The approximate simultaneity of environmental changes within a few million years of 8 Ma in and near Tibet with tectonic events in the same area has been interpreted to suggest that the Tibetan Plateau grew rapidly at or just before ~8 Ma, and that the rise of the plateau affected regional climate changes, including a strengthening of the Indian monsoon (e.g., Harrison et al. 1992, 1995; Molnar et al. 1993). Although some of us have dedicated much of our research effort to addressing both the mechanics of Tibet’s apparently rapid growth and the impact of its growth on regional climates, it might be appropriate to ask whether the facts motivating this effort merit consideration of such interconnected geodynamic and meteorological processes. In fact, new evidence gathered in the past 10 years has both added and removed support for such simultaneity and for an environmental response to a tectonic change. Hence, a review of such data is timely.
Will Downs and I became acquainted and then friends as I became interested in how tectonics, erosion, and climate change interact, and our interaction culminated in a collaborative paper that related results of erosion to climate change rather than tectonics (Zhang et al. 2001). Thus, a volume dedicated to him seems to be an appropriate place to discuss other data that bear on these interactions.
My main goal is to ask two specific questions. First, to what extent do paleoenvironmental data imply widespread environmental change in the Tibetan region near 8 Ma? Second, do tectonic and other observations imply a rapid change in Tibet’s growth at ~8 Ma? Before reviewing the relevant observations, however, let me summarize the basic scientific hypotheses derived from observations that motivate a review and depend on them.
If crust shortens horizontally and thickens, so that isostasy supports a high mountain belt or plateau, the underlying mantle lithosphere must also shorten. Suppose that shortening occurs by pure shear (at least approximately), and not by “intracontinental subduction,” the detachment of mantle lithosphere from the overlying crust (e.g., Mattauer 1986; Wellman 1979). Then, the shortening and thickening of cold, dense mantle lithosphere adds excess mass to the lithospheric column, with the likely result that its isostatic compensation supports a lower mountain belt or plateau (by ~1000 m) than if only crust thickened (England and Houseman 1989). Thickening of mantle lithosphere also requires downward advection of isotherms, so that cold material beneath the region of thickening is juxtaposed next to hotter material, a thermal state that facilitates convective instability, which ultimately can remove part, if not all, of the thickened mantle lithosphere (e.g., Houseman et al. 1981). Many studies of such instability suggest that at least the lower half of thickened mantle lithosphere should be removed in less than 10-20 Myr after it has been thickened roughly two times (e.g., Conrad and Molnar 1999; Houseman and Molnar 1997; Houseman et al. 1981; Molnar and Houseman 2004; Molnar and Jones 2004), although some (e.g., Buck and Toksöz 1983) disagree. Removal of the thickened mantle lithosphere would lighten the load on the base of the lithosphere beneath a mountain belt or high plateau, and that already-high terrain should rise somewhat higher (~1000-2000 m; England and Houseman 1989). This process not only does not, but cannot imply that the entire ~5000 m mean elevation of the plateau would occur as a result of removal of mantle lithosphere; most of the high terrain must be supported by thick crust, as Airy (1855) envisaged 150 years ago. Moreover, whether Tibet has grown upward en masse or outward over time remains another open question.
Steady winds blowing northeasterly along the coast of east Africa and Arabia toward India characterize the Indian monsoon. Warm rising air over northern India and Tibet, southward flow in the upper troposphere, and subsiding air south of the equator over the Indian Ocean complete return flow that feeds the monsoonal winds. In summer, the hottest part of earth’s upper troposphere lies above Tibet (e.g., Li and Yanai 1996; Webster et al. 1998). A high plateau contributes to this circulation because of the transparency of the atmosphere to both solar energy and infrared radiation. The air over the plateau warms and becomes warmer than air at the same pressure (and hence same altitude) over India, but the transparency of the atmosphere to infrared radiation prevents it from warming so much as to create thermal runaway (e.g., Molnar and Emanuel 1999). Most calculations of global climate using general circulation models of the atmosphere associate a higher Tibetan Plateau with a stronger Indian monsoon (e.g., Hahn and Manabe 1975; Kutzbach et al. 1989, 1993), and those that consider many possible mean heights show a monotonic increase in monsoon strength with height (e.g., Abe et al. 2003; Kitoh 2004).
For various reasons, calculations using general circulation models of the atmosphere do not demonstrate that a small change in height (of only 1000 m) should strengthen the monsoon by a large amount. Hence, perhaps more important than these studies of a realistic atmosphere are theoretical arguments for a threshold in the north-south temperature gradient necessary for meridional (monsoon-like) circulation (e.g., Emanuel 1995; Plumb and Hou 1992). These arguments allow for the possibility that a small change in Tibet’s height might be sufficient for the meridional temperature gradient to exceed the threshold needed for meridional circulation (e.g., Molnar et al. 1993).