Attribution: You must attribute the work in the manner specified by the author or licensor ( but no in any way that suggests that they endorse you or your use of the work).Noncommercial ‒ you may not use this work for commercial purpose.No Derivative works ‒ You may not alter, transform, or build upon this work.Sharing ‒ Individual scientists are hereby granted permission, without fees or further requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in other subsequent works and to make unlimited photo copies of items in this journal for noncommercial use in classrooms to further education and science.

The recent publication, “Present-day kinematics at the India-Asia collision zone” (Meade, 2007), attempts to address an ongoing controversy about the dynamics of the Tibetan Plateau by finding block models for the region that correctly reproduce the kinematics from many global positioning system (GPS) observations. A closely related paper (Thatcher, 2006) undertakes a similar exercise, using a subset of velocity observations and different assumptions about the elastic energy cycle in the uppermost crust. The conclusions of both authors note that these block models allow slow slip rates on some large Tibetan faults. They argue, therefore, that a regional velocity field dominated by the interactions among nearly rigid crustal blocks cannot be ruled out for Tibetan tectonics, even though these slow slip rates have previously been used to argue for continuous deformation of the plateau.

In fact, a comparison of both papers hints at an opposite possibility: that the bulk mean rheology of Tibetan lithosphere behaves viscously, and deformation of a highly viscous material is best able to reproduce Tibetan kinematics at large length scales. Such a continuous viscous material is simply a body with a continuously varying strain function (no discontinuities), which may or may not contain lateral strength variations; this treatment is appropriate for Tibet only if one considers the integrated anomalously thick crust (60–70 km), rather than just the seismogenic layer, which indisputably hosts discontinuous strain in the presence of faults.

Figure 1 shows an approximate superposition of the block models and velocity fields from Meade (2007) and Thatcher (2006). Two important points can be made by this figure: 1) the best agreement occurs at the Tarim basin, which has greater mechanical rigidity than the main plateau, and 2) the worst agreement occurs where the largest number of GPS observations are available (on the eastern plateau boundary). Indeed, Shen et al. (2005) provide yet another, even more complicated, block geometry for the eastern plateau.

In Meade's (2007) and in Thatcher's (2006) studies, the uppermost crust is certainly brittle, and deforms elastically during interseismic intervals. What matters for continental dynamics at the largest scales is whether the interaction between these brittle blocks or viscous flow at greater depth drives the surface velocity and deformation fields. In block models and in fluids with non-zero viscosity, point velocities are correlated with neighboring velocities relative to their distance. In a fluid, the correlation is a function of distance determined by the viscosity and the thickness of the elastic lid. Therefore, a kinematic model of rigid crustal blocks, where the block size is comparable to this correlation distance, and a continuous viscous model with the appropriate viscosity will both yield a good fit to discrete velocity data. For example, Flesch et al. (2005) provide a continuous kinematic model that fits the observed GPS velocities as well as the data of Meade (2007) and Thatcher (2006). In a block approximation of a continuous-velocity field, only the scale of blocks and not their exact geometry determines the goodness of fit. The block boundaries are non-unique (McKenzie and Jackson, 1986). The observation that two different configurations of blocks, both with comparable scaling, do good jobs of reproducing observed velocities in eastern Tibet (Fig. 1) suggests that the velocity field may be dominated by a highly viscous material, rather than the interaction of unique blocks. Because block models with small blocks approach a continuous approximation, and continuum models with very high bulk viscosity approach a block approximation, new and better velocity observations may not suffice to resolve this question.

No one disputes that the uppermost crust is rigid and brittle. Understanding the temporal and spatial distribution of strain at the scale of individual crustal structures mandates a rigid block formulation. The question is whether the strength of this uppermost crust dominates the total regional dynamics, or whether the rheology of the bulk lithosphere is more important.

Both Brendan Meade and Wayne Thatcher graciously provided figure materials and constructive comments on this manuscript. Roland Burgmann and an anonymous reviewer also suggested several critical revisions.