminated, or ‘varved’, sediments.
Varved sediments have been found in lakes on almost all
continents, and from the tropics to high-latitudes, as well as in
some restricted marine basins (eg, the Cariaco Basin and the Santa
Barbara Basin) and high-latitude fjords, where bottom-waters usually
low in oxygen inhibit bioturbation. There are many types of
varved sediment, some being biogenic in origin, others chemical
or physical (ie, sedimentological). In most cases where varves
have palaeoclimatic utility, they are made up of two or more seasonal,
or subseasonal, laminae that can be recognized as together
constituting an annual period of sediment accumulation. Varves
are rarely of use for reconstructing the high-resolution palaeoclimatology
of the last millennium unless both their chronology and
climate sensitivity is well understood.
In terms of chronology, the most important step in using varved
sediments is to document that the sediment laminations are indeed
reliable annual varves. Ideally, this takes the form of three mutually
confirming analyses (eg, Lamoureux, 2001; Shanahan et al.,
2008). First, multiple independent age estimates (eg, from 210
Pb,
14
C, bomb nuclides, tephras, etc.) need to be used to show that the
laminated sediments in question are annually laminated. Second,
it is also helpful when the nature of the sedimentation is well
understood and known to be consistent with a documented varve
formation mechanism. This can be explored in a great variety of
ways depending on the lake or marine basin setting, but usually
involves both a detailed subannual characterization (eg, biology,
chemistry, physical stratigraphy) of the varved sediments, as well
as knowledge of corresponding environmental variability in and
around the lake or basin; the two should be consistent in a welldocumented
fashion. Third, and where possible, a good correspondence
between reconstructed varved-based environmental
(ie, climatic) variability and that observed in an instrumental or
other independent record, should also be documented. Modern
varved sediment chronology building has also evolved to utilize
multiple replicate cores from a single basin, rather than only one
or two, as a way to reduce chronological uncertainty resulting
from lake, sediment and coring processes that can disturb sediments.
Even with such efforts, however, chronological error can
often be as much as 2–5%, perhaps more, so users of varved sediment
data should pay particular attention to making sure that they
do not overestimate temporal accuracy.
Once a varve-based chronology is available, then a wealth of
available sediment-based quantitative climate reconstruction
methods can be used to produce a varved-based palaeoclimate
record. As with chronology, the key is to confirm that the resulting
record represents what it is purported to represent, usually via
a time-series comparison with instrumental or other independent
climatic data. Some approaches to varve-based palaeoclimate
reconstruction use variations in the varves themselves calibrated
against instrumental data (eg, Hughen et al., 2000; Besonen et al.,
2008; Romero-Viana et al., 2008) or in the chemical and isotopic
variations of the sediments calibrated against instrumental data
(eg, Field and Baumgartner, 2000). A wider set of varved sediment
reconstructions have used spatial relationships between
proxy and climate to generate a ‘transfer-function’ analogue or
other means to convert temporal sequences of varved sediment
proxy data into quantitative time series (including errors) of
inferred palaeoclimate. This has been done with pollen data (eg,
Gajewski, 1988), diatoms (eg, Bradbury et al., 2002; Stewart
et al., 2008), foraminifera (eg, Field, D.B. et al., 2006), and other
biotic proxies. With these reconstructions, many of the usual concerns
must be evaluated (eg, Telford and Birks, 2005; Batterbee
et al., 2008), including the possibility that there is a temporal lag
(eg, Overpeck et al., 1990; Overpeck, 1996), or frequency bandpass,
bias between climate variability and the proxy response (eg,
vegetation as reflected by fossil pollen data) to that climate variability.
Single references focused on methods used with varved
sediments tend to evolve quickly given the rapidly evolving
nature of the field, so readers should consult the extensive varved
sediment literature regularly for the latest developments.
Speleothem archives
Speleothems are cave deposits, such as stalactites and stalagmites,
formed when calcium carbonate (usually calcite) precipitates from
degassing solutions seeping into limestone caves. Palaeoclimatic
reconstruction efforts typically focus on stalagmites, whose
growth patterns are more regular than stalactites. Growth rates
vary between ~ 0.05 and 0.4 mm/yr and provided annual bands are
present and well preserved, the combination of annual band counts
and Uranium/Thorium-series (U/Th) dating results in absolute
chronologies, but with age uncertainties (Fleitmann et al., 2004).
Dating errors arise from several sources, including analytical error
18 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
in U/Th dates, uncertainty in corrections for detrital thorium in
most speleothems, and the fact that minor hiatuses in stalagmite
growth often go unresolved. Chronological uncertainties can be
greatly improved by obtaining records from multiple overlapping
stalagmites (via crossdating), and the generation of multiple estimates
of contaminant thorium composition. The thickness of the
bands can be used to reconstruct climate as well as trace metal
concentrations and δ18
O and δ13
C isotopic ratios, with the latter
being the most frequently used climate proxies. Very few studies
(eg, Smith et al., 2006) have focused on the reconstruction of continental-scale
climate variability during the past 500 to 1000 years
using speleothems, mainly because of the difficulty in devising
high-resolution sampling strategies. There are a number of wellknown
reconstructions (Polyak and Asmerom, 2001; Fleitmann
et al., 2003, 2004, 2007; Proctor et al., 2004; Mangini et al., 2005;
Partin et al., 2007), but, at present, dating limitations appear to
preclude reliable comparison with higher-resolution proxies such
as tree-rings, corals and documentary data.
Lower-resolution proxy records
There are a wide range of terrestrial and palaeoclimate archives,
and two types provide the continuous and quantitative records of
use for reconstructing the high-resolution palaeoclimatology of
the last millennium: lake/wetland sediments and temperature–
depth profiles measured in boreholes. Because annually laminated,
varved sediments have the potential for annual, even
seasonal, resolution of both terrestrial and marine palaeoclimate,
these archives are discussed above in the section ‘Other potential
high-resolution proxies’.
Non-varved sediments are common from both lakes and wetlands
from all of the continents of Earth, and thus are extremely
valuable for reconstructing a wide range of terrestrial palaeoenvironmental
information. When it comes to the last millennium,
however, available time control limits the utility of these nonvarved
sediments. On the positive side, the combination of 210
Pb,
bomb-nuclide and other geochronological tools available for dating
the last c. 150 years of sediment accumulation make it possible
to generate proxy records that can be compared directly with
instrumental data for proxy calibration and verification purposes
(see section ‘Other potential high-resolution proxies’ for more
detail). However, below c. 150 years, the need to rely on radiocarbon
and other radiometric dating methods results in temporal
errors that likely exceed 10%, or even 20% in many cases.
Exceptions exist, for example, where volcanic tephra of known
ages can be used to constrain age models, but any use of sediments
for understanding the high-resolution palaeoclimatology of the
last millennium needs to provide clear evidence that the sediment
geochronology is sufficiently accurate, and that bioturbation is not
distorting the sediment record. For example, one way is to demonstrate
that the sediment-based reconstruction is capable of reproducing
climate variations observed in the instrumental record, and
in this case, not just a monotonic trend that is observed in the
instrumental record.
One new approach being attempted employs Bayesian hierarchical
methods that rigorously entrain explicit statistical models of
proxy–climate relationships across multiple-frequency bands (cf.
Ammann et al., 2007a).
Borehole archives
Within the context of millennial climate reconstructions, borehole
profiles have been one source of information that has significantly
contributed to our understanding of centennial temperature
changes. Temperature–depth profiles measured in boreholes contain
a record of temperature changes at the Earth’s surface. Various
approaches to deriving past temperature at local- and hemisphericscales
based on the information recorded in temperature logs have
become well established within the last decade (eg, Huang et al.,
2000; Harris and Chapman, 2001; Beltrami, 2002; Beltrami and
Bourlon, 2004; Pollack and Smerdon, 2004; Gonzalez-Rouco
et al., 2008). Reconstructions have been developed using the large
number of borehole logs and derived-temperature reconstructions
(Huang and Pollack, 1998) that in 2004 included 695 sites in the
NH and 166 in the SH (see Jansen et al., 2007: figure 6.11).
As the solid Earth acts as a low-pass filter to downward propagating
surface temperature signals, borehole reconstructions
lack annual resolution, so they typically portray multidecadal-tocentennial
timescale changes. Temperature reconstruction based
on borehole profiles depends on the assumption that surface air
temperature (SAT) changes are coupled to ground surface temperature
(GST) changes and propagate to the subsurface by thermal
conduction merging with the background geothermal
gradient field. The downward propagation of the temperature disturbance
is a function of the relatively small thermal diffusivity
of rock strata, so that the first few hundred metres below the surface
store the integrated thermal signature of the last millennium
(eg, Gonzalez-Rouco et al., 2008). Long-period events such as
the temperature changes associated with postglacial warming will
still produce significant signal amplitudes at depths greater than
1000 m. A palaeoclimatic reconstruction derived from borehole
temperatures is characterized by a decrease in temporal resolution
as a function of depth and time. Typically, individual climatic
events can be resolved from borehole temperatures only if
their duration was at least 60% of the time since its occurrence,
ie, an event that occurred 500 years before present will be seen as
a single event if it lasted for at least 300 years; otherwise events
are incorporated into the long-term average temperature change
(Gonzalez-Rouco et al., 2008).
Borehole reconstructions have an advantage in that they
arguably do not have to be calibrated against the instrumental
record because temperature itself is measured directly. However,
a disadvantage is that, as discussed above, they measure GST
rather than the desired quantity (SAT), and under certain conditions
there may be substantial differences between the them.
Important quantitative uncertainties exist, but general trends in
these reconstructions are likely robust. Estimates of the magnitude
of recent warming as hemispheric (continental) averages, or averages
over the middle latitudes, are approximately 0.7–0.9°C, from
the mid-nineteenth century to the late-twentieth century (Pollack
and Huang, 2000; Huang et al., 2000; Harris and Chapman, 2001),
which is similar to the increase estimated from the instrumental
record. Some borehole-based reconstructions also indicate an earlier
persistent but smaller warming of roughly 0.3°C from 1500 to
1850 (Pollack and Huang, 2000; Huang et al., 2000). Regional
estimates for the warming for the last 150 years range from 2°C to
4°C for northern Alaska (Lachenbruch and Marshall, 1986;
Pollack and Huang, 2000) to 0.5°C for Australia (Pollack and
Huang, 2000). Estimates for the western and eastern sectors of
North America are rather different, 0.4–0.6°C and 1.0–1.3°C,
respectively (Pollack and Huang, 2000). Such a reconstruction
(for the NH) is similar to the cooler multiproxy reconstructions in
the sixteenth and seventeenth centuries, but sits in the middle of
the multiproxy range in the nineteenth and early-twentieth centuries
(eg, Jansen et al., 2007).
Borehole reconstructions based on the available data base generally
yield somewhat muted estimates of the twentieth-century trend
because of the relatively sparse representation of borehole data
north of 60°N and the fact that many were logged more than 20
years ago. The assumption that the reconstructed GST history is a
good representation of the SAT history has been examined with
both observational data and model studies. The mean annual GST
differs from the mean annual SAT in regions where there is snow
cover and/or seasonal freezing and thawing (eg, Smerdon et al.,
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 19
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
2004; Taylor et al., 2006). The long-term coupling between SAT
and GST has been addressed by simulating both air and soil temperatures
in GCMs. Mann and Schmidt (2003) suggested that GST
estimates during the winter season are biased by seasonal influences
related to changing snow cover, and that less that 50% of the total
spatiotemporal variance in GST is explained by SAT variations during
the cold half of the year. Chapman et al. (2004) contest the
implications this has for recent temperature trends (however, see
Schmidt and Mann, 2004), and some long-term simulations (eg, by
González-Rouco et al., 2003, 2006, using the ECHO-G model) suggests
the possibility that seasonal differences in coupling might be
of little significance over long timescales as long as temperature
trends are similar in different seasons. However, in cases where
there are large seasonal differences in climate trends (a possibility
that remains for the past few centuries, see Mann et al., 2003) such
seasonal bias issues could lead to misleading inferences regarding
long-term SAT trends from indicators of past GST change.
Marine sedimentary archives
Palaeoclimatic reconstructions generated from marine sedimentary
archives offer a unique view of ocean climate change, even
extending to a three-dimensional view of past states of the ocean
through the use of depth transects across ocean basins. The greatest
progress has been achieved in the reconstruction of SSTs based
on proxies that make use of the chemical composition or abundance
of planktonic organisms. Depending on the region of interest,
seasonal temperature estimates can be obtained by using
different proxies (eg, alkenone unsaturation index, faunal assemblages
and Mg/Ca ratio of foraminiferal shells).
The classical approach has been to use the species abundance of
plankton, most commonly planktonic foraminifera and diatoms,
using transfer functions calibrated by statistically comparing the
surface sediment distributions to modern surface ocean hydrography.
A critical assessment of these spatially based transfer functions
(Telford and Birks, 2005) indicates that some early claims of
the accuracy of this approach were overstated. The optimal reproducibility
of transfer function estimates appears to be at best ±1°C
(Telford and Birks, 2005). Progress over the past years has been
particularly good concerning reconstruction of sea-surface temperatures
based the chemical composition of shells of planktonic
organisms (Cronin et al., 2003) or biomarker molecules originating
from specific plankton groups. Depending on the region of
interest and availability of microfossils in the sediments, seasonal
temperature estimates can also be obtained by using different
chemical proxies such as an alkenone unsaturation index or the
Mg/Ca ratio of foraminiferal shells. The uncertainty of the reconstructions
is on the order of ±0.5–1°C for these methods, but the
uncertainty increases with lower temperature and lower salinities.
Examples of high-resolution SST records with a resolution
between 2 and 30 years over recent millennia, based on transfer
functions and geochemical temperature proxies, include surface
layer temperatures and dynamics in the low- and high-latitude
regions of the North Atlantic (eg, Andersson et al., 2003;
Andersen et al., 2004, Black et al., 2007). Also recently published
records indicate an imprint of solar irradiance on sea-surface temperature
in the North Atlantic (Jiang et al., 2005) and upwelling
intensity variations off northwest Africa (McGregor et al., 2007).
High-resolution proxy records of sea-ice cover off Iceland are
published with decadal resolution, comparing well with historical
data (Moros et al., 2006). In general, major century-scale cold
episodes, apparent in Northern Hemisphere multiproxy temperature
reconstructions, are detected in the marine high-latitude North
Atlantic SST reconstructions (eg, a cooling centred around 1450,
cooling in the seventeenth century and the first half of the nineteenth
century, a generally warmer period 1000–1100 and the
twentieth century warming).
Changes in large-scale ocean circulation (eg, in the rate of the
Atlantic Meridional Overturning Circulation, also referred to as
Atlantic Thermohaline Circulation), and the associated poleward
heat transport are of primary interest for a comprehensive assessment
of natural climate variability. Using vertical profiles of
the oxygen-isotope composition of planktonic and benthic
foraminifera it is possible to infer changes in the density structure
of water masses. This approach has been successfully employed to
quantify past variations in Gulf Stream transport through the
Florida Strait during the last millennium (Lund et al., 2006). So
far this approach has only produced records of at best centuryscale
resolution, but it seems feasible to improve on this given
optimal conditions. Associated with the changes in Florida Strait
throughflow were changes in ocean salinity as recorded at multidecadal
resolution by the combination of Mg/Ca palaeothermometry
and δ18
O data. The results show that reduced Florida Strait
throughflow occurred with enhanced salinities of the Gulf Stream
waters (Lund et al., 2006).
Sedimentological proxies based on grain size analyses are being
used to estimate changes in deep-water flow speed. Although the
method awaits calibration to quantify changes in velocity, some
initial results are very promising. A bi-annual record covering the
last 230 years suggests variations in flow speed along the deepwater
pathway in the North Atlantic that parallel changes in the
phase of the North Atlantic Oscillation (Boessenkool et al., 2007).
A new method based on the protactinium–thorium ratio of
foraminifera shells (McManus et al., 2004) holds great promise
for assessing variations in large-scale mixing and overturning of
the ocean, which is of great relevance for oceanic heat transport as
well as the global carbon cycle. So far this method has not been
employed successfully on high-resolution sediments, and there is
a need for improved calibrations and refinement to be able to
detect smaller-scale changes in the overturning circulation than
those of the last deglaciation, and to detect changes occurring on
shorter timescales than a century.
In some areas marine sediments provide records of terrestrial
and atmospheric processes. The Cariaco basin in the waters off
Venezuela contains decadal-scale records of runoff variability
associated with north/south migrations of the Inter Tropical
Convergence Zone, obtained through the titanium content of the
sediments which are related to local riverine inputs (Haug et al.,
2001, 2003). Such records can be linked to terrestrial palaeoclimatic
records, documenting drought patterns in the Caribbean (eg,
Hodell et al., 2005).
In the Mediterranean Sea, considerable attention has been paid
recently to obtaining high-resolution records of SST, salinity and
water chemistry for the Holocene, using new archives such as vermetid
reefs (living reefs, with data spanning 500–600 years and
with resolution every 30–50 years), non-tropical corals (ie, living
Cladocora caespitosa for the last 100–150 years, eg, Silenzi et al.,
2004, 2005; Montagna et al., 2006 and references therein) and
deep-sea corals (ie, living Desmophyllum dianthus, Lophelia pertusa,
Madrepora oculata for last 100–150 years). Vermetids
are thermophile and sessile gastropods living in intertidal or
shallow subtidal zones, forming dense aggregates of colonial individuals.
These new archives potentially complement and improve
the information derived from the main climate indicators
such as foraminifera, alkenones, dinoflagellate cysts, calcareous
nanoplankton and, especially for the Mediterranean Sea, serpulid
overgrowth on submerged speleothems (Antonioli et al., 2001).
All these latter marine markers enable longer palaeoclimate reconstructions
but with a much coarser resolution (usually lower than
100–200 years). However, in areas of extremely high sedimentation
rates (>80 cm/ka), such as in the southern Levantine Basin, a
resolution of 40–50 years may be possible during the last millennium
(Schilman et al., 2001).
20 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Combining marine sedimentary proxy records with reconstructions
from tree rings, ice cores and corals can be very challenging
because of the uncertainty associated with the dating of the marine
records. The dating error is generally significantly larger than for the
other types of proxy archives. Nevertheless, significant progress has
been made using well-dated ash layers and 210
Pb dating for the most
recent time span as well as improved marine radiocarbon chronologies.
Future developments in producing records of the short-term
variations in the radiocarbon reservoir age of surface waters will
help improve the situation in coming years (eg, using schlerochronology,
Scourse et al., 2006). Owing to the dynamical nature
of the ocean surface and its variable seasonal stratification, major
uncertainties remain in terms of ascribing the proxy data to the correct
season and water depth when the carrier signal was derived. As
an example, studies of Holocene high-resolution records from the
same core have revealed major differences in how the proxies
responded to the increased summer insolation and seasonal contrasts
of the early Holocene. Some proxies, recording the temperature
during the summer season of the euphotic zone (typically the
upper 50 m) document strong warming in response to increases in
insolation, whereas proxies that derive their chemical signature
deeper in the surface layer did not respond to the insolation changes
(Jansen et al., 2008). Thus improved knowledge about the inclusion
of the climatic signal into the carrier organisms is required. On the
other hand, the differences point to improved possibilities to capture
seasonal differences using multiproxy approaches.
At present only a small number of marine sedimentary archives
have been retrieved that allow potential interannual resolution. An
increasing number of records with temporal control that are adequate
to assess multidecadal-to-century scale variations have been
produced and concerted efforts to further improve this situation are
underway in the marine palaeoclimate community. Detailed surveys
of the seafloor in recent years have revealed many potential
sites, where such high-resolution sedimentary archives could be
obtained. Acquiring such archives would be a major step towards
better integration of marine sedimentary proxy data with other
proxy data. To reduce the uncertainties associated with marine
proxies it will be necessary to conduct open-ocean culturing experiments
to optimize the calibration of palaeoenvironmental proxies.
A co-ordinated effort of the palaeoceanographic community, therefore,
could generate a marine equivalent to the land-based temperature
reconstructions for the last few millennia, developing much
improved records for integration with land-based records.
Combining proxies to reconstruct
large-scale patterns, continental
and hemispheric averages
The various approaches to combination
Most previous proxy-based reconstructions of large-scale climate
have targeted hemispheric or global average temperature, though
some studies have also attempted to reconstruct the underlying
spatial patterns of past surface temperature changes at global (eg,
MBH98/99; Rutherford et al., 2005; Wahl and Ammann, 2007;
Mann et al., 2007) and regional (eg, Fritts et al., 1979; Briffa
et al., 1988, 1994; Luterbacher et al., 2004, 2006, 2007; Casty
et al., 2005; 2007; Xoplaki et al., 2005; Meier et al., 2007;
Riedwyl et al., 2008a) scales, and other fields such as precipitation
(Casty et al., 2005; Luterbacher et al., 2006; Pauling et al., 2006),
Palmer Drought Severity Index (PDSI; eg, Cook et al., 2004b,
2007; Nicault et al., 2008), sea level pressure (Briffa et al., 1986;
Luterbacher et al., 2002b) and SST (eg, Evans et al., 2002; Wilson
et al., 2006). Still other studies have focused on the reconstruction
of particular climate indices that represent coherent large-scale
patterns of variability, such as the North Atlantic Oscillation
(NAO) and the related Northern Annular Mode (NAM), the
Pacific Decadal Oscillation (PDO), the Atlantic Multidecadal
Oscillation (AMO), the Southern Annular Mode (SAM) and
indices that describe the variability of the ENSO phenomenon. An
extensive review is provided by Jones and Mann (2004).
A comparison of methods: CPS and CFR
Most reconstructions of large-scale mean climate (eg, hemispheric-average
temperature) or of individual climate indices have
employed the ‘Composite Plus Scaling’ (CPS) method. A selection
of climate proxy records are first standardized, then averaged
(composited) and finally centred and scaled to provide an estimate
of the temperature series for the region or hemisphere represented
by the proxy records (the ‘target’ time series).
The CPS method has been implemented in a variety of ways,
contributing to the wide range of possible reconstructions. The
proxy-selection process may make use of information about the
correlation between individual proxy records and their local climate
(eg, Briffa et al., 2001; Mann and Jones, 2003; Osborn and
Briffa, 2006), or might be based on more general prior expectations
of the climate signal contained within the candidate proxy records
(eg, season and climate variable they represent) or of their retention
of low-frequency variability (eg, Esper et al., 2002). The averaging
stage may be unweighted (eg, Jones et al., 1998) or weighted, with
weights determined according to some intrinsic assessment of reliability
(eg, Briffa et al., 2001), according to an assumed area of
representation (eg, Esper et al., 2002; Mann and Jones, 2003) or in
some cases determined to give a best-fit to the target climate record
(eg, the inverse regression results of Juckes et al., 2007). In the latter
case, the CPS method may have some parallels with the climate
field reconstruction (CFR) methods discussed below, because in
some CFR techniques (eg, Briffa et al., 1986), the reconstructed
fields (and, therefore, their spatial average) can be expressed as a
weighted sum of the proxy records used in the reconstruction. A
worthwhile future endeavour would be to examine the relationships
between individual proxy records and large-scale climate patterns
that arise in the context of CFR-based approaches.
After forming the (weighted or unweighted) average of the individual
proxies, the composite time series is centred and scaled to
achieve a quantitative estimate of the target climate series in
appropriate physical units. The composite time series is centred by
the addition of a constant, chosen so that (over a defined period
during which they overlap) the time-mean of the composite time
series is equal to the time-mean of the instrumental target series.
Finally, the centred composite time series is multiplied by a scaling
coefficient. The scaling coefficient has been determined in a
variety of ways, including (i) regression of the proxy composite
onto the temperature (eg, Briffa and Osborn, 2002), (ii) regression
of the temperature onto the proxy composite (eg, the inverse
regression results of Juckes et al., 2007), (iii) matching the variance
of the composite to that of the temperature record (eg, Jones
et al., 1998; Crowley and Lowery, 2000; Mann and Jones, 2003;
Moberg et al., 2005; D’Arrigo et al., 2006b; among others) or,
more recently, (iv) by using total least squares regression and
apportioning the regression error between both the temperature
and proxy records (Hegerl et al., 2007). Another potentially useful
extension to the CPS method has been to combine high-pass
filtered tree-ring data with low-pass filtered information from
low-resolution (decadal- or centennial-scale) proxies that might be
better-suited to reconstructing low-frequency climate variability
(Moberg et al., 2005). This latter method does not, however, overcome
the limitation that low-resolution information cannot be
calibrated by temporal regression against the relatively short
instrumental temperature record. By applying the same scaling
coefficient to the filtered low-resolution composite as to the
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 21
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
high-pass filtered tree-ring data, Moberg et al. (2005) implicitly
rely on the standardizing process to ensure that the low-frequency
component is not artificially inflated relative to the higher-frequency
component (Mann et al., 2005).
An alternative to CPS is the climate field reconstruction (CFR)
approach, which assimilates proxy records into reconstructions of
the underlying spatial patterns of past climate change (eg, MBH98;
Luterbacher et al., 2002a, 2004; Casty et al., 2005, 2007;
Rutherford et al., 2005; Xoplaki et al., 2005; Pauling et al., 2006;
Ammann and Wahl, 2007; Mann et al., 2007; Riedwyl et al.,
2008a,b). A number of statistical methods have been used for CFR.
All of these methods make use of the covariance between proxy
series and instrumental series during a calibration period, but there
are key differences in how they use covariance within the proxy
data set or within the instrumental data set to combine records into
a smaller number of variables. Some examples are as follows.
(1) Luterbacher et al. (2004) implement separate multiple regression
equations between each leading principal component
(PC) time series of the proxy records and all the leading PC
time series of the instrumental data, thus explicitly making use
of covariance within the separate proxy and instrumental data
sets. This approach has been given various names, including
principal components regression, multivariate truncated-EOF
regression, or orthogonal spatial regression (Cook et al.,
1994). The climate field is then reconstructed by multiplying
each reconstructed instrumental PC time series (or RPC) by
the corresponding instrumental data EOF pattern, and then
summing the resulting patterns (termed ‘EOF expansion’).
(2) The core of the MBH98 method is based on separate multiple
regressions between each ‘individual’ proxy record and all
the retained PCs of the instrumental data. Covariance within
the proxy data set is utilized, both because some ‘individual’
proxy records are actually principal components of regional
tree-ring chronology networks and because the least squares
simultaneous solution of these multiple regressions implicitly
depends on covariability between proxy records. Once the
RPCs are obtained in this method, the climate field is reconstructed
from the instrumental EOF expansion as in (1) above.
(3) Cook et al. (2004b) use covariance within the tree-ring data to
define PCs of the proxy data, but then use these in separate
multiple regressions for each grid-box in the instrumental climate
domain (the point-by-point regression approach).
(4) The Regularized Expectation Maximization (RegEM,
Schneider, 2001) method makes use of covariance within the
proxy data, within the instrumental data and between proxy
and instrumental data, in the form of regression equations
between all proxy and instrumental variables considered. The
regression equations are fitted to the available data, taking
into account the covariance between variables and the estimated
covariance of the regression residuals. The regressions
are then used to reconstruct the earlier climate variables, and
then the procedure is iterated until convergence. The regularization
step is necessary to avoid overfitting these regression
equations. Rutherford et al. (2005) used ridge regression to
regularize the algorithm; effectively this downweights variability
that is present in relatively few series compared with
variations that are coherent across many series. Mann et al.
(2007) regularize the algorithm by substituting all proxy and
instrumental series by the leading PCs of the combined proxy
and instrumental data set, truncating to exclude less important
PCs. In this latter implementation of RegEM, the climate field
is not directly reconstructed, rather the RegEM algorithm is
used to determine the RPCs of the climate field. Once the
RPCs are obtained, the climate field is reconstructed from the
instrumental EOF expansion as in (1) above.
Compared with CPS reconstructions, CFRs allow a more detailed
investigation of past climate by highlighting spatial differences,
thereby also approaching the socially relevant local-to-regional
scale. Hemispheric or global averages, as well as any climate
indices of interest, can then be computed directly from the reconstructions
of the underlying spatial field. Another potential advantage
of the CFR approach is that the reconstructed spatial patterns
can be directly compared with model-predicted patterns of climate
responses to forcing (Shindell et al., 2001, 2004; Waple et al.,
2002), provided that the details that discriminate different patterns
can be reconstructed with sufficient skill. In this context, CFR methods
are used to validate model simulations while model simulations
are used as testing ground for the robustness and reliability of reconstructions
(pseudo-proxy experiments, see below).
CFR methods make use of both the local proxy–climate relationship
and non-local climate teleconnections by relating the
long-term proxy climate data to the temporal variations (PCs) in
the large-scale patterns of the spatial field of interest (generally
defined as the empirical orthogonal functions, or EOFs, of that
field). Examples of these teleconnected, or non-local, relationships
include the close tie between drought (and therefore
drought-sensitive tree rings) over the western USA and SST patterns
in the Pacific Ocean (eg, Stahle et al., 1998), or ice core
records from Greenland which are closely tied to the behaviour of
the NAO (Appenzeller et al., 1998; Vinther et al., 2003, 2006b),
which influences cold-season temperature patterns over North
America and Eurasia. The CFR approach takes fuller advantage of
the potential climate information in the proxy data set by exploiting
these relationships. Investigations using synthetic proxy data
(so called ‘pseudo-proxies’), suggest that these assumptions are
reasonable for the range of variability inferred over the past one or
two millennia (see section ‘Introduction to the pseudo-proxy
approach’). A more severe restriction of CFR (and also of the CPS
approach) is, however, the usually rapidly decreasing number of
proxies prior to the instrumental period (~1850), limiting the ability
of this approach to properly capture the entire climate field in
earlier years. A detailed discussion of this issue is also provided in
the section ‘Experiments with pseudo-proxies’.
For both CPS and CFR methods, the utility of the reconstructions
for, eg, detecting unusual climate change or comparing with
climate model simulations (Collins et al., 2002), is greatly
enhanced by the estimation of reconstruction uncertainties. The
magnitude of the uncertainties will generally vary with timescale,
although this is not always explicitly considered (exceptions are
the examination of the power spectrum of calibration residuals by
MBH99, and the presentation of uncertainties for different
timescales in figure 5 of Briffa et al., 2001). Often the uncertainties
are estimated from the residuals between the actual and reconstructed
values during the calibration period, though this will
likely be an underestimate of real uncertainties because (i) uncertainties
in the scaling coefficient (for CPS) or in the linear model
coefficients (for CPS with individual proxy weights determined
by regression against the climate data and for CFR methods) will
inflate the error outside of the calibration period (or, equivalently,
there will be a tendency to over-optimize the linear model coefficients
by using chance correlations between proxy errors and climate
data, thus artificially reducing the magnitude of the
residuals); and (ii) there may be some long-term inhomogeneities
in the proxy records or deterioration in the proxy reliability that
cannot be identified within the calibration period. The former
source of additional error can be estimated by using known properties
of the calibration algorithm (such as standard formulae for
estimating the standard error of regression coefficients – used, eg,
by Briffa et al., 2002a) though pseudo-proxy studies (see below)
can provide an alternative estimate of the total calibration
error and should be used more in future work. Both sources of
22 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
additional error can, at least partly, be addressed by derived error
estimates from residuals over an independent verification period
or cross-validation experiments (eg, Rutherford et al., 2005; Mann
et al., 2007). Finally, uncertainties can also be underestimated if
prior analysis steps – such as the selection of proxy records – have
already used the calibration (or verification) climate data (Bürger,
2007; Osborn and Briffa, 2007).
Introduction to the pseudo-proxy approach
Experiments involving synthetic climate proxy records, or
‘pseudo-proxies’, derived from climate model simulations (or, in
some cases, from instrumental climate data) can be used to investigate
many different aspects of reconstruction sensitivity, such as
the impact of different proxy networks (not just temporal and spatial
coverage – eg, Mann and Rutherford, 2002 and Rutherford
et al., 2003 – but also type of climate proxy – eg, Zorita and
González-Rouco, 2002 and Küttel et al., 2007), different levels of
individual proxy errors (eg, Mann et al., 2007; Lee et al., 2008),
and the behaviour/performance of different reconstruction algorithms
(eg, Bürger et al., 2006; Ammann and Wahl, 2007; Lee
et al., 2008; Riedwyl et al., 2008a,b; von Storch et al., 2008) at different
timescales and for different sets of calibration data (eg,
Mann et al., 2007; Lee et al., 2008). The pseudo-proxy approach
selects only a limited sample of grid boxes from the complete
global fields of relevant GCM integrations (principally those incorporating
all important forcings), often according to the location of
real proxy data, and degrades these grid-box time series by the
addition of noise, chosen to represent the unknown error in real
proxy data. In all the above cases, multiple repetitions may be
made with different realizations of random noise or even of random
pseudo-proxy selection. In each case the reconstruction target
is exactly known (being the model-generated full climate field or
the subset/average that is of interest). A much more detailed and
systematic investigation into the potential performance of climate
reconstructions can, therefore, be undertaken using the pseudoproxy
approach than can be obtained by simple calibration and verification
tests with real-world proxy and instrumental climate data.
Apparent success within the context of individual and specific
pseudo-proxy tests does not, of course, guarantee the reliability of
a particular method in all reasonable situations, including the realworld
situation. Two particular limitations of the pseudo-proxy
approach are that (i) our knowledge of the characteristics of real
proxy records, including a full understanding of their response to
climate variations and a full statistical model of their error, is still
incomplete, which prevents the generation of completely realistic
pseudo-proxies; and (ii) the climate model simulations may not
replicate real-world behaviour, either because of model deficiencies
or because of unrealistic forcing or initialization (Osborn
et al., 2006). The former limitation has been partly addressed by
using a wide range of noise models to generate multiple sets of
pseudo-proxies mimicking the specific temporal characteristics of
real world proxies (Mann et al., 2007; Moberg et al., 2008;
Riedwyl et al., 2008a,b). The latter limitation is being partly
addressed by using simulated data from more than one climate
model (eg, von Storch et al., 2004; Küttel et al., 2007; Mann et al.,
2007; Lee et al., 2008; Riedwyl et al., 2008a,b), though it has also
been argued (Osborn and Briffa, 2004) that even a partially unrealistic
model simulation is still a useful test-bed for pseudo-proxy
climate reconstructions – ie, a well-behaved reconstruction algorithm
should work in a relatively wide range of plausible situations.
The latter argument is less applicable if particular modes of
climate variation, such as ENSO, are of key importance to a particular
study and are poorly simulated.
The pseudo-proxy approach can, subject to these limitations,
provide useful insight into the relative performance of various
reconstruction techniques, under different conditions such as
calibration period and proxy quality. A current initiative of the
PAGES/CLIVAR Intersection is a ‘Paleoclimate Reconstruction
Challenge’ (Ammann, 2008; see also http://www.pages.unibe.
ch/science/prchallenge/index.html), which will allow multiple
groups to test different algorithms within the same pseudoproxy
setting. In particular, the core activity will involve a doubleblind
experiment, where those attempting to reconstruct the simulated
climate will be provided with pseudo-proxy and pseudoinstrumental
data, but will be unaware of the actual simulated climate
or the actual errors imposed on the pseudo-proxy data. This
mimics the information available when attempting real-world
climate reconstructions.
Before considering some specific results of recent pseudoproxy
studies, it is worth emphasizing that, while they can provide
important insights, they can never prove the veracity of a particular
real-world reconstruction; in particular, their results are
dependent on the quality of the pseudo proxies and the structure of
noise and bias in real-world proxies cannot be determined with
full confidence.
Experiments with pseudo-proxies
Pseudo-proxy intercomparison of multiple
reconstruction methods
Pseudo-proxy experiments have been used to test the performance
of both the CPS (Mann et al., 2005; Ammann and Wahl, 2007) and
CFR methods (Mann and Rutherford, 2002; Rutherford et al.,
2003; Zorita et al., 2003; von Storch et al., 2004; Mann et al.,
2005; Bürger et al., 2006; Ammann and Wahl, 2007; Smerdon and
Kaplan, 2007; Küttel et al., 2007; Mann et al., 2007; Riedwyl et al.,
2008b), though in many cases only a single reconstruction method
is considered (sometimes with a range of modifications). Lee et al.
(2008) compared the skill of several different CFR and CPS reconstruction
methods using millennial climate model simulations.
Their intercomparison included evaluation of CPS techniques
using simple variance matching, forward and inverse regression,
total least squares regression, and an elaboration of these techniques
based on a state-space time series model. The Lee et al.
(2008) evaluation also considered two CFR reconstruction
approaches: MBH98 and RegEM (Schneider, 2001; Rutherford
et al., 2005). Note, however, that regularization in RegEM was performed
by means of ridge regression (Rutherford et al., 2005),
rather than the truncated total least squares (TTLS) regression that
is used in Mann et al. (2007). Riedwyl et al. (2008b) compare multivariate
PC regression (used by Luterbacher et al., 2004, see also
section ‘Regional climate reconstructions’) and RegEM (with
TTLS regression) to reconstruct European summer and winter surface
air temperature over the past millennium. They use pseudoproxy
data derived from two climate model simulations (ECHO-G
4 and NCAR CSM 1.4, von Storch et al., 2004 and Ammann et al.,
2007b, respectively) to test the sensitivity and performance of the
two CFR techniques in a specific virtual experimental set-up.
Lee et al. (2008) assessed reconstruction technique performance
by constructing pseudo-proxies from these two long climate
model simulations and then comparing the reconstructed NH average
temperatures with the known values simulated by the climate
models, forced with reconstructions of historical anthropogenic
and natural external forcing. Tests considered randomly configured
15- and 100-location pseudo-proxy networks selected from
the NH locations for which proxies are currently available, so that
conclusions as to the relative performance of the different techniques
would be independent of decisions made by the practitioner
regarding the proxy-network configuration. Pseudo-proxies
were constructed using two signal-to-noise ratios (SNRs), and
either white noise or mildly persistent red noise. In addition, the
study also varied the length of the calibration period somewhat,
considered the effect of smoothing the proxies and target temperP.D.
Jones et al.: High-resolution palaeoclimatology of the last millennium 23
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
atures prior to calibration, and considered the effect of including a
detrending step prior to calibration. Performance was assessed
mostly using decadally smoothed series, though some assessment
at annual timescales was undertaken.
Figure 1 displays some of the results obtained by Lee et al.
(2008). The ranges and medians of root-mean-squared reconstruction
errors of hemispheric temperatures from multiple pseudoproxy
realizations, using white-noise (two different SNRs)
pseudo-proxies are shown. Figure 1 shows that conclusions
regarding the relative performance of the different techniques are
not affected by the SNR, size of network or particular choice of
millennial climate simulation. Conclusions were also not affected
by the length of the calibration period, by the use of red-noise
rather than white-noise proxies or if detrending was undertaken
prior to calibration (not shown here, but see Lee et al., 2008),
although in the case of the latter in particular, the reconstruction
quality was reduced in many cases, especially for the MBH
method (this issue was also considered by von Storch et al., 2006;
Wahl et al., 2006; and Mann et al., 2007).
The state–space model and RegEM approaches provided better
reconstructions at the annual timescale and for the Northern
Hemisphere, than the other techniques evaluated by Lee et al.
(2008), but most methods provided satisfactory and similar results
at decadal timescales. Exceptions were the MBH technique, and the
non-smoothed forward regression and non-smoothed variance
matching methods. The results comparing multivariate PC regression
(also referred to as forward multivariate truncated-EOF regression)
and RegEM (implemented with TTLS as described by Mann
et al., 2007, see section ‘A comparison of methods: CPS and CFR’)
for European summer and winter average temperature reconstructions
(Riedwyl et al., 2008a,b) show that more skilful results are
generally achieved with RegEM, as low frequency variability is better
preserved, particularly for summer. However, RegEM showed a
noted tendency to introduce strong low-frequency artefacts in the
winter average temperature reconstructions at the lowest signal-towhite
noise ratios and at the strongest red noise additions examined.
Focusing on reconstruction results of the whole climate field, differences
between the two CFR techniques are very small and barely
detectable (Riedwyl et al., 2008a,b). Note, however, that these are
general results representative of performance for arbitrarily selected
proxy networks. The pseudo-proxy network in Riedwyl et al.
(2008b) is chosen according to site locations of published European
data and series that will be potentially available from current
European research projects. Riedwyl et al. (2008b) argue that if the
techniques already fail, using input data covering the full length of
1000 years (as is assumed in the paper for the pseudo-proxies at
these locations), it is certain they will when the number and spatial
distribution are reduced and change through time. In this sense, performance
for a specific fixed proxy network, such as that used by
MBH, was not considered (see section ‘Pseudo-proxy experiments
to test single time series reconstructions’ and also Wahl and
Ammann, 2007 and Ammann and Wahl, 2007, for discussion of the
robustness of the MBH reconstruction). When analysed with the
decadally smoothed real-world palaeoclimate proxy data from
Hegerl et al. (2007), all CPS methods considered provided almost
identical results. This similarity in performance suggests that the
difference between many real-world reconstructions are more likely
to be due to the choice of the proxy series, their quality, spatial and
temporal distribution, or the use of different target seasons or latitudes
than to the choice of statistical reconstruction method (see
also Rutherford et al., 2005; Juckes et al., 2007).
A further extension and optimization of the CPS approach considered
by Lee et al. (2008) is the use of a Kalman filter and
smoother algorithm to incorporate a state equation into the reconstruction
paradigm. This approach combines the ‘observing equation’,
that relates the observed proxy composite to the hemispheric
mean temperature with a ‘state equation’ that describes the
evolution of the hemispheric mean temperature over time. The latter
can be either a simple autoregressive time series model describing
the temporal persistence of the hemispheric mean temperature,
or something more complex that describes the dynamics of the
hemispheric mean and how it responds to external forcing. The
advantage of this approach is that it allows explicit representation of
the source and nature of the error in the observed proxy composite,
together with a representation of the temporal behaviour of the
reconstruction target. A potential disadvantage is that reconstructions
that incorporate information about the response to estimated
forcing histories cannot be considered as truly independent for the
purposes of assessing climate model simulations that have been
driven by the same (or similar) forcings. Both state-space model
approaches provide better reconstructions than existing CPS methods
at annual timescales, though differences are insignificant at
longer timescales. Lee et al. (2008) show that the state-space model
approach does not produce substantially different reconstructions
when the estimated responses to prescribed external forcing time
series are included as extra information (see, for example, Figure 1).
Including forcing response terms in the state-space model does,
however, allow a simultaneous reconstruction and detection analysis,
and also enables an extension that allows forecasts of future climates
(given prescribed forcing). Consistent with the results of
Hegerl et al. (2003, 2007), Lee et al. (2008) detect the effects of
anthropogenic forcing (greenhouse gas and aerosol forcing combined)
and volcanic forcing in real-world palaeoclimatic proxy data.
Pseudo-proxy experiments to test single time series
reconstructions
There remains a need for a more systematic exploration of the performance
of reconstruction methods under a much wider spread of
conditions than has yet been attempted. To further this aim, we outline
here a sample protocol for organizing the wide variety of
pseudo-proxy tests to systematically investigate the potential
performance of climate reconstructions, and then illustrate this protocol
with example applications. Our example applications specifically
examine the performance of climate reconstruction methods
given a fixed proxy network. The test protocol nests the experiments
in a hierarchy, starting with temporal orderings, and progressing
to calibration periods and noise levels in turn. This
hierarchy was developed as an example of a systematic way to
organize testing of the kinds of parameters that have been studied
so far in pseudo-proxy experiments, and to provide a structure for
future studies. The protocol is not intended to be prescriptive: modifications
and extensions are anticipated as research progresses.
Separate, though very similarly structured, protocols are outlined
for testing single time series reconstructions and for testing
whole-field reconstructions. The reason for separating tests of single
time series from tests of entire fields is to allow a more engineering-oriented
testing approach to be taken in the single series
case (for both CPS and the spatial-mean of CFR approaches).
Engineering-oriented testing is designed to include extreme ‘endmember’
situations that enable examination of reconstruction performance
under both optimal and worst-case conditions beyond
what would occur in actual applications. What such end-member
cases might be is more easily defined in the single series case. The
example protocol for such tests is shown in Figure 2.
This protocol can be used to evaluate many aspects of climate
reconstructions; here we illustrate its application by focusing on
one crucial issue – how sensitive are reconstruction methods to the
extent to which the full climatological range of a variable being
reconstructed is represented in the calibration period. In a generalized
regression context it is well-established that forecasting (or
‘hindcasting’ in the case of climate reconstructions) is less wellconstrained
when the predictor values lie outside the range of data
24 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
employed in calibration. One way to test this sensitivity is to
explore specific ‘end-member’ sets of model simulated climates.
Such test sets would use re-orderings of model output by year (or
other time step) that proceed from lowest values to highest, from
highest values to lowest, and including both highest and lowest
values in the calibration period (cf. Figure 2). Re-ordering the
model output will severely modify climate persistence and change
over multiple years, but few CPS and CFR methods in fact make
use of such information. Methods that work with (for example)
decadal-mean values could still be tested, by re-ordering the
decadally averaged model output. Obviously, data re-ordering
does not affect the spatial relationships across all model grid boxes
within each year or other time step. It should be noted that some
reorderings of data will (by design) artificially diminish the range
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 25
Figure 1 Relative root mean squared error (RRMSE) of the reconstruction error, expressed relative to the variability of the simulated hemispheric
temperature. The median RRMSE from multiple realizations of randomly selected pseudo-proxy networks (100 repetitions for the CPS techniques,
40 for the CFR techniques because of the greater computational cost involved) is indicated with horizontal bars and the estimated 5–95% range of
the RRMSEs is shown with vertical lines. Results using the 1860–1970 calibration period with different SNRs and varying number of pseudo-proxies
are shown for two millennial climate model simulations: (a) CSM and (b) ECHO-G. From Lee et al. (2008: figure 2)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
of data variation during the calibration period, thus introducing a
far larger extrapolation error than is likely to be encountered in the
real world. Results of ‘end member’ experiments must therefore
be carefully interpreted in that context.
Examinations of this kind (Figure 3a,b and further experiments
not shown) demonstrate that both the truncated-EOF CFR method
used by MBH98 (employing inverse regression) and the simple
regression methods that have been widely used by many other
palaeoclimate researchers exhibit some degree of sensitivity to the
range of climatological information used during calibration. The
MBH98 method performs very well in reconstructing NH temperature
(Figure 3b) when the full climatological range over the reconstruction
period is represented in the calibration period, whereas it
systematically loses amplitude the further the range of temperature
in the hindcast period moves outside the range occurring within the
calibration period (Figure 3a). (Note that a simple CPS reconstruction,
not shown, exhibits no systematic loss in this situation.) Other
model-based testing in which significant portions of the full climatological
range occur outside the calibration period show qualitatively
similar results for this version of CFR methodology (von
Storch et al., 2006, version without trend removal; Bürger et al.,
2006, ‘MBH98 analogue’ version). Bürger et al. (2006) show a
variant of PC regression that has much less amplitude loss when
significant portions of the full climatological range occur outside
the calibration period (their version ‘11110’); note that this variant
predicts the mean NH temperature directly rather than the full climate
field, using an inverse regression of the pseudo-proxy data on
PCs and might be more sensitive to the particular proxy records
used than other methods (Juckes et al., 2007; their inverse regression
method shows such sensitivity and is similar, though not identical,
to the Bürger et al., 2006, version ‘11110’).
Küttel et al. (2007) also find a reduced amplitude reconstruction
for the multivariate PC regression CFR method (cf. Luterbacher
et al., 2004), which has been employed extensively for European
regional reconstructions (section ‘Regional climate reconstructions’).
This loss of amplitude was clearly found to be related to
the insufficient predictor network prior to ~1750 and to a much
lesser degree to the noise added. Riedwyl et al. (2008b) find that
26 The Holocene 19,1 (2009)
Protocol to Test Climate Reconstructions in
Climate Model Experimental Framework
Single Series Reconstructions
(in full Monte Carlo framework for proxies – potentially also for instrumental data)
Level 1 – By Cases
Extreme Cases
Highest-to-Lowest Rank Order (from
series end)
Lowest-to-Highest Rank Order (from
series end)
Highest and Lowest In Calibration
Level 1 – By Cases
Model “As Is” Case
“Custom” Orderings
Mann/Bradley/Hughes Rank Order
Other Orderings as Specified
-- cf. Osborn/Briffa, 2006
Level 2 – By Calibration Period *
20th century
19th century
“Custom” periods
(e.g., Maunder Minimum)
* Replicate all or chosen Level 1 experiments
Level 3 – By Noise Type & Level *
White Noise
Range of S/N ratios
* Replicate all or chosen Level 1 & 2
experiments
Level 3 – By Noise Type & Level *
Red Noise / Blue Noise
Range of S/N ratios
-- Range of rho values
* Replicate all or chosen Level 1 & 2
experiments
Special Cases –
e.g., by Frequency Band *
Specify filter band-pass level
* Replicate all or chosen Level 1, 2 & 3 experiments
Figure 2 Protocol for testing single-series climate reconstructions in model-based pseudo-proxy experiments
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 27
Figure3Sixexamplesofsingle-seriespseudo-proxyexperiments,illustratinghowreconstructionperformancedependsontherangeoftemperaturevariationwithinthecalibrationperiod,calibrationperiodlength
andspatialcoverage.Ineachexample,theblacklineistheactualmodel-simulatedNHtemperaturetimeseriesofthegridcellsusedforcalibrationandtheredlineisapseudo-proxyreconstructionusingtheWahl
andAmmann(2007)emulationoftheMBH98CFRmethod.Full1000-yrannualtimeseriesand30-yrsmoothedtimeseriesareshown:thereference(‘zero’)lineinallgraphsisthe1900–1980mean.Inallexam-
ples,simulateddataarefromtheNCAR-CSMsimulationofAmmannetal.(2007b),pseudo-proxiesaresampledfromthegridboxesandclimatevariablesthatmatchtheMBH981820–1980proxynetworkand
havewhitenoiseaddedtogiveSNR=0.25(intermsofrelativevariances).(a)Simulatedtemperaturesarere-orderedsothatNHaveragesurfacetemperaturesprogressfromcoldesttowarmestandthe1800–1980
calibrationperiodcontainsonlythewarmesttemperatures.(b)Simulatedtemperaturesarere-orderedsothatthehighestandlowesttemperaturesalloccurwithinthe1800–1980calibrationperiod.(c)–(f)Simulated
temperaturesarere-orderedtohavethesameyear-to-yearprogressionastheMBH99temperaturereconstruction,thusemulatingapossiblereal-worldsituationintermsofthetemperaturerangesampledwithinthe
calibrationperiodcomparedwiththerangethatmayhaveoccurredoutsidethecalibrationperiod(cf.AmmannandWahl,2007).Thedifferencesarethat(c)and(d)usecalibrationperiodsandspatialcoveragethat
couldbeimplementedusingrealdata(1850–1980usingthesparserMBH98verificationgridand1900–1980usingthefullMBH98calibrationgrid,respectively)while(e)and(f)demonstratethepotentialimprove-
mentthatmightbeachievedifthelonger1800–1980calibrationwerepossible(usingthesparserMBH98verificationgridandthefullMBH98calibrationgrid,respectively).Notethattherescaling,usedbyMBH98,
sothatthevarianceofthereconstructedPCsmatchesthatoftheactualtemperaturePCsisnotemployedhereandtestsindicatethatithaslittleeffectonthesepseudo-proxyreconstructions
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
rather high degrees of noise added to the pseudo-proxy signal lead
to underestimation of target European average temperature for
both this and the RegEM (implemented with TTLS) techniques,
though less for RegEM than for multivariate PC regression, particularly
for summer temperature. As noted, in this set of experiments,
RegEM introduces strong low-frequency artefacts for
winter temperature at the highest levels of added white and red
noise. Rutherford et al. (2005) used the RegEM algorithm to
reconstruct global temperature fields. Pseudo-proxy experiments
reported by Mann et al. (2005) appeared to show that this method
did not underestimate the amplitude of the reconstructed NH temperature
anomalies. Smerdon and Kaplan (2007), however, show
that this may have been a false negative result arising from differences
between the implementation of the RegEM algorithm in the
pseudo-proxy experiments and in the real-proxy reconstructions
(also noted by Lee et al., 2008). Mann et al. (2007; cf. their figures
3 and 4) demonstrate that a variant of the RegEM method,
that uses TTLS rather than ridge regression, produces an NH temperature
reconstruction whose amplitude fidelity does not exhibit
the calibration interval dependence of the previous implementation
by Mann et al. (2005), and yields reconstructions that do not
suffer from amplitude loss for a wide range of signal-to-noise
ratios and noise spectra (though Lee et al., 2008, suggest that an
appropriately implemented ridge regression can also produce
good results). With TTLS as implemented by Mann et al. (2007)
reconstructing NH temperature, RegEM performs without amplitude
loss in model-based tests (versions without trend removal),
including using the high-amplitude ECHO-G model output utilized
by Bürger et al. (2006), von Storch et al. (2006) and Küttel
et al. (2007) to examine truncated-EOF methods. When used to
reconstruct the regional-level Niño3 index, however, a longer calibration
period improves amplitude fidelity in these experiments
(Mann et al., 2007: auxiliary material, section 9).
A custom re-ordering of the data to follow the trajectory of the
MBH99 reconstruction of NH temperature (Figure 3c, d) can be
examined to determine if this theoretical potential for systematic
amplitude loss significantly impacts the MBH98/99 result in a
model simulating its real-world context (Ammann and Wahl,
2007). The proxy instrumental evidence available to MBH98/99
suggests that the coldest extended period during the 1000–1980
reconstruction time span occurred during the nineteenth century
and the warmest temperatures occurred during the latter twentieth
century (Ammann and Wahl, 2007). Thus, such a custom re-ordering
allows direct examination of whether extending the calibration
period from 1902–1980 to ~1850–1980 would make a difference
in terms of amplitude fidelity in the real-world instrumental and
proxy data setting actually faced by MBH98. Figure 3d demonstrates
that there is potential for a relatively small absolute amplitude
loss in the situation actually faced by MBH98 (using
pseudo-proxies with signal-to-noise and climate sensitivity characteristics
similar to the real proxies actually employed) when
calibration is limited to the twentieth century. Figure 3c suggests
that amplitude loss could be much reduced by extending the calibration
50 years earlier (to cover the period 1850–1980) to include
much better sampling of both ends of the full climatological range
of NH temperatures: systematic amplitude loss is eliminated in
parallel extension of the calibration period with the multivariate
PC regression method (not shown). Such an extension of the calibration
period in the real world means that the spatial coverage of
the instrumental data used in calibration would be restricted by a
factor of nearly five times in the MBH context, to the MBH98/99
verification data set. The question of whether the proxies used by
MBH98/99 (or other reconstruction studies) were themselves subject
to low-frequency amplitude limitations is not at issue in these
experiments, which only focussed on the reconstruction algorithms
per se.
The outcomes shown in Figure 3c, d are important for further
consideration of reconstruction optimization in the truncated-EOF
CFR method. Together they suggest that ensuring maximum inclusion
of the full climatological range of a time series being reconstructed
during calibration is more critical for amplitude fidelity
than the spatial coverage of the instrumental data being used (cf.
Ammann and Wahl, 2007), as long as the instrumental data field is
appropriately sampled, which appears to be so in the MBH98/99
case. This conclusion is reinforced by the outcomes shown in
Figure 3e, f. The only difference between couplets (c/e) and (d/f)
in Figure 3 is that experiments (e) and (f) are calibrated over the
longer period used for experiments (a) and (b), 1800–1980. In these
cases, the MBH-style truncated-EOF reconstructions show very
small (e) and essentially no (f) average amplitude loss (with the
exception of the sharp later-fifteenth century cooling). An additional
impact of using the spatially reduced verification grid (e) for
calibration shows up as a somewhat enhanced overall variation in
the high frequency domain, which is expected from sample richness
considerations (cf. model annual ‘target’ data, the black line
in Figure 3e). An interesting extension of these experiments would
be to include a replication of the period ~1450–1475 within the calibration
period, to include the lowest temperatures over the entire
model period in calibration. The outcome shown in Figure 3b suggests
that in this case, reconstruction of the later-fifteenth century
cold period would not be subject to any average amplitude loss. In
the real world, extending the length of the calibration data set
backwards in time necessarily means limiting the coverage of the
instrumental field being reconstructed (eg, to Europe in the lateeighteenth
century), so doing this is clearly undesirable when the
goal is whole-field reconstruction. For the truncated-EOF method
this situation would imply the need to examine possible tradeoffs
between the potential for amplitude loss and the spatial coverage of
reconstructed fields. The RegEM method appears to avoid this
trade-off at the scale of the NH temperature field; and in fact, Mann
et al. (2007) show that using RegEM with the same real-world
proxy data employed by MBH98 results in a nearly identical reconstruction
to the MBH98 original, suggesting that the general potential
for amplitude loss in the truncated-EOF method noted here
may have had no effect on the actual MBH outcome.
These considerations represent examples of Levels 1 and 2 in
the test protocol (Figure 2), in terms of the ordering of the model
output examined and then varying the calibration period. A third
level would be to generalize consideration of the noise level(s) and
noise model(s) being used to generate the pseudo-proxies (c.f.,
Mann et al., 2007; Lee et al., 2008; Riedwyl et al., 2008b).
Other potential methods for generating pseudo-proxies might
eliminate stochastic models altogether, instead using mechanistic
approaches to generate tree increment growth from model output
(cf. Vaganov, 1996; Fritts et al., 1999) for tree-ring proxy sites.
Finally, other kinds of analyses might be employed, including frequency
decompositions designed to examine whether specific frequency
bands are better reconstructed than others.
Pseudo-proxy experiments to test
whole-field reconstructions
A similar example protocol for testing whole-field reconstructions
is outlined in Figure 4, but with an important difference in Level 1.
In the case of whole-field reconstructions, end-member sets and
custom orderings of specific large-scale averages are not
employed, since the relationship between such artificial cases and
field responses is not as well-defined as in the case of examining
large-scale averages. Instead, Level 1 in this situation focuses on
the fidelity of field reconstructions at times of specific interest in
relation to anomalies of climate forcings (see section ‘Climate forcing
and histories’) and in relation to how climate forcings and
modes of climate variability relate to each other. It is recognized
28 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
that organizing Level 1 examinations in such a context is potentially
problematic, as a given model may not respond in dynamically/spatially
realistic ways to climate forcings (and with differing
degrees of realism to different forcings), or it may not exhibit realistic
modes of internal variability (eg, ENSO). However, an unrealistic
response to climate forcing changes by a model does not
mean, ipso facto, that good fidelity between a pseudo-proxy reconstructed
field and the corresponding model field is of little meaning
(cf. section ‘Introduction to the pseudo-proxy approach’).
In the context of these caveats, the example protocol suggests
using various kinds of analyses that either target composite differences
across time periods of interest in relation to forcing histories,
or that target the effects of specific forcings (eg, superposed
epoch analyses of volcanic forcing events; cf. Adams et al., 2003;
Fischer et al., 2007). Then, the further dimensions of altering calibration
periods, noise type and level, and other examinations of
specific interest are introduced, as in the protocol for testing
reconstructions of individual time series in Figure 2. As in the
individual time series case, the protocol for whole-field reconstructions
is not intended to be paradigmatic.
An example of testing model pattern fidelity is shown in Figure
5, for whole-field reconstructions of winter temperature anomalies
over Europe induced by large tropical volcanic eruptions. In this
examination, the climate model itself (NCAR-CSM 1.4; Ammann
et al., 2007b) exhibits a spatial pattern quite similar to that
reported by Fischer et al. (2007), based on a combination of
instrumental, documentary and other proxy information
(Luterbacher et al., 2004). The pseudo-proxy performance of the
MBH98-style inverse regression truncated-EOF CFR in the model
context is quite good in terms of capturing the overall spatial pattern
of the forcing effect, but it significantly underestimates the
amplitude of response. This amplitude loss is, at least in part,
related to the amount of noise added to create the pseudo-proxies
(not shown). Notably, in this example, the loss of amplitude for
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 29
Protocol to Test Climate Reconstructions in
Climate Model Experimental Framework
Whole Field Reconstructions
(in full Monte Carlo framework for proxies – potentially also for instrumental data)
Level 1 – By Forcings
Solar and Anthropogenic
• Difference Analyses across Time Periods –
as defined by forcing histories
Volcanic
• Superposed Epoch Analyses
• Composite + Compare to Reference Periods
-- focusing on target regions
Level 1 – By Modes
• Compositing by Time Periods related
to forcings --
focusing on target regions
• Compositing by Modal Index
Stratifications –
focusing on target regions
Level 2 – By Calibration Period *
20th century
19th century
“Custom” periods
(e.g., Maunder Minimum)
* Replicate all or chosen Level 1 experiments
Level 3 – By Noise Type & Level *
White Noise
Range of S/N ratios
* Replicate all or chosen Level 1 & 2
experiments
Level 3 – By Noise Type & Level *
Red Noise / Blue Noise
• Range of S/N ratios
-- Range of rho values
* Replicate all or chosen Level 1 & 2
experiments
Special Cases –
e.g., by Frequency Band *
• Specify filter band-pass level
* Replicate all or chosen Level 1, 2 & 3 experiments
Figure 4 Protocol for testing whole field climate reconstructions in model-based pseudo-proxy experiments
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
30 The Holocene 19,1 (2009)
70N
65N
60N
55N
50N
45N
40N
70N
60N
50N
40N
35N
25W 20W 15W
−2
−2 −1.5 −1 −0.5 0 0.5 1 1.5 2
−1.5 −1 −0.5 0 0.5 1 1.5 2
10W 5W
0.01
0.01
0.05
0.05
0.05
0.05
0.1
0.1
0
0
5E
T_anom CSM 1.4 / DJF Year 0
a
b
c
Temperature Anomalies CSM 1.4 Reconstructed / DJF
Anomalies Year 0
10E 15E 20E 25E 30E
30E
70N
60N
50N
40N
0 30E
35E 40E
Figure 5 Example of field reconstruction performance: composite winter temperature anomalies for Europe after 15 tropical volcanic events over
the past 500 years (cf. Fischer et al., 2007). (a) Anomalies reconstructed from instrumental, documentary and proxy records using a truncatedEOF
method (multivariate PC regression, Luterbacher et al., 2004). (b) Anomalies from NCAR-CSM model (Ammann et al., 2007) over the same
time period. (c) Pseudo-proxy reconstructed anomalies in same model context as (b), using the Wahl and Ammann (2007) emulation of the MBH98
truncated-EOF inverse regression method
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
the positive anomaly focused on Fennoscandia and northwestern
Russia might be sufficient to reverse the sign of the Europeanmean
composite response to volcanic forcing in reasonably conceivable
reconstruction situations (eg, low-enough proxy SNR
and/or short-enough calibration period). This indicates that for
some situations (eg, where a spatial-mean represents the average
of two similar-magnitude but opposite sign anomalies) reconstructing
climate field anomaly patterns may be a more robust
overall capacity of the truncated-EOF CFR methods than reconstructing
spatially averaged climate. The level of noise does not
have a direct relationship with loss of amplitude fidelity in a
whole-field reconstruction simulation undertaken with the RegEM
method for the post-Tambora global temperature field (Mann
et al., 2007) (if anything the relationship may vary inversely),
although additional noise does have a detrimental effect on the
ability to resolve fine details of the field.
Finally, we note that quantitatively measuring the significance
of differences between model and reconstructed fields (and this is
relevant both to the evaluation of pseudo-proxy experiments and
to the comparison of models with real-world reconstructions)
raises a number of issues that apply more generally to measuring
(and determining the significance of) differences in ≥2 dimensions.
Statistics such as Reduction of Error (RE) and Coefficient
of Error (CE) (eg, Cook et al., 1994; MBH98; Mann et al., 2007)
and the Mann-Whitney Rank Sum Test (eg, Fischer et al., 2007)
have been used for this purpose, both at the level of the entire field
and on a grid-box by grid-box basis. However, such tests are not
able to capture the potential differential importance of parts of a
reconstructed field in terms of sensitivity to the direct effects of
forcings and the indirect effects of forcing-induced dynamical
changes. It is beyond the scope of this paper to address this issue
in detail, but it is an important area of analysis that needs significant
attention in order to more fully operationalize the quantitative
examination of field reconstruction fidelity.
Other considerations for use of model-based
experiments
Model-based examinations make possible the creation of ‘perfectpseudo-proxy’
cases, ie, situations in which the simulated proxy
values for a given grid box have no noise added and thus by definition
have SNRs of ∞. In this kind of examination, once successful
performance is demonstrated for a reconstruction methodology
by an examination similar to the one shown in Figure 3b (ie, when
the full range of reconstructed variables is present in the calibration
period), the question being addressed reduces to fidelity
losses driven solely by the spatial coverage of the proxies. A combination
of noisy-proxy and perfect-proxy analyses with the custom
temporal ordering of model output in Figures 3e, f could be
used to separate the extent to which reductions of proxy richness
are related to the generation of amplitude loss and the extent to
which noise hastens the onset of this problem (ie, the extent to
which noise acts as a de facto reduction of proxy richness).
Examinations of this kind have already been started with the
RegEM CFR technique, showing that even a very reduced proxy
set leads to no low frequency amplitude loss for reconstruction of
the NH temperature series, although higher frequency variation is
significantly increased in this case (cf. Mann et al., 2007).
In a related vein, pseudo-proxy experiments can be used to
explore how the addition of one or more specific proxies can
affect reconstruction quality. Küttel et al. (2007), for example,
evaluated the impact of adding a single pseudo-proxy in northern
Finland on the amplitude fidelity for reconstruction of European
average winter surface temperatures. Their results show a strong
enhancement of reconstruction amplitude by addition of the single
predictor in this previously poorly reconstructed subregion, suggesting
that it would be an excellent investment of resources to
attempt to develop real-world ‘winter’ proxy data from this area.
Historical archives from southern Sweden (eg, Leijonhufvud
et al., 2008) or from the eastern Baltic (Tarand and Nordli, 2001)
might provide some of the documentary records being extended
back into the early sixteenth century, allowing the future development
of a southern Scandinavian or eastern Baltic winter temperature
reconstruction for the last approximately 500 years.
It is likely that many more applications may benefit from testing
within potentially realistic surrogate climates provided by
GCM simulations of recent centuries and millennia. GonzalezRouco
et al. (2006), for example, used the pseudo-proxy method
to evaluate the recovery of surface temperature variations from
vertical profiles of ground temperature anomalies (ie, pseudoborehole
temperatures) that were themselves generated using
models of vertical heat diffusion driven by GCM-simulated surface
temperatures. Their results indicate that the current distribution
of borehole temperature records may be sufficient to provide
useful estimates of surface temperature change over recent centuries,
though with expected attenuation at the shorter timescales
during earlier times (e.g. Gonzalez-Rouco et al., 2008 and references
therein). The impact of different sources and magnitudes of
noise, the possible underestimate of Mediaeval warmth in situations
with strong cooling between Mediaeval and ‘Little Ice Age’
periods (see Gonzalez-Rouco et al., 2006: figure 3c), and the precision
with which the timing of maximum ‘Little Ice Age’ cooling
can be estimated (again, see their figure 3c), all deserve further
investigation.
In addition to providing opportunities for testing statistical
reconstruction methods, information from climate model simulations
can be combined, via data assimilation techniques, with
information derived from proxy records, to yield improved estimates
of past climate change. This approach has proved to be very
successful for ‘reanalysis’ of atmosphere and ocean states over the
last 60 years. The few studies performed so far for the pre-instrumental
period have been mainly devoted to the testing of particular
techniques (Goosse et al., 2006a) or to specific short periods
(van der Schrier and Barkmeijer, 2005). Nevertheless, these initial
studies have shown the potential advantages of combining information
from forcings, models and data in this way, to yield physically
consistent reconstructions of variables that are difficult to
estimate on the basis of proxy records alone, such as the largescale
oceanic and atmospheric circulation. Data assimilation exercises
should be pursued further, with a focus on the last
millennium, to adapt and test assimilation techniques to the specific
problems that arise in the palaeoclimate context, such as
coarse time resolution and large uncertainties in proxy and forcing
data. This approach should thus be considered as a valuable complement
to statistical reconstruction of past climate.
Climate forcing and histories
Forcing and climate models
An important reason for improving climate reconstructions of the
past few millennia is that these reconstructions can help both evaluate
climate model responses and sharpen understanding of
important mechanisms and feedbacks. Therefore, a parallel task to
improving the reliability of proxy climate data and climate reconstructions
is to assess and independently constrain forcings of the
climate system over that period.
Forcings can generically be described as external effects (also
called exogenous effects in other modelling communities, such as
economics) on a specific system. Responses within that system that
also have an impact on its internal state are described as feedbacks.
For the atmosphere, sea surface temperature changes could therefore
be considered a forcing, but in a coupled ocean–atmosphere
model they could be a feedback to another external factor or be
intrinsic to the coupled system. Thus the distinction between forcP.D.
Jones et al.: High-resolution palaeoclimatology of the last millennium 31
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
ings and feedbacks is not defined a priori, but is a function of the
scope of the modelled system. This becomes especially important
when dealing with the biogeochemical processes in climate that
affect trace gas concentrations (CO2 and CH4) or aerosols. For
example, if a model contains a carbon cycle, then the CO2 variations
as a function of climate will be a feedback, but for a simpler
physical model, CO2 is often imposed directly as a forcing from
observations, regardless of whether in the real world it was a feedback
to another change, or a result of human industrial activity.
It is useful to consider the pre-industrial period (pre-1850) separately
from the more recent past, since the human influence on
many aspects of atmospheric composition has increased dramatically
in the twentieth century. In particular, aerosol and land-use
changes are poorly constrained prior to the late-twentieth century
and have large uncertainties. Note, however, there may be a role
for human activities even prior to the nineteenth century owing to
early agricultural activity (Ruddiman, 2003; Goosse et al., 2006b).
In pre-industrial periods, climate forcings can be separated into
purely external changes (variations of solar activity, volcanic eruptions,
orbital variations) and those that are intrinsic to the Earth system
(greenhouse gases, aerosols, vegetation, etc.). Such changes in
Earth system elements will occur predominantly as feedbacks to
other changes (whether externally forced or simply as a function of
internal climate ‘noise’). In the more recent past, the human role in
affecting atmospheric composition (trace gases and aerosols) and
land use have dominated over natural processes and so these
changes can, to large extent, be considered external forcings as well.
Traditionally, the ‘system’ that is most usually implied when
talking about forcings and feedbacks are the ‘fast’ components of
the atmosphere–land, surface–upper ocean system that, not coincidentally,
correspond to the physics contained within atmospheric
general circulation models (AGCMs) coupled to a slab
ocean. What is not included (and therefore considered as a forcing
according to the previous definition) are ‘slow’ changes in vegetation,
ice sheets or the carbon cycle. In the real world these features
will change as a function of other climate changes, and in
fact may do so on relatively ‘fast’ (ie, multidecadal) timescales.
Our choice then of the appropriate ‘climate system’ is slightly
arbitrary and does not give a complete picture of the long-term
Earth system sensitivity.
These distinctions become important because the records available
for atmospheric composition do not record the distinction
between feedback and/or forcing. They simply give, for instance,
the history of CO2 and CH4 (though for the industrial era, good
estimates of CO2 emissions are available, based on fossil fuel consumption).
Depending on the modelled system, those records will
either be a modelling input, or a modelling target.
While there are good records for some factors (particularly the
well-mixed greenhouse gases such as CO2 and CH4), records for
others are either relatively uncertain in magnitude (tropospheric
and volcanic aerosols, and solar activity), incomplete (dust, vegetation)
because of poor spatial or temporal resolution or non-existent
(eg, ozone). Estimates of the magnitude of these latter
forcings can only be made using a model-based approach. This
can be undertaken using GCMs that include more Earth system
components (interactive aerosols, chemistry, dynamic vegetation,
carbon cycles, etc.), but these models are still very much a work
in progress and have not been used extensively for palaeoclimatic
purposes. Some initial attempts have been made for selected feedbacks
and forcings (Gerber et al., 2003; Goosse et al. 2006b) but
a comprehensive assessment over the millennia prior to the preindustrial
does not yet exist.
Even for those forcings for which good records exist, there is a
question of how well they are represented within the models. This
is not so much of an issue for the well-mixed greenhouse gases
(CO2, N2O, CH4) since there is a sophisticated literature and
history of including them within models (though some aspects,
such as minor short-wave absorption effects for CH4 and N2O are
still not universally included; Collins et al., 2006).
Individual climate forcings
In this section, we discuss the major external forcing factors that
are important for the climate system and its palaeoclimatic modelling.
Human-caused changes to the atmosphere (well known for
greenhouse gases, but markedly less so for aerosols) are fully discussed
in detail by Forster et al. (2007). We later illustrate how the
various forcings are prescribed and/or determined by one particular
A/OGCM.
Solar irradiance
The most straightforward way of including solar irradiance effects
on climate is to change the solar ‘constant’ (more accurately
described as total solar irradiance – TSI). However, observations
show that solar variability is highly dependent on wavelength with
UV bands having about ten times as large amplitude changes than
TSI over a solar cycle (Lean, 2000). Thus including this spectral
variation for all solar changes allows for a slightly different behaviour
(eg, larger solar-induced changes in the stratosphere where
the UV is mostly absorbed). Additionally, the changes in UV
affect ozone production in both the stratosphere and troposphere,
and this mechanism has been shown to affect both the total radiative
forcing and dynamical responses (Haigh, 1996; Shindell
et al., 2001, 2006). Within a chemistry–climate model this effect
would potentially modify the radiative impact of the original solar
forcing, but could also be included as an additional (parameterized)
forcing in standard GCMs.
Reconstructions of solar variability in past millennia are usually
based on cosmogenic isotopes whose production is modulated by
solar magnetic activity and whose concentrations can be found in
ice cores (ie, 10
Be) or tree rings (14
C) (eg, Bard et al., 2000;
Solanki et al., 2004; Muscheler et al., 2005). Calibration of these
archives is based on the ~30 year series of satellite observations of
solar irradiance over three solar cycles and more recent observations
of the spectral character of that variability. The main uncertainty
is the magnitude of any long-term secular trend in solar
forcing that is unconnected with the direct effects of sunspots and
faculae (Foukal et al., 2006) and the potential for climatic
contamination of the proxy archives (Field, C.V. et al., 2006).
It has also been argued that an indirect effect of solar magnetic
variability on the shielding of cosmic rays may affect the production
of cloud condensation nuclei (Dickinson, 1975). There have
been no quantitative calculations of the magnitude of this effect
(which would require a full study of the relevant aerosol and cloud
microphysics), and so its impact on climate has yet to be included.
Explosive volcanicity
Large volcanic eruptions produce significant amounts of sulphur
dioxide (SO2). If this is injected into the tropical stratosphere during
a particularly explosive eruption, the resulting sulphate
aerosols can persist in the atmosphere for a number of years (eg,
Pinatubo in 1991). Less explosive, but more persistent eruptions
(eg, Laki in 1783) can still affect climate though in a more
regional way and for a shorter period (Oman et al., 2005). These
aerosols have both a shortwave (reflective) and longwave (absorbing)
impact on the radiation and their local impact on stratospheric
heating can have important dynamical effects. It is therefore better
to include the aerosol absorber directly in the radiative transfer
code. However, in less sophisticated models (eg, Crowley, 2000)
or experiments (eg, von Storch et al., 2004), the impact of the
aerosols has been parameterized as an equivalent decrease in TSI.
Reconstructions of the volcanic forcing history are based on welldated
sulphate layers in ice cores combined with calibration from
32 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
historically observed eruptions. Magnitudes of effects, hemispheric
distribution and details of the aerosol microphysics that might be
unique to each eruption are the main sources of uncertainty (Naveau
and Ammann, 2005). Combining information from multiple icecore
records, especially from both polar regions, is important to distinguish
between local and large-scale events. Recent work (Gao
et al., 2008) has attempted to further improve both the accuracy and
detail of the estimated forcing history by utilizing sulphate records
from 54 ice cores (more than double that used by previous studies).
Not only do the additional records reduce the overall uncertainty,
but they also allow information from the patterns of aerosol deposition
to be used in conjunction with analysis of seasonally dependent
atmospheric transport and deposition to estimate stratospheric sulphate
loading from ice-core deposition. The resulting data set provides
estimates of sulphate loading as a function of latitude, altitude
and month. This level of detail, though subject to a number of
caveats and uncertainties, is nevertheless useful for forcing the next
generation of GCM simulations.
Land surface characteristics and aerosols
Land-cover and land-use changes have occurred both due to deliberate
modification by humans (deforestation, imposed fire
regimes, agriculture) as well as a feedback to climate change (the
desertification of the Sahara c. 5500 yr ago). Changing vegetation
in a standard model affects the seasonal cycle of albedo, the surface
roughness, the impact of snow, evapotranspiration (through
different plant rooting depths), etc. However, modelling of the
yearly cycle of crops, or incorporating the effects of large scale
irrigation, are both still very much work in progress.
Anthropogenic aerosol effects (through industrial emissions or
biomass burning) are a critical feature of modern climate change.
Increases in reflective aerosols (sulphates and nitrates), as well as
absorbing aerosols (black carbon) have direct radiative effects
as well as indirect impacts on cloud formation and lifetime as
well as snow albedo. There are natural components to these
aerosols as well (sulphates via planktonic emission of dimethyl
sulphide, black carbon from natural fires, etc.) but aerosol changes
over the few millennia prior to industrialization are very poorly
constrained. Changes might have arisen from climatically or
human-driven changes in dust emissions, ocean biology feedbacks
on circulation change or climate impacts on the emission of
volatile organics from plants (which also have an impact on ozone
chemistry). Some work on modelling a subset of those effects has
been undertaken for the last glacial maximum or the 8.2 kyr event
(LeGrande et al., 2006), but there have been no quantitative estimates
for the late Holocene.
Combining natural and anthropogenic
forcings in GCM simulations
Owing to the relative expense of doing millennial simulations
with state-of-the-art GCMs, existing simulations have generally
only included the minimum required to incorporate relevant solar,
GHG and volcanic forcings (Jansen et al., 2007: table 6.2 identifies
which forcings were used in most of the recent modelling
studies). Progress can be expected soon on more sophisticated
treatments of those forcings and the first quantitative estimates of
additional effects.
The combined impact of the individual natural and anthropogenic
forcings described in the previous sections – together with other
important forcings that we have not considered in detail here, such as
orbital changes and anthropogenic tropospheric sulphate aerosols –
can be estimated either directly or indirectly. Direct estimates rely on
calculations of the impact of each mechanism on the radiation balance
of the Earth (eg, functions relating greenhouse gas concentrations
to radiative forcing), and these are then combined and provided
as input to some model simulations – especially for simpler models
(eg, Hegerl et al., 2003), because their formulation requires this. The
combined forcing can be estimated indirectly by forcing GCMs with,
for example, concentrations or emissions of the relevant forcing
agent, and then diagnosing the modification to the radiation balance
that the radiative transfer component of the GCM predicts. Figure 6
shows one such example, for the HadCM3 simulation of the last 250
years (Tett et al., 2007). In this case, for example, rather than prescribing
the volcanic forcing (eg, as an equivalent reduction in solar
irradiance as assumed by von Storch et al., 2004), time series of
volcanic aerosol loading were applied (with variations amongst four
latitudinal bands to allow for the different effects of high-latitude
and tropical eruptions). This approach is recommended for future
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 33
Figure 6 The latitude–time evolution of zonal-mean forcing (W/m2
, relative to the mean of the first 50 years, 1750–1799) diagnosed from the
ALL250 simulation with HadCM3 described by Tett et al. (2007). This simulation was forced by variations in solar irradiance, volcanic aerosol
loading, orbital forcing, land use, sulphate aerosol emissions and concentrations of well-mixed greenhouse gases and ozone. Values are smoothed
in the time dimension (by a 5-yr low-pass filter) but not in the latitude dimension
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
experiments, because more complex interactions with atmospheric
radiative transfer can be simulated, such as the stratospheric warming
caused by absorption of both short- and long-wave radiation by
the sulphate aerosol which might have dynamical responses in the
atmosphere.
This particular example (Figure 6, from HadCM3) also highlights
the potential importance of strong geographic variations in
forcing. The short-term impact of volcanic eruptions is clear,
influencing most latitudes, and there is a long-term trend towards
positive forcing arising mostly from greenhouse gas forcing (with
a small contribution from solar irradiance changes). The forcing
from sulphate aerosol emissions into the mid-latitudes of the
Northern Hemisphere, while rather uncertain in its magnitude, is
strong enough in this implementation to dominate the greenhouse
gas forcing throughout the simulation (even in the late-nineteenth
century, as pointed out by Tett et al., 2007), though the negative
forcing begins to reduce in recent decades. The negative forcing
from Antarctic stratospheric ozone depletion is also clear in the
last decade of this simulation. Forcings may have important seasonal
as well as geographic structure; Goosse et al. (2006b), for
example, suggest that land-use changes (with a minor influence
from orbital changes) might have contributed to warming during
the Mediaeval period (relative to the present day) but only in summer
and only for some regions in the mid- and high-latitudes of
the Northern Hemisphere.
Conclusions
This article has reviewed the characteristics and current research
status of documentary and high-resolution proxy climatic sources
and addressed the various approaches by which they may be combined
to provide field reconstructions or large-area averages. We
have also extensively discussed the use of climate model simulations
and how they can inform the debate and take forward research
into the optimal combination and interpretation of the various data.
Finally, we summarize our principal findings and recommendations
for the continued exploitation of palaeoclimatological data
for reconstructing large-scale averages and spatial patterns of climate
variability over recent millennia. These recommendations are
ordered according to the section order in the review:
(1) In the area of tree-ring research the number of available
long chronologies is expanding but remains small, and although
potential sites with known subfossil data are limited, the effort in
developing these is fully justified: there remain large areas of the
terrestrial world where chronology network development is in its
infancy (much of the lower-latitudes and virtually all of the SH).
(2) It was widely believed 20 years ago that cross-dating trees
in tropical regions was not possible. Recent work has led to a
growing number of cross-dated chronologies being developed,
principally in southeast Asia in regions of marked seasonality in
rainfall. It is important this work continues.
(3) There have been recent and continuing improvements in
statistical methods for producing long chronologies, which can be
shown to retain low-frequency climatic variability more realistically
than was previously the norm. Work is needed to further
assess the applicability of these methods in a wider range of situations
than have been explored to date.
(4) Significant continued effort is required to provide the high
degree of intrasite and importantly intersite (ie, regional scale)
sample replication needed to demonstrate reliable long-timescale
chronology expression. Local and regional-average chronologies
and regression-based estimates of climate variability should be
routinely presented with explicit indications of their separate
timescale and time-dependent confidence limits.
(5) There is pressing need for further study of the likely precedence
and causes of the apparent ‘divergence’ between instrumentally
recorded and some dendroclimatically estimated
temperature trends (typically some high-latitude NH regions) in
34 The Holocene 19,1 (2009)
1000 AD 1500 AD 1900 AD
Year
Figure 7 The black curve and the x- and y-axes are a redrawn version of figure 7.1c from the First Working Group of IPCC Report (Folland et al.,
1990). The y-axis originally had only unnumbered tick markings and was labelled ‘temperature scale’. The red curve is from Lamb (1982: figure 30,
the upper (annual) curve). The amplitude of this curve has been scaled to correspond to that of the black curve. The Lamb (1982) time series does have
an explicit temperature scale, and the best-fit scaling between this curve and the IPCC curve indicates that one tick-mark interval on the IPCC figure
corresponds almost exactly with 1°C. The degree of smoothing for both these curves is unknown, but Lamb (1982) states that the red curve is based on
50-yr means (supported by earlier publications). The blue curve is a smoothed version of the annual instrumental Central England Temperature record
from Manley (1974, updated) including the last complete year of 2007. This has been smoothed with a 50-yr Gaussian weighted filter with padding
(Mann, 2004). The blue curve is plotted with the same scaling as used for the red curve, further supporting the conclusions that the red curve is based
on the same data after the start of the instrumental record in 1659. The red and blue curves illustrate the differences that can occur between a filtered
curve and one composed of non-overlapping 50-yr averages, and also that recent measured warming may be comparable with presumed earlier warmth
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
recent decades. This emphasizes the priority requirement for systematic
updating of many existing tree-ring data, to continue in
parallel with efforts to expand the representation of data into new
areas.
(6) In coral proxy climate research, replication of important
single coral δ18
O and Sr/Ca records would allow quantification of
signal versus noise in coral reconstructions. Owing to the rarity of
long coral cores, however, this may only be possible over the latetwentieth
century.
(7) There should be greater effort to increase the number of
long coral climate reconstructions using both modern and subfossil
corals, taking full advantage of existing resources and by collecting
new materials.
(8) In situ monitoring of environmental variables (eg, SST,
salinity, seawater δ18
O etc.) is required to improve the interpretation
of coral δ18
O and Sr/Ca records with respect to regional climate
patterns.
(9) Standard calibration and verification procedures (as commonly
used in dendroclimatology) should be developed for coral
records, with a focus on the interannual-to-decadal timescale,
rather than seasonal, calibrations.
(10) Coral Sr/Ca ratios should be routinely measured with
δ18
O to improve the climatic interpretation of both geochemical
tracers.
(11) All corals used for climate reconstruction should be carefully
screened for diagenesis, and this information, along with
metadata related to the sampling location, methods and environment
should be made available through an established data centre.
(12) Cross-dating of the many Greenland ice cores is reducing
dating uncertainties and improving understanding of the factors that
cause variations at interannual timescales. Volcanic horizons are
particularly important in this regard, especially known events in the
historic past such as Icelandic eruptions and Vesuvius in AD 79. This
cross-dating of ice cores in Greenland needs to extend to the
Antarctic, even though it will be much harder to achieve.
(13) Both analyses of time series of ice-core isotope data and
modelling approaches have shown that the traditional spatial calibration
of the isotopic thermometer may be unsuitable in many
cases. It needs to be supplemented by an improved, quantitative
understanding of processes.
(14) Calibration of ice-core isotopic series and of coral
records should be undertaken at the appropriate timescale for the
problem being studied whenever possible. Attempts to reconstruct
the annual cycle may give a false sense of calibration skill.
(15) Further intercomparison and process studies using model
and meteorological data are required to improve understanding of the
calibration of the ice-core thermometer, its seasonal biases, temporal
stability and geographical applicability. The processes controlling
isotopic content in non-polar ice cores require particular attention.
(16) Changes in ice sheet elevation and changes in climatic
conditions upstream of an ice-core drill site can introduce nonclimatic
biases in isotopic series. Hence such effects should be
considered when interpreting isotopic records from ice cores.
(17) The full range of proxy information available from ice
cores has not been exploited. Approaches which use a wider range
should be pursued.
(18) Documentary data are limited to regions with long-written
histories, but archives from Turkey, Venice, the Vatican and
the Middle East have barely been exploited.
(19) In Europe, documentary information decreases significantly
once instrumental records commence. This is a severe
impediment to their use in CFR and CPS reconstructions, as the use
of degraded instrumental data to extend the series to the present
may give a false sense of their reliability.
(20) Extending early instrumental data is vital, particularly for
the calibration and verification of variability on decadal and
longer timescales. Longer instrumental records (than those readily
available in climatic data bases) can be found in many regions,
with some extending for more than 100 years before the founding
of NMSs.
(21) In Europe, there is a potential warm bias in pre-1860
summer temperatures (related to thermometer exposure), particularly
in central Europe and Scandinavia.
(22) Wind information from ship logbooks is a reliable source
for the reconstruction of past large-scale atmospheric circulation,
but has hardly been exploited. There are also numerous (particularly
British) logbooks yet to be digitized, which have the potential
to improve and extend re-analyses of air or sea temperature as
well as sea-level pressure further back in time.
(23) Varved sediments and speleothem records are finding
increased palaeoclimatic utility, but a greater focus on quantitative
documentation of chronological accuracy and climate sensitivity
of both types of records is needed.
(24) Externally forced GCM simulations of the last millennium
provide useful test beds for assessing the characteristics of
the range of CPS and CFR techniques now available. A range of
standard experiments should be developed to test these techniques,
perhaps using common sets of pseudo-proxy networks,
which could be made widely available for extensive testing of current
and new methods.
(25) CPS-based reconstructions of NH temperature averages
at the decadal timescale are relatively independent of the reconstruction
approach. CFR-based reconstructions of internally consistent
climate fields can offer key additional insights into spatial
climate processes, but their reliability must be carefully tested (eg,
the individual reliability of specific patterns and regions) and
linked to the underlying proxy records. Issues regarding the potential
underestimation of long-term variability in both CPS and CFR
reconstructions have not yet been fully resolved: however, initial
work in examining field fidelity for CFRs suggests that pattern
reproduction may be robustly resolvable even in the face of significant
amplitude loss.
(26) More realistic assessments of reconstruction uncertainty
are needed that consider all sources of error and which can powerfully
assess field fidelity. Methods need to be developed that incorporate
uncertainties in individual proxies, the effect of proxy
selection and uncertainties in the regression/scaling models. The
ability of simple residuals between reconstructions and observations
during calibration or verification to represent the total error needs to
be further assessed.
(27) Further simulations with a hierarchy of climate models
(GCMs, EMICs and EBMs) are needed to quantify the uncertainties
associated with past forcings (eg, using a range of possible past forcings,
perhaps with separate simulations for some individual forcings),
with internal variability (eg, using ensembles) and varying
climate processes (eg, using multiple models and by perturbing
physical parameters with a model).
(28) Comparisons between simulations and climate reconstructions
must take these forcing/model-related uncertainties into
consideration, in addition to errors in the climate reconstructions.
Only then can robust conclusions be drawn from such model–data
comparisons.
We noted in the Introduction the dramatic improvements in
late-Holocene palaeoclimatology made since the early 1990s.
Interest in the subject and the large-scale reconstructions has multiplied
over the same period. This development has, however, not
kept pace with the needs for a more reliable picture for the lateHolocene
climates. The questions that are now being asked are also
different and can only be adequately addressed with realistic consideration
of reconstruction uncertainty ranges. More realistic climate
models are providing multiple simulations at higher-spatial
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 35
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
resolution for assessing reconstructions and combination (CPS
and CFR) approaches, but improvements in reconstructions and
reductions in uncertainties in our understanding of late-Holocene
climate change will only come with better and more widespread
proxy climatic information from more diverse sources.
Acknowledgements
The workshop that was the genesis of this paper, entitled ‘Past
Millennia Climate Variability – Synthesis and Outlook’ was held
in Wengen, Switzerland from 7 to 10 June 2006 and was organized
by the international joint PAGES (Past Global Changes)/CLIVAR
(Climate Variability and Predictability) intersection working group
in concert with the PAGES office in Bern, Switzerland. The workshop,
which was funded by EPRI (Electric Power Research
Institute), PAGES, the Swiss NCCR-Climate Programme and CLIVAR,
represented the continued efforts of the international
PAGES/CLIVAR intersection to identify the key priority areas for
future international collaborative research to advance understanding
of the nature of natural climate variability and the extent to
which human activities are causing change. The authors thank
Henry Diaz, Frank Oldfield and an anonymous reviewer for thorough
reviews. PDJ has been supported by the Office of Science
(BER), US Department of Energy, Grant No. DE-FG02-
98ER62601). KRB and TJO also acknowledge funding from UK
Natural Environment Research Council (NER/T/S/2002/00440).
JE acknowledges funding from EC project Millennium (Grant No.
017008). TDvO acknowledges the support of the Australian
Government’s Cooperative Research Centres Programme, through
the Antarctic Ecosystems and Climate CRC. HG is a research associate
with Fonds National de la Recherche Scientific, Belgium.
MEM acknowledges support from the US National Science
Foundation (Grant No. 0542356). ERW acknowledges support for
new research included in this paper from the National Center for
Atmospheric Research (funded by the National Science
Foundation) and Alfred University, USA.
Appendix A
Figure 7.1c of IPCC (1990)
In the first report of the Intergovernmental Panel on Climate
Change (IPCC, 1990) a ‘schematic’ diagram representing temperature
variations over the last millennium was used (Folland et al.,
1990: figure 7.1c, p. 202). The caption of part (c) of the figure
reads: ‘Schematic diagram of global temperature variations for
the last thousand years. The dotted line represents conditions near
the beginning of the twentieth century’. In the Supplementary
IPCC Report in 1992 (Folland et al., 1992), the diagram had been
dropped and the need for more data that would allow for the spatial
aspects of past changes acknowledged. Subsequent IPCC
reports included some of the first hemispheric reconstructions
based on the burgeoning proxy archives (Bradley and Jones, 1993,
in Nicholls et al., 1996 (Second IPCC Assessment Report, SAR)
and MBH98, 1999; Jones et al., 1998; Briffa, 2000 and Crowley
and Lowery, 2000 in Folland et al., 2001 (Third IPCC Assessment
Report, TAR)). Hence the original ‘schematic’ 1990 diagram
appeared to have been confined to history by subsequent IPCC
reports, although this was never specifically stated. It has continued
to reappear in a number of guises – web pages, reports (eg,
Wegman et al., 2006), school teaching literature, sometimes with
phrases evoking reminders of warmer/colder periods in the past
(eg, vineyards in southern Britain, Vikings in Greenland in
Mediaeval times, Frost Fairs on the Thames and icebergs off
Norway in later centuries) – but as far as palaeoclimatologists
were concerned the diagram was nothing more than how it was
originally described in the caption: a schematic.
So where did the schematic diagram come from and who drew
it? It can be traced back to a UK Department of the Environment
publication entitled Global climate change published in 1989
(UKDoE, 1989), but no source for the record was given. Using
various published diagrams from the 1970s and 1980s, the source
can be isolated to a series used by H.H. Lamb, representative of
central England, last published (as figure 30 on p. 84) by Lamb
(1982). Figure 7 shows the IPCC diagram with the Lamb curve
superimposed – clearly they are the same curve. The ‘Central
England’ curve also appeared in Lamb (1965: figure 3 and 1977:
figure 13.4), on both occasions shown as an ‘annual’ curve
together with the extreme seasons: winter (December to
February) and high summer (July and August). The IPCC diagram
comes from the 1982 publication as the vertical resolution
of the annual plot is greater. The data behind the 1977 version are
given in table app. V.3 in Lamb (1977), but these are essentially
the same as previously given in Lamb (1965). All three versions
of the plot have error ranges (which are clearest in the 1982 version
and indicate the range of apparent uncertainty of derived versions).
The 1982 version dispenses with the three possible curves
evident in Lamb (1965, 1977) and instead uses a version which
accounts for the ‘probable under-reporting of mild winters in
Medieval times’ and increased summer temperatures to meet
‘certain botanical considerations’. Lamb (1965) discusses the latter
point at length and raised summer temperatures in his
Mediaeval reconstructions to take account of the documentary
evidence of vineyards in southern and eastern England. The
amount of extra warmth added during 1100–1350 was 0.3–0.4°C,
or about 30% of the range in the black curve in Figure 7. At no
place in any of the Lamb publications is there any discussion of
an explicit calibration against instrumental data, just Lamb’s
qualitative judgement and interpretation of what he refers to as
the ‘evidence’. Variants of the curves also appear in other Lamb
publications (see, eg, Lamb, 1969).
Many in the palaeoclimatic community have known that the
IPCC (1990) graph was not representative of global conditions
(even when it first appeared) and hence the reference to it as a
schematic. Lamb’s (1965, 1977, 1982) series has been used as one
of the series comprising the NH composite developed by Crowley
and Lowery (2000), representative of Central England. Various
authors (eg, Farmer and Wigley, 1984; Wigley et al., 1986;
Ogilvie and Farmer, 1997) have shown that such representativeness
is only really the case for the instrumental part of the record
from 1659 which is based on the well-known Manley (1974)
series. Greater amounts of documentary data (than available to
Lamb in the early 1970s) were collected and used in the Climatic
Research Unit in the 1980s. These studies suggest that the sources
used and the techniques employed by Lamb were not very robust
(see, eg, Ogilvie and Farmer, 1997).
In summary, we show that the curve used by IPCC (1990) was
locally representative (nominally of Central England) and not
global, and was referred to at the time with the word ‘schematic’.
References
Adams, J.B., Mann, M.E. and Ammann, C.M. 2003: Proxy evidence
for an El Niño-like response to volcanic forcing. Nature 426,
274–78.
Alibert, C. and McCulloch, M.T. 1997: Strontium/calcium ratios in
modern Porites corals from the Great Barrier Reef as a proxy for sea
surface temperature: calibration of the thermometer and monitoring of
ENSO. Paleoceanography 12, 345–63.
Alley, R.B., Gow, A.J., Johnsen, S.J., Kipfstuhl, J., Meese, D.A.
and Thorsteinsson, T. 1995: Comparison of deep ice cores. Nature
373, 393–94.
36 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Ammann, C. 2008: The paleoclimate reconstruction challenge.
PAGES News 16, 4.
Ammann, C. and Wahl, E. 2007: The importance of the geophysical
context in statistical evaluations of climate reconstruction procedures.
Climatic Change 85, 71–88.
Ammann, C.M., Nychka, D., Berliner, L.M., Hoar, T., Evans, M.,
Hughes, M. and Wahl, E. 2007a: CMG collaborative research: development
of Bayesian hierarchical models to reconstruct climate over the
past millennium. National Science Foundation, NSF 0724828.
Ammann, C.M., Joos, F., Schimel, D., Otto-Bliesner, B.L. and
Tomas, R. 2007b: Solar influence on climate during the past millennium:
results from transient simulations with the NCAR Climate
System Model. Proceedings of the National Academy of Science 104,
3713–18.
Andersen, A., Koç, N. and Moros, M. 2004: A highly unstable
Holocene climate in the subpolar North Atlantic: evidence from
diatoms. Quaternary Science Reviews 23, 2155–66.
Andersson, C., Risebrobakken, B., Jansen, E., Dahl, S.O. and
Meland, M. 2003: Late Holocene surface-ocean conditions of the
Norwegian Sea (Vøring Plateau). Paleoceanography 18, 1044, doi:
10.1029/2001PA000654.
Antonioli, F., Silenzi, S. and Frisia, S. 2001: Tyrrhenian holocene
palaoeclimate trends from spelean sepulids. Quaternary Science
Reviews 20, 1661–70.
Appenzeller, C., Stocker, T.F. and Anklin, M. 1998: North Atlantic
Oscillation dynamics recorded in Greenland ice cores. Science 282,
446–49.
Argollo, J., Soliz, C. and Villalba, R. 2004: Potencialidad dendrocronológica
de Polylepis tarapacana en los Andes Centrales de
Bolivia. Ecología en Bolivia 39, 5–24.
Bagnato, S., Linsley, B.K., Howe, S.S., Wellington, G.M. and
Salinger, J. 2004: Evaluating the use of the massive coral Diploastra
heliopora for paleoclimate reconstruction. Paleoceanography 19,
PA1032, doi:10.1029/2003PA000935.
Barber, V.A., Juday, P.G. and Finney, B.P. 2000: Reduced growth
of Alaskan white spruce in the twentieth century from temperatureinduced
drought stress. Nature 405, 668–73.
Barclay, D.J., Wiles, G.C. and Calkin, P.E. 1999: A 1119-year treering-width
chronology from western Prince William Sound, southern
Alaska. The Holocene 9, 79–84.
Bard, E., Raisbeck, G., Yiou, F. and Jouzel, J. 2000: Solar irradiance
during the last 1200 years based on cosmogenic nuclides. Tellus
Series B-Chemistry Physics Meteorology 52, 985–92.
Barlow, L.K., White, J.W.C., Barry, R.G., Rogers, J.C. and
Grootes, P.M. 1993: The North Atlantic Oscillation signature in deuterium
and deuterium excess signals in the Greenland Ice Sheet Project
2 ice core, 1840–1970. Geophysical Research Letters 20, 2901–904.
Bartholin, T.S., Berglund, B.E., Eckstein, D., Schweingruber, F.H.
and Eggertsson, O., editors 1992: Tree rings and environment: proceedings
of the international dendrochronological symposium, Ystad,
South Sweden. Lund University, Department of Quaternary Geology.
Batterbee, R.W., Monteith, D.T., Juggins, S., Simpson, G.L., Shilland,
E.W., Flower, R.J. and Kreiser, A.M. 2008: Assessing the accuracy of
diatom-based transfer functions in defining reference pH conditions for
acidified lakes in the United Kingdom. The Holocene 18, 57–67.
Beck, J.W., Edwards, R.L., Ito, E., Taylor, F.W., Recy, J.,
Rougerie, F., Joannot, P. and Henin, C. 1992: Sea-surface temperature
from coral skeletal strontium/calcium ratios. Science 257, 644–47.
Becker, M. 1989: The role of climate on present and past vitality of
silver fir forests in the Vosges mountains of northeastern France.
Canadian Journal of Forest Research 19, 1110–17.
Beltrami, H. 2002: Paleoclimate: Earth’s long-term memory. Science
297, 206–207.
Beltrami, H. and Bourlon, E. 2004: Ground warming patterns in the
Northern Hemisphere during the last five centuries. Earth and
Planetary Science Letters 227, 169–77.
Berlage, H. 1931: On the relationship between thickness of tree rings
of Djati and rainfall on Java. Tectona 24, 939–53.
Bertler, N.A.N., Barrett, P.J., Mayewski, P.A., Fogt, R.L., Kreutz,
K.J. and Shulmeister, J. 2004: El Niño suppresses Antarctic warming,
Geophysical Research Letters 31, L15207, doi:10.1029/2004
GL0207490.
Besonen, M.R., Patridge, W., Bradley, R.S., Francus, P., Stoner,
J.S. and Abbott, M.B. 2008: A record of climate over the last millennium
based on varved lake sediments from the Canadian High Arctic.
The Holocene 18, 169–80.
Betts, R.A., Cox, P.M., Collins, M., Harris, P.P., Huntingford, C.
and Jones, C.D. 2004: The role of ecosystem–atmosphere interactions
in simulated Amazonian precipitation decrease and forest dieback under
global climate warming. Theoretical Applied Climatology 78, 157–75.
Biondi, F. 2001: A 400-year tree-ring chronology from the tropical
treeline of North America. Ambio 30, 162–66.
Biondi, F. and Fessenden, J. 1999: Radiocarbon analysis of Pinus
lagunae tree rings: implications for tropical dendrochronology,
Radiocarbon 41, 241–49.
Biondi, F., Hartsough, P.C. and Galindo Estrada, I. 2005: Daily
weather and tree growth at the tropical treeline of North America.
Arctic, Antarctic, and Alpine Research 37, 16–24.
Black, D.E., Abahazi, M.A., Thunell, R.C., Kaplan, A., Tappa,
E.J. and Peterson, L.C. 2007: An 8-century tropical Atlantic SST
record from the Cariaco Basin: baseline variability, twentieth-century
warming, and Atlantic hurricane frequency. Paleoceanography 22,
PA4204, doi:10.1029/2007PA001427.
Boessenkool, K.P., Hall, I.R., Elderfield, H. and Yashayaev, I.
2007: North Atlantic climate and deep-ocean flow speed changes during
the last 230 years. Geophysical Research Letters 34, L13614,
doi:10.1029/2007GL030285.
Böhm, R., Jones, P.D., Hiebl, J., Brunetti, M., Frank, D. and Maugeri,
M. 2008: The early instrumental warm-bias: a solution for long central
European temperature series 1760–2007. Climate Change in press.
Borchert, R. 1995: Climatic periodicity, phenology and cambium
activity in tropical dry forest trees. IAWA Journal 20, 239–48.
Bradbury, P., Cumming, B. and Laird, K. 2002: A 1500-year record of
climatic and environmental change in Elk Lake, Minnesota – III: measures
of past primary productivity. Journal of Paleolimnology 27, 321–40.
Bradley, R.S. and Jones, P.D. 1993: ‘Little Ice Age’ summer temperature
variations: their nature and relevance to recent global warming
trends. The Holocene 3, 367–76.
––— 1995: Recent developments in studies of climate since A.D.
1500. In Bradley, R.S. and Jones, P.D., editors, Climate since A.D.
1500 2nd Edition. Routledge, 666–79.
Bradley, R.S., Vuille, M., Hardy, D.R. and Thompson, L.G. 2003:
Low latitude ice cores record Pacific sea surface temperatures.
Geophysical Research Letters 30, 1174–77.
Brázdil, R., Pfister, C., Wanner, H., von Storch, H. and
Luterbacher, J. 2005: Historical climatology in Europe – the state of
the art. Climatic Change 70, 363–430.
Brienen, R. and Zuidema, P. 2005: Relating tree growth to rainfall in
Bolivian rain forests: a test for six species using tree ring analysis.
Oecologia 146, 1–12.
Brienen, R.J.W. and Zuidema, P.A. 2006: The use of tree rings in
tropical forest management: projecting timber yields of four Bolivian
tree species. Forest Ecology and Management 226, 256–67.
Briffa, K.R. 1995: Interpreting high resolution proxy climate data –
the example of dendroclimatology. In von Storch, H. and Navarra, A.,
editors, Analysis of climate variability – applications of statistical
techniques. Springer, 77–94.
––— 2000: Annual climate variability in the Holocene: interpreting
the message of ancient trees. Quaternary Science Reviews 19, 87–105.
Briffa, K.R. and Jones, P.D. 1993: Global surface air temperature
variations during the twentieth century: part 2, implications for largescale
high-frequency palaeoclimatic studies. The Holocene 3, 77–88.
Briffa, K.R. and Melvin, T.M. 2008: A closer look at Regional
Chronology Standardisation of tree-ring records: justification of the
need, a warning of some pitfalls, and suggested improvements in its
application. In Hughes, M.K., Diaz, H.F. and Swetnam, T.W., editors,
Dendroclimatology: progress and prospects. Developments in
Paleoenvironmental Research. Springer, in press.
Briffa, K.R. and Osborn, T.J. 2002: Blowing hot and cold. Science
295, 2227–28.
Briffa, K.R., Jones, P.D., Wigley, T.M.L., Pilcher, J.R. and
Baillie, M.G.L. 1986: Climate reconstruction from tree rings: part 2,
spatial reconstruction of summer mean sea-level pressure patterns
over Great Britain. Journal of Climatology 6, 1–15.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 37
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Briffa, K.R., Jones, P.D. and Schweingruber, F.H. 1988: Summer temperature
patterns over Europe: a reconstruction to 1750 A.D. based on maximum
latewood density indices of conifers. Quaternary Research 30, 36–52.
Briffa, K.R., Jones, P.D., Bartholin, T.S., Eckstein, D.,
Schweingruber, F.H., Karlen, W., Zetterberg, P. and Eronen, M.
1992: Fennoscandian Summers from AD 500: temperature changes on
short and long timescales. Climate Dynamics 7, 111–19.
Briffa, K.R., Jones, P.D. and Schweingruber, F.H. 1994: Summer
temperatures across northern North-America – regional reconstructions
from 1760 using tree-ring densities. Journal of Geophysical
Research-Atmospheres 99, 25 835–44.
Briffa, K.R., Jones, P.D., Schweingruber, F.H., Karlen, W. and
Shiyatov, S.G. 1996: Tree-ring variables as proxy-climate indicators:
problems with low frequency signals. In Jones, P.D., Bradley, R.S.
and Jouzel, J., editors, Climatic variations and forcing mechanisms of
the last 2000 years. Springer-Verlag, 9–41.
Briffa, K.R., Schweingruber, F.H., Jones, P.D., Osborn, T.J.,
Shiyatov, S.G. and Vaganov, E.A. 1998: Reduced sensitivity of
recent tree-growth to temperature at high northern latitudes. Nature
391, 678–82.
Briffa, K.R., Osborn, T.J., Schweingruber, F.H., Harris, I.C.,
Jones, P.D., Shiyatov, S.G. and Vaganov, E.A. 2001: Low-frequency
temperature variations from a northern tree-ring-density network.
Journal of Geophysical Research 106, 2929–41.
Briffa, K.R., Osborn, T.J., Schweingruber, F.H., Jones, P.D.,
Shiyatov, S.G. and Vaganov, E.A. 2002a: Tree-ring width and density
data around the Northern Hemisphere: part 2, spatio-temporal
variability and associated climate patterns. The Holocene 12, 759–89.
––— 2002b: Tree-ring width and density data around the Northern
Hemisphere: part 1, local and regional climate signals. The Holocene
12, 737–57.
Briffa, K.R., Osborn, T.J. and Schweingruber, F.H. 2004: Largescale
temperature inferences from tree rings: a review. Global and
Planetary Change 40, 11–26.
Briffa, K.R., Shishov, V.V., Melvin, T.M., Vaganov, E.A., Grudd,
H., Hantemirov, R.M., Eronen, M. and Naurzbaev, M.M. 2007:
Trends in recent temperature and radial tree growth spanning 2000
years across northwest Eurasia. Philosophical Transactions of the
Royal Society B 362, doi:10.1098/rstb.2007.2199.
Brönnimann, S., Xoplaki, E., Casty, C., Pauling, A. and
Luterbacher, J. 2007: ENSO influence on Europe during the last centuries.
Climate Dynamics 28, 181–97.
Brook, E. and Wolff, E.W. 2006: The future of ice core science. EOS
87, 39.
Brunet, M., Saladie, O., Jones, P.D., Sigro, J., Aguilar, E.,
Moberg, A., Lister, D.H., Walther, A., Lopez, D. and Almarza, C.
2006: The development of a new dataset of Spanish daily adjusted
temperature series (SDATS) (1850–2003). International Journal of
Climatology 26, 1777–802.
Buckley, B.M., Barbetti, M., Watanasak, M., D’Arrigo, R.,
Boonchirdchoo, S. and Sarutanon, S. 1995: Dendrochronological
investigations in Thailand. IAWA Journal 16, 393–409.
Buckley, B.M., Tongjit, O., Pumijumnong, N. and Poonsri, R.
2001: Dendrometer band studies on Tectona grandis in northern
Thailand. The Palaeobotanist 50, 83–87.
Buckley, B.M., Wilson, R.J.S., Kelly, P.E., Larson, D.W. and
Cook, E.R. 2004: Inferred summer precipitation for southern Ontario
back to AD 610, as reconstructed from ring widths of Thuja occidentalis.
Canadian Journal of Forest Research-Revue Canadienne De
Recherche Forestiere 34, 2541–53
Buckley, B.M., Cook, B.I., Bhattacharyya, A., Dukpa, D. and
Chaudhary, V. 2005: Global surface temperature signals in pine ringwidth
chronologies from southern monsoon Asia. Geophysical
Research Letters 32, L20704, doi:10.1029/2005GL023745.
Buckley, B.M., Duangsathaporn, K., Palakit, K., Butler, S.,
Syhapanya, V. and Xaybouangeun, N. 2007a: Analyses of growth
rings of Pinus merkusii from Laos P.D.R. Forest Ecology and
Management doi:10.1016/j.foreco.2007.07.018.
Buckley, B.M., Palakit, K., Duangsathaporn, K., Sanguantham, P.
and Prasomsin, P. 2007b: Decadal scale droughts over northwestern
Thailand over the past 448 years: links to the tropical Pacific and
Indian Ocean sectors. Climate Dynamics 29, 63–71.
Büntgen, U., Frank, D.C., Nievergelt, D. and Esper, J. 2006:
Summer temperature variations in the European Alps, AD 755–2004.
Journal of Climate 19, 5606–23.
Büntgen, U., Frank, D.C., Grudd, H. and Esper, J. 2008: Eight
centuries of Pyrenees summer temperatures from tree-ring density.
Climate Dynamics 31, 615–31.
Bürger, G. 2007: Comment on ‘The spatial extent of 20th-century
warmth in the context of the past 1200 years’. Science 316, 1844a.
Bürger, G., Fast, I. and Cubasch, U. 2006: Climate reconstruction
by regression regression – 32 variations on a theme. Tellus 58A, 227–
35.
Calvo, E., Marshall, J.F., Pelejero, C., McCulloch, M.T., Gagan,
M.K. and Lough, J.M. 2007: Interdecadal climate variability in the
Coral Sea since 1708 A.D. Palaeogeography, Palaeoclimatology,
Palaeoecology 248, 190–201.
Camuffo, D. and Jones, P.D., editors 2002: Improved understanding
of past climatic variability from early daily European instrumental
sources. Kluwer Academic Publishers, 392 pp.
Carricart-Ganivet, J.P. and Merino, M. 2001: Growth responses of
the reef-building coral Montastrea annularis along a gradient of continental
influence in the southern Gulf of Mexico. Bulletin of Marine
Science 68, 133–46.
Casty, C., Wanner, H., Luterbacher, J., Esper, J. and Böhm, R.
2005: Temperature and precipitation variability in the European Alps
since 1500. International Journal of Climatology 25, 1855–80.
Casty, C., Raible, C.C., Stocker, T.F., Wanner, H. and
Luterbacher, J. 2007: A European pattern climatology 1766–2000.
Climate Dynamics 29, 791–805.
Chalker, B.E. and Barnes, D.J. 1990: Gamma densitometry for the
measurement of coral skeletal density. Coral Reefs 9, 11–23.
Chapman, D.S., Bartlett, M.G. and Harris, R.N. 2004: Comment on
‘Ground vs. surface air temperature trends: implications for borehole surface
temperature reconstructions’ by M.E. Mann and G. Schmidt.
Geophysical Research Letters 31, L07205, doi:10.1029/2003GL019054.
Charles, C., Rind, D., Jouzel, J., Koster, R. and Fairbanks, R.
1995: Seasonal precipitation timing and ice core records. Science 269,
247–48.
Christie, D., Lara, A., Barichivich, J., Villalba, R., Morales, M.S.
and Cuq, E. 2008: El Niño-Southern Oscillation signal in the world’s
highest-elevation tree-ring chronologies from the Altiplano, Central
Andes. Special Issue on Regional high-resolution multiproxy climate
reconstruction for South America: state of the art and perspectives.
Palaeogeography, Palaeoclimatology, Palaeoecology doi:
10.1016/j.palaeo.2007.11.013.
Chuine, I., Yiou, P., Viovy, N., Seguin, B., Daux, V. and Le Roy
Ladurie, E. 2004: Grape harvest dates and temperature variations in
Eastern France since 1370. Nature 432, 289–90.
Clark, D.A. and Clark, D.B. 1994: Climate induced annual variation
in canopy tree growth in a Costa Rican tropical rain forest. Journal of
Ecology 82, 865–72.
Clark, D.A., Piper, S.C., Keeling, C.D. and Clark, D.B. 2003:
Tropical rain forest tree growth and atmospheric carbon dynamics
linked to interannual temperature variation during 1984–2000.
Proceedings of the National Academy of Sciences 100, 5852–57.
Cobb, K.M., Charles, C.D., Cheng, H., Kastner, M. and Edwards,
R.L. 2003a: U/Th-dating living and young fossil corals from the central
tropical Pacific. Earth and Planetary Science Letters 210, 91–103.
Cobb, K.M., Charles, C.D., Cheng, H. and Edwards, R.L. 2003b:
El Niño Southern Oscillation and tropical Pacific climate during the
last millennium. Nature 424, 271–76.
Cohen, A.L., Owens, K.E., Layne, G.D. and Shimizu, N. 2002: The
effect of algal symbionts on the accuracy of Sr/Ca paleotemperatures
from coral. Science 296, 331–33.
Cole, J.E. 2003: Holocene coral records: windows on tropical climate
variability. In Mackay, A., Battarbee, R., Birks, J. and Oldfield, F.,
editors, Global change in the Holocene. Arnold, 168–84.
Cole, J.E., Fairbanks, R.G. and Shen, G.T. 1993: Recent variability
in the Southern Oscillation: isotopic results from a Tarawa Atoll
Coral. Science 260, 1790–93.
Cole, J.E., Dunbar, R.B., McClanahan, T.R. and Muthiga, N.A.
2000: Tropical Pacific forcing of decadal SST variability in the western
Indian Ocean over the past two centuries. Science 287, 617–19.
38 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Collins, M., Osborn, T.J., Tett, S.F.B., Briffa, K.R. and
Schweingruber, F.H. 2002: A comparison of the variability of a climate
model with palaeo-temperature estimates from a network of treering
densities. Journal of Climate 15, 1497–515.
Collins, W.D., Ramaswamy, V., Schwarzkopf, M.D., Sun, Y.,
Portmann, R.W., Fu, Q., Casanova, S.E.B., Dufresne, J.-L.,
Fillmore, D.W., Forster, P.M.D. Galin, V.Y., Gohar, L.K.,
Ingram, W.J., Kratz, D.P., Lefebvre, M.-P., Li, J., Marquet, P.,
Oinas, V., Tsushima, Y. Uchiyama, T. and Zhong, W.Y. 2006:
Radiative forcing by well-mixed greenhouse gases: estimates from
climate models in the IPCC AR4. Journal of Geophysical Research
111, D14317, doi: 10.1029/2005JD006713.
Cook, E.R. 1990: Bootstrap confidence-intervals for red spruce ringwidth
chronologies and an assessment of age-related bias in recent
growth trends. Canadian Journal of Forest Research 20, 1326–31.
Cook, E.R. and Kairiukstis, L.A. 1990: Methods of dendrochronology.
International Institute for Applied Systems Analysis, Kluwer
Academic Publishers.
Cook, E.R. and Peters, K. 1997: Calculating unbiased tree-ring
indices for the study of climatic and environmental change. The
Holocene, 7, 361–70.
Cook, E.R., Briffa, K.R. and Jones, P.D. 1994: Spatial regression
methods in dendroclimatology: a review and comparison of two techniques.
International Journal of Climatology 14, 379–402.
Cook, E.R., Briffa, K.R., Meko, D.M., Graybill, D.A. and Funkhouser,
G. 1995: The segment length curse in long tree-ring chronology development
for paleoclimatic studies. The Holocene 5, 229–37.
Cook, E.R., Esper, J. and D’Arrigo, R.D. 2004a: Extra-tropical
Northern Hemisphere land temperature variability over the past 1000
years. Quaternary Science Reviews 23, 2063–74.
Cook, E.R., Woodhouse, C.A., Eakin, C.M., Meko, D.M. and
Stahle, D.W. 2004b: Long-term aridity changes in the western United
States. Science 306, 1015–18.
Cook, E.R., Buckley, B.M., Palmer, J.G., Fenwick, P., Peterson,
M.J., Boswijk, G. and Fowler, A. 2006: Millennia-long tree-ring
records from Tasmania and New Zealand: a basis for modelling climate
variability and forcing, past, present and future. Journal of
Quaternary Science 21, 689–99.
Cook, E.R., Seager, R., Cane, M.A. and Stahle, D.W. 2007: North
American drought: reconstructions, causes and consequences. EarthScience
Reviews 81, 93–134.
Correge, T. 2006: Sea surface temperature and salinity reconstruction
from coral geochemical tracers. Palaeogeography, Palaeoclimatology,
Palaeoecology 232, 408–28.
Craig, H. 1961: Isotopic variations in meteoric waters. Science 133,
1702–703.
––— 1965: The measurement of oxygen isotope paleotemperatures. In
Tongiorgi, E., editor, Stable isotopes in oceanographic studies and
paleotemperatures. Consiglio Nazionale delle Richerche, Laboratorio
do Geologia Nucleare, 9–130.
Cronin, T.M., Dwyer, G.S. Kamiya, T., Schwede, S. and Willard,
D.A. 2003: Medieval Warm Period, Little Ice Age and 20th century
temperature variability from Chesapeake Bay. Global and Planetary
Change 36, 17–29.
Crowley, T.J. 2000: Causes of climate change over the past 1000
years. Science 289, 270–77.
Crowley, T.J. and Lowery, T.S. 2000: How warm was the Medieval
Warm Period? A comment on ‘Man-made versus natural climate
change’. Ambio 39, 51–54.
Cuffey, K.M., Clow, G.D., Alley, R.B., Stuiver, M., Waddington,
E.D. and Saltus, R.W. 1995: Large Arctic temperature change at the
Wisconsin–Holocene glacial transition. Science 270, 455–58.
Curran, M.A.J., van Ommen, T.D., Morgan, V.I., Phillips, K.L.
and Palmer, A.S. 2003: Ice core evidence for Antarctic sea ice decline
since the 1950s. Science 302, 1203–206.
Dansgaard, W. 1964: Stable isotopes in precipitation. Tellus 16, 436–
68.
Dansgaard, W., Clausen, H.B., Gundestrup, N., Hammer, C.U.,
Johnsen, S.J., Kristinsdottir, P.M. and Reeh, N. 1982: A new
Greenland deep ice core. Science 218, 1273–77.
Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D.,
Gundestrup, N., Hammer, C.U., Hvidberg, C.S., Steffensen, J.P.,
Sveinbjörnsdóttir, Á.E., Jouzel, J. and Bond, G. 1993: Evidence for
general instability of past climate from a 250-kyr ice-core record.
Nature 364, 218–20.
D’Arrigo, R. and Smerdon, J.E. 2008: Tropical climate influences
on drought variability over Java, Indonesia. Geophysical Research
Letters 35, L05707, doi:10.1029/2007GL032589.
D’Arrigo, R., Jacoby, G. and Krusic, P. 1994: Progress in dendroclimatic
studies in Indonesia. Terrestrial Atmospheric and
Oceanographic Science 5, 349–63.
D’Arrigo, R., Barbetti, M., Watanasak, M., Buckley, B., Krusic, P.,
Boonchirdchoo, S. and Sarutanon, S. 1997: Progress in dendroclimatic
studies of mountain pine in northern Thailand. IAWA Journal 18, 433–44.
D’Arrigo, R.D., Kaufmann, R.K., Davi, N., Jacoby, G.C.,
Laskowski, C., Myneni, R.B. and Cherubini, P. 2004: Thresholds
for warming-induced growth decline at elevational tree line in the
Yukon Territory, Canada. Global Biogeochemical Cycles 18,
GB3021, doi:10.1029/2004GB002249.
D’Arrigo, R., Wilson, R., Deser, C., Wiles, G., Cook, E., Villalba,
R., Tudhope, A., Cole, J. and Linsley, B. 2005: Tropical–north
Pacific climate linkages over the past four centuries. Journal of
Climate 18, 5253–65.
D’Arrigo, R.D., Wilson, R.J., Palmer, J., Krusic, P.J., Curtis, A.,
Sakulich, J., Bijaksana, S., Zulaikah, S. and Ngkoimani, L.O. 2006a:
Monsoon drought over Java, Indonesia, during the past two centuries.
Geophysical Research Letters 33, L04709, doi:10.1029/2005GL025465.
D’Arrigo, R., Wilson, R. and Jacoby, G. 2006b: On the long-term
context for late twentieth century warming. Journal of Geophysical
Research 111, D03103, doi:10.1029/2005JD006352.
D’Arrigo, R., Wilson, R., Liepert, B. and Cherubini, P. 2008: On
the ‘divergence problem’ in northern forests: a review of the tree-ring
evidence and possible causes. Global Planetary Change 60, 289–305.
DaSilva, R., Santos, J., Tribuzy, E., Chambers, J., Nakamura, S.
and Higuchi, N. 2002: Diameter increment and growth patterns for
individual tree growing in Central Amazon, Brazil. Forest Ecology &
Management 166, 295–301.
Davi, N.K., Jacoby, G.C. and Wiles, G.C. 2003: Boreal temperature
variability inferred from maximum latewood density and tree-ring
width data, Wrangell Mountain region, Alaska. Quaternary Research
60, 252–62.
Dean, J.S., Meko, D.M. and Swetnam, T.W., editors 1996: Tree
rings, environment, and humanity. Proceedings of the international
conference. University of Arizona, Department of Geosciences.
Della-Marta, P.M. and Wanner, H. 2006: A method of homogenizing
the extremes and the mean of daily temperature measurements.
Journal of Climate 19, 4179–97.
de Villiers, S., Nelson, B.K. and Chivas, A.R. 1995: Biological controls
on coral Sr/Ca and d18
O reconstructions of sea surface temperatures.
Science 269, 1247–49.
DeWitt, E. and Ames, M. 1978: Tree-ring chronologies of eastern
North America. Chronology Series IV, 1, Laboratory of Tree-Ring
Research, University of Arizona, 42 pp.
Dickinson, R.E. 1975: Solar variability and the lower atmosphere.
Bulletin of the American Meteorological Society 56, 1240–48.
Dixon, D., Mayewski, P.A., Kaspari, S., Sneed, S. and Handley, M.
2005: Connections between West Antarctic ice core sulfate and climate
over the last 200+ years. Annals of Glaciology 41, 155–56.
Dobrovolný, P., Brázdil, R., Valášek, H., Kotyza, Ol., Macková, J.
and Halí, M. 2008: New approach to climate reconstruction in historical
climatology: an example from the Czech Republic, AD 1718–
2006. International Journal of Climatology in press.
Draschba, S., Patzold, J. and Wefer, G. 2000: North Atlantic climate
variability since AD 1350 recorded in d18
O and skeletal density
of Bermuda corals. International Journal of Earth Sciences 88, 733–
41.
Driscoll, W.W., Wiles, G.C., D’Arrigo, R.D. and Wilmking, M.
2005: Divergent tree growth response to recent climatic warming,
Lake Clark National Park and Preserve, Alaska. Geophysical
Research Letters 32, L20703, doi:10.1029/2005GL024258.
Druckenbrod, D., Mann, M.E., Stahle, D.W., Cleaveland, M.K.,
Therrell, M.D. and Shugart, H.H. 2003: Late 18th century precipitation
reconstructions from James Madison’s Montpelier Plantation.
Bulletin of the American Meteorological Society 84, 57–71.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 39
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Druffel, E.R.M. and Griffin, S. 1993: Large variations of surface
ocean radiocarbon: evidence of circulation changes in the southwestern
Pacific Ocean. Journal of Geophysical Research 98, 20 249–59.
Dunbar, R.B., Wellington, G.M., Colgan, M.W. and Glynn, P.W.
1994: Eastern Pacific sea surface temperature since 1600 A.D.: the d18
O
record of climate variability in Galapagos corals. Paleoceanography 9,
291–315.
Dünisch, O., Ribeiro Montóia, V. and Bauch, J. 2003:
Dendroecological investigations on Swietenia macrophylla King and
Cedrela odorata L. (Meliaceae) in the central Amazon. Trees 17, 244–50.
Dupouey, J., Denis, J. and Becker, M. 1992: A new method of standardisation
for examining long term trends in tree-ring chronologies.
Lundqua Report 34, 85–88.
Enmar, R., Stein, M., Bar-Matthews, M., Sass, E., Katz, A. and
Lazar, B. 2000: Diagenesis in live corals from the Gulf of Aqaba. I.
The effct on paleo-oceanography tracers. Geochimica Cosmochimica
Acta 64, 3123–32.
Erlandsson, S. 1936: Dendrochronological studies. University of
Uppsala, 117 pp.
Esper, J., Cook, E.R. and Schweingruber, F.H. 2002: Low-frequency
signals in long tree-ring chronologies for reconstructing past
temperature variability. Science 295, 2250–53.
Esper, J., Frank, D.C., Wilson, R.J.S. and Briffa, K.R. 2005: Effect
of scaling and regression on reconstructed temperature amplitude for
the past millennium, Geophysical Research Letters 32, L07711,
doi:10.1029/2004GL021236.
Esper, J., Frank, D., Büntgen, U., Verstege, A., Luterbacher, J.
and Xoplaki, E. 2007a: Long-term drought severity variations in
Morocco. Geophysical Research Letters 34, L17702, doi:10.1029/
2007GL030844.
Esper, J., Frank, D.C. and Luterbacher, J. 2007b: On selected
issues and challenges in dendroclimatology. In Kienast, F., Wildi, O.
and Ghosh, S., editors, A changing world: challenges for landscape
research. Springer, 113–32.
Esper, J., Frank, D.C., Wilson, R.J.S., Büntgen, U. and Treydte,
K. 2007c: Uniform growth trends among central Asian low- and highelevation
juniper tree sites. Trees-Structure and Function 21, 141–50.
Evans, M.N. and Schrag, D.P. 2004: A stable isotope-based approach
to tropical dendroclimatology. Geochimica et Cosmochimica Acta 68,
3295–305, DOI: 10.1016/j.gca.2004.01.006.
Evans, M.N., Kaplan, A. and Cane, M.A. 1998: Optimal sites for
coral-based reconstruction of global sea surface temperature.
Paleoceanography 13, 502–16.
––— 2000: Intercomparison of coral oxygen isotope data and historical
sea surface temperature (SST): potential for coral-based SST field
reconstruction. Paleoceanography 15, 551–63.
––— 2002: Pacific sea surface temperature field reconstruction from
coral d18
O data using reduced space objective analysis.
Paleoceanography 17, 10.1029/2000PA000590.
Fairbanks, R.G., Evans, M.N., Rubenstone, J.L., Mortlock, R.A.,
Broad, K., Moore, M.D. and Charles, C.D. 1997: Evaluating climate
indices and their geochemical proxies measured in corals. Coral Reefs
16 Supplement, S93–S100.
Farmer, G. and Wigley, T.M.L. 1984: The reconstruction of
European climate in decadal and shorter time scales. Final Report to
the Commission of the European Communities under Contract No.
CL-029-81-UK(H). Unpublished report, Climatic Research Unit.
February, E.C. 1996: Plant xylem anatomy, dendrochronology and
stable isotopes as tools in rainfall reconstruction in southern Africa.
Ph.D. dissertation, The University of Cape Town, 97 pp.
Feeley, K.J., Wright, J.S., Supardi, N.M.N., Kassim, A.R. and
Davies, S.J. 2007: Decelerating growth in tropical forest trees.
Ecology Letters 10, 461–69.
Felis, T. and Pätzold, J. 2003: Climate records from corals. In Wefer,
G., Lamy, F. and Mantoura, F., editors, Marine science frontiers for
Europe. Springer-Verlag, 11–27.
Felis, T., Patzold, J. and Loya, Y. 2003: Mean oxygen-isotope signatures
in Porites spp. corals: inter-colony variability and correction
for extension-rate effects. Coral Reefs 22, 328–36.
Felis, T., Lohmann, G., Kuhnert, H., Lorenz, S.J., Scholz, D.,
Patzold, J., Al-Rousan, S.A. and Al-Moghrabi, S.M. 2004:
Increased seasonality in Middle East temperatures during the last
interglacial period. Nature 429, 164–68.
Field, C.V., Schmidt, G.A., Koch, D. and Salyk, C. 2006: Modeling
production and climate-related impacts on 10Be concentration in ice
cores. Journal of Geophysical Research 111, D15107, doi:10.1029/
2005JD006410.
Field, D.B. and Baumgartner, T.R. 2000: A 900 year stable isotope
record of interdecadal and centennial change from the California
Current. Paleoceanography 15, 695–708.
Field, D.B., Baumgartner, T.R., Charles, C.D., Ferreira-Bartrina,
V. and Ohman, M.D. 2006: Planktonic foraminifera of the California
Current reflect 20th-century warming. Science 311, 63–66.
Fischer, E.M., Luterbacher, J., Zorita, E., Tett, S.F.B., Casty, C.
and Wanner, H. 2007: European climate response to tropical volcanic
eruptions over the last half millennium. Geophysical Research Letters
34, L05707, doi:10.1029/2006GL027992.
Fisher, D.A., Reeh, N. and Clausen, H.B. 1985: Stratigraphic noise
in time series derived from ice cores. Annals of Glaciology 14, 76–86.
Fisher, D.A., Koerner, R.M. and Reeh, N. 1995: Holocene climatic
records from Agassiz ice cap, Ellesmere Island, NWT, Canada. The
Holocene 5, 19–24.
Fisher, D.A., Koerner, R.M., Kuivinen, K., Clausen, H.B.,
Johnsen, S.J., Steffensen, J.P., Gundestrup, N. and Hammer, C.U.
1996: Inter-comparison of ice core d18
O and precipitation records
from sites in Canada and Greenland over the last 3500 years and over
the last few centuries in detail using EOF techniques. In Jones, P.D.,
Bradley, R.S. and Jouzel, J., editors, Climatic variations and forcing
mechanisms of the last 2000 years. Springer-Verlag, 297–328.
Fleitmann, D., Burns, S.J., Mudelsee, M., Neff, U., Kramers, J.,
Mangini, A. and Matter, A. 2003: Holocene forcing of the Indian
monsoon recorded in a stalagmite from the Southern Ocean. Science
300, 1737–39.
Fleitmann, D., Burns, S.J., Neff, U., Mudelsee, M., Mangini, A. and
Matter, A. 2004: Palaeoclimatic interpretation of high-resolution oxygen
istotope profiles derived from annually laminated speleothems of highresolution
oxygen profiles derived from annually laminated speleothems
from Southern Oman. Quaternary Science Reviews 23, 935–45.
Fleitmann, D., Dunbar, R.B., McCulloch, M., Mudelsee, M.,
Vuille, M., McClanahan, T., Cole, J.E. and Eggins, S. 2007: East
African soil erosion recorded in a 300 year old coral colony from
Kenya. Geophysical Research Letters 34, L04401, doi:10.1029/2006
GL028525.
Folland, C.K., Karl, T.R. and Vinnikov, K.Ya. 1990: Observed climate
variations and change. In Houghton, J.T., Jenkins, G.J. and
Ephraums, J.J., editors, Climate change. The IPCC scientific assessment.
, Cambridge University Press, 195–238.
Folland, C.K., Karl, T.R., Nicholls, N., Nyenzi, B.S., Parker, D.E.
and Vinnikov, K.Ya. 1992: Observed climate variability and change.
In Houghton, J.T., Callander, B.A. and Varney, S.K., editors, Climate
change 1992. The supplementary report to the IPCC scientific assessment.,
Cambridge University Press, 135–70.
Folland, C.K., Karl, T.R., Christy, J.R., Clarke, R.A., Gruza,
G.V., Jouzel, J., Mann, M.E., Oerlemans, J., Salinger, M.J. and
Wang, S.-W. 2001: Observed climate variability and change. In
Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden,
P.J., Dai, X., Maskell, K. and Johnson, C.A., editors, Climate change
2001: the scientific basis. , Cambridge University Press, 99–181.
Forster, P.M., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R.,
Fahey, D.W., Haywood, J., Lean, J., Lowe, D.C., Myhre, G.,
Nganga, J., Prinn, R., Raga, G., Schulz, M. and Van Dorland, R.
2007: Changes in atmospheric constituents and in radiative forcing. In
Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt,
K.B., Tignor, M.and Miller, H.L., editor, Climate change 2007: the
physical science basis. Contribution of Working Group I to the Fourth
Assessment Report of the Intergovernmental Panel on Climate
Change, Cambridge University Press, 129–234.
Foukal, P., Fröhlich, C., Spruit, H. and Wigley, T.M.L. 2006:
Variations in solar luminosity and their effect on the Earth’s climate.
Nature 443, 161–66.
Frank, D., Esper, J. and Cook, E.R. 2007: Adjustment for proxy number
and coherence in a large-scale temperature reconstruction.
Geophysical Research Letters 34, L16709, doi:10.1029/2007GL030571.
Fritts, H.C. 1976a: The statistics of ring-width and climatic data. In
Fritts, H.C., Tree rings and climate. Academic Press, 246–311.
––— 1976b: Tree rings and climate. Academic Press.
40 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
––—1991: Reconstructing large-scale climatic patterns from tree-ring
data. The University of Arizona Press.
Fritts, H.C. and Swetnam, T.W. 1989: Dendroecology – a tool for
evaluation variations in past and present forest environments.
Advances in Ecological Research 19, 111–88.
Fritts, H.C., Lofgren, G.R. and Gordon, G.A. 1979: Variations in
climate since 1602 as reconstructed from tree rings. Quaternary
Research 12, 18–46.
Fritts, H.C., Shashkin, A.V. and Downes, G.M. 1999: A simulation
model of conifer ring growth and cell structure. In Wimmer, R. and Vetter,
R.E., editors, Tree-ring analysis. Cambridge University Press, 3–32.
Gaetani, G.A. and Cohen, A.L. 2006: Element partitioning during
precipitation of aragonite from seawater: a framework for understanding
paleoproxies. Geochimica et Cosmochimica Acta 70, 4617–34.
Gagan, M.K., Ayliffe, L.K., Beck, J.W., Cole, J.E., Druffel,
E.R.M., Dunbar, R.B. and Schrag, D.P. 2000: New views of tropical
paleoclimates from corals. Quaternary Science Reviews 19, 45–64.
Gajewski, K. 1988: Late Holocene climate changes in eastern North
America estimate from pollen data. Quaternary Research 29, 255–62.
Gallego, D., García-Herrera, R., Ribera, P. and Jones, P.D. 2005:
Seasonal mean pressure reconstruction for the North Atlantic (1750–
1850) based on early marine data. Climate of the Past 1, 19–33.
Gao, C., Robock, A. and Ammann, C. 2008: Volcanic forcing of climate
over the past 1500 years: an improved ice-core-based index for
climate models. Journal of Geophysical Research 113, doi: 10.1029/
2008JD010077.
García, R.R., Díaz, H., García Herrera, R., Eischeid, J., Prieto,
M.R., Hernández, E., Gimeno, L., Rubio, F. and Bascary, A.M.
2001: Atmospheric circulation changes in the tropical Pacific inferred
from the voyages of the Manila galleons in the 16th–18th centuries.
Bulletin of the American Meteorological Society 82, 2435–55.
García-Herrera, R., Können, G.P., Wheeler, D., Prieto, M.R.,
Jones, P.D. and Koek, F.B. 2005a: CLIWOC: a climatological database
for the world’s oceans 1750–1854. Climatic Change 71, 1–12.
García-Herrera, R., Gimeno, L., Ribera, P. and Hernandez, E.
2005b: New records of Atlantic hurricanes from Spanish documentary
sources. Journal of Geophysical Research 110, D03109, doi:10.1029/
2004JD005272.
García-Herrera, R., Diaz, H.F., Garcia, R.R., Prieto, M.R.,
Barriopedro, D., Moyano, R. and Hernández, E. 2008: A chronology
of El Niño events from primary documentary sources in Northern
Peru. Journal of Climate 21, 1948–62.
Ge, Q., Zheng, J., Fang, X., Man, Z., Zhang, X., Zhang, P. and
Wang, W.-C. 2003: Winter half-year temperature reconstruction for
the middle and lower reaches of the Yellow River and Yangtze River,
China, during the past 2000 years. The Holocene 13, 933–40.
Ge, Q.S., Zheng, J.Y., Hao, Z.X., Zhang, P.Y. and Wang, W.-C.
2005: Reconstruction of historical climate in China, high-resolution
precipitation data from Qing Dynasty archives. Bulletin of the
American Meteorological Society 86, 671–79.
Ge, Q., Zheng, J., Tian, Y., Wu, W., Fang, X. and Wang, W.-C. 2008:
Coherence of climatic reconstruction from historical documents in China
by different studies. International Journal of Climatology 28, 1007–24.
Gerber, S., Joos, F., Bruegger, P.P., Stocker, T.F., Mann, M.E. and
Sitch, S. 2003: Constraining temperature variations over the last millennium
by comparing simulated and observed atmospheric CO2.
Climate Dynamics 20, 281–99.
Glaser, R. 2001: Kilmageschichte Mitteleuropas. 1000 Jahr Wetter, Klima,
Katastrophen. Primus Verlag, Wissenschaftliche Buchgesellschaft, 227 pp.
Gong, D-Y. and Luterbacher, J. 2008: Variability of the low-level
cross- equatorial jet of the western Indian Ocean since 1660 as derived
from coral proxies. Geophysical Research Letters 35, L01705,
doi:10.1029/2007GL032409.
González-Rouco, F., von Storch, H. and Zorita, E. 2003: Deep soil
temperature as proxy for surface air-temperature in a coupled model
simulation of the last thousand years. Geophysical Research Letters
30, 2116, doi:10.1029/2003GL018264.
González-Rouco, J.F., Beltrami, H., Zorita, E. and von Storch, H.
2006: Simulation and inversion of borehole temperature profiles in surrogate
climates: spatial distribution and surface coupling. Geophysical
Research Letters 33, L01703, doi:10.1029/2005GL024693.
González-Rouco, J.F., Beltrami, H., Zorita, E. and Stevens, M.B.
2008: Borehole climatology: a discussion based on contributions from
climate modelling. Climate of the Past Discussions 4, 1–80.
http://www.clim-past-discuss.net/4/1/2008/cpd-4–1-2008.pdf
Goodwin, I.D., van Ommen, T.D., Curran, M.A.J. and Mayewski,
P.A. 2004: Mid latitude winter climate variability in the south Indian
and south-west Pacific regions since 1300 AD from the Law Dome ice
core record. Climate Dynamics 22, 783–94.
Goosse, H., Renssen, H., Timmermann, A., Bradley, R.S. and
Mann, M.E. 2006a: Using paleoclimate proxy data to select optimal
realisations in an ensemble of simulations of the climate of the
past millennium. Climate Dynamics 27, 165–84, doi:10.1007/s00382–
006–0128–6.
Goosse, H., Arzel, O., Luterbacher, J., Mann, M.E., Renssen, H.,
Riedwyl, N., Timmermann, A., Xoplaki, E. and Wanner, H. 2006b:
The origin of the European ‘Medieval Warm Period’. Climate of the
Past 2, 99–113.
Gourlay, I. 1995: Growth ring characteristics of some African Acacia
species. Journal of Tropical Ecology 11, 121–40.
Grabner, M., Wimmer, R., Gindl, W. and Nicolussi, K. 2001: A
3474-year alpine tree-ring record from the Dachstein, Austria. In Tree
rings and people. An international conference on the future of dendrochronology.
Davos, Switzerland, Abstracts, 252–53.
Graf, W., Oerter, H., Reinwarth, O., Stichler, W., Wilhelms, F.,
Miller, H. and Mulvaney, R. 2002: Stable isotope records from
Dronning Maud Land, Antarctica. Annals of Glaciology 35, 195–201.
Grootes, P.M. and Stuiver, M. 1997: Oxygen 18/16 variability in
Greenland snow and ice with 10-3
to 105
-year time resolution. Journal
of Geophysical Research 102, 26 455–70.
Grootes, P.M., Stuiver, M., Thompson, L.G. and MosleyThompson,
E. 1989: Oxygen isotope changes in tropical ice,
Quelccaya Peru. Journal of Geophysical Research 94, 1187–94.
Grottoli, A.G. and Eakin, M. 2007: A review of coral d18
O and D14
C
proxy records. Earth Science Reviews 81, 67–91.
Guiot, J. 1985: The extrapolation of recent climatological series with
spectral canonical regression. Journal of Climatology 5, 325–35.
Guiot, J., Nicault, A., Rathgeber, C., Edouard, J., Guibal, F.,
Pichard, G. and Till, C. 2005: Last millennium summer-temperature
variations in western Europe based on proxy data. The Holocene 15,
489–500.
Guzmán, H.M. and Tudhope, A.W. 1998: Seasonal variations in
skeletal extension rate and stable isotopic (13C/12C and 18O/16O)
composition in response to several environmental variables in the
Caribbean reef coral Siderastrea siderea. Marine Ecology Progress
Series 166, 109–18.
Haigh, J.D. 1996: The impact of solar variability on climate. Science
272, 981–84.
Hammer, C.U., Johnsen, S.J., Clausen, H.B., Dahl Jensen, D.,
Gundestrup, N. and Steffensen, J.P. 2001: The palaoclimatic record
from a 345m long ice core from the Hans Tausen Iskappe, Meddeleser
om Grønland. Geoscience 39, 87–95.
Hantemirov, R.M. and Shiyatov, S.G. 2002: A continuous multimillennial
ring-width chronology in Yamal, northwestern Siberia. The
Holocene 12, 717–26.
Harris, R.N. and Chapman, D.S. 2001: Mid-latitude (30°–60°N) climatic
warming inferred by combining borehole temperatures with surface
air temperatures. Geophysical Research Letters 28, 747–50.
Hart, S.R. and Cohen, A.L. 1996: Sr/Ca in corals: an ionprobe study
of annual cycles and microscale coherence with other trace elements.
Geochemica et Cosmochimica Acta 60, 3075–84.
Haug, G.H., Hughen, K.A., Peterson, L.C., Sigman, D.M. and
Röhl, U. 2001: Southward migration of the Intertropical Convergence
Zone through the Holocene. Science 293, 1304–308.
Haug, G.H., Günther, D., Peterson, L.C., Sigman, D.M., Hughen,
K.A. and Aeschlimann, B. 2003: Climate and the collapse of Maya
civilzation. Science 299, 1731–35.
Hegerl, G.C., Crowley, T.J., Baum, S.K., Kim, K.-Y. and Hyde,
W.T. 2003: Detection of volcanic, solar, and greenhouse gas signals
in paleo-reconstructions of Northern Hemispheric temperature.
Geophysical Research Letters 30, 1242 doi:10.1029/2002GL016635.
Hegerl, G.C., Crowley, T.J., Hyde, W.T. and Frame, D.J. 2006:
Climate sensitivity constrained by temperature reconstructions over
the past seven centuries. Nature 440, 1029–32.
Hegerl, G.C., Crowley, T.J., Allen, M.R., Hyde, W.T., Pollack,
H.N., Smerdon, J.E. and Zorita, E. 2007: Detection of human
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 41
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
influence on a new, validated 1500 year temperature reconstruction.
Journal of Climate 20, 650–66.
Helama, S., Lindholm, M., Meriläinen, J., Timonen, M. and
Eronen, M. 2005: Multicentennial ring-width chronologies of Scots
pine along a north–south gradient across Finland. Tree-Ring Research
61, 21–32.
Helmle, K.P., Dodge, R.E. and Ketcham, R.A. 2002: Skeletal architecture
and density banding in Diploria strigosa by x-ray computed
tomography. Proceedings 9th International Coral Reef Symposium,
Bali, Indonesia 23–27 October 2000 1, 365–71.
Helsen, M.M., van de Wal, R.S.W., van den Broeke, M.R.,
Masson-Delmotte, V., Meijer, H.A.J., Scheele, M.P. and Werner,
M. 2006: Modeling the isotopic composition of Antarctic snow using
backward trajectories: simulation of snow pit records. Journal of
Geophysical Research 111, D15109, doi:10.1029/2005JD006524.
Henderson, K.A., Thompson, L.G. and Lin, P-N. 1999: Recording
of El Niño in ice core d18
O records from Nevado Huascarán, Peru.
Journal of Geophysical Research 104, 31 053–65.
Hendy, E.J., Gagan, M.K., Alibert, C.A., McCulloch, M.T.,
Lough, J.M. and Isdale, P.J. 2002: Abrupt decrease in tropical
Pacific sea surface salinity at end of Little Ice Age. Science 295,
1511–14.
Hendy, E.J., Gagan M.K. and Lough, J.M. 2003: Chronological
control of coral records using luminescent lines and evidence for nonstationary
ENSO teleconnections in northeast Australia. The Holocene
13, 187–99.
Hodell, D.A., Brenner, M., Curtis, J.H., Medina-Gonzalez, R.,
Rosenmeier, M.F., Guilderson, T.P., Chan-Can, E.I. and
Albornaz-Pat, A. 2005: Climate change on the Yucatan Peninsula
during the Little Ice Age. Quaternary Research 63, 109–21.
Hua, Q., Barbetti, M., Worbes, M., Head, J. and Levchenko, V.A.
1999: Review of radiocarbon data from atmospheric and tree ring
samples for the period 1945–1997 AD. IAWA Journal 200, 261–83.
Huang, S.P. and Pollack, H.N. 1998: Global borehole temperature
database for climate reconstruction. IGBP PAGES/World Data
Center-A for Paleoclimatology Data Contribution Series #1998–044,
NOAA/NGDC Paleoclimatology Program.
Huang, S.P., Pollack, H.N. and Shen, P.Y. 2000: Temperature trends
over the past five centuries reconstructed from borehole temperatures.
Nature 403, 756–58.
Hughen, K.A., Overpeck, J.T. and Anderson, R.F. 2000: Recent
warming in a 500-year palaeotemperature record from varved sediments,
Upper Soper Lake, Baffin Island, Canada. The Holocene 10,
9–19.
Hughes, M.K. 2002: Dendrochronology in climatology – the state of
the art. Dendrochronologia 20, 95–116.
Hughes, M.K., Kelly, P.M., Pilcher, J.R. and LaMarche, V.C., editors
1982: Climate from tree rings. Cambridge University Press.
Intergovernmental Panel on Climate Change 1990: Climate
change. The IPCC scientific assessment, working group 1 report.
IPCC, WMO and UNEP, edited by Houghton, J.T., Jenkins, G.J. and
Ephraums, J.J. Cambridge University Press, 364 pp.
––— 1995: Climate change 1995; the science of climate change.
Contribution of working group 1 to the second assessment report of
the Intergovernmental Panel on Climate Change. Edited by
Houghton, J.T., Meira Filho, L.G., Callendar, B.A., Harris, N.,
Kattenbureg, A. and Maskell, K. Cambridge University Press, 572 pp.
––— 2001: Climate change 2001: the scientific basis. Contribution of
working group 1 to the third IPCC scientific assessment. Edited by
Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden,
P.J., Dai, X., Maskell, K. and Johnson, C.A. Cambridge University
Press, 881 pp.
—–– 2007: Climate change 2007: the physical science basis. Edited
by Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt,
K.B., Tignor, M. and Miller, H.L. Cambridge University Press,
996 pp.
Isdale, P. 1984: Fluorescent bands in massive corals record centuries
of coastal rainfall. Nature 310, 578–79.
Jacobeit, J., Glaser, R., Luterbacher, J. and Wanner, H. 2003:
Links between flood events in Central Europe since AD 1500 and
large-scale atmospheric circulation modes. Geophysical Research
Letters 30, 1172, doi:10.1029/2002GL016433.
Jacoby, G.C. 1989: Overview of tree-ring analysis in tropical regions.
IAWA Bulletin 10, 99–108.
Jacoby, G.C. and D’Arrigo, R.D. 1995: Tree-ring width and density
evidence of climatic and potential forest change in Alaska. Global
Biogeochemical Cycles 9, 227–34.
Jacoby, G.C., Lovelius, N.V., Shumilov, O.I., Raspopov, O.M.,
Karbainov, J.M. and Frank, D.C. 2000: Long-term temperature
trends and tree growth in the Taymir region of northern Siberia.
Quaternary Research 53, 312–18.
Jansen, E., Overpeck, J., Briffa, K.R., Duplessy, J.-C., Joos, F.,
Masson-Delmotte, V., Olago, D., Otto-Bliesner, B., Peltier, W.R.,
Rahmstorf, S., Ramesh, R., Raynaud, D., Rind, D., Solomina, O.,
Villalba, R. and Zhang, D. 2007: Palaeoclimate. In Solomon, S., Qin,
D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M.
and Miller, H.L., editors, Climate change 2007: the physical science
basis; contribution of working group I to the fourth assessment report
of the Intergovernmental Panel on Climate Change. Cambridge
University Press, 433–97.
Jansen, E., Andersson, C., Moros, M., Nisancioglu, K.H., Nyland,
B.F. and Telford, R.J. 2008: The early to mid-Holocene thermal optimum
in the North Atlantic. In Battarbee, R. and Binney, H.A., editors,
Natural climate variability and global warming: a Holocene perspective.
Blackwell, 123–37.
Jiang, H., Eiriksson, J., Schulz, M., Knudsen, K.-L. and
Seidenkrantz, M.S. 2005: Evidence for solar forcing of sea-surface
temperature on the North Icelandic shelf during the late Holocene.
Geology 33, 73–76.
Johnsen, S.J. 1977: Stable isotope homogenization of polar firn and
ice. In Proceedings of the symposium on isotopes and impurities in
snow and ice. IUGG XVI, General Assembly, Grenoble, 1975.
International Association of Hydrological Sciences, 18, 210–19.
Johnsen, S.J. and Vinther, B.M. 2007: Stable isotope records from the
Greenland ice cores. In Elias, S.A., editor, Encyclopedia of Quaternary
Science. Elsevier, 1250–58, doi:10.1016/B0-444-52747-8/00345-8.
Johnsen, S.J., Dansgaard, W., Clausen H.B. and Langway, C.C.,
Jr, 1972: Oxygen isotope profiles through the Antarctic and
Greenland ice sheets. Nature 235, 429–34.
Johnsen, S.J., Clausen, H.B., Dansgaard, W., Gundestrup, N.S.,
Hansson, M., Jonsson, P., Steffensen, J.P. and Sveinbjörnsdóttir,
Á.E. 1992: A deep ice core from east Greenland. Meddelelser om
Grønland 29, 3–29.
Johnsen, S.J., Dahl-Jensen, D., Dansgaard, W. and Gundestrup,
N.S. 1995: Greenland palaeotemperatures derived from GRIP bore
hole temperature and ice core isotope profiles. Tellus 47B, 624–29.
Johnsen, S.J., Clausen, H.B., Dansgaard, W., Gundestrup, N.S.,
Hammer, C.U., Andersen, U., Andersen, K.K., Hvidberg, C.S.,
Dahl-Jensen, D., Steffensen, J.P., Shoji, H., Sveinbjörnsdóttir,
Á.E., White, J., Jouzel, J. and Fisher, D. 1997: The d18
O record
along the Greenland Ice Core Project deep ice core and the problem of
possible Eemian climatic instability. Journal of Geophysical Research
102, 26 397–410.
Johnsen, S.J., Clausen, H.B., Cuffey, K.M., Hoffmann, G.,
Schwander, J. and Creyts, T. 2000: Diffusion of stable isotopes in
polar firn and ice: the isotope effect in firn diffusion. In Hondoh, T.,
editor, Physics of ice core records. Hokkaido University Press,
121–40.
Jones, P.D. and Briffa, K.R. 1996: What can the instrumental record
tell us about longer timescale paleoclimatic reconstructions? In Jones,
P.D., Bradley, R.,S. and Jouzel, J., editors, Climatic variations and
forcing mechanisms of the last 2000 years. Springer, 625–43.
––— 2006: Unusual climate in northwest Europe during the period
1730 to 1745 based on instrumental and documentary data. Climatic
Change 79, 361–79.
Jones, P.D. and Mann, M.E. 2004: Climate over past millennia.
Review of Geophysics 42, RG2002, doi: 10.1029/2003RG000143.
Jones, P.D. and Salmon, M.I. 2005: Preliminary reconstructions of
the North Atlantic Oscillation and the Southern Oscillation index from
wind strength measures taken during the CLIWOC period. Climatic
Change 73, 131–54.
Jones, P.D. and Thompson, R. 2003: Instrumental records. In
Mackay, A., Batterbee, R., Birks, J. and Oldfield, F., editors,Global
change in the Holocene. Arnold, 140–58.
42 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Jones, P.D., Briffa, K.R., Barnett, T.P. and Tett, S.F.B. 1998: Highresolution
palaeoclimatic records for the last millennium: integration,
interpretation and comparison with General Circulation Model control
run temperatures. The Holocene 8 455–71.
Jones, P.D., Briffa, K.R. and Osborn, T.J. 2003: Changes in the
Northern Hemisphere annual cycle: implications for paleoclimatology?
Journal of Geophysical Research 108, 4588, doi:10.1029/2003
JD003695.
Jouzel, J., Alley, R.B., Cuffey, K.M., Dansgaard, W., Grootes, P.,
Hoffmann, G., Johnsen, S.J., Koster, R.D., Peel, D., Shuman,
C.A., Stievenard, M., Stuiver, M. and White, J. 1997: Validity of
the temperature reconstruction for water isotopes in ice cores. Journal
of Geophysical Research 102, 26 471–87.
Jouzel, J., Vimeux, F., Caillon, N., Delaygue, G., Hoffmann, G.,
Masson-Delmotte, V. and Parrenin, F. 2003: Magnitude of isotope/temperature
scaling for interpretation of central Antarctic ice
cores. Journal of Geophysical Research 108, 4361, doi:10.1029/2002
JD002677,
Jouzel, J., Masson-Delmotte, V., Cattani, O., Dreyfus, G., Falourd,
S., Hoffmann, G., Minster, B., Nouet, J., Barnola, J.M., Chappellaz,
J., Fischer, H., Gallet, J.C., Johnsen, S., Leuenberger, M.,
Loulergue, L., Luethi, D., Oerter, H., Parrenin, F., Raisbeck, G.,
Raynaud, D., Schilt, A., Schwander, J., Selmo, E., Souchez, R.,
Spahni, R., Stauffer, B., Steffensen, J.P., Stenni, B., Stocker, T.P.,
Tison, J.L., Werner, M. and Wolff, E.W. 2007: Orbital and millennial
Antarctic climate variability over the last 800 000 years. Science 317,
793–96.
Juckes, M.N., Allen, M.R., Briffa, K.R., Esper, J., Hegerl, G.C.,
Moberg, A., Osborn, T.J. and Weber, S.L. 2007: Millennial temperature
reconstruction intercomparison and evaluation. Climate of
the Past 3, 591–609.
Kaspari, S., Mayewski, P.A., Dixon, D., Spikes, V.B., Sneed, S.B.,
Handley, M.J. and Hamilton, G.S. 2004: Climate variability in West
Antarctica derived from annual accumulation rate records from
ITASE firn/ice cores. Annals of Glaciology 39, 585–94.
Kaspari, S., Mayewski, P.A., Dixon, D., Sneed, S.B. and Handley,
M.J. 2005: Sources and transport pathways for marine aerosol species
into West Antarctica. Annals of Glaciology 42, 1–9.
Kinsman, D.J.J. and Holland, H.D. 1969: The co-precipitation of
cations with CaCO3 – IV. The co-precipitation of Sr2+
with aragonite
between 16°C and 96°C. Geochimica et Cosmochimica Acta 33, 1–17.
Kleypas, J.A., McManus, J.W. and Menez, L.A.B. 1999:
Environmental limits to coral reef development: where do we draw the
line? American Zoologist 39, 146–59.
Klingbjer, P. and Moberg, A. 2003: A composite monthly temperature
record from Tornedalen in northern Sweden. 1802–2002.
International Journal of Climatology 23, 1465–94.
Kohn, M.J. and Welker, J.M. 2005: On the temperature correlation
of d18
O in modern precipitation. Earth and Planetary Science Letters
231, 87–96.
Können, G.P., Zaiki, M., Baede, A.P.M., Mikami, T., Jones, P.D.
and Tsukuhara, T. 2003: Pre-1872 extension of the Japanese instrumental
meteorological observation series back to 1819. Journal of
Climate 16, 118–31.
Krinner, G. and Werner, M. 2003: Impact of precipitation seasonality
chnages on isotopic signals in polar ice cores: a multi-model analysis.
Earth and Planetary Science Letters 216, 525–38.
Knuston, D.W., Buddemeier, R.W. and Smith, S.V. 1972: Coral
chronometers: seasonal growth bands in reef corals. Science 177,
270–72.
Küttel, M., Luterbacher, J., Zorita, E., Xoplaki, E., Riedwyl, N.
and Wanner, H. 2007: Testing a European winter surface temperature
reconstruction in a surrogate climate. Geophysical Research Letters
34, L07710. doi:10.1029/2006GL027907.
Lachenbruch, A.H. and Marshall, B.V. 1986: Changing climate:
geothermal evidence from permafrost in the Alaskan Arctic. Science
234, 689–96.
Lamb, H.H. 1965: The earlier medieval warm epoch and its sequel.
Palaeogeography, Palaeoclimatology, Palaeoecology 1, 13–37.
––— 1969: Climatic fluctuations. In Landsberg, H.E., editor, World
survey of climatology. Volume 2: General climatology, edited by
Flohn, H. Elsevier, 173–249.
––— 1977: Climate, present, past and future: volume 2: climatic history
and the future. Methuen, 835 pp.
––— 1982: Climate: history and the modern world. Methuen, 433 pp.
Lamoureux, S.F. 2001: Varve chronology techniques. In Last, W.M.
and Smol, J.P., editors, Developments in paleoenvironmental research
(DEPR) volume 2 – tracking environmental change using lake sediments:
physical and chemical techniques. Kluwer, 247–60.
Langway, C.C., Jr 1967: Stratigraphic analysis of a deep ice core
from Greenland. CRREL, Research Report 77, 130 pp.
Langway, C.C., Jr., Oeschger, H. and Dansgaard, W. 1985: The
Greenland Ice Sheet Program in perspective. Geophysical Monographs
33, 1–8.
Lean, J. 2000: Evolution of the Sun’s spectral irradiance since the
Maunder Minimum. Geophysical Research Letters 27, 2425–28.
Le Bec, N., Juilett-Leclerc, A., Correge, T., Blamart, D. and
Delcroix, T. 2000: A coral d18
O record of ENSO driven sea surface
salinity variability in Fiji (south-western tropical Pacific).
Geophysical Research Letters 27, 3897–900.
Lee, T.C.K., Zwiers, F.W. and Tsao, M. 2008: Evaluation of proxybased
millennial reconstruction methods. Climate Dynamics 31,
263–81.
LeGrande, A.N., Schmidt, G.A., Shindell, D.T., Field, C., Miller,
R.L., Koch, D., Faluvegi, G. and Hoffmann, G. 2006: Consistent
simulations of multiple proxy responses to an abrupt climate change
event. Proceedings of the National Academy of Science 103, 837–42,
10.1073pnas.0510095103.
Leiberman, V. 2003: Strange parallels: southeast Asia in global context
c. 800–1830. Volume 1: integration on the mainland. Cambridge
University Press, 484 pp.
Leijonhufvud, L., Wilson, R. and Moberg, A. 2008: Documentary data
provide evidence of Stockholm average winter to spring temperatures in
the eighteenth and nineteenth centuries. The Holocene 18, 333–43.
Le Quesne, C., Stahle, D.W., Cleaveland, M.K., Therrell, M.D.,
Aravena, J.C. and Barichivich, J. 2006: Ancient Austrocedrus treering
chronologies used to reconstruct Central Chile precipitation variability
from A.D. 1200 to 2000. Journal of Climate 19, 5731–44.
Le Quesne, C., Acuña, C., Boninsegna, J., Rivera, A. and
Barichivich, J. 2008: Long-term glacier variations in the Central
Andes of Argentina and Chile, inferred from historical records and
tree-ring reconstructed precipitation. Special Issue on Regional highresolution
multiproxy climate reconstruction for South America: state
of the art and perspectives. Palaeogeography, Palaeoclimatology,
Palaeoecology doi: 10.1016/j.palaeo.2008.01.039.
Le Roy Ladurie, E. 2004: Histoire humaine et comparée du climate.
Canicules et glaciers XIIIème–XVIIIème siècles, Editions Fayard, 740 pp.
Lewis, S.L., Phillips, O.L. and Baker, T.R. 2006: Impacts of global
atmospheric change on tropical forests. Trends in Ecology and
Evolution 21, 173–74.
Lilly, M.A. 1977: An assessment of the dendrochronological potential
of indigenous tree species in South Africa. Department of Geography
and Environmental Studies, University of the Witwatersrand,
Johannesburg, Occasional Paper No. 18.
Linsley, B.K., Dunbar, R.B., Wellington, G.M. and Mucciarone,
D.A. 1994: A coral-based reconstruction of intertropical convergence
zone variability over central America since 1707. Journal
Geophysical Research 99, 9977–94.
Linsley, B.K., Wellington, G.M. and Schrag, D.P. 2000: Decadal
sea surface temperature variability in the subtropical south Pacific
from 1726 to 1997 A.D. Science 290, 1145–48.
Linsley, B.K., Wellington, G.M., Schrag, D.P., Ren, L., Salinger,
M.J. and Tudhope, A.W. 2004: Geochemical evidence from corals for
changes in the amplitude and spatial pattern of South Pacific interdecadal
climate variability over the last 300 years. Climate Dynamics 22, 1–11,
Lloyd, A.H. and Fastie, C.L. 2002: Spatial and temporal variability
in the growth and climate response of treeline trees in Alaska.
Climatic Change 52, 481–509.
Lough, J.M. 2004: A strategy to improve the contribution of coral
data to high-resolution paleoclimatology. Palaeogeography,
Palaeoclimatology, Palaeoecology 204, 115–43.
––— 2007: Tropical river flow and rainfall reconstructions from coral
luminescence: Great Barrier Reef, Australia. Paleoceanography 22,
PA2218, doi: 10.1029/2006PA001377.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 43
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Lough, J.M. and Barnes, D.J. 2000: Environmental controls on
growth of the massive coral Porites. Journal of Experimental Marine
Biology and Ecology 245, 225–43.
Luckman, B.H. 2007: Dendroclimatology. In Elias, S., editor,
Encyclopedia of Quaternary science. Elsevier, 465–76.
Luckman, B.H. and Wilson, R.J.S. 2005: Summer temperatures in
the Canadian Rockies during the last millennium: a revised record.
Climate Dynamics 24, 131–44.
Lund, D.C., Lynch-Stieglitz, J. and Curry, W.B. 2006: Gulf Stream
density structure and transport during the past millennium. Nature
444, 601–604.
Luterbacher, J., Xoplaki, E., Dietrich, D., Jones, P.D., Davies,
T.D., Portis, D., González-Rouco, F.J., von Storch, H., Gyalistras,
D., Casty, C. and Wanner, H. 2002a: Extending North Atlantic
Oscillation reconstructions back to 1500. Atmospheric Science Letters
2, 114–24 (doi:10.1006/asle.2002.0047).
Luterbacher, J., Xoplaki, E., Dietrich, D., Rickli, R., Jacobeit, J.,
Beck, C., Gyalistras, D., Schmutz, C. and Wanner, H. 2002b:
Reconstruction of sea level pressure fields over the eastern North
Atlantic and Europe back to 1500. Climate Dynamics 18, 545–61.
Luterbacher, J., Dietrich, D., Xoplaki, E., Grosjean, M. and
Wanner, H. 2004: European seasonal and annual temperature variability,
trends, and extremes since 1500. Science 303, 1499–503.
Luterbacher, J., Xoplaki, E., Casty, C., Wanner, H., Pauling, A.,
Kuettel, M., Rutishauser, T., Broennimann, S., Fischer, E.,
Fleitmann, D., Gonzalez-Rouco, F.J., Garcia-Herrera, R.,
Barriendos, M., Rodrigo, F.S., Gonzalez-Hidalgo, J.C., Saz, M.A.,
Gimeno, L., Ribera, P., Brunet, M., Paeth, H., Rimbu, N., Felis, T.,
Jacobeit, J., Dünkeloh, A., Zorita, E., Guiot, J., Türkes, M.,
Alcoforado, M.J., Trigo, R., Wheeler, D., Tett, S., Mann, M.E.,
Touchan, R., Shindell, D.T., Silenzi, S., Montagna, P., Camuffo,
D., Mariotti, A., Nanni, T., Brunetti, M., Maugeri, M., Zerefos,
C.S., De Zolt, S., Lionello, P., Rath, V., Beltrami, H., Garnier, E.
and Le RoyLadurie, E. 2006: Mediterranean climate variability over
the last centuries: a review. In Lionello, P., Malanotte-Rizzoli, P. and
Boscolo, R., editors, The Mediterranean climate: an overview of the
main characteristics and issues. Elsevier, 27–148.
Luterbacher, J., Liniger, M.A., Menzel, A., Estrella, N., Della-Marta,
P.M., Pfister, C., Rutishauser, T. and Xoplaki, E. 2007: The exceptional
European warmth of Autumn 2006 and Winter 2007: historical
context, the underlying dynamics and its phenological impacts.
Geophysical Research Letters 34, L12704, doi:10.1029/2007GL029951.
MacFarling Meure, C., Etheridge, D., Trudinger, C., Steele, P.,
Langenfelds, R., van Ommen, T., Smith, A. and Elkins, J. 2006:
Law Dome CO2, CH4 and N2O ice core records extended to 2000 years
BP. Geophysical Research Letters 33, L14810, doi:10.1029/2006
GL026152.
Mangini, A., Spötl, C. and Verdes, P. 2005: Reconstruction of temperature
in the central Alps during the last 2000 years from a d18
O stalagmite
record. Earth and Planetary Science Letters 234, 741–51.
Manley, G. 1974: Central England temperatures: monthly means
1659 to 1973. Quarterly Journal of the Royal Meteorological Society
100, 389–405.
Mann, M.E. 2004: On smoothing potentially non-stationary climate
time series. Geophysical Research Letters 31, L07214, doi:10.1029/
2004GL019569.
Mann, M.E. and Jones, P.D. 2003: Global surface temperatures over
the past two millennia. Geophysical Research Letters 30, 1820,
doi:10.1029/2003GL017814.
Mann, M.E. and Rutherford, S. 2002: Climate reconstruction using
‘pseudoproxies’. Geophysical Research Letters 29, 1501, doi:10.1029/
2001GL014554.
Mann, M.E. and Schmidt, G.A. 2003: Ground vs. surface air temperature
trends: implications for borehole surface temperature reconstructions.
Geophysical Research Letters 30, 1607, doi:10.1029/
2003GL017170.
Mann, M.E., Bradley, R.S. and Hughes, M.K. 1998: Global-scale
temperature patterns and climate forcing over the past six centuries.
Nature 392, 779–87.
––— 1999: Northern Hemisphere temperatures during the past millennium:
inferences, uncertainties and limitations. Geophysical
Research Letters 26, 759–62.
Mann, M.E., Gille, E., Overpeck, J., Gross, W., Bradley, R.S.,
Keimig, F.T. and Hughes, M.K. 2000: Global temperature patterns in
past centuries: an interactive presentation. Earth Interactions 4, 1–29.
Mann, M.E., Rutherford, S., Bradley, R.S., Hughes, M.K. and
Keimig, F.T. 2003: Optimal surface temperature reconstructions
using terrestrial borehole data. Journal of Geophysical Research 108,
4203, doi: 10.1029/2002JD002532.
Mann, M.E., Rutherford, S., Wahl, E. and Ammann, C. 2005:
Testing the fidelity of methods used in proxy-based reconstructions of
past climate. Journal of Climate 18, 4097–107.
Mann, M.E., Briffa, K.R., Jones, P.D., Kiefer, T., Kull, C. and
Wanner, H. 2006: Past millennia climate variability. EOS 87,
526–27.
Mann, M.E., Rutherford, S., Wahl, E. and Amman, C. 2007:
Robustness of proxy-based climate field reconstruction methods.
Journal of Geophysical Research 112, D12109, doi:10.1029/2006
JD008272.
Mariaux, A. 1967: Les cernes dans les bois tropicaux africains, nature
et periodicite. Bois et Forets des Tropiques 113–114, 3–14.
Mayewski, P.A. and Goodwin, I.D. 1997: International transAntarctic
scientific expedition (ITASE), ‘200 years of past Antarctic
climate and environmental change’, science and implementation plan,
1996. PAGES Workshop Report, Series 97-1, 48 pp.
Mayewski, P.A., Frezzotti, M., Bertler, N., van Ommen, T.,
Hamilton, G., Jacka, T.H., Welch, B., Frey, M., Qin, D., Ren, J.,
Simoes, J., Fily, M., Oerter, H., Nishio, F., Isaksson, E., Mulvaney,
R., Holmlund, P., Lipenkov, V. and Goodwin, I. 2005: The international
trans-Antarctic scientific expedition (TASE): an overview.
Annals of Glaciology 41, 180–85.
McCarroll, D. and Loader, N.J. 2004: Stable isotopes in tree rings.
Quaternary Science Reviews 23, 771–801.
McConnaughey, T.A. 1989a: 13
C and 18
O isotopic disequilibrium in
biological carbonates: I. Patterns. Geochimica et Cosmochimica Acta
53, 151–62.
––— 1989b: 13
C and 18
O disequilibrium in biological carbonates: II. In
vitro simulation of kinetic isotope effects. Geochimica et Cosmochimica
Acta 53, 163–71.
––— 2003: Sub-equilibrium oxygen-18 and carbon-13 levels in biological
carbonates: carbonate and kinetic models. Coral Reefs 22,
316–27.
McConnell, J.R., Lamorey, G.W. Lambert, S.W. and Taylor, K.C.
2002: Continuous ice-core chemical analyses using inductively coupled
plasma mass spectrometry. Environmental Science & Technology
36, 7–11.
McCulloch, M.T., Fallon, S., Wyndham, T., Hendy, E., Lough, J.
and Barnes, D. 2003: Coral record of increased sediment flux to the
inner Great Barrier Reef since European settlement, Nature 421,
727–30.
McGregor, H.V., Dima, M., Fischer, H.W. and Mulitza, S. 2007:
Rapid 20th-century increase in coastal upwelling off northwest Africa.
Science 315, 637–39.
McManus, J.F., Francois, R., Gherardi, J.-M., Keigwin, L.D. and
Brown-Leger, S. 2004: Collapse and rapid resumption of Atlantic
meridional circulation linked to deglacial climate changes. Nature
428, 834–37.
McMorrow, A., van Ommen, T.D., Morgan, V. and Curran,
M.A.J. 2004: Ultra-high-resolution seasonality of trace-ion species
and oxygen isotope ratios in Antarctic firn over four annual cycles.
Annals of Glaciology 39, 34–40.
Meier, N., Rutishauser T., Pfister C., Wanner H. and Luterbacher,
J. 2007: Grape harvest dates as a proxy for Swiss April to August temperature
reconstructions back to AD 1480. Geophysical Research
Letters 34, L20705, doi:10.1029/2007GL031381.
Melvin, T.M. and Briffa, K.R. 2008: A ‘signal-free’ approach to dendroclimatic
standardisation. Dendrochronologia 26, 71–86.
Meyerson, E.A., Mayewski, P.A., Kreutz, K.J., Meeker, L.D.,
Whitlow, S.I. and Twickler, M.S. 2002: The polar expression of
ENSO and sea-ice variability as recorded in a South Pole ice core.
Annals of Glaciology 35, 430–36.
Mikami, T. 1999: Quantitative reconstruction in Japan based on historical
documents. Bulletin of the National Museum of Japanese
History 81, 41–50.
44 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
––— 2002: Summer and winter temperature reconstructions in Japan. In
Pfister, C., Wanner, H., Kull, C. and Alverson, K., editors. PAGES (Past
Global Changes) Newsletter on Documentary Evidence 10, 17–18.
Min, R.G., Edwards, R.L., Taylor, F.W., Recy, J., Gallup, C.D.
and Beck, J.W. 1995: Annual cycles of U/Ca in coral skeletons and
U/Ca thermometry. Geochimica et Cosmochimica Acta 59, 2025–42.
Mitsuguchi, T., Matsumoto, E., Abe, O., Uchida, T. and Isdale, P.J.
1996: Mg/Ca thermometry in coral skeletons. Science 274, 961–63.
Moberg, A., Alexandersson, H., Bergström, H. and Jones, P.D.
2003: Were south Swedish summer temperatures before 1860 as warm
as measured? International Journal of Climatology 23, 1495–521.
Moberg, A., Sonechkin, D.M., Holmgren, K., Datsenko, N.M. and
Karlen, W. 2005: Highly variable Northern Hemisphere temperatures
reconstructed from low- and high-resolution proxy data. Nature 433,
613–17.
Moberg, A., Mohammad, R. and Mauritsen, T. 2008: Analysis of
the Moberg et al. (2005) hemispheric temperature reconstruction.
Climate Dynamics 31, 957–71.
Montagna, P., McCulloch, M., Taviani, M., Mazzoli, C. and
Vendrell, B. 2006: Seawater nutrient chemistry phosphorus in coldwater
corals. Science 312, 1788–1791. DOI: 10.1126/science.1125781.
Morales, M.S., Villalba, R., Grau, H.R. and Paolini, L. 2004:
Rainfall-controlled tree growth in high-elevation subtropical treelines.
Ecology 85, 3080–89.
Morgan, V.I. and van Ommen, T.D. 1997: Seasonality in lateHolocene
climate from ice core records. The Holocene 7, 351–54.
Moros, M., Andrews, J.T., Eberl, D.D. and Jansen, E. 2006:
Holocene history of drift ice in the northern North Atlantic: evidence
for different spatial and temporal modes. Paleoceanography 21,
PA2017, doi:10.1029/2005PA001214.
Mosley-Thompson, E. 1992: Paleoenvironmental conditions in
Antarctica since A.D. 1500: ice core evidence. In Bradley, R.S. and
Jones, P.D., editors, Climate since A.D. 1500. Routledge, Chapman
and Hall, 572–91.
Mosley-Thompson, E., Thompson, L.G., Grootes, P. and
Gundestrup, N. 1990: Little Ice Age (Neoglacial) paleoenvironmental
conditions at Siple Station, Antarctica. Annals of Glaciology 14,
199–204.
Mosley-Thompson, E., McConnell, J.R., Bales, R.C., Li, Z., Lin,
P.-N., Steffen, K., Thompson, L.G., Edward, R. and Bathke, D.
2001: Local to regional scale variability of annual net accumulation on
the Greenland ice sheet from PARCA cores. Journal of Geophysical
Research 106, 33 839–51.
Mosley-Thompson, E., Readinger, C.R., Craigmile, P.,
Thompson, L.G. and Calder, C.A. 2005: Regional sensitivity of
Greenland precipitation to NAO variability. Geophysical Research
Letters 32, L24707, doi:10.1029/2005GL024776.
Mosley-Thompson, E., Thompson, L.G. and Lin, P.-N. 2006: A
multi-century perspective on 20th century climate change with new
contributions from high Arctic and Greenland (PARCA) cores. Annals
of Glaciology 43, 42–48.
Mudelsee, M., Börngen, M., Tetzlaff, G. and Grünewald, U. 2003:
No upward trends in the occurrence of extreme floods in central
Europe. Nature 425, 166–69.
Muscheler, R., Joos, F., Mueller, S.A. and Snowball, I. 2005: How
unusual is today’s solar activity? Nature 436, E3–E4.
Muller, A., Gagan, M.K. and McCulloch, M.T. 2002: Early marine
diagenesis in corals and geochemical consequences for paleoceanographic
reconstructions. Geophysical Research Letters 28, 4471–74.
National Research Council of the US National Academy of
Sciences 2006: Surface temperature reconstructions for the last 2000
years. National Academies Press, 145 pp.
Naurzbaev, M.M., Vaganov, E.A., Sidorova, O.V. and
Schweingruber, F.H. 2002: Summer temperatures in eastern Taimyr
inferred from a 2427-year late-Holocene tree-ring chronology and earlier
floating series. The Holocene 12, 727–36.
Naveau, P. and Ammann, C.M. 2005: Statistical distributions of ice
core sulfate from climatically relevant volcanic eruptions. Geophysical
Research Letters 32, L05711, doi:10.1029/2004GL021732.
Neter, J, Wasserman, W. and Kutner, M.H. 1990: Applied linear
statistical models. Regression, analysis of variance and experimental
designs. Third edition. Irwin, 1181 pp.
Nicault, A., Alleaume, S., Brewer, S., Carrer, M., Nola, P. and
Guiot, J. 2008: Mediterranean drought fluctuation during the last
500 years based on tree-ring data. Climate Dynamics 31, 227–45.
Nicholls, N., Gruza, G.V., Jouzel, J., Karl, T.R., Ogallo, L.A. and
Parker, D.E. 1996: Observed climate variability and change. In
Houghton, J.T., Meira Filho, L.G., Callander, B.A., Harris, N.,
Kattenberg, A. and Maskell, K., editors, The IPCC second scientific
assessment. Cambridge University Press, 133–92.
Nicolussi, K. and Schiessling, P. 2001: Establishing a multi-millennial
Pinus cembra chronology for the central Eastern Alps. In Tree rings
and people. An international conference on the future of dendrochronology.
Davos, Abstracts 87.
Nobuchi, T., Ogata, Y. and Siripatanadilok, S. 1995: Seasonal characteristics
of wood formation in Hopea odorata and Shorea henryana.
IAWA Journal 16, 361–69.
North Greenland Ice Core Project (NGRIP) Members 2004: Highresolution
record of Northern Hemisphere climate extending into the
last interglacial period. Nature 431, 147–51.
Nyberg, J. 2002: Luminescence intensity in coral skeletons from
Mona Island in the Caribbean Sea and its link to precipitation and
wind speed. Philosophical Transactions of the Royal Society London
A 360, 749–66.
Ogilvie, A.E.J. 1992: Documentary evidence for changes in the climate
of Iceland, A.D. 1500–1800. In Bradley, R.S. and Jones, P.D.,
editors, Climate since A.D. 1500. Routledge, 92–117.
––— 1996: Sea ice conditions off the coasts Iceland A.D. 1601–1850
with special reference to part of the Maunder Minimum period (1675–
1715). AmS-Varia 25, 9–12.
Ogilvie, A.E.J. and Farmer, G. 1997: Documenting the Medieval climate.
In Hulme, M. and Barrow, E., editors, Climates of the British
Isles: present, past and future. Routledge, 112–33.
Ogilvie, A.E.J. and Jónsdóttir, I. 1996: Sea ice incidence off the
coasts of Iceland: evidence from historical data and early sea ice
maps. In 26th international Arctic workshop, Arctic and Alpine environments,
past and present program with abstracts. INSTAAR,
Boulder CO, 14–16 March 1996, 109–10.
Oman, L., Robock, A., Stenchikov, G., Schmidt, G.A. and Ruedy,
R. 2005: Climatic response to high-latitude volcanic eruptions.
Journal of Geophysical Research 110, D13103, doi:10.1029/2004
JD005487.
Ortlieb, L. 2000: The documentary historical record of El Niño
events in Peru: an update of the Quinn record (sixteenth through nineteenth
centuries). In Diaz, H.F. and Markgraf, V., editors, El Niño and
the Southern Oscillation: multiscale variability and its impacts on
natural ecosystems and society. Cambridge University Press, 207–95.
Osborn, T.J. and Briffa, K.R. 2000: Revisiting timescale-dependent
reconstruction of climate from tree-ring chronologies.
Dendrochronologia 18, 9–25.
––— 2004: The real color of climate change? Science 306, 621–22.
––— 2006: The spatial extent of 20th-century warmth in the context
of the past 1200 years. Science 311, 841–44.
––— 2007: Response to the comment on ‘The spatial extent of
20th-century warmth in the context of the past 1200 years’. Science
316, 1844b.
Osborn, T.J., Briffa, K.R. and Jones, P.D. 1997: Adjusting variance
for sample-size in tree-ring chronologies and other regional timeseries.
Dendrochronologia 15, 89–99.
Osborn, T.J., Raper, S.C.B. and Briffa, K.R. 2006: Simulated climate
change during the last 1000 years: comparing the ECHO-G general
circulation model with the MAGICC simple climate model.
Climate Dynamics 27, 185–97.
Overland, J.E. and Wood, K. 2003: Accounts from 19th century
Canadian Arctic explorers’ logs reflect present climate conditions.
Eos 84, 410–12.
Overpeck, J.T. 1996: Varved sediment records of recent seasonal to
millennial-scale environmental variability. In Jones, P.D., Bradley,
R.S. and Jouzel, J., editors, Climatic variations and forcing mechanisms
of the last 2000 years. Springer-Verlag, 479–598.
Overpeck, J.T., Rind, D. and Goldberg, R. 1990: Climate-induced
changes in forest disturbance and vegetation. Nature 343, 51–53.
Palmer, A.S., van Ommen, T.D., Curran, M.A.J., Morgan, V.I.,
Souney, J.M. and Mayewski, P.A. 2001: High precision dating of
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 45
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
volcanic events (AD 1301–1995) using ice cores from Law Dome,
Antarctica. Journal of Geophysical Research 106, 28 089–96.
Palmer, W.C. 1965: Meteorological drought. US Department of
Commerce Weather Bureau, Research Paper No. 45.
Partin, J.W., Cobb, K.M., Adkins, J.F., Clark, B. and Fernandez,
D.P. 2007: Millennial-scale trends in west Pacific warm pool hydrology
since the Last Glacial Maximum. Nature 449, 452–55.
Pauling, A., Luterbacher, J. and Wanner, H. 2003: Evaluation of proxies
for European and North Atlantic temperature field reconstructions.
Geophysical Research Letters 30, 1787 (doi 10.1029/2003GL017589).
Pauling, A., Luterbacher, J., Casty, C. and Wanner, H. 2006: 500
years of gridded high-resolution precipitation reconstructions over
Europe and the connection to large-scale circulation. Climate
Dynamics 26, 387–405.
Peel, D.A. 1992: Ice core evidence from the Antarctic Peninsula
region. In Bradley R.S. and Jones, P.D., editors, Climate since A.D.
1500. Routledge, Chapman and Hall, 549–71.
Peel, D.A., Mulvaney, R. and Davison, B.M. 1988: Stableisotope/air-temperature
relationships in ice cores from Dolleman
Island and the Palmer Land Plateau, Antarctic peninsula. Annals of
Glaciology 10, 130–36.
Pfister, C. 1992: Monthly temperature and precipitation patterns in
Central Europe from 1525 to the present. A methodology for quantifying
man-made evidence on weather and climate. In Bradley, R.S.
and Jones, P.D., editors, Climate since A.D. 1500. Routledge,
Chapman and Hall, 118–42.
Pfister, C., Luterbacher, J., Schwarz-Zanetti, G. and Wegmann,
M. 1998: Winter air temperature variations in western Europe during
the Early and High Middle Ages (AD 750–1300). The Holocene 8,
535–52.
Pfister C., Brázdil, R. and Glaser, R., guest editors 1999: Climatic
variability in sixteenth century Europe and its social dimension: a synthesis.
Special Volume. Climatic Change 1/43.
Pfister, C., Weingartner, R. and Luterbacher, J. 2006:
Reconstruction of extreme hydrological winter droughts over the last
450 years in the Upper Rhine basin. A methodological approach.
Hydrological Science Journal 51, 966–85.
Pisaric, M.F.J., Carey, S.K., Kokelj, S.V. and Youngblut, D. 2007:
Anomalous 20th century tree growth, Mackenzie Delta, Northwest
Territories, Canada. Geophysical Research Letters 34, L05714,
doi:10.1029/2006GL029139.
Pollack, H.N. and Huang, S.P. 2000: Climate reconstruction from
subsurface temperatures. Annual Review of Earth and Planetary
Science 28, 339–65.
Pollack, H.N. and Smerdon, J.E. 2004: Borehole climate reconstructions:
spatial structure and hemispheric averages. Journal of
Geophysical Research 109, D11106, doi:10.1029/2003JD004163.
Polyak, V.J. and Asmerom, Y. 2001: Late Holocene climate and cultural
changes in the Southwestern United States. Science 294, 148–51.
Popa, I. and Kern, Z. 2008: Long-term summer temperature reconstruction
inferred from tree-ring records from the Eastern Carpathians.
Climate Dynamics DOI:10.1007/s00382–008–0439-x.
Poussart, P.F. and Schrag, D.P. 2005: Seasonally-resolved stable
isotope chronologies from northern Thailand deciduous trees. Earth
and Planetary Science Letters 235, 752–65.
Poussart, P.F., Evans, M.N. and Schrag, D.P. 2004: Resolving seasonality
in tropical trees: multi-decade, high-resolution oxygen and
carbon isotopic records from Indonesia and Thailand. Earth and
Planetary Science Letters 218, 301–16.
Poussart, P.F., Myneni, S.C.B. and Lanzirotti, A. 2006: Tropical
dendrochemistry: a novel approach to estimate age and growth from
ringless trees. Geophysical Research Letters 33, L17711,
doi:10.1029/2006GL026929.
Prieto, M.R. and Herrera, R. 1999: Austral climate and glaciers in
the 16th century through the observations of the Spanish Navigators.
Quaternary of South America and Antarctic Peninsula 11, 153–79.
Prieto, M.R., Herrera, R. and Dussel, P. 2000: Archival evidence for
some aspects of historical climate variability in Argentina and Bolivia during
the 17th and 18th centuries. In Volkheimer, W. and Smolka, P., editors,
Southern Hemisphere paleo- and neoclimates. Springer-Verlag, 127–42.
Prieto, M.R., García-Herrera, R. and Hernandez Martin, E. 2004:
Early records of icebergs in the South Atlantic Ocean from Spanish
documentary sources. Climate Change 66, 29–48.
Proctor, C.J., Baker, A., Barnes, W.L. and Gilmour, M.A. 2004: A
thousand year speleolthem proxy record of North Atlantic climate
from Scotland. Climate Dynamics 16, 815–20.
Pumijumnong, N., Eckstein D. and Sass, U. 1995: Tree-ring
research on Tectona grandis in northern Thailand. IAWA Journal
16, 385–92.
Qian, W-H. and Zhu, Y-F. 2002: Little Ice Age climate near Beijing,
China, inferred from historical and stalagmite records. Quaternary
Research 57, 109–19.
Qian, W., Hu, Q., Zhu, Y. and Lee, D.-K. 2003: Centennial-scale
dry-wet variations in East Asia. Climate Dynamics 21, 77–89.
Quinn, T.M. and Sampson, D.E. 2002: A multiproxy approach to
reconstructing sea surface conditions using coral skeleton geochemistry.
Paleoceanography 17, 1062, doi:10.1029/2000PA000528.
Quinn, T.M. and Taylor, F.W. 2006: SST artefacts in coral proxy
records produced by early marine diagenesis in a modern coral from
Rabaul, Papua New Guinea. Geophysical Research Letters 33,
L04601, doi:10.1029/2005GL024972.
Quinn, T.M., Taylor, F.W., Crowley, T.J. and Link, S.M. 1996:
Evaluation of sampling resolution in coral stable isotope records: a
case study using records from New Caledonia and Tarawa.
Paleoceanography 11, 529–42.
Quinn, W.H. and Neal, V.T. 1992: The historical record of El Niño
events. In Bradley, R.S. and Jones, P.D., editors, Climate since A.D.
1500. Routledge, 623–48.
Rasmussen, S.O., Andersen, K.K., Svensson, A.M., Steffensen,
J.P., Vinther, B.M., Clausen, H.B., Siggaard-Andersen, M.-L.,
Johnsen, S.J., Larsen, L.B., Bigler, M., Röthlisberger, R., Fischer,
H., Goto-Azuma, K., Hansson, M.E. and Ruth, U. 2006: A new
Greenland ice core chronology for the last glacial termination. Journal
of Geophysical Research 111, doi:10.1029/2005JD006079.
Reeh, N., Johnsen, S.J. and Dahl-Jensen, D. 1985: Dating the Dye-
3 ice core by flow model calculations. AGU Geophysical Monograph,
33, 57–65.
Riedwyl, N., Luterbacher, J. and Wanner, H. 2008a: An ensemble
of European summer and winter temperature reconstructions back to
1500. Geophysical Research Letters 35, L20707, doi: 10.1029/2008
GL035395.
Riedwyl, N., Küttel, M., Luterbacher, J. and Wanner, H. 2008b:
Comparison of climate field reconstruction techniques: application to
Europe. Climate Dynamics DOI:10.1007/s00382–008–0395–5.
Rivera, G., Elliott, S., Caldas, L.S., Nicolossi, G., Coradin, V.T.R.
and Borchert, R. 2002: Increasing day-length induces spring flushing
of tropical dry forest trees in the absence of rain. Trees 16, 445–56.
Robock, A. and Free, M.P. 1995: Ice cores as an index of global volcanism
from 1850 to the present. Journal of Geophysical Research
100, 11 549–67.
Rogers, J.C., Bolzan, J.F. and Pohjola, V.A. 1998: Atmospheric circulation
variability associated with shallow-core seasonal isotopic
extremes near Summit, Greenland. Journal of Geophysical Research
107, 11 205–19.
Romero-Viana, L., Julia, R., Camacho, A., Vicente, E. and
Miracle, M.R. 2008: Climate signal in varve thickness: Lake La Cruz
(Spain), a case study. Journal of Paleolimnology 40, 703–14.
Rozanski, K., Araguás-Araguás, L. and Gonfiantini, R. 1992:
Relation between long-term trends of oxygen-18 isotope composition
of precipitation and climate. Science 258, 981–84.
––— 1993: Isotopic patterns in modern global precipitation. In Swart,
P.K., Lohmann, K.C., McKenzie, J. and Savin, S., editors, Climate
change in continental isotope records. AGU Geophysical Monograph,
78, American Geophysical Union, 1–37.
Rozanski, K., Johnsen, S.J., Schotterer, U. and Thompson, L.G.
1997: Reconstruction of past climates from stable isotope records of
palaeo-precipitation preserved in continental archives. Hydrological
Sciences – Journal-des Sciences Hydrologiques 42, 725–45.
Ruddiman, W.F. 2003: The anthropogenic greenhouse era began
thousands of years ago. Climate Change 61, 261–93.
Rutherford, S., Mann, M.E., Delworth, T.L. and Stouffer, R. 2003:
Climate field reconstruction under stationary and nonstationary forcing.
Journal of Climate 16, 462–79.
Rutherford, S., Mann, M.E., Osborn, T.J., Bradley, R.S., Briffa,
K.R., Hughes, M.K. and Jones, P.D. 2005: Proxy-based Northern
Hemisphere surface temperature reconstructions: sensitivity to
46 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
methodology, predictor network, target season and target domain.
Journal of Climate 18, 2308–329.
Rutishauser, T., Luterbacher, J., Defila, C., Frank, D. and Wanner,
H. 2008: Swiss spring plant phenology 2007: extremes, a multicentury
perspective and changes in temperature sensitivity. Geophysical
Research Letters 35, L05703. doi:10.1029/2007GL032545.
Saleska, S.R., Didan, K., Huete, A.R. and da Rocha, H.R. 2007:
Amazon forests green-up during 2005 drought. Science 318, 612.
Sano, M., Buckley, B.M. and Sweda, T. 2008: Tree-ring based
hydroclimate reconstruction over northern Vietnam from Fokienia
hodginsii: 18th century mega-drought and tropical Pacific influence.
Climate Dynamics doi: 10.1007/s00382-008-0454-y.
Schilman, B., Bar-Matthews, M., Almogi-Labin, A. and Luz, B.
2001: Global climate instability reflected by Eastern Mediterranean
marine records during the late Holocene. Palaeogeography,
Palaeoclimatology, Palaeoecology 176, 157–76.
Schmidt, G.A. and Mann, M.E. 2004: Reply to comment on ‘Ground
vs. surface air temperature trends: implications for borehole surface
temperature reconstructions’ by D. Chapman et al. Geophysical
Research Letters 31, L07206, doi: 10.1029/2003GL0119144.
Schmidt, G.A., LeGrande, A.N. and Hoffmann, G. 2007: Water isotope
expressions of intrinsic and forced variability in a coupled ocean–
atmosphere model. Journal of Geophysical Research 112, D10103,
doi:10.1029/2006JD007781.
Schneider, D.P., Steig, E.J., van Ommen, T.D., Dixon, D.A.,
Mayewski, P.A., Jones, J.M. and Bitz, C.M. 2006: Antarctic temperatures
over the past two centuries from ice cores. Geophysical
Research Letters 33, L16707, doi:10.1029/2006GL027057.
Schneider, T. 2001: Analysis of incomplete climate data: estimation
of mean values and covariance matrices and imputation of missing
values. Journal of Climate 14, 853–71.
Schöngart, J., Piedade, M., Wittmann, F., Junk, W. and Worbes, M.
2004a: Wood growth patterns of Macrolobium acaciifolium(Benth.)
Benth.(Fabaceae) in Amazonian black-water and white-water floodplain
forests. Oecologia 145, 454–61.
Schöngart, J., Junk, W.J., Piedade, M.T.F., Ayres, J.M.,
Hüttermann, A. and Worbes, M. 2004b: Teleconnections between
tree growth in the Amazonian floodplains and the El Niño-Southern
Oscillation effect. Global Change Biology 10, 683–92.
Schrag, D.P. 1999: Rapid determination of high-precision Sr/Ca
ratios in corals and other marine carbonates. Paleoceanography 14,
97–102.
Schweingruber, F.H. 1996: Tree rings and environment dendroecology.
Paul Haupt.
Schweingruber, F. and Briffa, K.R. 1996: Tree-ring density networks
for climatic reconstruction. In Jones, P.D., Bradley, R.S. and
Jouzel, J., editors, Climatic variations and forcing mechanisms of the
last 2000 years. NATO Series, Springer-Verlag, Vol. I 41, 44–66.
Schweingruber, F.H., Briffa, K.T. and Jones, P.D. 1991: Yearly
maps of summer temperatures in Western-Europe from AD 1750 to
1975 and western North-America from 1600 to 1982 – results of a
radiodensitometrical study on tree rings. Vegetatio 92, 5–71.
Severinghaus, J., Sowers, T., Brook, E.J., Alley, R.B. and Bender,
M. 1998: Timing of abrupt climate change at the end of the Younger
Dryas interval from thermally fractionated gases in polar ice. Nature
391, 141–46.
Scourse, J., Richardson, C., Forsythe, G., Harris, I., Heinemeier,
J., Fraser, N., Briffa, K.R. and Jones, P.D. 2006: First crossmatched
floating chronology from the marine fossil record: data from
growth lines of long-lived bivalve mollusc Arctica islandica. The
Holocene 16, 967–74.
Shanahan, T.M., Overpeck, J.T., Beck, J.W., Wheeler, C.W.,
Peck, J.A., King, L.W. and Scholz, C.A. 2008: The formation of biogeochemical
laminations in Lake Bosumtwi, Ghana, and their usefulness
as indicators of past environmental changes. Journal of
Paleolimnology 40, 339–55.
Shen, G.T., Linn, L.J., Campbell, T.M., Cole, J.E. and Fairbanks,
R.G. 1992: A chemical indicator of trade-wind reversal in corals from
the western tropical pacific. Journal of Geophysical Research-Oceans
97, 12 689–97.
Shindell, D.T., Schmidt, G.A., Mann, M.E., Rind, D. and Waple,
A. 2001: Solar forcing of regional climate change during the Maunder
Minimum. Science 294, 2149–52.
Shindell, D.T., Schmidt, G.A., Mann, M.E. and Faluvegi, G. 2004:
Dynamic winter climate response to large tropical volcanic eruptions
since 1600. Journal of Geophysical Research 109, D05104, doi:
10.1029/2003JD004151.
Shindell, D.T., Faluvegi, G., Miller, R.L., Schmidt, G.A., Hansen,
J.E. and Sun, S. 2006: Solar and anthropogenic forcing of tropical
hydrology. Geophysical Research Letters 33, L24706, doi:
10.1029/2006GL027468
Shulmeister, J., Goodwin, I., Renwick, J., Harle, K., Armand, L.,
McGlone, M.S., Cook, E., Dodson, J., Hesse, P.P., Mayewski, P.A.
and Curran, M. 2004: The Southern Hemisphere westerlies in the
Australasian sector over the last glacial cycle: a synthesis. Quaternary
International 118–119, 23–53.
Shuman, C.A., Alley, R.B., Fahnestock, M.A., Bindschadler, R.A.,
White, J.W.C., McConnell, J.R. and Winterle, J. 1998:
Temperature history and accumulation timing for the snow pack at
GISP2, central Greenland. Journal of Glaciology 44, 21–30.
Sidorova, O.V., Naurzbaev, M.M. and Vaganov, E. 2006: Climatic
changes in subarctic Eurasia based on millennial tree ring chronologies.
Abstract at Holivar2006, available at http://www.holivar2006.
org/abstracts/viewabstract_search.php?
Silenzi, S., Antonioli, F. and Chemello, R. 2004: A new marker for
sea surface temperature trend during the last centuries in temperate
areas: Vermetid reef. Global Planetary Change 40, 105–14.
Silenzi, S., Bard, E., Montagna, P. and Antonioli, F. 2005: Isotopic
and elemental records in a non-tropical coral (Cladocora caespitosa):
discovery of a new highresolution climate archive for the Mediterranean
Sea. Global Planetary Change 49, 94–120.
Smerdon, J.E. and Kaplan, A. 2007: Comments on ‘Testing the
fidelity of methods used in proxy-based reconstructions of past climate’:
the role of the standardization interval, by M.E. Mann, S.
Rutherford, E. Wahl, and C. Ammann. Journal of Climate 20, 5666–70.
Smerdon, J.E., Pollack, H.N., Cermak, V., Enz, J.W., Kresl, M.,
Safanda, J. and Wehmiller, J.F. 2004: Air-ground temperature coupling
and subsurface propagation of annual temperature signals.
Journal of Geophysical Research 109, D21107, doi:10.1029/2004J
D005056.
Smith, C.L., Baker, A., Fairchild, I.J., Frisia, S. and Borsato, A.
2006: Reconstructing hemispheric-scale climates from multiple stalagmite
records. International Journal of Climatology 26, 1417–24.
Smith, S.V., Buddemeier, R., Redalje, R. and Houck, J. 1979:
Strontium-calcium thermometry in coral skeletons. Science 204,
404–406.
Solanski, S.K., Usoskin, I.G., Kromer, B., Schüssler, M. and Beer, J.
2004: Unusual activity of the Sun during recent decades compared to
the previous 11,000 years. Nature 431, 1084–87.
Solíz, C., Villalba, R., Argollo, J., Morales, M.S., Christie, D.A.,
Moya, J. and Pacajes, J. 2008: Spatio-temporal variations in
Polylepis tarapacana radial growth across the Bolivian Altiplano during
the 20th century. Special Issue on Regional high-resolution multiproxy
climate reconstruction for South America: state of the art and
perspectives. Palaeogeography, Palaeoclimatology, Palaeoecology
doi: 10.1016/j.paleo.2008.07.025.
Song, J. 1998: Changes in dryness/wetness in China during the last
529 years. International Journal of Climatology 18, 1345–55.
––— 2000: Reconstruction of the Southern Oscillation from dryness/wetness
in China for the last 500 years. International Journal of
Climatology 20, 1003–16.
Souney, J.M., Mayewski, P.A., Goodwin, I.D., Meeker, L.D.,
Morgan, V., Curran, M.A.J., van Ommen, T.D. and Palmer, A.S.
2002: A 700-year record of atmospheric circulation developed from
the Law Dome ice core, East Antarctica. Journal of Geophysical
Research 107, 4608, doi:10.1029/2002JD002104.
Stahle, D.W., Cleaveland, M.K., Haynes, G.A., Klinowicz, J.,
Muskove, P., Nyrvenya, P. and Nicholson, P. 1997: Development of a
rainfall-sensitive tree-ring chronology in Zimbabwe. Eighth symposium
on global change studies. American Meteorological Society, 205–11.
Stahle, D.W., D’Arrigo, R.D., Krusic, P.J., Cleaveland, M.K.,
Cook, E.R., Allan, R.J., Cole, J.E., Dunbar, R.B., Therrell, M.D.,
Gay, D.A., Moore, M.D., Stokes, M.A., Burns, B.T., VillanuevaDiaz,
J. and Thompson, L.G. 1998: Experimental dendroclimatic
reconstruction of the Southern Oscillation. Bulletin of the American
Meteorological Society 79, 2137–52.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 47
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Stahle, D.W., Mushove, P.T., Cleaveland, M.K., Roig, F. and
Haynes, G.A. 1999: Management implications of annual rings in
Pterocarpus angolensis from Zimbabwe. Forest Ecology and
Management 124, 217–29.
Steffensen, J.P., Clausen, H.B. and Christensen, J.M. 1997: On the
spatial variability of impurity content and stable isotope composition
in recent Summit snow. In Wolff, E.W. and Bales, R.C., editors,
Chemical exchange between the atmosphere and polar snow. NATO
ASI Ser I 43, 607–15.
Steig, E.J., Mayewski, P.A., Dixon, D.A., Kaspari, S.D., Frey,
M.M., Schneider, D.P., Arcone, S.A., Hamilton, G.S., Spikes, V.B.,
Albert, M., Meese, D., Gow, A.J., Shuman, C.A., White, J.W.C.,
Sneed, S., Flaherty, J. and Wumkes, M. 2005: High-resolution ice
cores from US-ITASE (West Antarctica): development and validation
of chronologies and determination of precision and accuracy. Annals
of Glaciology 41, 77–84.
Stewart, K.A., Lamoureux, S.F. and Finney, B.P. 2008: Multiple
ecological and hydrological changes recorded in varved sediments
from Sanagak Lake, Nunavut, Canada. Journal of Paleolimnology 40,
217–33.
Tarand, A. and Nordli, Ø. 2001: The Tallinn temperature series
reconstructed back half a millennium by use of proxy data. Climatic
Change 48, 189–99.
Tarhule, A. and Hughes, M.K. 2002: Tree-ring research in semi-arid
West Africa: need and potential. Tree-Ring Research 58, 31–46.
Taylor, A.E., Wang, K., Smith, S.L., Burgess, M.M. and Judge,
A.S. 2006: Canadian arctic permafrost observatories: detecting contemporary
climate change through inversion of subsurface temperature
time-series. Journal of Geophysical Research 111, B02411,
doi:10.1029/2004JB003208.
Telford, R.J. and Birks, H.J.B. 2005: The secret assumption of transfer
functions: problems with spatial autocorrelation in evaluating
model performance. Quaternary Science Reviews 24, 2173–79,
doi:10.1016/j.quascirev.2005.05.001
Tett, S.F.B., Betts, R., Crowley, T.J., Gregory, J., Johns, T.C.,
Jones, A., Osborn, T.J., Ostrom, E., Roberts, D.L. and Woodage,
M.J. 2007: The impact of natural and anthropogenic forcings on climate
and hydrology since 1550. Climate Dynamics 28, 3–34.
Therrell, M.D., Stahle, D.W. and Acuna Soto, R. 2004: Aztec
drought and the curse of One Rabbit. Bulletin of the American
Meteorological Society 85, 1263–72.
Therrell, M.D., Stahle, D.W., Ries, L.P. and Shugart, H.H. 2006:
Tree-ring reconstructed rainfall variability in Zimbabwe. Climate
Dynamics 26, 677–85.
Thomas, R.H. and PARCA Investigators 2001: Program for Arctic
regional climate assessment (PARCA): goals, key findings, and future
directions. Journal of Geophysical Research 106, 33 691–705.
Thompson, L.G. and Davis, M.E. 2005: Stable isotopes through the
Holocene as recorded in low-latitude, high-altitude ice cores. In
Aggarwal, P.D., Gat, J.R. and Froehlich, K.F.O., editors, Isotopes in
the water cycle. Springer, 321–40.
Thompson, L.G., Peel, D.A., Mosley-Thompson, E., Mulvaney, R.,
Dai, J., Lin, P.N., Davis, M.E. and Raymond, C.F. 1994: Climate
since A.D. 1510 on Dyer Plateau, Antarctic Peninsula: evidence for
recent climate change. Annals of Glaciology 20, 420–26.
Thompson, L.G., Mosley-Thompson, E. and Henderson, K.A. 2000:
Ice core paleoclimate records in tropical South America since the Last
Glacial Maximum. Journal of Quaternary Science 15, 377–94.
Thompson, L.G., Mosley-Thompson, E., Davis, M.E., Lin, P.-N.,
Henderson, K. and Mashiotta, T.A. 2003: Tropical glacier and ice
core evidence of climate change on annual to millennial time scales.
Climate Change 59, 137–55.
Thompson, L.G., Mosley-Thompson, E., Brecher, H., Davis, M.E.,
Leon, B., Les, D., Mashiotta, T.A., Lin, P.-N. and Mountain, K.
2006: Evidence of abrupt tropical climate change: past and present.
Proceedings of the National Academy of Science 103, 10 536–43.
Treydte, K.S., Schleser, G.H., Helle, G., Frank, D.C., Winiger, M.,
Haug, G.H. and Esper, J. 2006: The twentieth century was the
wettest period in northern Pakistan over the past millennium. Nature
440, 1179–82.
Treydte, K., Frank, D., Esper, J., Andreu, L., Bednarz, Z.,
Berninger, F., Boettger, T., D’Alessandro, C.M., Etien, N., Filot,
M., Grabner, M., Guillemain, M.T., Gutierrez, E., Haupt, M.,
Helle, G., Hilasvuori, E., Jungner, H., Kalela-Brundin, M.,
Krapiec, M., Leuenberger, M., Loader, N.J., Masson-Delmotte,
V., Pazdur, A., Pawelczyk, S., Pierre, M., Planells, O., Pukiene, R.,
Reynolds-Henne, C.E., Rinne, K.T., Saracino, A., Saurer, M.,
Sonninen, E., Stievenard, M., Switzur, V.R., Szczepanek, M.,
Szychowska-Krapiec, E., Todaro, L., Waterhouse, J.S., Weigl, M.
and Schleser, G.H. 2007: Signal strength and climate calibration of a
European tree-ring isotope network. Geophysical Research Letters 34,
L24302, doi:10.1029/2007GL031106.
Tudhope, A.W., Chilcott, C.P., McCulloch, M.T., Cook, E.R.,
Chappell, J., Ellam, R.M., Lea, D.W., Lough, J.M. and Shimmield,
G.B. 2001: Variability in the El Niño-Southern Oscillation through a
glacial–interglacial cycle. Science 291, 1511–17.
UK Department of the Environment 1989: Global climate change.
Report by the Department of the Environment in association with the
Meteorological Office, 24 pp.
Urban, F.E., Cole, J.E. and Overpeck, J.T. 2000: Influence of mean
climate change on climate variability from a 155-year tropical Pacific
coral reef. Nature 407, 989–93.
Vaganov, E.A. 1996: Mechanisms and simulation of tree ring information
in conifer wood. Lesovedenie (Russian Journal of Forestry) 1,
3–15 (in Russian).
Vaganov, E.A., Hughes, M.K., Kirdyanov, A.V., Schweingruber,
F.H. and Silkin, P.P. 1999: Influence of snowfall and melt timing on
tree growth in subarctic Eurasia. Nature 400, 149–51.
van der Schrier, G. and Barkmeijer, J. 2005: Bjerknes’ hypothesis
on the coldness during AD 1790–1820 revisited. Climate Dynamics
24, 355–71.
van Engelen, A.F.V., Buisman J. and IJnsen, F. 2001: A millennium
of weather, winds and water in the low countries. In Jones, P.D.,
Ogilvie, A.E.J., Davies, T.D. and Briffa, K.R., editors, History and
climate: memories of the future? Kluwer Academic/Plenum
Publishers, 101–24.
van Ommen, T.D. and Morgan, V.I. 1997: Calibrating the ice core
paleothermometer using seasonality. Journal of Geophysical
Research 102, 9351–57.
Vetter, R.E. and Wimmer, R. 1999: Remarks on the current situation
of tree-ring research in the tropics. In Wimmer, R. and Vetter, R.E.,
editors, Tree-ring analysis: biological, methodological, and environmental
aspects. CABI Publishing, 131–37.
Vinther, B.M., Johnsen, S.J., Andersen, K.K., Clausen, H.B. and
Hansen, A.W. 2003: NAO signal recorded in the stable isotopes of
Greenland ice cores. Geophysical Research Letters 30, 1387,
doi:10.1029/2002GL016193.
Vinther, B.M., Clausen, H.B., Johnsen, S.J., Rasmussen, S.O.,
Steffensen, J.P., Andersen, K.K., Buchardt, S.L., Dahl-Jensen, D.,
Seierstad, I.K., Siggaard-Andersen, M-L., Svensson, A.M., Olsen,
J. and Heinemeier, J. 2006a: A synchronized dating of three
Greenland ice cores throughout the Holocene. Journal of Geophysical
Research 111, D13102, doi:10.1029/2005JD006921.
Vinther, B.M., Andersen, K.K., Jones, P.D., Briffa, K.R. and
Cappelen, J. 2006b: Extending Greenland temperature records into
the late eighteenth century. Journal of Geophysical Research 111,
D11105, doi:10.1029/2005JD006810.
Vinther, B.M., Jones, P.D., Briffa, K.R., Clausen, H.B., Andersen,
K.K., Dahl-Jensen, D. and Johnsen, S.J. 2008: Climatic signals in
multiple highly resolved stable isotope records from Greenland.
Quaternary Science Reviews in press.
von Storch, H., Zorita, E., Jones, J.M., Dimitriev, Y., GonzalezRouco,
F. and Tett, S.F.B. 2004: Reconstructing past climate from
noisy data. Science 306, 679–82.
von Storch, H., Zorita, E., Jones, J.M., Gonzalez-Rouco, F. and
Tett, S.F.B. 2006: Response to comment on ‘Reconstructing past climate
from noisy data’. Science 312, 529c.
von Storch, H., Zorita, E. and González-Rouco, F. 2008:
Assessement of three temperature reconstruction methods in the virtual
reality of a climate simulation. International Journal of Earth Science
(Geologische Rundschau) DOI:10.1007/s00531–008–0349–5.
Vuille, M., Bradley, R.S., Healy, R., Werner, M., Hardy, D.R.,
Thompson, L.G. and Keimig, F. 2003: Modeling d18
O in precipitation
over the tropical Americas: 2. Simulation of the stable isotope
signal in Andean ice cores. Journal of Geophysical Research 108,
4175, doi:10.1029/2001JD002039.
48 The Holocene 19,1 (2009)
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from
Wahl, E.R. and Ammann, C.M. 2007: Robustness of the Mann,
Bradley, Hughes reconstruction of surface temperatures: examination
of criticisms based on the nature and processing of proxy climate evidence.
Climatic Change 85, 33–69.
Wahl, E.R., Ritson, D.M. and Ammann, C.M. 2006: Comment on
‘Reconstructing past climate from noisy data’. Science 312, 529b.
Wang, S.-W. and Zhao, Z.-C. 1981: Droughts and floods in China
1470–1979. In Wigley, T.M.L., Ingram, M.J. and Farmer, G., editors,
Climate and history. Cambridge University Press, 271–88.
Wang, S.-W., Gong, D. and Zhu, J. 2001: Twentieth-century climatic
warming in China in the context of the Holocene. The Holocene
11, 313–21.
Waple, A., Mann, M.E. and Bradley, R.S. 2002: Long-term patterns
of solar irradiance forcing in model experiments and proxy-based surface
temperature reconstructions. Climate Dynamics 18, 563–78.
Watanabe, T., Gagan, M.K., Correge, T., Scott-Gagan, H.,
Cowley, J. and Hantoro, W.S. 2003: Oxygen isotope systematics in
Diploastrea heliopora: new coral archive of tropical paleoclimate.
Geochimica Cosmochimica Acta 67, 1349–58.
Wegman, E.J., Scott, D.W. and Said, Y.H. 2006: Report of the US
House Committees on Energy and Commerce and the sub-committee
on Oversight and Investigations on ‘The “Hockey Stick” global climate
reconstructions’. Available at http://energycommerce.house.gov/
reparchives/108/home/07142006_Wegman_Report.pdf
Werner, M. and Heimann, M. 2002: Modeling interannual variability
of water isotopes in Greenland and Antarctica. Journal of
Geophysical Research 107, 4001, doi:1029/2001JD900253.
Werner, M., Mikolajewicz, U., Heimann, M. and Hoffmann, G.
2000: Borehole versus isotope temperatures on Greenland: seasonality
does matter. Geophysical Research Letters 27, 723–26.
Wheeler, D. 2005: An examination of the accuracy and consistency
of ships’ logbook weather observations and records. Climatic Change
73, 97–116.
White, J.W.C., Barlow, L.K., Fisher, D., Grootes, P., Jouzel, J.,
Johnsen, S.J., Stuiver, M. and Clausen, H. 1997: The climate signal
in the stable isotopes of snow from Summit, Greenland: results of
comparisons with modern climate observations. Journal of
Geophysical Research 102, 26 425–39.
Wigley, T.M.L., Briffa, K.R. and Jones, P.D. 1984: On the average
value of correlated time series with applications in dendroclimatology
and hydrometeorology. Journal of Climate & Applied Meteorology
23, 201–13.
Wigley, T.M.L., Farmer, G. and Ogilvie, A.E.J. 1986: Climatic
reconstruction using historical sources. In Ghazi, A. and Fantecchi,
R., editors, Current issues in climatic research. Proceedings of the EC
Climatology Programme Symposium, Sophia Antipolis, France, 2–5
October 1984. Commission of the European Communities, 97–100.
Williams, L.D. and Wigley, T.M.L. 1983: A comparison of evidence
for Late Holocene summer temperature variations in the Northern
Hemisphere. Quaternary Research 20, 286–307.
Wilmking, M., Juday, G.P., Barber, V.A. and Zald, H.S.J. 2004:
Recent climate warming forces contrasting growth responses of white
spruce at treeline in Alaska through temperature thresholds. Global
Change Biology 10, 1724–36.
Wilson, R., Tudhope, A., Brohan, P., Briffa, K.R., Osborn, T.J.
and Tett, S.F.B. 2006: 250-years of reconstructed and modeled tropical
temperatures. Journal of Geophysical Research 111, C10007,
doi:10.1029/2005JC003188.
Wilson, R., D’Arrigo, R., Buckley, B., Büntgen, U., Esper, J.,
Frank, D., Luckman, B., Payette, S., Vose, R. and Youngblut, D.
2007: A matter of divergence: tracking recent warming at hemispheric
scales using tree ring data. Journal of Geophysical ResearchAtmospheres
112, D17103, doi:10.1029/2006JD008318.
Wolodarsky-Franke, A. 2002: Reconstruction of the paleo-environments
in Volcán Apagado from Fitzroya cupressoides tree-rings.
M.Sc. Thesis Ecology, Universidad Austral de Chile.
Worbes, M. 1995: How to measure growth dynamics in tropical trees:
a review. IAWA Journal 16, 227–351.
Worbes, M. and Junk, W. 1989: Dating tropical trees by means of
14C from bomb tests. Ecology 70, 503–507.
Wright, S.J. 2006: Response to Lewis et al.: The uncertain response
of tropical forests to global change. Trends in Ecology and Evolution
21, 174–75.
Xoplaki, E., Luterbacher, J., Paeth, H., Dietrich, D., Steiner, N.,
Grosjean, M. and Wanner, H. 2005: European spring and autumn
temperature variability and change of extremes over the last half
millennium. Geophysical Research Letters 32, L15713 doi:10.1029/
2005GL023424.
Yang, B., Braeuning, A., Johnson, K.R. and Yafeng, S. 2002:
General characteristics of temperature variation in China during the
last two millennia. Geophysical Research Letters 29, 1234,
doi:10.1029/ 2001GL014485.
Yao, T., Thompson, L.G., Mosley-Thompson, E., Zhihong, Y.,
Xingping, Z. and Lin, P-N. 1996: Climatological significance of d18
O
in north Tibetan ice cores. Journal of Geophysical Research 101,
29 531–37.
Zaiki, M., Können, G.P., Tsukahara, T., Jones, P.D., Mikami, T.
and Matsumoto, K. 2006: Recovery of 19th century Tokyo/Osaka
meteorological data in Japan. International Journal of Climatology
26, 399–423.
Zhang, J. and Crowley, T.J. 1989: Historical climate records in China
and reconstruction of past climates. Journal of Climate 2, 833–49.
Zorita, E. and González-Rouco, F. 2002: Are temperature-sensitive
proxies adequate for North Atlantic Oscillation reconstructions?
Geophysical Research Letters 29, 1703, doi:10.1029/2002
GL015404.
Zorita, E., González-Rouco, F. and Legutke, S. 2003: Testing the
Mann et al. (1998) approach to paleoclimate reconstructions in the
context of a 1000-yr control simulation with the ECHO-G coupled
climate model. Journal of Climate 16, 1378–90.
P.D. Jones et al.: High-resolution palaeoclimatology of the last millennium 49
at GEOBOTANISCHES INSTITUT on January 12, 2009http://hol.sagepub.comDownloaded from