MOM5
MOM5 copied to clipboard
mom-ocean
Problems identified with bathymetry file used by several experiments
The bathymetry file used by several experiments may contain anomalous 'pits' on the shelves. The extent of this problem is not know yet.
Some more details about the history. The original dataset was based on that provided to Griffies by David Webb from the UK's OCCAM project. This dataset merged a few products to produce a 1/10th degree global bathymetry product. It was the only global product of its time 20 years ago, and widely used by MOM-type modellers interested in global model building with anything related to a fine grid. Unfortunately, it appears that the quality control for that OCCAM dataset was poor. As Russ Fiedler conjectures, there may have been something wrong with the indexing or the like. The main point of this history is to note that there seems to be nothing wrong with the regridding tools used to map onto the MOM025 grid. Instead, it is a problem with the starting OCCAM data.
The problem is present in many of the GFDL models, such as CM2.5 and CM2.6. It is also a problem in the ACCESS 1 degree model. There may be others. Work will be needed to resolve this issue.
Hi, this is Bin Zhao from GMAO. I'd like to point out another issue with this CM2.5 grid, not directly related to the bathymetry, but maybe worthy of discussion. Immediately to the north of the Hudson bay, there is a large area of open ocean filled by land mask. Please see the attached ppt file. Is this purposely done so?
Update:
I now understand that the masked region is related to the model CFL and time step. Please ignore this issue.
There's an attempt to make improved bathymetry underway - see https://github.com/COSIMA/access-om2/issues/158#issuecomment-669624837
Can we close this issue? - this has been fixed in /g/data/ik11/inputs/access-om2/input_20201022/ - see https://github.com/COSIMA/access-om2/issues/158#issuecomment-698758041