A Deep Learning Model for the Thermospheric Nitric Oxide Emission

• Published in: Space Weather Vol. 19; no. 3
• Main Authors: Chen, Xuetao, Lei, Jiuhou, Ren, Dexin, Wang, Wenbin
• Format: Journal Article
• Language: English
• Published: Washington John Wiley & Sons, Inc 01.03.2021
• Subjects: Artificial neural networks; Atmospheric models; Broadband; Climate variations; Context; Datasets; Deep learning; Electrodynamics; Emission analysis; Feature extraction; General circulation models; geomagnetic storm; Geomagnetic storms; Geomagnetism; Infrared radiation; Ionosphere; Magnetic storms; Missing data; Morphology; Neural networks; Nitric oxide; nitric oxide emission; Radiation; Radiometry; Satellites; Space weather; theoretical model; Thermosphere; Weather forecasting
• ISSN: 1542-7390; 1539-4964; 1542-7390
• DOI: 10.1029/2020SW002619

Nitric oxide (NO) infrared radiation is an essential cooling source for the thermosphere, especially during and after geomagnetic storms. An accurate representation of the three‐dimension (3‐D) morphology of NO emission in models is critical for predicting the thermosphere state. Recently, the deep‐learning neural network has been widely used in space weather prediction and forecast. Given that the 3‐D image of NO emission from the Sounding of the Atmosphere using Broadband Emission Radiometry (SABER) onboard the Thermosphere Ionosphere Energetics and Dynamics satellite contains a large amount of missing data which is unobserved, a context loss function is applied to extract the features from the incomplete SABER NO emission images. A 3‐D NO emission model (referred to as NOE3D) that is based on the convolutional neural network with a context loss function is developed to estimate the 3‐D distribution of NO emission. NOE3D can effectively extract features from incomplete SABER 3‐D images. Additionally, NOE3D has excellent performance not only for the training datasets but also for the test datasets. The NO emission climate variations associated with solar activities have been well reproduced by NOE3D. The comparison results suggest that NOE3D has better capability in predicting the NO emission than the Thermosphere‐Ionosphere Electrodynamics General Circulation Model. More importantly, NOE3D is capable of providing the variations of NO emission during extremely disturbed times. Key Points The convolutional neural network (CNN) with a context loss function can well capture the variations in the incomplete images of the observed nitric oxide (NO) emission The CNN‐based NO emission model performs better in reproducing the observed 3‐D distribution of NO emission than the theoretical model does The CNN‐based NO emission model captures the main features of NO emission that changes dramatically during disturbed periods