Gisp2 Ice Core Temperature Data

Figure 15.1. The environmental record during the Bolling-AIlerod/Younger Dryas period, as revealed by the GISP 2 ice core data (adapted from Lorius and Oeschger, 1994).

Figure 15.1. The environmental record during the Bolling-AIlerod/Younger Dryas period, as revealed by the GISP 2 ice core data (adapted from Lorius and Oeschger, 1994).

Gisp Ice Core Data

2800 2850 2900 m

Figure 15.2. Comparison, over the last 140 lea, of various climatic records covering the last glacial-interglacial cycle (adapted from Jouzcl, 1994). (a) The bcnthic <5180 record from VI9-30 showing continental ice volume variations, (b) The Yostok deuterium profile, (c) The GISP2 <51X0 record using a depth scale for the lower part of the record for which the GISP2 and GRIP records diverge, (d) The GRIP 5ls0 record, (e) Records obtained from North Atlantic deep-sea cores with indication of the I leinrich layers. The dashed lines indicate correspondences between climatic-events (Bender ct al., 1994; Jouzcl, 1994).

via atmosphere improbable. It points instead to a dominant role of the ocean controlling the past climate of both regions (Blunier et al., 1998).

One of the most fruitful interactions between the modern and paleo scientific communities is indeed the study of multiple states of the thermo-haline circulation that are quite probably associated with these rapid climatic changes, which occurred during the last glacial period and the following climatic transition (see Stocker, Chapter 9 in this volume). This characteristic feature of coupled ocean-atmosphere models would have been left aside without the numerous evidences gathered by glaciologists, palco-occanographers, and palcoclimatologists. General circulation models (GCMS) are now used to quantify the rapidity of swings between modes (Rahmstorf, 1994). Documenting the variability of the thermo-haline circulation on decadal-to-centennial time scales through the construction of high-resolution time series will lead to progress in our understanding of the thermo-haline circulation and will be useful for forecasting its future behavior (Stocker and Schmitter, 1997).

15.3 Climate Variability During Warm Periods

The study of climate variability during periods as warm or warmer than today focuses on the Holocene (the current interglacial) and on the Eemian (the last inter-glacial). Relevant to future climate change is the Eemian, which was probably slightly warmer than today with higher sea level (although this climate should not be taken as an analog of a future warm climate, inasmuch as climate forcings clearly differ). Study of the corresponding part of the GRIP and GISP2 records revealed rapid iso-topic and chemical changes that have raised questions about the stability of this period (GRIP project members, 1993). Records of atmospheric composition in the air bubbles trapped in ice have later shown that such instabilities observed for the lowest part of both Greenland cores (GRIP project members, 1993; Grootes et al., 1993, Fig. 2) do not correspond to climatic instabilities during the last interglacial period (Bender et al., 1994; Fuchs and Leuenberger, 1996; Chappellaz et al., 1997). On the other hand, the Vostok record (Petit et al., 1999) provides a very detailed and undisturbed record of the Eemian (and of the two previous interglacial periods, stages 7.5 and 9.3).

This Antarctic record illustrates that the first part of the last interglacial was warmer than the Holocene. This first period of ka (Figure 15.2) is then followed by a relatively rapid cooling and then a slower temperature decrease somewhat parallel to pollen records from Western Europe (Chcddadi et al., 1998) and to some North Atlantic deep-sea core records (Cortijo et al., 1994; Jouzel, 1994). Paleo-oceanographers have now examined the Eemian in close detail. From a Bermuda Rise high-resolution sediment record, Adkins et al. (1997) showed that this period was relatively stable but began and ended with abrupt changes in deep water flow with transitions occurring in less than 400 years. Interestingly, during this "warm" period, the "conveyor belt" circulation, which today carries heat north from the Tropics and warms much of Europe, remained strong and relatively steady.

One of the striking features of the Vostok climatic record is that the Holocene is, by far, the longest (Ml ka) stable warm period recorded in Antarctica over the past

420 ka. The above comparison with North Atlantic and European records suggests that this may be true worldwide, with profound implications for the evolution and the development of civilization (Petit et al., 1999). It now appears that this relatively stable Holocene was indeed marked by a millenial-scale cycle observed both in ice core and in deep-sea core records (O'Brien ct al., 1995; F. Yiou ct al., 1995,1997; Bond ct al., 1997). As noted by Bond et al. (1997), such variability was demonstrated from measurements of soluble impurities in Greenland ice, showing that Holocene atmospheric circulation above the ice cap was punctuated by a series of millenial-scale shifts. Such Holocene events have now been documented in North Atlantic sediments, where they make up a series of climate shifts with a cyclicity close to 1470 ± 500 years (Bond et al., 1997).

The most prominent event seen in the Greenland record over this period was the cooling that occurred ^8200 years ago. This 200-ycar-long cooling is also seen in western Europe (von Grafenstcin ct al., 1998; see Figure 15.3). Widespread proxy records from the Tropics to the polar regions now show that this short-lived cooling event possibly had a worldwide extension (Gasse and van Gampo, 1994; Alley ct al., 1998; Stager and Mayewski, 1997). This abrupt cooling may have been triggered by a sudden freshwater pulse from the collapse of the Hudson Ice Dome, although there is no clear evidence for such an event at this time (von Grafenstcin et al., 1998). Alternatively, Alley et al. (1998) assumed that the climatic oscillation was induced by a reduced conveyor belt strength, caused by weak changes in the freshwater supply to the North Atlantic. Such a weak forcing scenario would support models of the North Atlantic thcrmo-haline

Figure 15.3. Variation, over the last 16 ka, of the Greenland ice oxygen 18 record (GRIP) and of the oxygen 18 content of shells of two different species of ostracods sampled in a core drilled in Ammersee Lake in Havana (adapted from von Grafenstcin et al., 1998).

111111H11111 »11111 n 11111II111111| 11111 ■ 1111N111111| 11111H

2000 4000 6000 8000

cal. years BP

Figure 15.3. Variation, over the last 16 ka, of the Greenland ice oxygen 18 record (GRIP) and of the oxygen 18 content of shells of two different species of ostracods sampled in a core drilled in Ammersee Lake in Havana (adapted from von Grafenstcin et al., 1998).

circulation to small changes in the freshwater influx, which could occur in response to global warming (Alley et al., 1998).

15.4 The Last Millenium

We now have a better, but still insufficient, knowledge of the climate variability over the past few centuries, knowledge that is important for climate change detection. It is also necessary in order to provide a comprehensive record of natural (non-anthropogenicallv forced) seasonal-to-interdecadal variability and to put the past 100 years, well documented by meteorological information, in the context of the last mil-lenium (CLIVAR, 1997). Mann et al. (1998) recently inferred spatially resolved global reconstructions of annual surface temperature patterns over the past six centuries based on the multivariate calibration of w idely distributed proxy climate indicators. Ice core data contribute to such global reconstructions, but their geographical coverage is obviously limited (Greenland and Arctic ice caps and tropical glaciers from the Andes, from the Himalayas, and, potentially, from Antarctica).

Despite such geographical limitations, ice core data should play a key role in our effort to better assess climate variability over the last millenium. First, they record changes of climate patterns over large regions and, second, they give unique access to relevant climate forcings.

There is a clear impact of events such as El Nino Southern Oscillation (ENSO) on the regime of precipitation in the Andes, and studies of ice cores such as that from the Quelccava ice cap in the Southern Andes of Peru provide information on ENSO events over the last 1500 years (Thompson et al., 1984). More surprising are the striking similarities between the accumulation histories revealed by this Andean record and a record obtained in the Guliya ice cap in Tibet, inasmuch as the two sites are 20,000 km apart, lying on opposite sides of the Pacific Basin (Thompson, 1996). During the overlapping 1500 years the major periods of drought and wetness appear contemporaneous, suggesting teleconnections for ENSO events as well as for lower-frequency events.

It is also a record of accumulation change that shows the best potential for reconstructing a North Atlantic Oscillation (NAO) index from Greenland ice cores (Appenzeller et al., 1998) based on the connection between snow accumulation in western Greenland and mean pressure at sea level. This index extends back to a.d. 1650, indicating that the NAO is an intermittent climate oscillation with temporally active and passive phases. Greenland isotopic records provide another way to get information on conditions prevailing in the North Atlantic (Figure 15.4). One example is given by the deuterium-excess record at the GRIP site. The deuterium excess (a linear combination of deuterium and oxygen 18 ratios in water) is influenced by the temperature and humidity prevailing in the source regions for Greenland precipitation. The anomalous low deuterium-excess values in the 18th and 17th centuries (Figure 15.5) are interpreted as reflecting cooler conditions in the subtropical North Atlantic and thus correspond to a cool period in Europe known as the Little Ice Age (Hoffmann et al., unpublished).

Was this article helpful?

0 0

Post a comment