Closed jagoosw closed 8 months ago
@jagoosw my plate is a bit full now... But could try to look at this in a week or so..
looks ok, I don't know how deep I can review this. is there a test that tests this? or a validation that now works and before it wasn't?
Thank you for reviewing!
The reason I made these changes is that I'm using a sediment model (not BGC but this was the easiest interface for it) that depended on the surface radiation which I have as a time-dependent function, and I also realised it was more generally consistent with other tendency calculations that the time be included.
I then moved the kernels to be functions because (I think) you can't externally define new methods for @kernel
s. If a user wanted to implement their own sediment model I think this would be essential.
I've come across an instance where I'm trying to write a different sediment model and need the forcings to have a time dependence so need to pass the time to
_calculate_tendencies!
, and modifies sediment tendency kernel to make new method definition possible.