2026-08-30 –, 601
A hands-on workshop on cloud-native raster analytics using open standards and scalable Python tools. Discover data with STAC, stream Cloud-Optimized GeoTIFFs, model data cubes with xarray, and perform parallel geospatial analysis using Dask.
Modern satellite archives are massive, multi-temporal, and increasingly hosted in cloud environments. Traditional file-based GIS workflows struggle when working with large raster collections, global time series, or multi-band imagery.
This hands-on workshop introduces a practical, open-source workflow for scalable raster analytics using STAC, Cloud-Optimized GeoTIFF (COG), xarray, rioxarray, and Dask.
Participants will:
- Discover satellite imagery from public STAC catalogs
- Stream raster data directly from the cloud without downloading entire scenes
- Structure multi-temporal imagery into a labeled data cube
- Perform scalable computations using chunked, parallel processing
- Generate derived products such as vegetation indices and temporal summaries
Workshop materials, including content, notes, and code examples, are available at:
https://amanchry.github.io/scalable-raster-analytics/
Participants must have:
- Python 3.10 or newer installed
- Conda (Miniconda or Anaconda) installed for environment setup
- Basic Python programming knowledge
- Be comfortable running Python scripts or Jupyter notebooks
- Have basic understanding of raster data (bands, CRS, GeoTIFF)
- Be familiar with general GIS concepts
Aman Chaudhary is a Geospatial Software Developer. His work focuses on building scalable, open-source geospatial infrastructure that supports decision-making, monitoring, and planning across diverse domains, including water resources, agriculture, climate, urban development, disaster risk, and sustainability.
Website: https://www.amanchaudhary.com/