climate cycles can be seen remarkably well: the d
18 O c is
systematically depleted by *4‰ during periods of strong
summer monsoon, mainly due to the effect of mass (the d
18
O
of the rain is inversely proportional to the volume of rainfall)
but also due to changes in the source (Wang et al. 2008).
Thanks to the many U-Th datings carried out on several
stalagmites in caves in Sanbao, Hulu and Dongge (25° N to
32° N, China), it has been shown that the d
18
O c of the calcite
had a periodicity of 23 ka and was directly correlated with
changes in insolation at 65° N, demonstrating that variations
in the monsoons were caused by orbital changes (Fig. 14.2)
(Wang et al. 2008). Superimposed on these large climate
variations, millennial climate events were also recorded and
linked to Dansgaard-Oeschger events (D/O) detected in the
Greenland ice, which shows the strong connection between
the climate systems of South Asia and the North Atlantic.
This interconnection between the monsoon regime and the
North Atlantic region was confirmed by one of the longest
and most precise paleoclimate recordings produced by the
study of Chinese speleothems (Cheng et al. 2016). The
variations in monsoon intensity were reconstructed for the
past 640,000 years with an unmatched precision. The multiplicity of samples, precise U-Th dating and the very high
resolution of isotopic analyses suggest new theories on the
causes of the major glacial-interglacial climate cycles and
millennial variability. One of these theories posits that the
period between terminations (glacial-interglacial transitions)
is equal to a multiple of the length of the precession cycles;
moreover, a very strong teleconnection between the
dynamics of the ice caps of the northern hemisphere, the
circulation of the Atlantic Ocean and the Asian monsoon is
implied. Similarly, the variation in monsoon intensity in
South America was recorded in stalagmites from southern
Brazil (24° S to 27° S), at the orbital and the millennial scale
(Cruz et al. 2005). In this case, the latitudinal changes in the
position of the ITCZ (intertropical convergence zone) seem
to be the root cause of these variations.
At higher latitudes, in Europe for example, the d
18 O c of
calcite has a less pronounced response to changes in climate,
probably because of the opposite effects controlling the
d
18 O c of calcite (e.g. external temperature impacts on the
d
18 O of the rain while cave temperature impacts on d
18 O c of
calcite). However, the study of several European samples
has revealed a certain logic within the paleoclimate chronicles of varying appearance: there is a variable gradient in the
d
18 O c speleothems during the Holocene along a west-east
transect (McDermott et al. 2011). This is related to differences in warming during the Holocene depending on longitude and also on different atmospheric circulations. This
logic shows that even if speleothems do not precipitate at
isotopic equilibrium, they contain valuable information on
atmospheric paleocirculation. As a result of the influence of
these multiple climatic, hydrological and kinetic factors, the
evolution of the d
18 O c of the speleothem calcite is not uniform from one region to another: in general the d
18
O
decreases when the climate becomes warm and humid (in
temperate regions for example), while this trend may be
reversed (d
18 O increases as the temperature increases) at
certain altitudes as has been observed in the Alps (Boch et al.
2011; Moseley et al. 2014) and in eastern Europe on the
coast of the Black Sea (Fleitmann et al. 2009).
The d
13 C c of calcite may also react to climate changes,
and sometimes in a more obvious way than d
18 O c . The
carbon atoms of the calcite molecule, CaCO 3 , which makes
up the speleothems come from two main sources: CO 2 from
the soil and CaCO 3 from the surrounding rock. Soil CO 2 ,
produced by plant roots and microbial activity, has a d
13 C c
close to −24‰ (for C3 type plants, most frequently found in
temperate areas), CaCO 3 from marine limestone has a d
13 C c
of between *−2 and +2‰. It has been shown that the main
source of carbon in speleothems is the CO 2 from the soil
which can represent up to 90% of C contained in the CaCO 3
of speleothems (Genty et al. 1998). In several sites in the
South of France, the carbon from the dissolution of limestone (also called dead carbon because it contains no
14 C)
represents, in this particular case, only 15 ± 5% of the C in
the speleothems. Consequently, any change in the vegetation
above a cave brought about by climate change, such as
proportion of type C3 to C4 or vegetation density, will have
an impact on the d
13 C of the CO 2 in the soil and thus on the
d
13 C c of the speleothems. In summary:
d
13 C c = f (type of vegetation, density of vegetation,
hydrology, isotopic equilibrium).
The d
13
C c in the stalagmites in the Villars cave (South-West
France) shows abrupt changes from −2‰ to −5‰ over the last
80 ka. These have been linked to Dansgaard-Oeschger events
recorded in Greenland ice cores and to temperature reconstructions using analyzes of pollen from lakes and marine cores
(Genty et al. 2003).
The value of this comparison is, firstly, to test the
chronology from other archives against the absolute
chronology provided by speleothems, and if necessary, to
adjust it. Refining the age of an abrupt climatic transition,
such as the one which occurred at the beginning of D/O 12
(Fig. 14.3), is important in order to find out its cause by
comparing a sequence of climate events with the external
forcings and with archives from other latitudes. Comparison
with pollen reconstructions shows the close link between
changes in vegetation brought on by variations in temperature and humidity, and the d
13
C c recorded in speleothems.
Other examples show that the d
13
C c recorded the last
deglaciation, in New Zealand as well as in Europe (Genty
et al. 2006; Moreno et al. 2010). Thus, a stalagmite from the
Chauvet cave (Ardèche, France) records the climate events
(e.g. Bølling-Allerød, Intra Allerød Cold Period,
Younger-Dryas) that punctuated the last deglaciation with a
172
D. Genty and A. Moreno
18 O c is
systematically depleted by *4‰ during periods of strong
summer monsoon, mainly due to the effect of mass (the d
18
O
of the rain is inversely proportional to the volume of rainfall)
but also due to changes in the source (Wang et al. 2008).
Thanks to the many U-Th datings carried out on several
stalagmites in caves in Sanbao, Hulu and Dongge (25° N to
32° N, China), it has been shown that the d
18
O c of the calcite
had a periodicity of 23 ka and was directly correlated with
changes in insolation at 65° N, demonstrating that variations
in the monsoons were caused by orbital changes (Fig. 14.2)
(Wang et al. 2008). Superimposed on these large climate
variations, millennial climate events were also recorded and
linked to Dansgaard-Oeschger events (D/O) detected in the
Greenland ice, which shows the strong connection between
the climate systems of South Asia and the North Atlantic.
This interconnection between the monsoon regime and the
North Atlantic region was confirmed by one of the longest
and most precise paleoclimate recordings produced by the
study of Chinese speleothems (Cheng et al. 2016). The
variations in monsoon intensity were reconstructed for the
past 640,000 years with an unmatched precision. The multiplicity of samples, precise U-Th dating and the very high
resolution of isotopic analyses suggest new theories on the
causes of the major glacial-interglacial climate cycles and
millennial variability. One of these theories posits that the
period between terminations (glacial-interglacial transitions)
is equal to a multiple of the length of the precession cycles;
moreover, a very strong teleconnection between the
dynamics of the ice caps of the northern hemisphere, the
circulation of the Atlantic Ocean and the Asian monsoon is
implied. Similarly, the variation in monsoon intensity in
South America was recorded in stalagmites from southern
Brazil (24° S to 27° S), at the orbital and the millennial scale
(Cruz et al. 2005). In this case, the latitudinal changes in the
position of the ITCZ (intertropical convergence zone) seem
to be the root cause of these variations.
At higher latitudes, in Europe for example, the d
18 O c of
calcite has a less pronounced response to changes in climate,
probably because of the opposite effects controlling the
d
18 O c of calcite (e.g. external temperature impacts on the
d
18 O of the rain while cave temperature impacts on d
18 O c of
calcite). However, the study of several European samples
has revealed a certain logic within the paleoclimate chronicles of varying appearance: there is a variable gradient in the
d
18 O c speleothems during the Holocene along a west-east
transect (McDermott et al. 2011). This is related to differences in warming during the Holocene depending on longitude and also on different atmospheric circulations. This
logic shows that even if speleothems do not precipitate at
isotopic equilibrium, they contain valuable information on
atmospheric paleocirculation. As a result of the influence of
these multiple climatic, hydrological and kinetic factors, the
evolution of the d
18 O c of the speleothem calcite is not uniform from one region to another: in general the d
18
O
decreases when the climate becomes warm and humid (in
temperate regions for example), while this trend may be
reversed (d
18 O increases as the temperature increases) at
certain altitudes as has been observed in the Alps (Boch et al.
2011; Moseley et al. 2014) and in eastern Europe on the
coast of the Black Sea (Fleitmann et al. 2009).
The d
13 C c of calcite may also react to climate changes,
and sometimes in a more obvious way than d
18 O c . The
carbon atoms of the calcite molecule, CaCO 3 , which makes
up the speleothems come from two main sources: CO 2 from
the soil and CaCO 3 from the surrounding rock. Soil CO 2 ,
produced by plant roots and microbial activity, has a d
13 C c
close to −24‰ (for C3 type plants, most frequently found in
temperate areas), CaCO 3 from marine limestone has a d
13 C c
of between *−2 and +2‰. It has been shown that the main
source of carbon in speleothems is the CO 2 from the soil
which can represent up to 90% of C contained in the CaCO 3
of speleothems (Genty et al. 1998). In several sites in the
South of France, the carbon from the dissolution of limestone (also called dead carbon because it contains no
14 C)
represents, in this particular case, only 15 ± 5% of the C in
the speleothems. Consequently, any change in the vegetation
above a cave brought about by climate change, such as
proportion of type C3 to C4 or vegetation density, will have
an impact on the d
13 C of the CO 2 in the soil and thus on the
d
13 C c of the speleothems. In summary:
d
13 C c = f (type of vegetation, density of vegetation,
hydrology, isotopic equilibrium).
The d
13
C c in the stalagmites in the Villars cave (South-West
France) shows abrupt changes from −2‰ to −5‰ over the last
80 ka. These have been linked to Dansgaard-Oeschger events
recorded in Greenland ice cores and to temperature reconstructions using analyzes of pollen from lakes and marine cores
(Genty et al. 2003).
The value of this comparison is, firstly, to test the
chronology from other archives against the absolute
chronology provided by speleothems, and if necessary, to
adjust it. Refining the age of an abrupt climatic transition,
such as the one which occurred at the beginning of D/O 12
(Fig. 14.3), is important in order to find out its cause by
comparing a sequence of climate events with the external
forcings and with archives from other latitudes. Comparison
with pollen reconstructions shows the close link between
changes in vegetation brought on by variations in temperature and humidity, and the d
13
C c recorded in speleothems.
Other examples show that the d
13
C c recorded the last
deglaciation, in New Zealand as well as in Europe (Genty
et al. 2006; Moreno et al. 2010). Thus, a stalagmite from the
Chauvet cave (Ardèche, France) records the climate events
(e.g. Bølling-Allerød, Intra Allerød Cold Period,
Younger-Dryas) that punctuated the last deglaciation with a
172
D. Genty and A. Moreno