20
Air-Ice Interface: Tropical Glaciers
Françoise Vimeux
Just as the polar caps are excellent archives of information
on past climates, so too are the high-altitude tropical glaciers. Indeed, the prevailing temperature and humidity conditions there usually ensure very good preservation of
chemical and isotopic tracers. These archives have therefore
been used for the past 25 years to study the variability of the
tropical climate over past centuries and millennia. They are
mostly found in the Andes in South America between 0° and
20° S (Vimeux 2009; Vimeux et al. 2009), although drilling
has been conducted on Kilimanjaro (Thompson et al. 2002)
and the southern Himalayas (Thompson et al. 2000, 2006).
The focus of this chapter is on climate information taken
from the Andean glaciers.
The rapid dynamic of these glaciers, the high level of snow
accumulation per year (0.5–1 m) and reduced ice thickness
(100–150 m) mean that it is not possible to access climate
periods as old as those possible in polar cores. The oldest cores
date back to the last glacial maximum, about 20 000 years ago,
with the Sajama core (Bolivia, 6542 m, Western Cordillera,
18° 06′ S, 68° 53′ W) dating back to * 25,000 years
(Thompson et al. 1998). On the other hand, the tropical ice
allows our climate to be studied with a very good temporal
resolution, at the seasonal level for recent centuries.
Paleoclimate Markers
Several types of quantifiable climate variables can be
extracted from tropical glaciers: the net accumulation, the
temperature in the borehole and regional precipitation. We
propose to review these and to present the main results in
terms of climate variability. Although complementary, we
will not discuss here the qualitative results from the chemical
analysis of ice, which essentially informs us about changes
in the environmental, atmospheric transport processes and
air pollution.
Over recent centuries, the seasonal cycles of the chemical
elements and stable isotopes in precipitation allow us to date
the annual layers with a relatively good degree of accuracy
(±5 to 10 years around 1900). It is thus possible to calculate
the annual net accumulation. To correct for the effects of
snow compaction in the depths, an ice flow model is applied,
or alternatively, the thinning of the layers is corrected by
directly observing the relationship between the annual
thickness and the depth, along the core. The latter method
cannot however be used to discuss climate trends which are,
in principle, corrected using the same method. The question
that then arises is: what does the net accumulation represent
given that it is actually the combination of the total accumulation (controlled by the precipitation and wind) minus
removal (sublimation, erosion by wind), and that the latter
processes can have different seasonalities? If one considers
sites where accumulation is low (0.31 m of water/year) with
a high level of sublimation throughout a long dry season,
such as at Cerro Tapado (Chile, arid diagonal, 5550 m, 30°
08′ S, 69° 55′ W), it is difficult to use this parameter as a
marker of the amount of deposited precipitation. This is less
the case for sites such as the one at Illimani (Bolivia, eastern
cordillera, 6350 m, 16° 37′ S, 67° 46′ W) where the rainfall
season is longer, the annual snow accumulation heavier
(0.58 m of water/year) and the sublimation is concentrated
over a shorter time of the year which is different to the
accumulation season (Ginot et al. 2006). Only the end of the
rainy season which represents about 10% of the annual
accumulation may be truncated in the records. In that case,
we can we assume that large variations in the net accumulation are representative of the amount of precipitation. The
accumulation estimates made on the Quelccaya core (Peru,
5670 m, 13° 56′ S, 70° 50′ W, the only core with dating by
F. Vimeux (&)
Laboratoire HydroSciences Montpellier (HSM), Institut de
Recherche pour le Développement (IRD),
Montpellier, 34095, France
e-mail: Francoise.Vimeux@lsce.ipsl.fr
Laboratoire des Sciences du Climat et de l’Environnement (LSCE),
Institut Pierre Simon Laplace (IPSL),
91191 Gif-sur-Yvette, France
© Springer Nature Switzerland AG 2021
G. Ramstein et al. (eds.), Paleoclimatology, Frontiers in Earth Sciences,
https://doi.org/10.1007/978-3-030-24982-3_20
219
Air-Ice Interface: Tropical Glaciers
Françoise Vimeux
Just as the polar caps are excellent archives of information
on past climates, so too are the high-altitude tropical glaciers. Indeed, the prevailing temperature and humidity conditions there usually ensure very good preservation of
chemical and isotopic tracers. These archives have therefore
been used for the past 25 years to study the variability of the
tropical climate over past centuries and millennia. They are
mostly found in the Andes in South America between 0° and
20° S (Vimeux 2009; Vimeux et al. 2009), although drilling
has been conducted on Kilimanjaro (Thompson et al. 2002)
and the southern Himalayas (Thompson et al. 2000, 2006).
The focus of this chapter is on climate information taken
from the Andean glaciers.
The rapid dynamic of these glaciers, the high level of snow
accumulation per year (0.5–1 m) and reduced ice thickness
(100–150 m) mean that it is not possible to access climate
periods as old as those possible in polar cores. The oldest cores
date back to the last glacial maximum, about 20 000 years ago,
with the Sajama core (Bolivia, 6542 m, Western Cordillera,
18° 06′ S, 68° 53′ W) dating back to * 25,000 years
(Thompson et al. 1998). On the other hand, the tropical ice
allows our climate to be studied with a very good temporal
resolution, at the seasonal level for recent centuries.
Paleoclimate Markers
Several types of quantifiable climate variables can be
extracted from tropical glaciers: the net accumulation, the
temperature in the borehole and regional precipitation. We
propose to review these and to present the main results in
terms of climate variability. Although complementary, we
will not discuss here the qualitative results from the chemical
analysis of ice, which essentially informs us about changes
in the environmental, atmospheric transport processes and
air pollution.
Over recent centuries, the seasonal cycles of the chemical
elements and stable isotopes in precipitation allow us to date
the annual layers with a relatively good degree of accuracy
(±5 to 10 years around 1900). It is thus possible to calculate
the annual net accumulation. To correct for the effects of
snow compaction in the depths, an ice flow model is applied,
or alternatively, the thinning of the layers is corrected by
directly observing the relationship between the annual
thickness and the depth, along the core. The latter method
cannot however be used to discuss climate trends which are,
in principle, corrected using the same method. The question
that then arises is: what does the net accumulation represent
given that it is actually the combination of the total accumulation (controlled by the precipitation and wind) minus
removal (sublimation, erosion by wind), and that the latter
processes can have different seasonalities? If one considers
sites where accumulation is low (0.31 m of water/year) with
a high level of sublimation throughout a long dry season,
such as at Cerro Tapado (Chile, arid diagonal, 5550 m, 30°
08′ S, 69° 55′ W), it is difficult to use this parameter as a
marker of the amount of deposited precipitation. This is less
the case for sites such as the one at Illimani (Bolivia, eastern
cordillera, 6350 m, 16° 37′ S, 67° 46′ W) where the rainfall
season is longer, the annual snow accumulation heavier
(0.58 m of water/year) and the sublimation is concentrated
over a shorter time of the year which is different to the
accumulation season (Ginot et al. 2006). Only the end of the
rainy season which represents about 10% of the annual
accumulation may be truncated in the records. In that case,
we can we assume that large variations in the net accumulation are representative of the amount of precipitation. The
accumulation estimates made on the Quelccaya core (Peru,
5670 m, 13° 56′ S, 70° 50′ W, the only core with dating by
F. Vimeux (&)
Laboratoire HydroSciences Montpellier (HSM), Institut de
Recherche pour le Développement (IRD),
Montpellier, 34095, France
e-mail: Francoise.Vimeux@lsce.ipsl.fr
Laboratoire des Sciences du Climat et de l’Environnement (LSCE),
Institut Pierre Simon Laplace (IPSL),
91191 Gif-sur-Yvette, France
© Springer Nature Switzerland AG 2021
G. Ramstein et al. (eds.), Paleoclimatology, Frontiers in Earth Sciences,
https://doi.org/10.1007/978-3-030-24982-3_20
219