Feature request: Experiment setup for TIEGCM

[hkershaw-brown]

Use case

What are you trying to accomplish? Sometimes there may already be a way to do what you need.

Set up experiments for the NASA-TIEGCM team: https://ccmc.gsfc.nasa.gov/ PMO, filter for a given date range.

There is https://github.com/NCAR/DART/issues/469 hanging out which should be pulled into the experiment setup.

Is your feature request related to a problem?

We don't really have out-of-the box setups that I can point people to when setting up a cycling experiment for a large model. Particularly for PMO. There is https://github.com/NCAR/DART/tree/main/observations/obs_converters/even_sphere which is for synthetic radiosondes

Describe your preferred solution

Python to set up and run the experiment:

• setup experiment directory
• setup assimilation time windows
• synthetic observation sequence files
  • create batch scripts for the model, and for PMO, filter
  • submit the jobs ( step through each assimilation cycle)

Describe any alternatives you have considered

It would nice to not have the setup scripts be model specific. For now, focusing on the TIEGCM-NASA deliverable.

[hkershaw-brown]

./obs_seq_to_netcdf
loc = ncread('obs_epoch_001.nc', location);
scatter(loc(1,:), loc(2,:), loc(3,:))

GPS profile locations:

[mgharamti]

These look like the olympic rings to me. I don't know how to interpret them. @hkershaw-brown is there any issue with 'obs_epoch_001.nc' or is this the behavior you expect?

[hkershaw-brown]

@mgharamti no issue I just did a 2d scatter plot instead of 3d and you get a ripple which entertained me. The correct plots are in there too.

[hkershaw-brown]

This is up and running. There is no perturbing.

Met with CCMC and CU Boulder people. Just getting the original TIEGCM-DART shell scripts for cycling filter as a deliverable, so not putting these python setup scripts into DART and closing this issue for now.

https://github.com/NCAR/DART/tree/nasa-cycle Archived here also https://github.com/hkershaw-brown/DART/tree/nasa-cycle

May resurrect some of this for CROCODILE.