Strange proViscosity leap under lithostatic condition

[longgangfan]

test.py.txt From step 0 to step 1 (100000 year), the proViscosity of the crust get a big change, it is quite hard to understand.

I run the model in the UWgeodynamics downloaded from dockehub lastest (2.10.5).

Any suggestion? Thank you.

[PatriceFRey]

Yep it does. A time step 0 the boundary conditions are not yet applied. So the viscosity is simply function of the temperature. At time step 1 the boundary conditions are applied, and the viscosity, which probably depends on strain rate in you model, increases.

[longgangfan]

Yep it does. A time step 0 the boundary conditions are not yet applied. So the viscosity is simply function of the temperature. At time step 1 the boundary conditions are applied, and the viscosity, which probably depends on strain rate in you model, increases.

My boundary conditions are as below:

Temperature

Model.set_temperatureBCs(top=293.15 u.degK, bottom=893.15 u.degK,materials=[(stickyair, 293.15 * u.degK)])

velocity

Model.set_velocityBCs(left=[0, None], right=[0, None], top=[None, None], bottom=[None, 0])

1e27 seems to be too large?

[PatriceFRey]

Try to add a bit of velocity (2, 3 cm both sides) and see the result on the viscosity.

[longgangfan]

Try to add a bit of velocity (2, 3 cm both sides) and see the result on the viscosity.

I think it will be OK, if I add a bit of velocity. But why the proViscosity is so large, when velocity = 0?

[PatriceFRey]

viscosity = (deviatoric stress) / (strain_rate)

[longgangfan]

viscosity = (deviatoric stress) / (strain_rate)

Thank you. Forgive my stupidity. But what make the viscosity larger than geological common-sense in my boundary conditions? There is no situation where the velocity = 0 on earth? or these is no such low strain_rate?

[longgangfan]

Thank you, I think I know the point. I will close this issue.