A Machine‐Learning Approach to Derive Long‐Term Trends of Thermospheric Density. Issue 6 (11th March 2020)

• Record Type: Journal Article
• Authors: Weng, Libin; Lei, Jiuhou; Zhong, Jiahao; Dou, Xiankang; Fang, Hanxian
• Is Part Of: Geophysical research letters. Volume 47:Issue 6(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-03-11
• Subjects: thermospheric density -- Long‐term trend -- Artificial Neural Network method -- solar activity; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1029/2020GL087140
• Languages: English
• ISSNs: 0094-8276

Abstract

Key Points: The Artificial Neural Network Model (ANNM) has good capability of characterizing the variability of the satellite drag‐derived densities The density data during either 1967–2005 or 1967–2013 intervals are used to give a similar long‐term trend from the ANNM The thermospheric density trend relative to the ANNM from 250 to 575 km is about −1.5% to −2.0% per decade without obvious solar flux dependence Abstract: In this study, we revisit the long‐term trend of thermospheric density by using a Machine‐Learning approach. Our Artificial Neural Network Model (ANNM) can better capture the variations in the satellite drag‐derived densities than earlier empirical models, especially during extremely solar minimum period of 2007–2009. The long‐term trends from ANNM are similar when density data during either 1967–2005 or 1967–2013 intervals are used in the calculation. In addition, our trend estimates relative to the ANNM are −1.5% to −2.0% per decade from 250 to 575 km without obvious solar activity dependence. The Machine‐Learning approach provides a good way to give stable long‐term trend estimates of thermospheric density.