How "plastic" do you think heated rocks become?

[PinkGeologist]

Consider a magma chamber in the Earth's crust ... models show that 1-2 km surrounding this chamber are heated above the brittle-ductile transition zone.

Let's say that you know the yield strength of the crust in general (not heated) is 15 MPa ... and after that, rocks under heavy confining pressure behave plastically anyway. Let's say they behave as an ideal plastic.

If you want to create a plastic model of the area, how would you define the extra plasticity of that hot zone? A lower yield strength? How much lower? Is there a function of temperature you know of as a standard?

 

[Dr. Courtney]

How do you even know that plastic is the right model (rather than visco-elastic)?

How many of the rock materials have been tested on an Instron (or similar) to map out their stres-strain relationships over the range of temperatures?

Without data, a lot of these models are little more than guesses.

 

[PinkGeologist]

Many rocks of the composition I'm interested in have undergone stress-strain testing in laboratory conditions that show plastic behaviors under the high confining pressure we see in the Earth's crust ... and for confining pressures up to nearly 100 MPa the behavior is nearly ideally plastic ... after that, there is some significant hardening.

I am just interested in how I'd treat the heated rock ... would the only way to get even a hand-waving handle on the parameters be to have empirical lab test results? Well, I know that is a loaded question ... but I guess I just want to know if there are reasonable methods to relate the plasticity and temperature in the material.

 

[Dr. Courtney]

I am just interested in how I'd treat the heated rock ... would the only way to get even a hand-waving handle on the parameters be to have empirical lab test results? Well, I know that is a loaded question ... but I guess I just want to know if there are reasonable methods to relate the plasticity and temperature in the material.

I've seen too much bogus modeling claiming to be "reasonable" without anything like real experimental validation.

Lab testing under temperature would be one way to gain confidence in the parameter behaviors.

Another approach might be to conduct modeling beforehand with some guesses for the parameters and then conduct some in situ testing to see if the guesses about parameters yield reasonable results.

This approach runs the risk of having too many adjustable parameters for the experiments to be a valid test of the modelling efforts.