Unsupervised Learning on GOLD N2 Lyman-Birge-Hopfield Measurements to Characterize Thermal Structure Changes in the Thermosphere
Abstract
Far-ultraviolet observations by the NASA Global-scale Observations of the Limb and Disk (GOLD) mission provide unprecedent global measurements of Earth's airglow. The N2 Lyman-Birge-Hopfield (LBH) band emissions that GOLD measures over Earth's disk are sensitive to thermospheric temperature. An exploratory unsupervised machine learning technique is applied here to better isolate the temperature signal within the GOLD LBH Level 1C emission data and to develop a low-dimensional representation of the LBH spectrum that captures the relevant geophysical information. This temperature signal is characterized in terms of the dominant modes of LBH emission variability that relate to the morphology of vibrational emission features. Various factors which contribute to changes in these dominant modes, such as solar zenith angle and geomagnetic activity, are investigated using principal components and their time-varying magnitudes. This paper presents how this low-dimensional, succinct representation of GOLD data can track the observed changes of the thermal structure in the thermosphere.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMSA11A..05C
- Keywords:
-
- 0355 Thermosphere: composition and chemistry;
- ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 0358 Thermosphere: energy deposition;
- ATMOSPHERIC COMPOSITION AND STRUCTURE;
- 3369 Thermospheric dynamics;
- ATMOSPHERIC PROCESSES;
- 7949 Ionospheric storms;
- SPACE WEATHER