GSoC 2025: Week 6 Report: AI-ready Dataset Metadata as a Service using ZOO-Project

Progress Report for Week 6 (July 7 – July 13)

1. What did I get done this week?

  • Explored and analyzed the NASA POWER Zarr timeseries datacube using AWS S3 and Xarray.
  • Loaded and inspected the datacube structure, including dimensions, variables, and metadata.
  • Developed and validated functions to convert the Zarr datacube into GeoCroissant JSON-LD, with type inference and metadata cleaning.
  • Converted both the full datacube and a targeted 2020 T2M subset into GeoCroissant format, adding detailed monthly descriptions.
  • Validated the GeoCroissant metadata using mlcroissant tools to ensure compliance.
  • Processed the 2020 T2M timeseries into monthly subsets and created visualizations of monthly and annual patterns.

2. Plan for Next Week ( July 14 - July 20):

  • UMM to GeoCroissant conversion support.

3. Am I blocked on anything?

  • No, I am not currently blocked on anything.

Links to Work Done:

Best Regards,
Harsh Shinde