Surface diffusion

ASPECT
1

Hi all,

I am using ASPECT 2.3.0 to simulate crustal deformation in 3D in presence of compression and extension, and for some of my simulations I would like to introduce erosion. As pointed out in the manual, erosion is reproduced by diffusing the surface, and the diffusion rate depends on a hillslope transport coefficient k that has a default value of 10e-6.

I have already run my models by adopting a free surface, and everything works as expected. However, when I switch from free surface to diffusion, I always obtain a flat model surface, regardless of the adopted value for k (I tested values ranging from 10e-4 to 10e-30). In addition, when I implement time dependent, compression-to-extension velocity boundary conditions, the solver does not converge anymore when the change of velocity at the boundaries occurs.

Can anyone tell me a little bit more about how the surface diffusion works, and why I do not see any relief in the model surface?

Thank you very much.
Alessio

2

Hi Alessio,

Thank you for posting the question on the forum and happy holidays!

The short answer is that you will need to specify that the top boundary is both a free surface, which is subject to diffusion.

Here is an example of how to specify this in the PRM file:

subsection Mesh deformation
  set Mesh deformation boundary indicators = top: free surface, top: diffusion
  subsection Free surface
    set Surface velocity projection = normal
  end
  subsection Diffusion
     # Diffusivity term. Increasing this value will result
     # in a smoother free surface and lower topography
     # amplitudes.
     set Hillslope transport coefficient = 1.e-6
  end
end

I hope this is helpful and please let us know if you have any additional questions!
John

3

Dear John,
thank you very much for your quick reply. I modified the .prm file according to your suggestion, and everything seems to be working fine now. I will update the topic when I get the confirmation that the code also converge. Happy holidays to you!
Alessio