Julia-Earth-Observation-Workshop

Link to this webpage: https://tinyurl.com/JuliaEO21-code

Before the workshop

• It is important to install all the software before the workshop (see below).
• To get some sample data, you need to register at these sites and have your username and password with you for the workshop:
  • CMEMS: https://data.marine.copernicus.eu/register
  • NASA Earth data: https://urs.earthdata.nasa.gov/users/new

One can either user Docker or the native julia package manager to install all software dependencies.

Workshop material

• Download and uncompress the code in this repository (or use git clone https://github.com/gher-uliege/Julia-Earth-Observation-Workshop). The notebooks are already included in the shared USB drive.

Installation with Docker

• Install Docker
• Open a terminal (Windows users, please use a PowerShell terminal)

Using docker without fast internet

• Locate the file julia-gher.tar.gz (1.8 GB) on the folder copied from the USB drive
• Load it with the following command

docker load -i julia-gher.tar.gz

Using docker with fast internet

• Run in a terminal run the following command:

docker pull abarth/julia-gher

Start the container

• Go to the directory contaning the notebooks

cd "replace this with the directory name with the notebook"

• Run in a terminal run the following commands (for Windows user this needs to be a PowerShell terminal):

docker run -p 8888:8888 -v ${PWD}:/home/jovyan/data abarth/julia-gher

where ${PWD} is automatically replaced by your current directory which is made available inside the docker container under the /home/jovyan/data. If you see the error Error starting userland proxy: listen tcp4 0.0.0.0:8888: bind: address already in use., it means that the port 8888 is already taken. Either you use a different port, or stop the other program currently using port 8888.

• Look for these lines:

    To access the server, open this file in a browser:
        file:///home/jovyan/.local/share/jupyter/runtime/jpserver-7-open.html
    Or copy and paste one of these URLs:
        http://1a8fae3e9eb9:8888/lab?token=SOME-LONG-TOKEN
     or http://127.0.0.1:8888/lab?token=SOME-LONG-TOKEN

Open the link http://127.0.0.1:8888/lab?token=SOME-LONG-TOKEN with your 48 characters long token.

Installation directly with julia's package manager

• Install julia
• On Linux, install also matplotlib (e.g. runnningsudo apt install python3-matplotlib in terminal for Debian/Ubuntu)
• Activate and instantiate the environement and start jupyter notebook:

using Pkg
cd("this_directory")
Pkg.activate(".") # needs to be repeated for every session
Pkg.instantiate() # install all package
using IJulia
notebook(dir=pwd())

where "this_directory" is the directory containing the ipynb files. Note, in Windows a path C:\Users\Foo\Bar should be written as raw"C:\Users\Foo\Bar" or "C:\\Users\\Foo\\Bar".

Now open the first notebooks file 01-DIVAnd-data-preparation.ipynb in the jupyter interface.