Glacio-isostatic subsidence of the Tibetan plateau below the ELA under ice pressure is interpreted to be one important factor for deglaciations. The resulting albedo drop is a plausible mechanism to pace a self-organizing path towards ice-free conditions. The high uplift rates of 12 mm/a measured in the region of the Tibetan plateau (Hsu et al. 1998) may be a first indication of glacio-isostatic recovery. The author's own observations of moraine deposits on the northern slope of Shisha Pangma, in the border area between the Himalayas and the Tibetan plateau, are consistent with this (Kuhle 1988c). These extensive pedestal moraines were left by the ice during the late Late Glacial.
Kaufmann and Lambeck (1997), Kaufmann (2003) have shown that, on the basis of secular changes in geoid anomaly and free-air gravity anomaly, it is possible to distinguish the amount of glacio-isostatic uplift from uplift caused by tectonic movements. The predictable effects of the melting of an up-to-2 km thick incland-ice on the Tibetan plateau are so profound that the current satellite missions CHAMP and GRACE would be able to identify them.
A further issue is the extent of the influence of the Tibetan ice sheet on the heat balance of the atmosphere, particularly whether this was marked enough to have a decisive initial impact on the pattern of global ice ages. Only modeling results can provide an answer.
Using an atmospheric general circulation model (GCM), Verbitsky and Oglesby (1992) studied the evolution of ice sheets in relation to the atmospheric CO2 concentration (i.e. temperature change) using a computed "glaciation sensitivity" index. Their results showed that Tibet and Siberia were more likely to develop an ice sheet than Canada or Fenno-Scandinavia.
Similar conclusions are drawn by Marsiat (1994), who used a 3-D climate model to simulate global ice growth generated by temperature changes caused by orbital variations. Here too, the first large ice sheet formed on the Tibetan plateau, followed by Siberia. In contrast to Verbitsky and Oglesby, Marsiat sees a strong glaciation tendency in Alaska too.
Both models thus confirm the hypothesis that the Tibetan ice sheet played a foremost role in terms of time.
The latest modeling results presented by Kutzbach et al. (1998) point into the same direction. However, these climate and biome simulations carried out using the Community
Climate Model, Version 1 (CCM1) do not investigate the pattern of ice build-up but are based on the reconstructed glaciation conditions described in CLIMAP (1981), i.e. the Tibetan plateau is assumed to be non-glaciated.
As model results show, there are still "small areas of permanent snow-cover over non-glaciated areas of western Canada-Alaska, northern Eurasia between the Eastern Siberian and Western Eurasian ice sheets and over Tibet at 21 ka. This result indicates that the model would develop a larger area of glaciated land than actually appeared to have been present at 21 ka. The simulation of permanent snow cover over Tibet contributed to the reduction of the Asian monsoon at 21 ka." (Kutzbach et al. 1998, p.496).
That a Tibetan ice sheet not only influenced the monsoon but also had a global impact, is demonstrated by a sensitivity experiment conducted by Marsiat (1994). By reducing the global albedo of snow-covered mountain areas Marsiat attempts to prevent what he considers to be the unrealistic formation of ice sheets on the Tibetan plateau and the Rocky Mountains: "although the mountainous areas were covered by ice after some period, the perturbation occuring at the beginning of the glacial cycle influences the remainder of the simulation, showing lower ice volumes during the entire glacial cycle." This suggests that the albedo effect of the Tibetan ice sheet has a global impact. Thus, unintended, the results of Marsiat (1994) are consistent with the field data: Without tuning the global albedo the model produces the inland-ice on the Tibet-Plateau. Exactly this was observed.
In order to investigate this impact in greater depth, it is appropriate to use sensitivity experiments with circulation models that consider the Tibetan ice sheet to be a realistic possibility rather than a simulation error. It would make sense to place alternative simulations side by side, showing, on the one hand, the climatic effects of a glaciated Tibetan plateau - as reconstructed by the author on the basis of his empirical fieldwork - and, on the other, simulations based on the old view (Wissmann 1959) of a non- or only slightly glaciated plateau, which is still being advocated by Derbyshire and Shi Yafeng (1991), among others. Sensitivity experiments with atmospheric circulation models could also be used to quantify the differential influence of the albedo of the various ice sheets on the global cooling balance.
The crucial role played by the Tibetan plateau for our understanding of Quaternary climatic change is undisputed (Hughes 1998). Due consideration of the Tibetan ice sheet is expected to significantly increase the realism of climatic simulations.
Was this article helpful?