NOC-MSM / pyBDY

pyBDY: a Python based regional NEMO model configuration toolbox.
GNU General Public License v3.0
7 stars 7 forks source link

Add option to keep the vertical grid of the source data #158

Open rdPatmore opened 4 months ago

rdPatmore commented 4 months ago

There is currently no option to retain the vertical grid of the original source data. It would be usefull to have this as a namelist option.

jdha commented 4 months ago

Hi Ryan - I've been using this option for many years - it was introduced when I wrote the interpolation-on-the-fly in the core NEMO code ... so there is an option somewhere - but it maybe a little opaque! I've avoided pyBDY for too long now. I'll clear the schedule tomorrow and focus on getting pyBDY tasks in order. Apologies.

rdPatmore commented 4 months ago

Hmm I couldn't see anything obvious. The only thing I found in the code is the isslab flag. If you could point me to the option, that would be great.

The motivation for using this option is that I'm finding strange results in my forcing whereby there is a jump in Temp and Salinity at the bounday (see image). I have AMM15's sigma grid as the destination. I am using pyBDY's zps option but feeding it gdept_0 as the depths, so I thought that would be fine (the sco option does not work as the variables required appear to be relevant to a different version of NEMO). My solution was to avoid vertical interpolation and let NEMO do it on the fly.

Screenshot 2024-04-10 at 10 00 34

jdha commented 4 months ago

Yeah - I wrote the bdy-on-the-fly in NEMO 6-7 years ago and haven't used vert interp in pyBDY since then ... will look tomorrow ... in the meantime I guess you could feed the sn_src_zgr to sn_dst_zgr .... not too sure whether the code will broadcast 1d->3d ok in this instance though