Himalayan–Tibetan uplift and global climate change

Himalayan–Tibetan uplift and global climate change During the late 1980s and early 1990s much attention was focused on the potential role of late Cenozoic uplift of the Himalayas and Tibetan plateau in driving global cooling and the growth of large continental ice sheets during Late Tertiary and Quaternary times. The formation of the Himalayas and the Tibetan plateau is thought to have altered global atmospheric circulations and to have changed the concentrations of gases in the atmosphere as a result of changes in the biogeochemical cycles associated with the increased weathering of newly exposed rock surfaces, which in turn led to global cooling.

There is unequivocal evidence to show that the Indian and Asian continental plates collided between 54 and 49 Ma (million years) ago. Since then, India has continued to move northwards into Asia at a rate of between 40 and 50 mm a year⁻¹. This has resulted in folding and thrusting of the Indian crust in the Himalayas and southern Tibet. The homogeneous deformation of Tibet entailed shortening and doubling the thickness of the crust, elevating the Himalayas and Tibetan plateau to an average height of between 5000 to 5500 m above sea level. The tectonic forces resulting from the northward movement of India were not, however, in the opinion of some geologists great enough to support the thickened crust, and the plateau began to spread under its own weight. Currently, the plateau is spreading, extending east–west on normal faults, by about 1 per cent of its extent every million years.

This rise in elevation may have altered climate in a number of ways. First, the rising plateau would have deflected and blocked regional air systems, causing jet streams to meander, and facilitating the southward migration of cold polar air into key locations in the northern hemisphere, particularly North America and north-west Europe, the sites of ice sheet development. This in turn would have affected atmospheric circulations on a global scale. Secondly, the elevated region would have enhanced temperature-driven atmospheric flows as higher and lower atmospheric pressure systems developed over the plateau during the winter and summer, respectively. This would have intensified the Indian monsoon, leading to heavier rainfall along the frontal ranges of the Himalayas, while the aridity across the Tibet plateau and central Asia would have increased as northward-moving monsoon winds lost their moisture as they were forced over the Himalayas. In Central Asia, the increased aridity would have led to higher temperatures during the summers and extremely low temperatures during the winters. In addition, the plateau would have cooled as it rose higher into the atmosphere, and glaciers would have been able to develop. These glaciers would have provided positive feedback, reflecting incoming solar radiation and leading to further cooling. Thirdly, the greater availability of new exposed rock surfaces and detritus produced by denudation processes would have allowed more chemical weathering to take place. During chemical weathering atmospheric carbon dioxide (CO₂) reacts with rock-forming silicate minerals to produce bicarbonates, which are washed into the oceans where they eventually form carbonate rocks. Because CO₂ is an important greenhouse gas, helping to warm the atmosphere, a decrease in the amount of atmosphere CO₂ would lead to global cooling. It has been argued that this process resulted in decreased CO₂ over the past 40 Ma leading to an icehouse effect and the onset of the Quaternary ice age.

Throughout the Cenozoic, other major tectonic events also took place, which could have forced climate change. Continental displacements led to changes in the configurations of the oceans, and seaways opened and closed. These would have given rise to major changes in global and regional oceanic circulations, altering the exchange of heat throughout the globe. The rate of sea-floor spreading also decreased during early Tertiary times, reducing the supply of CO₂ from volcanic eruptions to the atmosphere. As sea-floor spreading was reduced, the oceanic spreading ridges cooled and subsided, lowering sea levels. This produced more land available to be weathered and allowed more plants to grow, storing CO₂ as organic carbon, further reducing the amounts of atmospheric CO₂. The importance of each of these processes, and of mountain uplift, in causing global cooling throughout the Tertiary, and ultimately the onset of the Quaternary ice age, is difficult to prove and assess.

A variety of proxy data have been presented to demonstrate the connection between Tibetan uplift and climate change on regional and global scales. The ratio of oxygen isotopes (¹⁸O/¹⁶O) in the calcareous shells of foraminifera (unicellular marine organisms) preserved in marine sediments essentially records past ice-sheet volumes and is a proxy for ocean temperatures. The oxygen isotope record for the past 55 Ma shows a long-term increase in δ¹⁸O values (δ¹⁸O expresses the changes in the ¹⁸O/¹⁶O ratio), indicating progressive cooling, punctuated by times of rapid cooling (Fig. 1a). The first major increase in δ¹⁸O values (indicating cooling) occurred about 35 Ma ago and probably represents the initiation of the growth of the Antarctic ice sheet. This was followed by 20 Ma of cooling. Further increases in δ¹⁸O values occurred during mid-Miocene times (approximately 15–13 Ma ago) and late Pliocene times (approximately 4–2 Ma ago); these increases probably represent further growth of Antarctic ice and the build-up of the northern hemisphere ice sheets, respectively. Tills, moraines, ice-rafted sediments and palaeontological evidence provide additional support for this cooling trend and the onset of glaciation in the Antarctic at about 35 Ma ago and in the Northern Hemisphere at approximately 4.5 and 2.5 Ma ago.

⁸⁷Sr/⁸⁶Sr isotope ratios recorded in marine carbonates provide an important proxy for estimating the rates of chemical weathering. A marked increase in ⁸⁷Sr/⁸⁶Sr ratios after Late Eocene times (approximately 38 Ma ago: Fig. 1c) may be attributed to increased weathering or a change in the type of rock being weathered (erosion of rocks supplies radiogenic strontium from the land) or a decrease in sea-floor hydrothermic supply. The latter is controlled by changes in the rates of sea-floor spreading. Because sea-floor spreading rates changed little during the past 30–40 Ma, the changes in the ⁸⁷Sr/⁷⁶Sr ratios may be attributed to increased weathering. Today, the Ganges–Bramaputra river systems have unusually high ⁸⁷Sr/⁸⁶Sr isotope ratios, and it is therefore argued that the late Cenozoic ⁸⁷Sr/⁸⁶Sr ratios may be attributed to enhanced weathering in the Himalayas. Others argue, however, that the increased ratios may reflect enhanced supply by glacial erosion associated with the growth of the Antarctic ice sheet.

Such an increase in weathering would deplete the atmosphere of all its CO₂ within a few million years. Negative feedback must therefore operate to return CO₂ to the atmosphere. One possible mechanism is related to imbalances in the organic carbon subcycle. ¹³C/¹²C ratios in marine carbonates reflect changes in the size of the organic carbon reservoir. If, for example, the burial of organic matter decreases, then δ¹³C would decrease (a decrease in the ¹³C/¹²C ratio). δ¹³C values in marine carbonates showed a marked decrease during the Cenozoic (Fig. 1c), indicating that carbon burial also decreased. This in turn could have caused additional CO₂ to be released into the oceans and atmosphere. Because oxygen is necessary to oxidize organic matter and release CO₂, the decrease in buried organic carbon may reflect an increase in dissolved oxygen concentrations in the oceans. Oxygen solubility increases with decreasing temperature. The dissolution of oxygen into the oceans would therefore have increased as the oceans began to cool. This would lead to the oxidation of organic matter and a greater release of CO₂ into the atmosphere. The residual times of CO₂ (less than 1 Ma) and O₂ (c. 10 Ma) in the atmosphere would not lead to a perfect balance of these processes. Rather, the balance of the organic subcycle would become less effective with time. This may have led to more intense forcing of cooling in late Cenozoic times as CO₂ became progressively depleted. Other possible negative feedback mechanisms that could release CO₂ into the atmosphere include the precipitation of silicate minerals in the deep ocean, and sea-floor weathering of basalt, but these processes are poorly understood.

It is difficult to assess the relative importance of all these complex processes in driving climate change. Attempts have been made, however, to model the evolution of Cenozoic climate using computer models of the Earth's atmosphere (General Circulation Models: GCM). Using GCMs, American Earth scientists have investigated the atmospheric changes that would be associated with the uplift of the Himalayas and Tibetan plateau. Their initial results have shown that many of the global changes in precipitation and temperature that were calculated by the models are consistent with those observed in the geological record over the past 40 Ma. Additional experiments have extended the GCMs to examine the sensitivity of the Indian monsoon. The results showed that strong monsoons, similar to those of today, could have been induced by strong solar forcing when the elevation was only half its present height. The question then arises as to when was Tibet high enough to initiate the monsoons, and was this coincident with the proxy evidence for global cooling and increased monsoon activity at around 7 to 8 Ma ago?

The timing of uplift is one of the major uncertainties in attributing uplift to climate change. With regard to the timing of uplift, there are three contrasting schools of thought. Primarily on the basis of mammalian fauna, palaeokarst, and geomorphology, the first school advocates very rapid late Pliocene– Pleistocene uplift. These changes in fauna, palaeokarst, and geomorphology could equally, however, be attributed to climate change. The second school suggests that uplift has occurred rather gradually since Early Eocene times, with substantial elevations being reached by late Eocene times. This is based on lithological variations in lacustrine rocks from eastern China, but there is a large degree of uncertainty with this hypothesis.

The third school, and the one whose view is most generally accepted, considers that rapid uplift occurred after about 25 Ma ago, and that the plateau attained its present elevation by about 14 to 15 Ma ago. This is based on the assumption that the present elevation and extensional deformation of the plateau probably resulted from uplift owing to convective thinning of the underlying lithospheric mantle. Post mid-Miocene lavas from the northern part of the Tibetan plateau, derived from lithospheric mantle, owe their origins to the thinning of the lithospheric mantle. These lavas have been dated at about 13 Ma. Furthermore, extensional faulting in the Thakkola graben in Nepal, which is the result of the extensional collapse of the plateau, has been dated at 14 Ma. These examples support the view that maximum elevation must have been achieved prior to 14 Ma. A reassessment of the sediments deposited on the Indo-Gangetic Plain and the Bengal fan has shown that there was a reduction in sediment supply from the Himalayas 8 Ma ago. This could be attributed to reduced tectonic activity after 8 Ma, further supporting the view that plateau uplift occurred more than about 8 Ma ago. Alternatively, reduced denudation could be related to major vegetational changes which also occurred at about 8 Ma ago. These uplift dates are essential for modelling and understanding the role of uplift on climate change.

An alternative view on uplift and climate change proposes that the Quaternary glaciations in the Himalayas could have enhanced uplift. This theory suggest that when the Earth's surface is evenly eroded, and is therefore unloaded, a corresponding readjustment will occur raising the crust to approximately 5/6th of its original height. If, however, the crust was not evenly eroded, for example by glaciers which deeply incised into the landscape, but the same amount of material was removed, as was the case with even erosion, then the rebound would help produce higher mountain peaks. This process may have helped produce the high peaks in the Himalayas. While this process may provide an alternative mechanism for uplift, the geological data suggest, however, that much of the Himalayas obtained their height before the onset of major glaciation. Unfortunately, however, there are no reliable dates for the oldest glaciations in the Himalayas and Tibet, and the extent of glaciation is poorly constrained. It is difficult therefore to evaluate the connection between glaciation and uplift.

To conclude, it is not possible to unequivocally attribute late Cenozoic climate change to uplift of the Himalayas and Tibetan plateau. Researchers are only beginning to try to understand the complex processes and feedback mechanisms that drove the Late Cenozoic climate change. More proxy data for climate change in the Himalayas and Tibetan regions, and elsewhere in the world, are needed to substantiate the timing and nature of tectonic and climatic changes.

Lewis A. Owen

Bibliography

Raymo, M. E. and and Ruddiman, W. F. (1992) Tectonic forcing of late Cenozoic climate. Nature, 359, 117–22.
Ruddiman, W. F. and and Kutzbach, J. E. (1991) Plateau uplift and climatic change. Scientific American, 264, (3) 42–51.
Ruddiman, W. F. (1997) Tectonic uplift and climate change. Plenum Press, New York.