For those who might not be familiar, Lagrangian datasets track the movement of sensing platforms or particles (or in the case below, atmospheric phenomena like cyclones) through space and time, providing a detailed path of their journey. This is in contrast to Eulerian datasets, which focus on changes in a fixed location.
The new dataset we've incorporated is the HURDAT2 dataset [0] , offering access to cyclone track data from 1852 to 2022 for both the Atlantic and Pacific Ocean basins. This dataset is provided as an xarray [1] Dataset utilizing CF (Climate and Forecast) conventions [2] for data variables that are organized as a Ragged Arrays [3].
To get started with `clouddrift` and the HURDAT2 dataset, check out our example Jupyter notebooks [4].
If you'd like to receive regular updates on each release or just want to get an idea of features we've been working on checkout our latest [5].
I'd love to hear from HN. How do you see clouddrift working with ragged array datasets of any kind you may be working with? Are there any challenges created by working with ragged arrays and/or Lagrangian datasets you have faced which clouddrift may not be accounting for? Any and all feedback is welcome as we look to make this tool as useful as possible for the scientific community!
[0] https://github.com/Cloud-Drift/clouddrift/blob/main/clouddri...
[1] https://docs.xarray.dev/en/stable/
[2] https://cfconventions.org/
[4] https://github.com/Cloud-Drift/hurdat2-get-started
[5] https://mailchi.mp/3d21f0d9d6b6/clouddrift-release-v0330-115...