Home▸AI & Machine Learning▸Satellite AI Emission Tracker

Satellite AI Emission Tracker

A Gaussian-plume dispersion model of a facility's methane/CO2 plume, viewed as a simulated satellite scene: an AI mass-balance algorithm integrates noisy downwind concentration readings to estimate the true emission rate, exactly as GHGSat, Sentinel-5P and MethaneSAT retrieval pipelines do.

AI & Machine Learning3DAdvanced60 FPS📱 Mobile-adapted

This simulator renders a real Gaussian atmospheric-dispersion plume drifting downwind from a facility stack, then runs the same mass-balance retrieval that AI-assisted satellite missions (GHGSat, Sentinel-5P TROPOMI, MethaneSAT) use to turn a noisy image of a gas plume into an estimated emission rate — without ever measuring the source directly. Adjust the true emission rate, wind speed and atmospheric stability class to reshape the physical plume, then tune the sensor noise and the AI classifier's detection threshold to see how instrument noise and threshold clipping bias the recovered estimate away from the ground truth, exactly as they do in operational satellite emission-monitoring pipelines.

⚙ Under the hood

A Gaussian-plume dispersion model of a facility's gas emissions, retrieved by a simulated AI satellite pipeline: adjust wind, stability, sensor noise and detection threshold to see how a mass-balance algorithm estimates the true emission rate from a noisy downwind image, just like GHGSat, Sentinel-5P and MethaneSAT.

AIremote sensingclimatesatelliteatmospheric dispersionemissions

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install