Ed Doddridge
Ed Doddridge
Ok, that's good to know, and I think it further justifies spending our efforts on implementing an external solver.
They are enforced. I guess the point of this would be to check that they've been done correctly.
I've been working on this in the background, and have found that in n+1/2 layer mode the model receives 1% more momentum than I expect it to. I've gone through...
I agree - I don't think this is a numerical instability. For larger forcings the momentum increases more than it should, but also sporadically decreases back towards the expected value....
One thing I've noticed while doing this is that the convergence criterion used for the linear solve makes quite a difference to the solution. This may be what caused the...
That's a good thought. We can definitely use sponges to mop up edge effects.
I have some code on the `pysics-test` branch that I think closes #99 and provides more detail for #100. But, because of my earlier, rather cavalier, approach to merging the...
I don't know of any compiler switches that would let us do this, but we might be able to make it happen through the `parameter.in` file. See [here](http://fortranwiki.org/fortran/show/Real+precision) for a...
This is a good idea. I think we can use some plots of the reproductions for this purpose. The simulations will likely take too long to spin up from rest...
As to what the graphic should be, I think a 3D plot in which the layer interfaces are plotted, and perhaps coloured according to some other variable, would be visually...