Abstract
The Mohr-Coulomb criterion for failure, modified in light of the concept of effective stress, is where τc is the critical shear stress at failure, S the normal pressure, and p the pore pressure across the plane of internal slippage at failure, ϕi the internal friction angle, and τ0 an empirical constant, commonly referred to as the cohesive strength. Experiments showed that the τ0 for sedimentary rocks is about 200 bars.
Hubbert and Rubey (1959) assumed that once a fracture is started, TO is eliminated and further movement results when They proceed, however, to use this formula for the frictional sliding of cohesionless block as the criterion of failure of large thrusts, after they assumed that τ0 could be eliminated through a concentration of stress. This assumption led to their conclusions that very long overthrust blocks are possible and that such blocks may have moved by gravitational sliding along very gentle slopes.
I present arguments to show that their assumption of zero τ0 was based upon a faulty argument and to point out that the τ0 term should not be omitted unless it could be proved the moving block slid along an already existing fracture plane.
The first part of this paper consists mainly of conclusions based on computations. Clearly, an unjustified omission of a 200-bar cohesive strength would lead to erroneous and misleading results; particularly, gravitational sliding cannot be an important mechanism if such a cohesive strength has not been eliminated during overthrust faulting.
The second part presents evidence to distinguish between movements of cohesively bound blocks and cohesionless blocks. The Glarus overthrust, characterized by presence of a ductilely deformed limestone layer within the thrust zone, is considered a typical example of thrusting of cohesively bound blocks. The Heart Mountain thrust, characterized by a shattering of the “upper plate” and absence of a weak layer above the thrust contact, is interpreted as an example of thrusting of cohesionless blocks. The former is compared to slowly creeping slides moving at rates of centimeters or less per year, and the latter with catastrophic landslides (such as the Flims, Goldau, and Vaiont slides) moving at speeds of many meters per second.
Third, the conclusion of Raleigh and Griggs (1963) that large thrusts can only form when a toe of the thrust is continually eroded is also traced to the assumption of zero cohesive strength along thrust plane. Otherwise, the toe effect would produce a zone of imbrication at the front of overthrust blocks, particularly those sliding downslope under their own weight.
