The source of ASPECT sediment？

ASPECT
1

Hello everyone,

I am currently using Fastscape to simulate the surface processes of ASPECT, aiming to understand the erosion-deposition processes in orogenic belts. I have a question regarding the model results.

The thickness of the sediment on the surface in ASPECT does not match the thickness of the total_erosion generated by Fastscape. The sediment in ASPECT is noticeably thicker and continuously increasing, which clearly violates the law of conservation of mass.

To figure out if the issue was with the Boundary composition model, I turned off sediment_1&2. However, ASPECT no longer had any sediment-related items, meaning the erosion produced by Fastscape was not reflected in ASPECT.

subsection Boundary composition model
  set Model name = function
  set Fixed composition boundary indicators = top, bottom
  set Allow fixed composition on outflow boundaries = true
  
  subsection Function
    set Coordinate system   = cartesian
    set Variable names      = x,y,t
    set Function constants  = tsed=2e6 # duration of sedimentation switch time intervall 
    set Function expression           = if(y>580e3, t/1e6, 0); \
                                        0; \
					0; \
                                        0; \
                                        if(y>580e3 && sign( sin( t/tsed * pi) ) == 1,1,0); \
                                        if(y>580e3 && sign( sin( t/tsed * pi) ) == -1,1,0); \
                                        if(y>580e3 && t==0, 1,0); \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0  
  end
end

After:

subsection Boundary composition model
  set Model name = function
  set Fixed composition boundary indicators = top, bottom
  set Allow fixed composition on outflow boundaries = true
  
  subsection Function
    set Coordinate system   = cartesian
    set Variable names      = x,y,t
    set Function constants  = tsed=2e6 # duration of sedimentation switch time intervall 
    set Function expression           = 0; \
                                        0; \
			0; \
                                        0; \
                                        0; \
                                        0; \
                                        if(y>580e3, 1,0); \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0; \
                                        0  
  end
end

I would like to inquire about the data transfer between Fastscape and ASPECT software. I understand that ASPECT provides conditions such as velocity and elevation to Fastscape. Apart from returning the deformation of the ASPECT grid, will Fastscape also return the thickness of the eroded material and how can I obtain the correct deposition model?

this is my current .prm file:

0316test1.prm (33.4 KB)

Thank you very much for taking your time to read this and for your valuable guidance.:blush:

Bests，

Yushun

2

Hi Yushun,

Thanks for posting this on the forum! I’ll admit I’m still relatively new to using ASPECT/FastScape so @jbnaliboff and @Djneu may have more to add but my understanding of the way that ASPECT and FastScape interface is that the exact values of deposited sediment determined in fastscape are not sent to ASPECT, but instead the change in the topography that FastScape calculates is what is communicated. This change in topography is converted to a velocity within ASPECT, that displaces the mesh, and with the right composition boundary conditions this leads to “deposition” on outflow boundaries (e.g. when the ASPECT topography increases, due to deposition). The outline of the steps is as follows:

  1. The Stokes equations are solved within ASPECT’s domain, and the velocities on the surface of ASPECT are sent to FastScaspe
  2. FastScape uses the velocity from ASPECT to move the topography, and determines the change in topography due to surface processes
  3. FastScape sends to ASPECT what the new topography is
  4. The mesh within ASPECT is advected to match the topography calculated by FastScape
  5. If the ASPECT mesh is deformed positively (i.e. in the opposite direction of gravity), this is “outflow” and compositions will be added to the ASPECT mesh there based on the boundary composition model and if set Allow fixed composition on outflow boundaries = true.

In theory, this should capture the sedimentation predicted by FastScape because the ASPECT mesh will move positively in areas where FastScape calculates deposition. But in practice, because the cell size within ASPECT is likely quite a bit larger than the amount of sediment being deposited, and because of numerical diffusion, the amount added to ASPECT will not match the amount calculated in FastScape.

I hope that this helps, and it’s actually amazing timing that you posted this because I’ve been diving into the way that ASPECT and FastScape communicate composition information over the last couple of weeks myself. Don’t hesitate to ask more questions!

Cheers,

Daniel

3

Hi Daniel，

Thanks for your professional and detailed response! This has clarified my confusion and is very helpful to me.I will continue to study the FastScape code to understand the specific transfer function. I will contact you if I have any findings or questions. Thank you again for your prompt reply.

Best，

Yushun