 |
 |
I’m afraid I must take issue with some of John Platt’s views. There is no way a narrow fault (as little as a centimetre) that is accommodated by high temperature
plastic flow in the granulite facies can be regarded as brittle, yet such things are quite common. The mode of failure is localized (commonly called a plastic shear zone though still a fault) but the mechanism is plastic. Also , cataclastic flow in nature
can be by distributed mesofracture, such as accommodates large-scale folding of limestones in central Italy. Multi-cm cracks allow small displacements and dilatancy that allows long-wavelength bending to be accommodated. The mechanism is brittle fracture but
the mode of failure is a ductile flow. We have not (yet) been able to reproduce this in lab experiments. On the other hand, cataclastic flow in experiments is seen in porous rocks, where pore collapse permits local hardening that causes the deformation to
be distributed. Exactly the same kind of flow occurs when a reservoir collapse as a result of pore fluid withdrawal, so cataclastic flow is definitely not restricted to lab experiments.
So death to the BDT in the way it is sometimes used, and I don’t think the alternatives can be pulled apart as easily as John believes!
Ernie Rutter
From: Tectonics & structural geology discussion list [mailto:[log in to unmask]]
On Behalf Of John Platt
Sent: 07 September 2011 15:31
To: [log in to unmask]
Subject: Re: The ficticious brittle/ductile transition
As someone who has been guilty of using the term brittle-ductile transition (BDT) - ugh - I would like to leap to its defence as a useful practical term for conveying the concept of a level in the crust above which deformation occurs mainly
by discontinuous brittle faulting, and below which deformation takes place by a variety of ductile processes without loss of continuity at the scale of observation. And that's the key - the scale of observation. Consider this: flow of a viscous fluid
(air, say) can be described at the microphysical level entirely by the elastic and frictional interactions among molecules: they bounce off each other and swap neighbours on picosecond timescales. The deformation is pressure-dependent too: the viscosity
scales linearly with pressure. And at the other extreme, since the insightful work of David Elliott, Bill Chapple, and others, we have learnt to treat large-scale tectonic features such as thrust belts as macroscopically ductile structures - we use the formalism
of continuum mechanics, and a constitutive relationship (Coulomb plasticity) to describe their bulk deformation. This is valid at a scale of observation larger than the individual faults, just as crystal plasticity is valid at scales larger than the discontinuities
(dislocations). The concept of the BDT depends on the scale of observation, which is the human one: if we can see and measure a fault displacement, we call it brittle.
The issue of cataclastic flow as seen at the grain-scale in rock-mechanics experiments, is, in my opinion, a red herring. In nature, cataclastic flow is almost entirely limited to narrow zones directly associated with large displacement
brittle faults. There is no evidence in nature of a transition from dicontinuous faulting to continuous cataclastic flow with depth: major faults rupture right through the BDT into the underlying plastic zone.
So long live the BDT as a useful, if slightly fuzzy, way of conveying a simple concept. If we want to pick nits, we can pull apart all the alternatives just as easily.
John Platt.
--
Professor J.P. Platt
Department of Earth Sciences
University of Southern California
3651 Trousdale Parkway
Zumberge Hall 315
Los Angeles, CA 90089-0740
email: [log in to unmask]
phone: +1-213-821-1194