Ice sheet grow too fast on the boundary

Schofield

Hi, everyone,
I am using ISSM to calculate Antarctica ice sheet thickness from a bare base. The problem is there are some point on the boundary grow faster than the surrounding points, and the velocity calculating blows up, because these point will grow very high latter.
The thickness before blowing up and base is shown here.

I use HO flow equation.

mathieumorlighem

Hi,
it is impossible to tell what's happening without more details and screenshots. Your thickness looks ok. Do you model the whole ice sheet or just a basin? You should definitely check your boundary conditions if this is were the problem happens.
Mathieu

Schofield

Thanks for reply!
I model the whole ice sheet with a .exp file depict a domain which is a little small than the whole land. The base is set close to zeros on the boundary. The topg&temp is shown here. It is paleo condition of Antarctica.
https://raw.githubusercontent.com/Mike000-ai/Schofield/master/Antarctica_ISSM/fig/base&temp.png

the boundary condition I set is

% Thermal
md.thermal.spctemperature(uppervertex) = InterpFromGridToMesh(x1,y1,temp,md.mesh.x2d,md.mesh.y2d,0);
md.thermal.spctemperature(othervertex) = NaNones(8md.mesh.numberofvertices2d,1);

% Stress balance
md.friction.coefficient=10ones(md.mesh.numberofvertices,1);
md.friction.p=ones(md.mesh.numberofelements,1);
md.friction.q=ones(md.mesh.numberofelements,1);
md.stressbalance.spcvx=NaNones(md.mesh.numberofvertices,1);
md.stressbalance.spcvy=NaNones(md.mesh.numberofvertices,1);
md.stressbalance.spcvz=NaNones(md.mesh.numberofvertices,1);
md.stressbalance.spcvx=NaN*ones(md.mesh.numberofvertices,1);

initial conditions is

%Thermal
md.initialization.temperature=InterpFromGridToMesh(x1,y1,temp,md.mesh.x,md.mesh.y,0);

% Stress balance
md.initialization.vx=zeros(md.mesh.numberofvertices,1);
md.initialization.vy=zeros(md.mesh.numberofvertices,1);
md.initialization.vz=zeros(md.mesh.numberofvertices,1);
md.initialization.vel=zeros(md.mesh.numberofvertices,1);

Schofield

I used pdd method to calculate mass balance

mathieumorlighem

Hi,
Send me your model by email when you get a chance (save a .mat file with the model after running a transient solution, and send me a Dropbox/GoogleDrive link mmorligh@uci.edu) so that I can try to figure out what's going on.
Best
Mathieu