eScience 2024 course

• Jupyterhub access
  • Creating Feide OpenIdP account:
  • Joining eScience2024 Feide group:
  • Accessing Jupyterhub:
• Workflow for eScience2024 course
  • Basic setup
    • 1. Fork projects repository
    • 2. Get GitGub Access token
    • 3. Setup git and clone on jupyterhub
    • 4. Sharing your work within the group
  • Basic Rules for the course
  • Sharing data
• Jupyterhub basic usage
  • Starting and stopping your server
  • The JupyterLab Interface
    • Menu Bar
    • Left Sidebar
    • Tips
    • Create a new python 3 notebook
      • Using debugger
  • Using different kernels in notebooks
    • Available kernels
    • Using the conda environments outside of the course
  • JupyterLab settings
  • Styling warnings in notebooks
• Data
  • Acessing model data using intake-esm
    • Reading and browsing the catalog
      • Browsing the catalog: Comparing the change in AOD over the historical period across CMIP6 models
    • Loading the data and plotting
      • Exporting a subset of the catalog
  • Local shared data
    • Direct local data access
  • ACTRIS data
    • Access via thredds server from NILU
    • Check availability via EBAS web interface
• Learning Corner
  • Brief introduction to Dask
  • Xarray + Dask
    • Working with chunks
    • Accessing the dask array
  • Numba
  • Using xarray to read EBAS data
    • Example 2 aerosol sulphate data
  • Interpolate hybrid sigma pressure coordinates to pressure
    • Open CMIP6 online catalog
    • Search corresponding data
    • Create dictionary from the list of datasets we found
    • Sub select time period (to make calculations faster):
  • Making a nice table
    • Create dictionary from the list of datasets we found
    • simplify keys:
    • Compute global mean (replace by whatever you want to calculate)
    • Extract units
  • Mask land
    • Reading data (CMIP-PPE)
      • Open the overview over the parameters in the CAM6 CESM PPE
      • Change longitude to -180 to 180
    • Open shapefile
    • Alternative using a gridded land sea dataset:
      • Interpolate to grid
  • Using masks and computing weighted average
    • Import python packages
    • Search data
    • Create dictionary from the list of datasets we found
    • Compute weighted mean
    • Compute weighted average over the entire globe
    • Comparison with unweighted mean
    • Compute Weigted arctic average
  • Using cartopy and projections for plotting
    • Open ERA5 dataset
    • Reading and browsing the catalog
      • Browsing the catalog: Compering the change in AOD over the historical period across CMIP6 models
    • Get metadata
      • Get variable metadata (air_temperature_at_2_metres)
    • Select time
    • Visualize data
      • Simple visualization from xarray plotting method
        • Same plot but with local file
      • Customize plot
        • Set the size of the figure and add coastlines
        • Change plotting projection
        • Choose extent of values
      • Combine plots with different projections
  • Read CMIP-PPE data and emulate with Gaussian Processor
    • Open the overview over the parameters in the CAM6 CESM PPE
      • Open CMIP6 online catalog
      • Search corresponding data
      • Create dictionary from the list of datasets we found
    • Global mean GP Model
    • Calibrate
  • Some tips with xarray and pandas
    • What are pandas and xarray?
      • Pandas
      • Xarray
    • 1. Read in CMIP6 data: We will skip this next part, but you can check it later to read data:
      • Import python packages
      • Search corresponding data
      • Create dictionary from the list of datasets we found
    • 1.1 Reading in the data from file:
      • Open dataset
        • Opening multiple files:
    • 2. Check how your dataset looks
      • Different types of information/data:
    • 3. Sometimes we want to do some nice tweaks before we start:
      • Selecting data and super quick plotting:
        • See here for nice overview
      • Slice in time: the sel method
        • isel, sel: index selecting
    • 3.2 Calculates variables and assign attributes!
      • 3.3 Convert longitude:
    • 4. The easiest interpolation: select with ‘nearest’ neighboor
      • Super quick averaging etc
    • 5. Mask data and groupby: pick out seasons
      • Tip: Might want to load or compute.
    • 6. Controle the plot visuals:
    • 7. Plotting with cartopy
    • My headline
  • comment number two
    • 8. Global average: weighted means!
    • 9. Convert to pandas & do some random fun stuff:
    • 10 Other stuff
      • Pick out station:
      • Resample:
      • Obs: might want to convert to pandas
      • lets do something unnecesarily complicated :D
      • Convert to pandas dataframe
      • qcut, cut
      • Convert back to xarray if we need:
      • mask by category
      • Improve and reduce risk of typo:
  • Regridding model data with xESMF
    • Import python packages
    • Set path to save data:
    • Open CMIP6 online catalog
    • Get data in xarray
      • Search od550aer variable dataset
      • Create dictionary from the list of datasets we found
    • Select model and visualize a single date
      • Plot on NorthPolarStereo and set the latitude limit
      • Get attributes (unique identifier)
    • Regrid CMIP6 data to common NorESM2-LM grid
    • Save regridded data intermediately
    • Concatenate all models
    • Compute seasonal mean of all regridded models
    • Save seasonal mean in a new netCDF file
    • Visualize final results (seasonal mean for all models)
  • Useful links
    • Some useful links
    • CMIP6 links:
    • Documentation links
  • Previous eScience Courses
• Using version control on Jupyterhub:
  • Starting with git on the jupyterhub
    • Adding your git username and email
    • Using git GUI extension:
    • Using git CLI
  • Setting up remotes
    • Adding remotes
    • Checking out remote branches
  • Pull requests and Issues
  • Issues
    • Pull requests
• Contribute to this page
  • The recommended workflow for development
    • Locally build this page
  • After you are done with your changes:
  • Reviewing PRs
• Credit
  • Sponsors: