Skip to content
Oceanum.io

Oceanum Notebooks

Oceanum Notebooks is a JupyterLab environment that runs entirely in your web browser at notebook.oceanum.io. There is nothing to install and no server to start: the Python kernel is Pyodide, a build of Python for WebAssembly, and it runs inside the browser tab.

Sign in with your Oceanum.io account and your Datamesh credentials are passed into the kernel for you, so a connector needs no token:

from oceanum.datamesh import Connector


Connector().get_catalog(search="era5", limit=5)

What is included

Section titled “What is included”

  • A kernel with the scientific Python stack already in it — the oceanum package, xarray, pandas, numpy, dask, matplotlib, scipy, cartopy, plotly and seaborn are bundled with the site. Packages are downloaded the first time they are imported; there is no pip install step.

  • Datamesh access from your sign-in — the kernel is given your credentials when you sign in, and again each time they are refreshed.

  • Notebooks saved to your account — Save to Oceanum stores a notebook on Oceanum.io rather than in the browser, so it follows you between machines and can be shared. See Saving and sharing.

  • The Datamesh panel and AI chat — the Oceanumlab extension is part of the site, so you can search the Datamesh and drop queries into a notebook without leaving JupyterLab.

What runs where

Section titled “What runs where”

Cells run on your own machine, in the browser; only the data you query crosses the network. That shapes what the environment is good for:

  • The kernel is single-threaded and limited to the memory the browser gives one tab. Prototype here and query subsets; run long or heavy processing on the platform.

  • Packages load on demand. The first import of the Datamesh connector downloads about 44 MB, which the browser then caches.

  • netCDF4 is not available in the browser. netCDF data is read through h5netcdf.

  • Files you create in JupyterLab are held in that browser’s own storage. Another machine will not see them, and clearing site data for notebook.oceanum.io deletes them — which is what Save to Oceanum is for.

Next steps

Section titled “Next steps”

  • Getting started — open the site, sign in and run your first Datamesh query.

  • Saving and sharing — keep notebooks in your account and share them with colleagues.

  • Your notebooks — find your saved notebooks on the Oceanum.io dashboard.