oceanhackweek / ohw22-proj-gcmfilters

Mapping eddy flow structures using GCM-Filters

ohw22-proj-gcmfilters

Mapping eddy flow structures using GCM-Filters

Important links:

• Pangeo Data Catalog
• GCM-filters documentation

Below we have our plan for the week!

Goals for Tuesday:

• Make sure everyone can open and is comfortable with jupyterhub - ok
• Run some examples of GCM-Filters on the Hub
• Decide what data set to use from the Pangeo Data Catalog - ok (NCAR CESM Global High Resolution Ocean Model Fields)

Goals for Wednesday:

• Run the example "Filtering on CPUs vs GPUs" from GCM-Filters;
• Explore the data set, examine regions of interest and try to apply the GCM-Filters using the examples;
• Update the code for using Laplacians to filter vector fields in the NCAR CESM POP model.

Instructions for running a GCM-filters environment

• Make sure you have pulled (git pull) the updated version of this repo so that you have the file gcm-filters_env.yaml in your JupyterHub. You can take a look at the .yaml file and see the packages listed that we will have access to within our soon-to-be created environment.
• Open a Terminal on your Hub.
• In terminal, type:

conda env create -f gcm-filters_env.yaml

This creates a new environment from the file (hence the -f) named gcm-filters_env.yaml.

• You will have to wait several minutes while the environment is created.
• At the end, you should see something like the following:

Preparing transaction: done
Verifying transaction: done
Executing transaction: done
#
# To activate this environment, use
#
#     $ conda activate gcm-filters-env
#
# To deactivate an active environment, use
#
#     $ conda deactivate

You do not need to do anything else at the command line if you get to this stage. (That is, you do not need to activate the environment, though are welcome to if you'd like.)

• Optional: In case we are missing a package, we can add it to the gcm-filters_env.yaml file later and then type

conda env update --file gcm-filters_env.yaml --prune

• The last step is to start a new notebook with the new environment we just created! To do this, you will need to open a new landing page (click the "+" in the upper left of the Hub). You should see an option to open a notebook in the top row with the new environment called "gcm-filters-env". NOTE: it may take a couple minutes for the environment to appear.

ohw22-proj-gcmfilters's Issues

Plot figures with cartopy

Just a reminder to myself that I would like to try using cartopy to make cool figures (if there is time) 😄

How to convert the filter_scale to the geographic distance

Hi, I see your beautiful work about filtering the eddy flow through the gcm_filter package. But I have a question here. The filter_scale only represents severel grid points. How to explain the filter_scale as the precise scale in unit of kilometers?

Merge the B-grid vector Laplacian into GCM-filters

@NoraLoose and I will tie up loose ends and get this PR merged so we can use it in our project!

Make movies

I saw that @NoraLoose did a pull request to add examples how to make movies with xmovies.
At the same time I made a video by saving the images and then using ffmpeg. What I got:

movie_SSH_scale10.mp4

Run "Filtering on CPUs versus GPUs" with data from pangeo catalog

I am running this example but loading the NCAR data from pangeo catalog. Insted of filtering SST I filtered SSH just to test something different. That is what I got until now:

I think the next step is to select the area before filtering to save computional effort(?)

Filtering the velocity fields and looking at the EKE

Creating an issue so we can post our results when are ready.