Hosted by Department of Energy
Lawrence Livermore National Laboratory
Powered by ESGF logo
Welcome, Guest. | Login | Create Account
You are at the ESGF@DOE/LLNL node

Use Cases

The following two use cases use Jupyter Notebooks to analyze and visualize reanalysis data.

The Jupyter Notebook is an open-source web application that allows you to create and share documents that contain live code, equations, visualizations and narrative text. Jupyter Notebook requires Python 2.7, or Python 3.3 or greater. Jupyter strongly recommends installing Python and Jupyter using the Anaconda Distribution, which includes Python, the Jupyter Notebook, and other commonly used packages for scientific computing and data science.

Use case #1: Global Monthly Mean Jupyter Notebook

Download or View the Global Monthly Mean Jupyter Notebook.

The notebook in this use case opens and extracts the surface air temperature from two different reanalyses (MERRA2 and ERA-Interim in this example), calculates the global monthly mean, and plots a graph for the reanalysis period.

This notebook is referenced in:

Potter, G., L. Carriere, J. Hertz, M. Bosilovich, D. Duffy, T. Lee, and D. Williams,
2017: Enabling Reanalysis Research Using the Collaborative Reanalysis Technical Environment (CREATE). 
Bull. Amer. Meteor. Soc. doi:10.1175/BAMS-D-17-0174.1

This use case also requires Community Data Analysis Tools (CDAT), a front-end to a rich set of visual-data exploration and analysis capabilities well-suited for data analysis problems. CDAT can be installed easily using Anaconda.

If you already have Jupyer Notebook installed, download the Global Monthly Mean Jupyter Notebook.

Installation steps (links open in a new tab)

  1. Install Anaconda:

    Download and install the Anaconda 5.1 distribution for Windows, macOS or Linux. Python 2.7 and Python 3.6 are available. Installation instructions are linked on the download page just below the Download buttons.

  2. Install CDAT:

    Note: Two installation methods are described on the page. We've experienced minor issues using the environment files provided, you may prefer installing from the conda channels if you're not an experienced Python user. When the installation is complete continue with these steps:

  3. Start Jupyter Notebook. Type on a command line:
    A browser window showing the Jupyter home page will open. If a browser is not yet installed (which may happen in fresh distribution installations) refer to your particular Linux distribution's instructions for installing a browser.                                                                                
  4. Run the notebook
    • Download the Global Monthly Mean Jupyter Notebook from within the browser or download and move the notebook to a location in the file system accessible to the Jupyter notebook.
    • Open the notebook by clicking on Multiple_Anomaly_DEMO_plots_for_paper.ipynb.
    • Kernel not found dialog window may appear after opening the notebook, asking you to select a kernel. Choose Python [conda env:cdat8] from the pull-down menu and choose Set Kernel. The current kernel is displayed at the top far right of the notebook page. The kernel can be changed from the menu, select Kernel and Change kernel for a list of available kernels.
    • Click inside the cell/shaded area then choose Kernel -> Restart & Run All from the drop-down menu at the top of the page.
    • Choose the Restart and Run All Cells button. Depending on your system it may take a few minutes for the plot to appear.

An abbreviated set of these steps are listed in the Global Monthly Mean Jupyter Notebook. Contact with questions or for support.

Editing the notebook

If you're new to Jupyter Notebooks and want to try out simple changes to the output plot, see the comments in the notebook to locate the latitude range, time range or the reanalyses being compared. Browse the CREATE-IP catalog for available reanalyses and the CREATE-IP Global Reanalyses page for variables and time range information.

For more resources go to all things CDAT for information on the project, tutorials and support.

Use case #2: Examples using the Earth Data Analytics Service (EDAS)

The NCCS has built a server side analytics tool, EDAS, to allow researchers to leverage our compute power to analyze large datasets located at the NCCS through a web-based interface, thereby eliminating the need to download the data. The CREATE data is available through EDAS and there are multiple example Jupyter Notebooks available in the Example Code section of the page.


Last Update: June 17, 2020, 8:06 a.m. by Vincent Wild
Browse Projects
Start typing, or use the 'Delete' key to show all available tags.
create-ip Tags: None