season). Moreover, many proxies are not sensitive to a single
atmospheric variable but to the combination of effects related
to temperature and hydrology. The combined use of multiple
markers within a single medium or multiple archives from
the same site allows these effects to be separated out. Finally,
comparison between proxies, paleoclimate reconstructions,
modeling of climate and proxies all improve our understanding of how climate dynamics and proxies operate.
The dynamics of the atmosphere can also be estimated
from the continental paleoclimate records. During the
instrumental period, it is possible to determine how certain
modes of atmospheric circulation such as the El Niño-Southern Oscillation (ENSO), the North Atlantic Oscillation
(NAO), the Southern Annular Mode (SAM) or the
Pacific-North American oscillation (PNA) modulate the
spatial response of proxies, just as they affect the spatial
distribution of temperature and rainfall. Provided that the
spatial distribution of proxy records is sufficient in key areas,
and depending on the stability of these tele-connections
through time, this fingerprint can then be used to estimate
past inter-annual variations in pressure indices characteristic
of patterns of atmospheric circulation in recent centuries
(PNA, NAO, ENSO, SAM).
Over large time scales, loess deposits and dunes reflect
the prevailing wind direction. Similarly, concentrations of
marine and continental aerosols, the size distribution of
continental dust particles in polar ice, reflect changes in the
aridity of the regions of origin as well as changes in the
efficiency of transport of aerosols in the atmosphere. However, quantitative estimates of the intensity of surface winds
remain a challenge. Past changes in other atmospheric
parameters such as cloudiness are difficult to determine from
proxy records in natural archives.
Continental paleoclimatology can also help to characterize the frequency and intensity of ‘extreme’ events. The
intensity and amplitude of past droughts have been estimated
using dendrochronological databases in North America and
Europe. Sedimentary and geochemical markers from lake
sediments are used to determine the intensity and occurrence
of flooding by the great rivers. High-resolution analysis of
lagoon sediments and the isotopic composition of tree rings
or speleothems are currently being used in an attempt to
characterize past variations in the activity (trajectories,
intensity, frequency) of tropical cyclones and extratropical
storms.
We have briefly presented the atmospheric variables that
can be estimated from continental paleoclimate records. In
the following chapters, we will explain the reconstruction
methods used; the assumptions upon which they are based;
their limitations and uncertainties; and finally, we will present several specific examples reflecting the diversity of
continental archives (lakes, vegetation, ice) whose dating
techniques were described in Part II. We will present some
archives and some proxies particularly well suited to each of
the interfaces under consideration. We have considered high
latitudes and low latitudes separately. Several archives (lacustrine cores, speleothems) and proxies (pollen, diatoms)
are present in both temperate and tropical regions, but their
interpretation is specific to the particular geographical area.
In the high latitudes of both hemispheres, polar ice is of
paramount importance in paleoclimatology because it
records both climate forcings and some local and global
climatic variations. Sedimentary archives (from lakes and
bogs) with their pollen records have long been studied in
paleoclimatology and cover almost all of the continents.
They were initially understood to mostly reflect a local
signal. However, the comparison with ice and marine
records revealed the broad geographical spread of many
events known to palynologists, such as the Younger Dryas
cold episode which lasted for close to a millennium and was
felt throughout the Atlantic area of the northern hemisphere.
The loess covers a significant area of the continents; their
sequences are excellent indicators of atmospheric circulation. On a shorter timescale, some archives provide information with annual or near annual resolution. Among these,
tree rings provide a wealth of information on temperate
regions. They are supplemented by multi-centennial archives
from historical written documents, such as wine harvesting
dates, in Europe, or cherry blossom dates, in Japan.
At low latitudes, sedimentological tracers collected in
tropical lakes, complemented by biological proxies, such as
the abundance of diatoms or pollen, contribute to a better
understanding of the functioning of the major inter-tropical
climate systems and give an insight into sometimes discontinuous records. Diatoms are good indicators of the
characteristics of lake water and, complemented by adequate
hydrological modeling, they allow an assessment of water
resources in watersheds. At high altitudes, tropical glaciers
are very sensitive to climate variations in the long and
medium term.
Cave records, for instance from speleothems, offer
well-dated, albeit discontinuous, records particularly sensitive to changes in the hydroclimate and vegetation cover
above the caves, sometimes with very high temporal
resolution.
Interpretation of Records, Limitations
and Uncertainties
Paleoclimatology draws its information mainly from two
types of approaches with their advantages and limitations.
The first approach is the most basic and consists of using
simple equations to interpret a climate signal from a univariate series. This is the preferred approach for geochemistry which often uses scaling to transform an isotopic signal
138
V. Masson-Delmotte and J. Guiot
atmospheric variable but to the combination of effects related
to temperature and hydrology. The combined use of multiple
markers within a single medium or multiple archives from
the same site allows these effects to be separated out. Finally,
comparison between proxies, paleoclimate reconstructions,
modeling of climate and proxies all improve our understanding of how climate dynamics and proxies operate.
The dynamics of the atmosphere can also be estimated
from the continental paleoclimate records. During the
instrumental period, it is possible to determine how certain
modes of atmospheric circulation such as the El Niño-Southern Oscillation (ENSO), the North Atlantic Oscillation
(NAO), the Southern Annular Mode (SAM) or the
Pacific-North American oscillation (PNA) modulate the
spatial response of proxies, just as they affect the spatial
distribution of temperature and rainfall. Provided that the
spatial distribution of proxy records is sufficient in key areas,
and depending on the stability of these tele-connections
through time, this fingerprint can then be used to estimate
past inter-annual variations in pressure indices characteristic
of patterns of atmospheric circulation in recent centuries
(PNA, NAO, ENSO, SAM).
Over large time scales, loess deposits and dunes reflect
the prevailing wind direction. Similarly, concentrations of
marine and continental aerosols, the size distribution of
continental dust particles in polar ice, reflect changes in the
aridity of the regions of origin as well as changes in the
efficiency of transport of aerosols in the atmosphere. However, quantitative estimates of the intensity of surface winds
remain a challenge. Past changes in other atmospheric
parameters such as cloudiness are difficult to determine from
proxy records in natural archives.
Continental paleoclimatology can also help to characterize the frequency and intensity of ‘extreme’ events. The
intensity and amplitude of past droughts have been estimated
using dendrochronological databases in North America and
Europe. Sedimentary and geochemical markers from lake
sediments are used to determine the intensity and occurrence
of flooding by the great rivers. High-resolution analysis of
lagoon sediments and the isotopic composition of tree rings
or speleothems are currently being used in an attempt to
characterize past variations in the activity (trajectories,
intensity, frequency) of tropical cyclones and extratropical
storms.
We have briefly presented the atmospheric variables that
can be estimated from continental paleoclimate records. In
the following chapters, we will explain the reconstruction
methods used; the assumptions upon which they are based;
their limitations and uncertainties; and finally, we will present several specific examples reflecting the diversity of
continental archives (lakes, vegetation, ice) whose dating
techniques were described in Part II. We will present some
archives and some proxies particularly well suited to each of
the interfaces under consideration. We have considered high
latitudes and low latitudes separately. Several archives (lacustrine cores, speleothems) and proxies (pollen, diatoms)
are present in both temperate and tropical regions, but their
interpretation is specific to the particular geographical area.
In the high latitudes of both hemispheres, polar ice is of
paramount importance in paleoclimatology because it
records both climate forcings and some local and global
climatic variations. Sedimentary archives (from lakes and
bogs) with their pollen records have long been studied in
paleoclimatology and cover almost all of the continents.
They were initially understood to mostly reflect a local
signal. However, the comparison with ice and marine
records revealed the broad geographical spread of many
events known to palynologists, such as the Younger Dryas
cold episode which lasted for close to a millennium and was
felt throughout the Atlantic area of the northern hemisphere.
The loess covers a significant area of the continents; their
sequences are excellent indicators of atmospheric circulation. On a shorter timescale, some archives provide information with annual or near annual resolution. Among these,
tree rings provide a wealth of information on temperate
regions. They are supplemented by multi-centennial archives
from historical written documents, such as wine harvesting
dates, in Europe, or cherry blossom dates, in Japan.
At low latitudes, sedimentological tracers collected in
tropical lakes, complemented by biological proxies, such as
the abundance of diatoms or pollen, contribute to a better
understanding of the functioning of the major inter-tropical
climate systems and give an insight into sometimes discontinuous records. Diatoms are good indicators of the
characteristics of lake water and, complemented by adequate
hydrological modeling, they allow an assessment of water
resources in watersheds. At high altitudes, tropical glaciers
are very sensitive to climate variations in the long and
medium term.
Cave records, for instance from speleothems, offer
well-dated, albeit discontinuous, records particularly sensitive to changes in the hydroclimate and vegetation cover
above the caves, sometimes with very high temporal
resolution.
Interpretation of Records, Limitations
and Uncertainties
Paleoclimatology draws its information mainly from two
types of approaches with their advantages and limitations.
The first approach is the most basic and consists of using
simple equations to interpret a climate signal from a univariate series. This is the preferred approach for geochemistry which often uses scaling to transform an isotopic signal
138
V. Masson-Delmotte and J. Guiot