Other dating methods

Another approach to the dating of ice cores is the use of radioactive decay methods, but their usefulness is limited by factors such as the half-life, the low concentration of trace substances and gases in the ice as well as complexities concerning sampling and time resolution. The reader is referred to the literature, for example, the review paper by Stauffer et al. (1989). Such methods can, however, serve as a rough verification of the model or seasonal chronologies, or can be used where other methods are not applicable. Radioactive dating also can be of use to ice-core chronologies in an indirect way, especially for the older part of the ice cores, which is difficult to date directly. If an event in the ice core can be reconciled with an event in marine or terrestial records and if the latter can be dated by a radiometric method the date can be transferred to the ice core.

This leads us into the use of reference horizons, which holds a great potential for tying together the chronologies of ice cores, terrestial and marine records, for example, by mutual volcanic events. The final goal is to obtain precise and accurate chronologies for all palaeoclimatic records in order to compare their information over time.

The chronological work on ice cores is presently in an interesting and exciting phase, which in my opinion makes it possible to see faintly the future chronological framework of ice cores. The framework could be viewed in the following way.

1 A 'master chronology' spanning the past 100,000 yr is constructed based on the Greenland deep ice cores from Dye 3, GRIP and North GRIP.

2 Based on common major equatorial volcanic eruptive signals in Greenland and Antarctic ice cores the two polar regions are locked into a common chronology representing the past 100,000 yr. The use of volcanic signals is important, because it offers a high number of common events, thus limiting the time periods for which interpolation between reference horizons are needed. If the master chronology has a high precision and accuracy so will the dates of the reference horizons.

3 For ice-core layers older than 100,000 yr it will be necessary to use common reference horizons between ice cores, marine and terrestrial sediments.

In the following I shall concentrate on the first point, because most ice-core chronologies deal with ages younger than 100,000 yr and a common masterchronology for these ages has not yet been agreed upon. All dates given as BP will refer to ad 1950.

78.3 Holocene chronology

During the years 1979-1982 the South Greenland Dye 3 ice core was drilled in a high accumulation area, i.e. some 1.25 m of annual snow deposition corresponding to 0.56 m of ice equivalent. Such a high annual snow deposition combined with a high number of major snowfalls per year ensures that the seasonal isotopic composition of the ice could be used as a dating technique back to approximately 8000 yr BP. The accuracy of the dating was improved by using the continuous acidity profile, electrical conductivity method (ECM) (Hammer, 1980) and in some cases also a detailed dust concentration profile, as a cross-check of the iso-topic cycles.

The ECM and dust profiles were measured with a much higher resolution along the core than the isotopic profile and the diffusive processes obliterating the seasonal signal with depth are much smaller than for the corresponding isotopic profile. The iso-topic seasonal changes are, however, connected to the seasonal cycle of the atmospheric circulation in a more regular way than the changes in the ECM signal and the dust concentration. Irrespective of this the two profiles were quite helpful in resolving some of the more subtle seasonal variations in the isotopic profile.

The resulting time-scale was checked by historical dated volcanic reference horizons for the past 1000 yr, which indicated an accuracy of ±3yr for a 1000-yr-old ice layer. Recently verification of the Dye 3 Holocene dating was extended 900 more years back in time when the volcanic signal of the ad 79 Vesuvius eruption was identified in the GRIP ice layer of ad 79 (ad 80 in the Dye 3 core) (Clausen et al., 1997; Barbante et al., in preparation), i.e. only one year from the expected year ad 80. As the stratigraphi-cal dating of the Dye 3 core between ad 80 and 8000 yr BP was not basically different from its dating between ad 1980 and ad 80 it can be concluded that the Dye 3 dating can serve as a master chronology back to some 8000yrBP with a tree-ring-like precision. This was further confirmed after the GRIP deep drilling, when the major volcanic signals in the Dye 3 and GRIP ice cores were compared, based on the independent stratigraphical dating of the two cores back to 4000yrBP (Clausen et al., 1997).

The GRIP core was drilled during the years 1989-1992. The annual snow deposition on the GRIP site is only some 40% of the deposition in the Dye 3 area, which makes seasonal isotopic dating more difficult and limits a fairly accurate dating to 4000yr BP. The Holocene part of the GRIP core, back to 8000yrBP, therefore has been dated by using the several common volcanic signals, with the Dye 3 core as reference horizons: only unambiguous volcanic signals were used in order to avoid misinterpretations. This could be accomplished because some of the major volcanic signals in Greenland ice cores are characterized by an acid deposition, which is clearly revealed by signal strength (acidity), duration of deposition and depositional pattern over time (see Hammer, 2002). Figure 78.1 shows how the various Greenland deep ice cores were tied together around 8000 yr BP. At deeper strata the Dye 3 core is less relevant for accurate dating owing to the ice flow in the Dye 3 region: fast thinning of the annual layers and problems close to the bedrock. To proceed further back in time we need to consider the GRIP deep ice core from central Greenland.

Between 8000yrBP and the Younger Dryas (YD) the seasonal stratigraphy of the GRIP core was superior to the Dye 3 core, because the annual layer thicknesses were four to six times larger than in the Dye 3 core. Also the analytical techniques have improved considerably since the Dye 3 drilling; this made the interpretation of the very detailed nitrate and dust concentration profiles fairly simple. The counting of annual years was checked by several independent countings and the end of the YD was established as 11,510 ± 40yrBP with a deviance of only 5yr between the various countings over the period 8000 yr BP to the end of the YD. The accuracy of this date is not only a function of the interpretation of seasonal variations in the Dye 3 and GRIP core, but must also include the uncertainty of defining the end of the YD: the latter has been chosen here as the time when the dust concentrations prevailing during the YD fall dramatically. This happens almost at the same time as the isotopic values of the ice are changing, but is seen much more clearly in the dust concentration profile.

The same dating technique was used to take the dating further back in time, but already during the in situ measurements of the nitrate and dust concentrations over the YD period it became clear that the nitrate and dust variations were not independent. The two methods therefore could not be used to cross-date the core. Hence the dating of the last glacial stage had to be performed with only one seasonal dating parameter—the dust concentration. The dust concentration was chosen because it was less apt to diffusion and could be measured with a very high resolution along the core, i.e. a few millimetres. The high resolution of the measurements made it in theory possible to date the core as long as the annual layer thicknesses were larger than 1 cm. By using a special technique it even could be used as long as the annual layers were larger than a few millimetres, but in the latter case the measurements were very time consuming.

78.4 Chronology of the last glacial stage

The GRIP and GISP 2 drillings at Summit in central Greenland (1989-1993) retrieved two ice cores to bedrock and 28 km apart, which came to play an essential role in the palaeoclimatic research of the Late Quaternary. Not only were the in situ measurements and sample collections more extensive than for any previous ice-core drilling, but for the first time ice deposits covering the last glacial stage became available with an unprecedented time resolution. Hundreds of papers have been published dealing with the results obtained on the Summit ice cores and the readers who want to know more are recommended to seek information in a special issue of the Journal of Geophysical Research (1997) on the Greenland Summit Ice Cores. Information on the chronology of the GISP 2 core can be found in, for example, Alley et al. (1993, 1997a) and Meese et al. (1997).

From a chronological point of view of the last glacial stage the Summit drillings offered 1200-m-long ice-core records covering the time between the end of the Eemian to the end of the YD. Such long records made it in principle possible to date the records by seasonal stratigraphy back to the Eemian. Technically this was done for both ice cores, although the techniques used were different.

Before the Summit drillings the Byrd ice core of West Antarctica had actually been dated by seasonal stratigraphy back to 50kyrBP (Hammer et al., 1994) using the ECM technique. The successful dating of the Byrd core owed its origin to the seasonal production of sulphur-rich products in the ocean around West Antarctica (the measurements were actually performed during the 1980s). This eventually led to a very strong seasonal acid variation in the ice layers. It was, however, a problem that the Byrd ice-core chronology could not be compared, in a strict sense, with

Figure 78.1 The pre-Holocene GRIP isotopic profile and the corresponding dust concentration profile. The time-scale shown to the right is the ss09 (see also main text).

Figure 78.1 The pre-Holocene GRIP isotopic profile and the corresponding dust concentration profile. The time-scale shown to the right is the ss09 (see also main text).

%o Concentration [mg/kg]

other chronologies of a similar accuracy. Actually the Byrd chronology was not verified until Blunier et al. (1998) proved that the abrupt climate changes (DO events) during the last glacial were not in phase between the two hemispheres. Nevertheless the findings of Blunier et al. did not depend on absolute dating and it was still a problem that the deep ice cores from Antarctica and Greenland presented various chronologies: no agreement existed on a master chronology.

In 1996 a new deep drilling, North GRIP, started in northeast Greenland and in 1995 the EPICA drilling started in East Antarctica. These two deep drillings have now recovered two new deep ice cores and in the coming years several papers on the results will appear. Both drillings were characterized by an almost comprehensive use of experimental techniques both in situ and in the laboratory. The chronology of these cores will extend ice-core chronology far back in time and with new precision. Their chronology will be the result of all the new data and the knowledge already obtained on the previous deep ice cores. Before all these data become available it will be worthwhile to consider the chronologies of several deep ice cores as the situation was in 2002. Schwander et al. (2001) compared the various time-scales in order to discuss and show the differences. The result of this comparison over the various chronologies of the last glacial stage clearly illustrated the overall problem: the time-scales deviated up to 5kyr! This is problematic, because the important climatic events observed in various palaeorecords must be on a more or less accurate time-scale in order to reveal the climatic teleconnections correctly. The time-scales can be criticized, but what we need is not just critiscism but verifications or rejections of the various time-scales.

I have in the following chosen to present some data on the GRIP ice core, which in my opinion strongly indicate that it is indeed possible to establish a master chronology for the last glacial stage.

Was this article helpful?

0 0

Post a comment