NEMO-surge AMM7: Generate model harmonic tide prediction datasets

[ukmo-ansaulter]

Generating a surge residual by subtracting a harmonic tide prediction based on a long run of the surge model significantly reduces systematic biases in the modelled surge.

This enhancement generates a database of mode harmonic predictions information that can be used to derive the surge residual from the NEMO-surge AMM7 model.

[jpolton]

I shared AMM15 and AMM7 boundary files, generated from FES2014, and passed them to Clare O'N on 8th Feb

Jeff said:

Following up on some of my homework, here are the tidal constituents from FES2014:

For AMM15:
https://gws-access.jasmin.ac.uk/public/jmmp/AMM15/FORCING/TIDES/FES2014/
These are actually 2D fields that are mapped, and rotated, onto the AMM15 grid.
This requires a namelist edit for the 2D thing to work:

&nambdy_tide
filtide='./FORCING/TIDES/FES2014/bdry_tide_34csts_AMM15'
ln_bdytide_2ddta=.true.,
ln_bdytide_conj=.false.,

I don’t think they need a coordinates.bdy.nc file. If they do I will sort you out with one, quick as a flash. (FYI Sarah made these).

For AMM7_surge we are using the “traditional” boundary vectors.
I just created you a fresh batch with coordinates.bdy.nc file (38 constituents) and tested M2 only aginst other known datasets.
I tar’red and gzipped it separately:
tar -cvf FES2014_amm7_surge.tar outputs/
then
gzip FES2014_amm7_surge.tar
Attached, with the corresponding coordinates.bdy.nc file