Snakes in the Microwaves: How Python is Powering the Golden Age of SAR
summary
Alaska Satellite Facility manages ~30 PB archive in AWS US West 2 and expects >100 PB as calibrated NISAR arrives, turning decade‑long Sentinel‑1 records and new NISAR frames into continuous Earth streams. The stack uses NumPy, Xarray, Zarr, Jupyter with Flask APIs, Docker containers, Step Functions orchestration and on‑demand processing (credit allotment ~8,000 per EarthData user) to produce serverless services including ITS_LIVE outputs and about 40 million granules of velocity estimates accessible as virtualized data cubes.
AI-written summary · about
chapters (10)
- 0:00Introduction and context
- 3:35What is SAR and why it matters
- 10:18Challenges in SAR data interpretation
- 17:04Analysis ready SAR data explained
- 21:23Discovering SAR data with Python
- 25:38On-demand processing with Python
- 30:10NISAR and higher-level products
- 33:21Cloud optimization and data cubes
- 35:10Real-time applications and future work
- 39:04Q&A session
related talks
34mGeoscience Big Data Computing with Python
16mThe Open Data Cube and working towards a Digital Earth
39mSatellite Imagery to Machine-Readable Intelligence
14mSatellite data monitoring for analytical models
05mAlex Kudriashova Interview
21mRealizing the Full Potential of NOAA’s Open Data
10mPaco Nathan (Liber 118) Interview - OSCON 2014
21mAquawatch Data Systems a cloud solution for water quality monitoring and forecasting