Post-Cutoff.com
  1. Home
  2. Timeline
  3. 2026
  4. Google and NASA JPL release MAPL-EMIT, an AI model that…

Google and NASA JPL release MAPL-EMIT, an AI model that maps methane plumes worldwide from the EMIT space instrument (PNAS)

★★★after cutoffscienceGoogle ResearchNASA JPLconfidence: high

On Sept 9, 2026 Google Research and NASA's Jet Propulsion Laboratory published MAPL-EMIT (Methane Analysis and Plume Localization with EMIT) in PNAS. The deep-learning model detects, quantifies and localizes methane plumes in hyperspectral data from NASA's EMIT instrument on the ISS. It finds 50% more plumes than human experts and more than 23,000 additional plumes globally, and the database, models and code are open.

Key facts

Science result

Field
climate-weather / remote sensing of greenhouse gases
Problem
Global detection and quantification of methane super-emitters from orbital hyperspectral imagery
Result
Model detects 50% more plumes than human experts and adds 23,000+ plumes to the global record
AI system
MAPL-EMIT
Human role
Human-led with AI tools: Google/JPL researchers built and validated the model
Verification
Peer-reviewed in PNAS
Status
confirmed

What happened

EMIT was built to map surface minerals in dust-source regions, but its hyperspectral bands also pick up methane's spectral signature. Google trained a detector on simulated plumes and ran it over the whole archive.

Why it matters

Methane is a strong short-term warming gas (Google cites about 30x the warming potential of CO2 over 100 years). Automated global plume mapping with open data lets regulators and companies find and fix leaks, landfills and super-emitters at scale.

Changelog

  • 2026-10-01: created (leads run, from the Google blog audit)

Sources (5)

id: 2026-09-09-google-nasa-jpl-mapl-emit-methane · updated 2026-10-01 · open in the interactive timeline