Long term relationships of electron density with solar activity

Abstract

Greenhouse gases such as carbon dioxide and methane that are causing climate change may cause long term trends in the thermosphere and ionosphere. The paper aims to contribute to explore long term effects in the ionosphere focusing on the impact of solar activity changes. Peak electron density data derived from vertical sounding measurements covering 65 years at the ionosonde stations Juliusruh (JR055), Boulder (BC840) and Kokubunji (TO536), have been utilized to estimate the long-term behavior of daytime ionospheric F2 layer ionization in relation to the solar 10.7 cm radio flux index F10.7. In parallel, Global Navigation Satellite System (GNSS) based vertical total electron content (TEC) data over the ionosonde stations in combination with the peak electron density data have been used to derive the equivalent slab thickness τ for estimating long-term behavior in the time period 1996-2022. A new approach has been developed for deriving production and loss term proxies for studying long-term ionization effects from F2 layer peak electron density and TEC data. The derived coefficients allow estimating the long-term variation of atomic oxygen and molecular nitrogen concentrations including their ratio during winter months. The noon-time slab thickness values over Juliusruh correlate well with the decrease of F10.7 and the F2 layer peak height and enable estimating the neutral gas temperature. The equivalent slab thickness decreases by about 20 km per decade in the period 1996-2022, indicating a thermospheric cooling by about 100 K per decade for Juliusruh. Whereas the oxygen concentration decreases, the loss term, considered as a proxy for molecular components of the neutral gas, in particular N2, increases with the long-term solar activity variation. Considering 11 years averages of the production and loss terms under wintertime conditions, the long-term study reveals for the O/N2 ratio a percentage decrease of 5% per decade and for F10.7 about 3.1% per decade in a linear approach referred to the year 1970. Linear models of 11 years averaged NmF2 and foF2 from corresponding F10.7 show a very close correlation with the temporal variation of F10.7 until about 1990. The root mean square errors are in the order of 1.0 -1.3 ‧1010m-3 for NmF2 and 0.03-0.05 MHz for foF2. After 1990 the linear models clearly deviate from F10.7 at all selected ionosonde stations indicating a non-local effect.