Installation

CLASS4GL has been tested on GNU/Linux environments, including Cent OS and Ubuntu. Although it is not mandatory, a supercomputer environment with qsub functionality Is highly recommended. Other operating systems such as MAC OS X or Windows are not supported yet, but should also work. In case of problems, please contact hendrik.wouters at ugent.be for support.

STEP 1: install Python 3.X

Most GNU/Linux distributions have Python and Ruby environments pre-installed. Please be sure that it is a Python 3.X environment. Python 2.X environments will not work. Please check your Python version as follows:

python --version

In case you don’t have a Python 3.X environment, you can easily install it with Anaconda, see: https://docs.anaconda.com/anaconda/install/linux.

STEP 2: install Python libraries

Before installing CLASS4GL, you need to install its Python dependencies. In case you are using Anaconda-based Python, you can install the dependencies with:

conda install pandas xarray beautifulsoup4 pyyaml netcdf4 seaborn pip ipython
pip install pysolar

STEP 3: install CLASS4GL

Go to (or create) a directory in which you want to install CLASS4GL, for example:

cd mkdir ~/software 
cd software

Get the CLASS4GL package through git and go into the directory:

wget https://files.pythonhosted.org/packages/3a/eb/7906bb1f5e6c6e8ae482a490458613dc572671f120e54a11a6de73d7e2d6/class4gl-1.0.3.tar.gz
tar -xzvf class4gl-1.0.3.tar.gz
cd class4gl-1.0.3/class4gl

Set the $CLASS4GL environment variable to the current directory where the CLASS4GL software is installed:

CLASS4GL=`pwd`

This environmental variable will be used further. You can verify it with:

echo $CLASS4GL

You can always check for a version update of the package at https://pypi.org/project/class4gl/.

STEP 4: download the reference data library

In order to be able to run CLASS4GL, you need to install the input reference library. First, create a directory where you want to store the class4gl data (eg. <code>$HOME/data/class4gl/</code>):

mkdir $HOME/data/class4gl/
cd $HOME/data/class4gl/

Next, download the reference input library as follows:

CLASS4GL_DATA=`pwd`
mkdir -p $CLASS4GL_DATA/forcing/
cd $CLASS4GL_DATA/forcing/
wget http://users.ugent.be/~hpwouter/class4gl/IGRA_PAIRS_20190515.tar.gz
tar -xzvf IGRA_PAIRS_20190515.tar.gz

In order to perform simulations that are based ERA5 evaporative fraction (see Section 2.3 of Wouters et al., 2019), you need to install an additional input data package (a tutorial will be added very soon to use this data):

wget http://users.ugent.be/~hpwouter/class4gl/IGRA_PAIRS_20190515_BASE_ITER_EFERA5.tar.gz
tar -xzvf IGRA_PAIRS_20190515_BASE_ITER_EFERA5.tar.gz

STEP 5: make environment variables permanent

Make the $CLASS4GL and $CLASS4GL_DATA environment variables permanent by activating it at every time you login to your system or open a terminal:

printf "\n# location of CLASS4GL software\nCLASS4GL='$CLASS4GL'\n" >> $HOME/.bashrc
printf "\n# location of CLASS4GL data library\nCLASS4GL_DATA='$CLASS4GL_DATA'\n" >> $HOME/.bashrc

The environmental variables should be at the bottom of your ~/.bashrc file. You can verify it with:

cat $HOME/.bashrc

You can now run your first experiment by following the steps here.

Please note that you will need to update the above environmental variables in case you install a new version.