Large Scale Sea level visualizations of unstructured mesh data

This is a developmental prototype for visualizing large scale results of hydrodynamic simulations.

Thalassa is powered by

Instalation

Start by cloning the repository:

Option 1: Use a virtualenv

Your system needs to have:

python>=3.8
geos
gdal=3.2.1
proj <8
poetry

There are multiple ways to satisfy these requirements, e.g. using your distro's package manager, compiling from source, etc. An easy way to proceed is to create a conda environment like the following:

conda create -n Thalassa pip python=3.8 geos gdal=3.2.1 proj=7 poetry

Afterwards, activate the new conda environment, create a virtualenv and install the dependencies using poetry:

conda activate Thalassa
python3 -m venv .venv
source .venv/bin/activate
poetry install

You are ready to go!

Option 2: Use conda

Install the dependencies in a conda environment with:

conda env create -f binder/environment.yml

Option3: Use docker

docker/build.sh

This will create a docker image

Obtaining Data

You will need some data. If you don't have any you can download a sample dataset from here:

wget -O data/animation.mp4 https://static.techrad.eu/thalassa/animation.mp4
wget -O data/dataset.nc    https://static.techrad.eu/thalassa/dataset.nc
wget -O data/stations.csv  https://static.techrad.eu/thalassa/stations.csv
wget -O data/stations.zip  https://static.techrad.eu/thalassa/stations.zip
wget -O data/thalassa.png  https://static.techrad.eu/thalassa/thalassa.png

Running Thalassa

Conda or virtualenv

If you used conda or virtualenv, you can launch the Thalassa web server with:

thalassa serve --websocket-origin='localhost:9000' --port 9000

An image should open on your visit http://localhost:9000

Conda

If you used conda you can also use:

panel serve --show Thalassa.ipynb

it will open in your default browser.

docker

If you build the docker image, execute:

docker/run.sh

This will start a webserver listening on port 61112. So visit: http://localhost:61112

NOTE: If you want to deploy this on a server, you will probably want to change the websocket-origin in docker/run.sh to something more secure (e.g. to a subdomain).

License

The project is released under the EUPL v1.2 license.

moghimis/Thalassa