User Interface
This page highlights some key aspects of the VCDAT 2.0 user interface.
The VCDAT development team created VCDAT 2.0 as a JupyterLab extension so it makes use of all the generic features available via the JupyterLab interface, but adds additional functionality needed to perform VCDAT functions like selecting a subset of data from a NetCDF file and plotting that subset of data.
When you launch JupyterLab for the first time, you will likely see several Python icons in the Launcher tab of the JupyterLab interface as shown below. Depending on how you launched the JupyterLab interface (on your own computer/network or via Cori, acme1 or crunchy) you may or may not see files in the lefthand pane. The screenshot from acme1 below does not show any files in the left panel.
After installing and launching the JupyterLab interface, it is simplest to open a blank Jupyter notebook. There are two ways to do this. If your screen looks something like this (with a "Launcher" tab in the main window):
If you don't see the "Launcher" tab, go to File > New > Notebook as shown below.
If it asks you to select a Kernel, choosing Python 3 is a good option:
Once you've opened a blank notebook, you can run the commands on the Access Data page to download some sample data. The notebook is automatically saved at frequent intervals, but it is always a good idea to save it yourself by clicking on the disc icon on the notebook menubar, though you might want to rename it by right clicking on the file name in the left panel and choosing Rename OR right clicking on the Notebook Tab in the main/right panel and choosing Rename Notebook as shown here:
To save, click on the disc icon:
The CDAT logo 
on the left side of your screen will open the VCDAT controls. This is where you can load and plot variables.
Next you'll likely want to work with some data (see the Access Data page), load some variables, and plot the data.
For more detailed information on the user interface, see the JupyterLab interface documentation.