WRF Workshop 2017

Welcome wrf-python tutorial attendees!

The instructions below should be completed prior to arriving at the tutorial. There will not be enough time to do this during the tutorial.

Prerequisites

This tutorial assumes that you have basic knowledge of how to type commands in to a command terminal using your preferred operating system. You should know some basic directory commands like cd, mkdir, cp, mv.

Regarding Python, to understand the examples in this tutorial, you should have some experience with Python basics. Below is a list of some Python concepts that you will see in the examples, but don’t worry if you aren’t familiar with everything.

  • Opening a Python interpreter and entering commands.
  • Importing packages via the import statement.
  • Familiarity with some of the basic Python types: str, list, tuple, dict, bool, float, int, None.
  • Creating a list, tuple, or dict with “[ ]”, “( )”, “{ }” syntax (e.g. my_list = [1,2,3,4,5]).
  • Accessing dict/list/tuple items with the “x[ ]” syntax (e.g. my_list_item = my_list[0]).
  • Slicing str/list/tuple with the ”:” syntax (e.g. my_slice = my_list[1:3]).
  • Using object methods and attributes with the “x.y” syntax (e.g. my_list.append(6)).
  • Calling functions (e.g. result = some_function(x, y))
  • Familiarity with numpy would be helpful, as only a very brief introduction is provided.
  • Familiarity with matplotlib would be helpful, as only a very brief introduction is provided.

If you are completely new to Python, that shouldn’t be a problem, since most of the examples consist of basic container types and function calls. It would be helpful to look at some introductory material before arriving at the tutorial. If you’ve programmed before, picking up Python isn’t too difficult.

Here are some links:

https://www.learnpython.org/

https://developers.google.com/edu/python/

Step 1: Open a Command Terminal

To begin, you will first need to know how to open a command line terminal for your operating system.

For Windows:

WINDOWS + r
type cmd in the run window

For Mac:

Finder -> Applications -> Utilities -> Terminal

For Linux:

Try one of the following:

CTRL + ALT + T
CTRL + ALT + F2

Step 2: Download Miniconda

For this tutorial, you will need to download and install Miniconda. We are going to use Python 2.7, but it should also work with Python 3.5+. However, due to limitations with open source compilers on conda-forge, only Python 2.7 is available for Windows.

Please use the appropriate link below to download Miniconda for your operating system.

Note

64-bit OS recommended

Win64

Mac

Linux

For more information, see: https://conda.io/miniconda.html

Note

What is Miniconda?

If you have used the Anaconda distribution for Python before, then you will be familiar with Miniconda. The Anaconda Python distribution includes numerous scientific packages out of the box, which can be difficult for users to build and install. More importantly, Anaconda includes the conda package manager.

The conda package manager is a utility (similar to yum or apt-get) that installs packages from a repository of pre-compiled Python packages. These repositories are called channels. Conda makes it easy for Python users to install and uninstall packages, and also can be used to create isolated Python environments (more on that later).

Miniconda is a bare bones implementation of Anaconda and only includes the conda package manager. Since we are going to use the conda-forge channel to install our scientific packages, Miniconda avoids any complications between packages provided by Anaconda and conda-forge.

Step 3: Install Miniconda

Windows:

  1. Browse to the directory where you downloaded Miniconda2-latest-Windows-x86_64.exe.
  2. Double click on Miniconda2-latest-Windows-x86_64.exe.
  3. Follow the instructions.
  4. Restart your command terminal.

Mac and Linux:

For Mac and Linux, the installer is a bash script.

  1. Using a terminal, you need to execute the bash shell script that you downloaded by doing:

    bash /path/to/Miniconda2-latest-MacOSX-x86_64.sh [Mac]
    
    bash /path/to/Miniconda2-latest-Linux-x86_64.sh [Linux]
    
  2. Follow the instructions.

  3. At the end of the installation, it will ask if you want to add the miniconda2 path to your bash environment. If you are unsure what to do, you should say “yes”. If you say “no”, we’re going to assume you know what you are doing.

    If you said “yes”, then once you restart your shell, the miniconda2 Python will be found instead of the system Python when you type the “python” command. If you want to undo this later, then you can edit either ~/.bash_profile or ~/.bashrc (depending on OS used) and comment out the line that looks similar to:

    # added by Miniconda2 4.1.11 installer
    export PATH="/path/to/miniconda2/bin:$PATH"
    
  4. Restart your command terminal.

  5. [Linux and Mac Users Only] Miniconda only works with bash. If bash is not your default shell, then you need to activate the bash shell by typing the following in to your command terminal:

    bash
    
  6. Verify that your system is using the correct Python interpreter by typing the following in to your command terminal:

    which python
    

    You should see the path to your miniconda installation. If not, see the note below.

    Note

    If you have already installed another Python distribution, like Enthought Canopy, you will need to comment out any PATH entries for that distribution in your .bashrc or .bash_profile. Otherwise, your shell environment may pick to wrong Python installation.

    If bash is not your default shell type, and the PATH variable has been set in .bash_profile by the miniconda installer, try executing “bash -l” instead of the “bash” command in step 5.

Step 4: Set Up the Conda Environment

If you are new to the conda package manager, one of the nice features of conda is that you can create isolated Python environments that prevent package incompatibilities. This is similar to the virtualenv package that some Python users may be familiar with. However, conda is not compatible with virtualenv, so only use conda environments when working with conda.

The name of our conda environment for this tutorial is: tutorial_2017.

Follow the instructions below to create the tutorial_2017 environment.

  1. Open a command terminal if you haven’t done so.

  2. [Linux and Mac Users Only] The conda package manager only works with bash, so if bash is not your current shell, type:

    bash
    
  3. Add the conda-forge channel to your conda package manager.

    Type or copy the command below in to your command terminal. You should run this command even if you have already done it in the past. This will ensure that conda-forge is set as the highest priority channel.

    conda config --add channels conda-forge
    

    Note

    Conda-forge is a community driven collection of packages that are continually tested to ensure compatibility. We highly recommend using conda-forge when working with conda. See https://conda-forge.github.io/ for more details on this excellent project.

  4. Create the conda environment for the tutorial.

    Type or copy this command in to your command terminal:

    conda create -n tutorial_2017 python=2.7 matplotlib=1.5.3 cartopy netcdf4 jupyter git ffmpeg wrf-python
    

    Type “y” when prompted. It will take several minutes to install everything.

    This command creates an isolated Python environment named tutorial_2017, and installs the python interpreter, matplotlib, cartopy, netcdf4, jupyter, git, ffmpeg, and wrf-python packages.

Note

When the installation completes, your command terminal might post a message similar to:

If this is your first install of dbus, automatically load on login with:

mkdir -p ~/Library/LaunchAgents
cp /path/to/miniconda2/envs/tutorial_test/org.freedesktop.dbus-session.plist ~/Library/LaunchAgents/
launchctl load -w ~/Library/LaunchAgents/org.freedesktop.dbus-session.plist

This is indicating that the dbus package can be set up to automatically load on login. You can either ignore this message or type in the commands as indicated on your command terminal. The tutorial should work fine in either case.

Note

In this tutorial, we need to use matplotlib v1.5.3 due to some issues with cartopy, which should be fixed in a later version of cartopy. Be sure to supply the version number as indicated in the command above.

  1. Activate the conda environment.

    To activate the tutorial_2017 Python environment, type the following in to the command terminal:

    For Linux and Mac (using bash):

    source activate tutorial_2017
    

    For Windows:

    activate tutorial_2017
    

    You should see (tutorial_2017) on your command prompt.

    To deactivate your conda environment, type the following in to the command terminal:

    For Linux and Mac:

    source deactivate
    

    For Windows:

    deactivate tutorial_2017
    

Step 5: Download the Student Workbook

The student workbook for the tutorial is available on GitHub. The tutorial_2017 conda environment includes the git application needed to download the repository.

These instructions download the tutorial in to your home directory. If you want to place the tutorial in to another directory, we’re going to assume you know how to do this yourself.

To download the student workbook, follow these instructions:

  1. Activate the tutorial_2017 conda environment following the instructions in the previous step (source activate tutorial_2017 or activate tutorial_2017).

  2. Change your working directory to the home directory by typing the following command in to the command terminal:

    For Linux and Mac:

    cd ~
    

    For Windows:

    cd %HOMEPATH%
    
  3. Download the git repository for the tutorial by typing the following in to the command terminal:

    git clone https://github.com/NCAR/wrf_python_tutorial.git
    
  4. There may be additional changes to the tutorial after you have downloaded it. To pull down the latest changes, type the following in to the command terminal:

    For Linux and Mac:

    source activate tutorial_2017
    
    cd ~/wrf_python_tutorial
    
    git pull
    

    For Windows:

    activate tutorial_2017
    
    cd %HOMEPATH%\wrf_python_tutorial
    
    git pull
    

    Note

    If you try the “git pull” command and it returns an error indicating that you have made changes to the workbook, this is probably because you ran the workbook and it contains the cell output. To fix this, first do a checkout of the workbook, then do the pull.

    git checkout -- wrf_workshop_2017.ipynb
    git pull
    

Step 6: Verify Your Environment

Verifying that your environment is correct involves importing a few packages and checking for errors (you may see some warnings for matplotlib or xarray, but you can safely ignore these).

  1. Activate the tutorial_2017 conda environment if it isn’t already active (see instructions above).

  2. Open a python terminal by typing the following in to the command terminal:

    python
    
  3. Now type the following in to the Python interpreter:

    >>> import netCDF4
    >>> import matplotlib
    >>> import xarray
    >>> import wrf
    
  1. You can exit the Python interpreter using CTRL + D

Step 7: Obtain WRF Output Files

For this tutorial, we strongly recommend that you use your own WRF output files. The tutorial includes an easy way to point to your own data files. The WRF output files should all be from the same WRF run and use the same domain. If your files are located on another system (e.g. yellowstone), then copy 2 or 3 of these files to your local computer prior to the tutorial.

If you do not have any of your own WRF output files, then you can download the instructor data files from a link that should have been provided to you in an email prior to the tutorial.

If you are using the link provided in the email for your data, you can follow the instructions below to place your data in the default location for your workbook.

  1. The link in the email should take you to a location on an Amazon cloud drive.
  2. If you hover your mouse over the wrf_tutorial_data.zip file, you’ll see an empty check box appear next to the file name. Click this check box.
  3. At the bottom of the screen, you’ll see a Download button next to a cloud icon. Click this button to start the download.
  4. The download was most likely placed in to your ~/Downloads folder [%HOMEPATH%\Downloads for Windows]. Using your preferred method of choice for unzipping files, unzip this file in to your home directory. Your data should now be in ~/wrf_tutorial_data [%HOMEPATH%\wrf_tutorial_data for Windows].
  5. Verify that you have three WRF output files in that directory.

Getting Help

If you experience problems during this installation, please send a question to the Google Group support mailing list.

We look forward to seeing you at the tutorial!