Modeling ionospheric <I>fo</I>F2 by using empirical orthogonal function analysis

Abstract

Abstract.A similar-parameters interpolation method and an empirical orthogonal function analysis are used to construct empirical models for the ionospheric foF2 by using the observational data from three ground-based ionosonde stations in Japan which are Wakkanai (Geographic 45.4 • N, 141.7 • E), Kokubunji (Geographic 35.7 • N, 140.1 • E) and Yamagawa (Geographic 31.2 • N, 130.6 • E) during the years of [1971][1972][1973][1974][1975][1976][1977][1978][1979][1980][1981][1982][1983][1984][1985][1986][1987]. The impact of different drivers towards ionospheric foF2 can be well indicated by choosing appropriate proxies. It is shown that the missing data of original foF2 can be optimal refilled using similar-parameters method. The characteristics of base functions and associated coefficients of EOF model are analyzed. The diurnal variation of base functions can reflect the essential nature of ionospheric foF2 while the coefficients represent the long-term alteration tendency. The 1st order EOF coefficient A 1 can reflect the feature of the components with solar cycle variation. A 1 also contains an evident semi-annual variation component as well as a relatively weak annual fluctuation component. Both of which are not so obvious as the solar cycle variation. The 2nd order coefficient A 2 contains mainly annual variation components. The 3rd order coefficient A 3 and 4th order coefficient A 4 contain both annual and semi-annual variation components. The seasonal variation, solar rotation oscillation and the small-scale irregularities are also included in the 4th order coefficient A 4 . The amplitude range and developing tendency of all these coefficients depend on the level of solar activity and geomagnetic activity. The reliability and validity of EOF model are verified by comparison with observational data and with International Reference Ionosphere (IRI). The agreement between observations and EOF model is quite well, indicating that the EOF model can reflect the major changes and the temporal distribution characteristicsCorrespondence to: D.-H. Zhang (zhangdh@pku.edu.cn) of the mid-latitude ionosphere of the Sea of Japan region. The error analysis processes imply that there are seasonal anomaly and semi-annual asymmetry phenomena which are consistent with pre-existing ionosphere theory.

Topside correction of IRI by global modeling of ionospheric scale height using COSMIC radio occultation data

“…This method has already been widely used in analyzing meteorology essential factors, atmospheric vertical structures, and layering vertical dynamic models. In the ionospheric community, many researchers attempted to apply EOF analysis to construct ionospheric and thermospheric empirical models [Dvinskikh, 1988;Daniell et al, 1995;Marsh et al, 2004;Zhao et al, 2005;A et al, 2011;Zhang et al, 2010;Lin et al, 2014]. Dvinskikh [1988] firstly introduced EOF analysis into the empirical modeling of the ionospheric parameters; Daniell et al [1995] applied EOF analysis to present the altitude profiles of ion concentration in their parameterized model of ionosphere; Marsh et al [2004] present a three-dimensional nitric oxide empirical model that parameterizes the spatial distribution of NO in the lower thermosphere based on EOF method; Zhao et al [2005] used EOF to analyze and model the topside ionospheric plasma density N i at middle to equatorial latitudes; a similar parameters interpolation method and empirical orthogonal function analysis are used to construct empirical models for the ionospheric f o F 2 by using the observational data from ionosonde stations [A et al, 2011]; global modeling of M(3000)F 2 and h m F 2 based on three alternative EOF expansion methods is Journal of Geophysical Research: Space Physics 10.1002/2016JA022785 described in the work of Zhang et al [2010];and Lin et al [2014] constructed a global ionospheric peak electron density height model based on EOF analysis using GNSS ionospheric radio occultation measurements during the years 2002-2011.…”

“…Therefore, E will reflect the latitudinal dependence, and T indicate the diurnal and solar cycle dependence of Hsc. It has been widely accepted that the first few orders of base functions and associated coefficients are able to reproduce almost the entire matrix Y with all the variations preserved [Lin et al, 2014;A et al, 2011]. This property results in a quick convergence speed of EOF analysis and leads to a great extent of computation reduction in specifying modeling parameters without obvious accuracy loss.…”

A global model: Empirical orthogonal function analysis of total electron content 1999–2009 data

“…The advantages of the EOF analysis method have made it one of the preferred methods in the meteorological field to analyze the dominant components of temporal and spatial variations [Lorenz, 1956]. The EOF analysis method has also been used for empirical modeling of ionospheric and thermospheric variability [e.g., Dvinskikh, 1988;Singer and Dvinskikh, 1991;Daniell et al, 1995;Matsuo et al, 2002Matsuo et al, , 2005Marsh et al, 2004;Zhao et al, 2005;Zapfe et al, 2006;Liu et al, 2008;Mao et al, 2008;Zhang et al, 2009Zhang et al, , 2010Matsuo and Forbes, 2010;A et al, 2011;Wan et al, 2012]. From the mathematical perspective, the orthogonal base functions, also called EOFs, are discrete functions of discrete variables.…”

A new global empirical NmF2 model for operational use in radio systems

“…Oyeyemi and McKinnell [2008] claim that their foF2 model can be used as a replacement option for the URSI and CCIR maps within the IRI model. Another approach of ionospheric parameter modeling based on empirical orthogonal function analysis has been studied by Dvinskikh [1988], Liu et al [2008], Zhang et al [2009], andA et al [2011].…”