Reply to: “Comments on: ‘6000-year climate records in an ice core from the Høghetta ice dome in northern Spitsbergen’”

Sir,

First, we appreciate the comments and suggestions by Dowdeswell and others (1990) on our chronology and palaeoclimatic interpretation of the ice cores obtained in northern Spitsbergen (Fujii and others, 1990). Since reliable chronology of the ice core obtained from the superimposed ice zone, as at our research site (Høghetta) is difficult because of the disappearance of seasonal and even annual stratification by melting, percolation and/or wash-out, there might be a case for some chronological interpretation on the basis of the analysed data.

We should like to describe our chronological interpretation of the ice core giving some recent data and other evidence additional to that already described. In table II of the paper there is a misprint: 1770 in column “Period” should be read as 1700 as appears in the text. Therefore, we should read 1770 in the comments by Dowdeswell and others as 1700.

Core chronology for recent decades

The ice-coring site at Høghetta, Spitsbergen, is located in a typical superimposed ice zone as no firn was observed in the core. Percolation of melt water to the lower layers is not so important in superimposed ice because the water channels developed at grain boundaries only in the upper 20 cm, as was observed in the cob-webbed air-bubble layer. So the annual characteristics of the ice are considered to be preserved when the annual balance is positive, even though percolation occurs.

The tritium profile shown in figure 3 of the paper (Fujii and others, 1990) is, therefore, thought to reflect the annual characteristics when mass balance was positive. The variation pattern is compared with the recent results of tritium analyses (Izumi, in press) for firn cores obtained at site J (67.5° N, 43.5° W), southern Greenland (the project has been outlined by Watanabe and Fujii (1990)), in 1989 where the mass balance was positive and the seasonal stratigraphic features were preserved. The variation pattern of the Høghetta cores is very similar to that of the site J core. The tritium-concentration peaks of the site J core between depths of 22 and 16 m which are layers of the 1950s and 1960s correspond well to those of the Høghetta core which, however, appear near the surface between depths of 3 and 0.5m (Fig. 1). This suggests frequent occurences of negative balance or a small positive balance after the later 1960s.

Fig. 1. Comparison of tritium–concentration profiles of site [Inline 1], Greenland (a), and Høghetta, Spitsbergen (b) (Izumi, in press).

A decreased mass balance after 1963, at the latest, at Høghetta can be supported by the continuous negative mass balance of Austre Broggerbreen, western Spitsbergen (Hagen and Liestøl, 1990) during the whole observation period of 1986–88 even though there was a cold climate in the 1960s. Furthermore, as the water equivalent of the winter balance of 1986–87 was only 15.2g (54 cm depth), as shown in figure 3 of the paper (Fujii and others, 1990), the annual balance from the end of the 1986 summer should have been at most several centimeters of water, taking account of the ablation of snow during the summer of 1987.

Chronology of the upper part of the ice core

Recently, we have obtained new results of 210-Pb measurements, as shown in Figure 2 (personal communication from T. Suzuki), which suggest 16 cm of water (or about 18 cm of ice) as an average annual accumulation rate for the upper 10 m of the ice core. The accumulation is rounded to “about 20 cm of ice”, which was used in the paper.

The pH profile shown in figure 4 in the paper (Fujii and others, 1990) was compared with those obtained from Austfonna, Svalbard, by Zagorodonov and Arkhipov (1990) and at site J, Greenland. These three pH profiles, together with the chronological time-scale determined independently by each other, show common variation patterns. As for the chronology for the site J 205 m ice core, several different methods using the seasonal cycle of ECM records (by F. Nishio), interpretation of the pH signals (by K. Kamiyama and Y. Fujii), the tritium profile (by K. Izumi) and the density-depth profile (by T. Kameda and H. Shoji) show an average annual accumulation rate of 38–40 cm of water, which agrees with the accumulation rate of 37 cm of ice at site 2 (Clausen and others, 1988), where it is about 50 km from site J. Coincidence of the pH-variation pattern between Høghetta, Spitsbergen, and site J, Greenland, cores suggests the rationality of the chronology of the upper part of the Høghetta ice core.

The highest pH value of 8.3 at a depth of 32.5 m shown in figure 2 of the paper (Fujii and others, 1990) may be comparable to the highest Cl⁻-concentration peak of 7.73 mgl⁻¹ at 76.1 m depth which corresponded to some year in the early AD 1800 s (personal communication from V. Zagorodnov). On the other hand, the age of the highest pH peak in the Høghetta ice core is estimated to be around 1810. This could support the chronology completed for the upper part of the Høghetta ice core.

Possibility of an hiatus prior to AD 1700

Recent studies by Zagorodnov and Arkipov (1990) on an ice core from the summit of Austfonna, Nordaustlandet, in Svalbard indicate evidence of a very warm climate from depths of 110 to 90 m during the Little Ice Age. They have deduced the warm climate from the concentration and thickness of infiltration ice layers. As they have proposed two time-scales for the ice core, the warm period can be interpreted approximately from c. 400 to c. 300 years BP, after the colder period began 1200–1300 years BP.

Fig. 2. The vertical profile of Pb-210 activity in the ice core from Høghetta, Spitsbergen (measured by T. Suzuki).

According to the oxygen-isotope profile obtained for an ice core from Vestfonna, Nordaustlandet (Vaykmae and others, 1985), the warm period continued from c. AD 1200 to c. AD 1600 prior to the cold period from c. AD 1600 to c. AD 1900.

Prior to AD 1700, there may have been a negative mass balance during these and earlier warm periods.

In conclusion, we suppose the following glaciological history at Høghetta in northern Spitsbergen: that the formation of glacier ice re-commenced at around AD 1700 during the Little Ice Age on stagnant ice after an hiatus of about 4000 years and the annual mass balance has been reduced since the middle 1960s.

Acknowledgements

I thank T. Suzuki, K. Izumi, F. Nishio, H. Shoji, K. Kamiyama, T. Kameda and V. Zagorodnov for providing unpublished data.

Yoshiyuki Fujii

National Institute of Polar Research, 9–10 Kaga 1-Chome, Itabashiku, Tokyo 173, Japan

2 November 1990