Fully Automatic Trace Gas Plume Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911647022448640 |
|---|---|
| author | Růžička, Vít Thompson, David R. Fahlen, Jay E. Lopez, Amanda M. Lu, Steven Xiang, Chuchu Bender, Holly Jensen, Daniel Brodrick, Philip G. Lee, Jake Bue, Brian Cusworth, Daniel H. Guanter, Luis Chlus, Adam Thorpe, Andrew Green, Robert O. |
| author_facet | Růžička, Vít Thompson, David R. Fahlen, Jay E. Lopez, Amanda M. Lu, Steven Xiang, Chuchu Bender, Holly Jensen, Daniel Brodrick, Philip G. Lee, Jake Bue, Brian Cusworth, Daniel H. Guanter, Luis Chlus, Adam Thorpe, Andrew Green, Robert O. |
| contents | Future imaging spectrometers will increase data volumes by orders of magnitude, requiring automated detection of trace gas point sources. We present a fully automated framework that combines machine learning-based morphological analysis with physics-based spectroscopic fitting to detect plumes without human participation. Applied to EMIT imaging spectrometer data, the system operates in two modes: "daily digest" that runs automatically on all downlinked data, flagging the largest events for immediate response, and a retrospective analysis that identifies plumes missed by prior human review. The daily digest demonstrates that a significant fraction of the largest plumes can be detected automatically with negligible false positives, while retrospective analysis suggests at least 25% of plumes may have been overlooked. In addition to the previously observed methane point sources, we extend detection to three understudied trace gases: NH3, NO2 and the first observations of carbon monoxide (CO) plume in EMIT imagery. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_03372 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Fully Automatic Trace Gas Plume Detection Růžička, Vít Thompson, David R. Fahlen, Jay E. Lopez, Amanda M. Lu, Steven Xiang, Chuchu Bender, Holly Jensen, Daniel Brodrick, Philip G. Lee, Jake Bue, Brian Cusworth, Daniel H. Guanter, Luis Chlus, Adam Thorpe, Andrew Green, Robert O. Machine Learning Future imaging spectrometers will increase data volumes by orders of magnitude, requiring automated detection of trace gas point sources. We present a fully automated framework that combines machine learning-based morphological analysis with physics-based spectroscopic fitting to detect plumes without human participation. Applied to EMIT imaging spectrometer data, the system operates in two modes: "daily digest" that runs automatically on all downlinked data, flagging the largest events for immediate response, and a retrospective analysis that identifies plumes missed by prior human review. The daily digest demonstrates that a significant fraction of the largest plumes can be detected automatically with negligible false positives, while retrospective analysis suggests at least 25% of plumes may have been overlooked. In addition to the previously observed methane point sources, we extend detection to three understudied trace gases: NH3, NO2 and the first observations of carbon monoxide (CO) plume in EMIT imagery. |
| title | Fully Automatic Trace Gas Plume Detection |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.03372 |