ODD PI 26.1 Objective 1: 🗺️ Add dynamic tiling and timeseries support for Virtual Zarr Stores

Motivation

Virtual Zarr provides uniform access across NASA datasets. Using virtual Zarr, copies of the data do not need to be maintained in VEDA. Virtual Zarr has been shown to deliver 15x improvements in performance when it comes to analytical queries such as timeseries.

Description

Building upon the existing titiler ecosystem, this objective will allow performant map and timeseries visualization of data stored as archival file formats (e.g., GRIB, HDF5/NetCDF4, TIFF) using virtual Zarr stores and titiler-multidim.

Acceptance Criteria

• Integrate datasets: Demonstrate + finalize architecture for how tiles and timeseries APIs can be served from virtual zarr stores using titiler-multidim and integrated with the VEDA UI, likely leveraging the STAC catalog. Current plan is to demonstrate for NLDAS for the water insight portal.
• Stretch: Develop new features: Develop the timeseries/statistics endpoint in titiler-multidim. Identify additional I/O parameters to allow for per-dataset optimizations.
• Stretch - technical maintenance: Reduce technical maintenance burden by homogenizing the xarray-based I/O approaches across titiler extensions and applications (i.e., titiler.xarray, titiler-multidim, and titiler-cmr)