3 ms·
Show HN: I built a satellite forensic engine to detect fraud in Carbon Markets
I got tired of reading PDF reports for Carbon Offsets that claimed to be "Protected Forests" but looked suspicious on Google Maps.
So I built Silent Earth—a standalone forensic engine (Python/React) to audit them mathematically.
The Stack:
* Ingestion: Scrapes Verra/Gold Standard registries for project coordinates.
* Telemetry: Queries the Sentinel-2 Satellite Constellation (via Microsoft Planetary Computer).
* Analysis: Calculates NDVI (Biomass Density) and performs spectral analysis on the claimed area.
* AXIOM Kernel: Cross-references industrial "shutdown" claims against Tropospheric NO2 levels (Sentinel-5P).
The Result: I routinely find "Verified" assets with an NDVI < 0.2 (Bare Soil).
The tool is currently local-only for privacy, but I'm planning to move the data stream to Snowflake.
Happy to answer questions about the geospatial stack (Rasterio/Shapely) or the satellite APIs.
- kccanarch 9mo agoOP here. One of the wildest things I found was a project in Sub-Saharan Africa claiming dense forestry that had a heat signature (NO2) matching a light industrial zone. If anyone is working with Sentinel-5P data, I'd love to compare notes on cloud-masking algorithms.
- polishdude20 9mo agoAre the satellite API's free?
- kccanarch 9mo agoYes. The raw telemetry comes from the Sentinel-2 constellation, which is open access However, querying the raw hub is slow :( API hosts the same data but indexes it via the STAC standard. You don't pay for data, you only pay for the compute to process the pixels.