mathieumorlighem
There is something I do not fully understand.
If I run a steady-state thermal simulation before any transient simulation, the steady-state simulation does take into account the initial temperature values when it calculates the results.
This brings up my following question: in this case where I was running a steady-state thermal simulation in between two transient runs, how does the code "decide" not to take into account the initial temperature values (when for the previous case mentioned it does)?
Moreover, if you take a look at the code I provided before, although I delete the results from the previous transient run and then overwrite md.initialization.temperature with the new initial temperature value I want for the next transient run, the code does still somehow use the temperature values/results from the last step of the previous transient run when it computes the steady-state thermal simulation results. How does it do that? Does it take these values internally from the c-code?
I want to assign new temperature values to a model in between two transient runs. For instance, for the first transient simulation (0-10yrs), I want a 150K surface temperature value. For the second transient simulation (10-20yrs), I want a 170K surface temperature value. What I am doing - and I might be wrong - is to run a steady-state thermal simulation before the transient simulation in order to have proper temperature values all over the glacier, let's say 150K at the top and 165K-ish at the bottom instead of just 150K everywhere. I am able to do this before the first transient run, but when I try to do this again before the second transient run - or in between both runs - the steady-state thermal simulation does not take into account the new initial temperature values I set and somehow uses the results from the last step of the first run after I cleared and overwrote them out.
I would like to understand, if possible, how the code works and what determines it to take or not into account initial values depending on the situation. Moreover, I would appreciate if you could tell me which is the best way code-wise to tackle this situation.
Note: md.thermal.spctemperature is set to NaN for all the vertices of the mesh.
Thanks a lot for your patience and time.