2022 EarthCube Annual meeting

Accelerating Lagrangian analyses of oceanic data: benchmarking typical workflows

Philippe Miron¹, Shane Elipot², Rick Lumpkin³, Bertrand Dano²

• ¹ Florida State University, USA
• ² University of Miami, USA
• ³ NOAA’s Atlantic Oceanographic and Meteorological Laboratory, USA

For data, “Lagrangian” refers to oceanic and atmosphere information acquired by observing platforms drifting with the flow they are embedded within, but also more broadly refers to the data originating from uncrewed platforms, vehicles, and animals that gather data along their unrestricted but complicated paths. Because such paths traverse both spatial and temporal dimensions, Lagrangian data often convolve spatial and temporal information that cannot always and readily be organized, cataloged, and stored in common data structures and file formats with the help of common libraries and standards. As such, for both data generators and data users, Lagrangian data present challenges that the CloudDrift project aims to overcome.

This Notebook, consists of systematic comparisons and evaluations of workflows for Lagrangian data, using as a basis the velocity and sea surface temperature dataset emanating from the drifting buoys of the Global Drifter Program. Specifically, we consider the interplay between diverse storage file formats (NetCDF, Parquet) and the data structure associated with common existing libraries in Python (xarray, pandas, and awkward) in order to test their adequacies for performing three common Lagrangian tasks:

1. binning of a variable on an Eulerian grid (e.g. mean temperature map);
2. extracting data within given geographical and/or temporal windows;
3. analyses per trajectory (e.g. single statistics, Fast Fourier Transforms).

Because the CloudDrift project aims at accelerating the use of Lagrangian data for atmospheric, oceanic, and climate sciences, we hope that this notebook will incite the community of users to eventually test it on their own platform and thus provide us with feedback on its ease of use and the intuitiveness of the proposed methods and guide future developments.

Please cite this individual notebook as:

Miron, Philippe, Elipot, Shane, Lumpkin, Rick, & Dano, Bertrand. (2022). Accelerating Lagrangian analyses of oceanic data: benchmarking typical workflows (ec2022v4). Zenodo. https://doi.org/10.5281/zenodo.6781037.

This Notebook is part of a collection of notebooks that form the 2022 EarthCube call for Notebooks published with DOI 10.5281/zenodo.6792049.

How to run the Notebook on your own computer ?

In a terminal window, clone and build the environment using Anaconda or Miniconda, then launch jupyter lab:

git clone https://github.com/Cloud-Drift/earthcube-meeting-2022
cd earthcube-meeting-2022
conda env create -f environment.yml
conda activate earthcube
jupyter lab

The last command should open the jupyter lab interface in your web browser. Once completed, click on the Notebook PM_05_Accelerating_Lagrangian_analyses_of_oceanic_data_benchmarking_typical_workflows.ipynb, and start exploring!