Stratigraphic timescale between 11510 and 60000 yr BP

The stratigraphical dating of the GRIP core has hitherto been presented only in an internal report (Hammer et al., 1997), but in the following I will try to show why this dating of the GRIP core is probably the most accurate dating of the last glacial stage back to 60ka. The most important dates are presented in Table 78.1.

As mentioned earlier only the dust concentration profile was available to date the GRIP core prior to 11,550yrBP. This immediately raises the questions 'How sure can we be that the interpretations of the seasonal variations are correct during the glacial cycle and can we verify the corresponding time-scale?'.

The interpretations of the seasonal cycles are based on the following:

1 the seasonal variations in the GRIP core are clearly annual to some extent, because there are no other reasonable explanations for the observed consistent variations;

2 a reasonable explanation of the seasonal dust concentration during the last glacial stage is the presence of extreme dryness between winter and spring and more wet summer and autumn conditions;

3 the corresponding ice time-scale can be verified by comparison with other types of well-dated records.

Table 78.1 GRIP time-scale between Preboreal and 60kyr BP. The dates refer to the start of the stated climate periods. The IS numbers refer to the interstadials as given in Dansgaard et al. (1993) and the C numbers to the preceding cold stadials

Start of period (before AD 1950)

Climate period

Start

Preboreal

11510

Younger Dryas

12704

Allerod

14136

Bolling

14872

C1

23615

IS2(a + b)

24089

C2

28130

IS3

28495

C3

29407

IS4

29799

C4

33490

IS5

33853

C5

34936

IS6

35248

C6

36152

IS7

36964

C7

38098

IS8

39846

C8

41442

IS9

41583

C9

42358

IS10

42854

C10

43492

IS11

44482

C11

47072

IS12

49101

C12

51125

IS13

51706

C13

54382

IS14

57319

C14

58400

IS15

58807

C15

59874

The first two points are based on reasonable assumptions, but their correctness cannot be proven without involving the dating of other kinds of records, hence the third point is important. In order to verify the stratigraphical dating of the GRIP core it will be helpful first to consider the transition from the Last Glacial Maximum (LGM) to the end of the YD. This is a period worth special attention as many kinds of palaeoclimatic data cover this period and it represents very abrupt and strong climate changes. The climate changes can be seen in the isotopic profile (Fig. 78.2), but are even more clearly observed as annual layer changes; these changes directly reflect the changing climate. The isotopic profile is only a quantitative indicator of palaeotemperatures if a calibration curve exists. The calibration curve will depend on the seasonal distribution of the snowfall and the height of the old ice surface; information on these parameters is difficult to obtain, although a rough calibration can be made by means of borehole temperatures (see Cuffey et al., 1995; Dahl-Jensen et al., 1998; Johnsen et al., 2001).

Figure 78.2 The chronological link between the Dye 3 core, the Camp Century and GRIP core. The isotopic minimum in all three records (asterisk) and a common major well-identified volcanic signal (triangle) tie the chronology of the Camp Century (A), Dye 3 (B) and the GRIP core (C) together around 8000yr BP. All records are plotted on a depth scale, but only the GRIP depth scale is shown (to the right). Profile D indicates the corresponding annual layer thickness of the GRIP core.

Figure 78.2 The chronological link between the Dye 3 core, the Camp Century and GRIP core. The isotopic minimum in all three records (asterisk) and a common major well-identified volcanic signal (triangle) tie the chronology of the Camp Century (A), Dye 3 (B) and the GRIP core (C) together around 8000yr BP. All records are plotted on a depth scale, but only the GRIP depth scale is shown (to the right). Profile D indicates the corresponding annual layer thickness of the GRIP core.

Figure 78.3 shows the annual layer thickness a(z) in the GRIP core over the transition from pre-B0lling to a little after the end of the YD. Even though this ice core sequence is 130 m long the change in a(z) due to the ice flow is only around 10%, i.e. the variation in a(z) first of all reflects the relative change in the original surface accumulation A(t). The changes are dramatic both in the values of a(z), nearly a factor of two, and in the rapidity of the changes. Further an accurate tree-ring date exists for the end of the YD (Spurk et al., 1998) and a similar accurate U/Th date for the onset of the B0lling (Bard et al., 1993). The agreement between the dates is quite satisfactory: the ice chronology indicates 11,510 ± 40yrBP for the end of the YD and the tree-ring date 11,500 ± 30yrBP. In the case of the onset of the B0lling the ice-core date is 14,872 ± 80yrBP and the U/Th dating 15,000 ± 100yrBP. With this verification it can be concluded that the GRIP stratigraphical dating is accurate enough to serve as a master chronology over the transition.

With a GRIP master chronology at hand it is of interest to compare some transitional GRIP events to similar events in other palaeorecords. Some obvious events could be: the icelandic Vedde volcanic ash layer has been dated to ca. 12,000 yr BP (Gr0nvold et al., 1995) and is observed in the GRIP ice core at 12,008yrBP; the Laacher See volcanic ash from the Eifel mountains in Germany is generally placed at 12,800yrBP and a possible ash layer in the GRIP core exists at 12,800yrBP (not analysed yet): the rapid decline in a(z) around 12,600yrBP concurs with the estimated date of 11,00014CyrBP for the onset of the YD (see W. Broecker's review, 2003). The review by Broecker also discusses the chronological problems associated with the Lake Agassiz catastrophic flood and the onset of the YD. If the Laacher See tephra can be identified in the GRIP core around 12,800yrBP (GRIP stratigraphical date) it would prove that there was indeed a 200yr period between the Laacher See eruption and the onset of the YD. In any case a 130-m-long, well dated ice-core sequence exists from which year by year information on atmospheric changes can be inferred over this unusual climatic transition.

The ss09 model time-scale over the transition differs from the stratigraphical time-scale as it places the start of B0lling at 14,700 yrBP. The most likely explanation is that the empirical relation between the isotopic values of the ice and the surface accumula

Figure 78.3 The annual layer thickness as a function of time over the Last Glacial Maximum-Holocene transition. The plot is a smoothed version of the more detailed layer counting. The arrows indicate various 'events' that can be used as reference points in order to validate the time-scale by comparing it to other types of dated records: end of Younger Dryas (a); Vedde ash layer (b); Lake Agassiz meltwater pulse (c); Lacher See ash layer? (d); start of Bolling (e).

Figure 78.3 The annual layer thickness as a function of time over the Last Glacial Maximum-Holocene transition. The plot is a smoothed version of the more detailed layer counting. The arrows indicate various 'events' that can be used as reference points in order to validate the time-scale by comparing it to other types of dated records: end of Younger Dryas (a); Vedde ash layer (b); Lake Agassiz meltwater pulse (c); Lacher See ash layer? (d); start of Bolling (e).

tion overestimates the surface accumulation; but whatever the reason only the seasonal stratigraphical chronology fits the U/Th date for the start of B0lling.

What about the the pre-B0lling chronology?

Was this article helpful?

0 0

Post a comment