2. Analysis of the Available Models

[4]  Belikovich et al. [1983] used as initial data for creation of the model the original results of the measurements by the PR method for the period 1969-1980 and also the results of the measurements by the PR and cross-modulation methods, as well as in rocket experiments published in literature (totally about 500 measurements). The variations in the electron concentration with solar zenith angle ( c ), season, and solar activity (the Wolf number, W ) were described by where the initial height is h₀ = 60 km, p = 0.5-0.9, a = -0.008 0.012 depending on h, m is the number of days counted from 21 March. The coefficient k determining the seasonal variations of n_e(h) is positive at heights of 60-80 km and negative at heights of 85-95 km. The model is applicable in the latitude range j = 30^o-60^o at altitudes 60-95 km in the sunlit period of a day, and in quiet geophysical conditions. The Danilov et al. [1991, 1995] model was created on the basis of the database including 276 profiles with n_e values at altitudes of 60-85 km obtained by various rocket methods during the period from 1948 to 1978 in the latitude interval 81^oN-30^oS. The model makes it possible to calculate the values of n_e(h) in the lower ionosphere depending on c, season (S) and geomagnetic index ( Kp ). The model takes into account the meteorological effects: the winter anomaly events and stratospheric warmings according to where A₀-A₅ are the coefficients different for different altitudes (with a step of 5 km) determining variations depending on each of the parameters taken into account. The global model by Chasovitin [1988] is created for the 80-600 km range, however below 200 km the electron concentration is calculated using the Chasovitin [1983] model. The latter may be used for the conditions of moderate solar activity ( 100 F_10.7 150 ) and quiet geomagnetic situation ( Kp 4 ). The model is presented in the form of tables calculated for each variation in the input parameters. The Bilitza et al. [1981] model was a basis for the description of the dependency of the undisturbed lower ionosphere structure on the time of a day (day-night), solar zenith angle, and solar activity in the models IRI 1979, 1990, 1995/96, and 2001. The model considers only rocket measurements (about 80). The vertical profile of ne is approximated by the polynomial of the third degree centered above and below the inflection point located at a height of h = 81 km. Two latitudinal intervals are considered: the lower and middle (however the differences between them are negligible). It is shown that the value of n_e in the inflection point may be given by where R₁₂ is the running mean (over 12 years) Wolf number. The Pancheva and Mukhtarov [1997] model was created on the basis of the measurements of the radio wave absorption obtained by the A3 method. The data obtained only in Bulgaria were used, so the model does not provide the global structure of the lower ionosphere. Nevertheless the model analyzes in detail diurnal and seasonal variations of n_e. The model profile n(h) is presented in some set of characteristic points connected by cubic splines to provide the continuity of the profile. The values of n_e are determined unambiguously only in the characteristic points and the variations of the n_e(h) profile with c, F_10.7, and season are modeled changing the values of n_e in these characteristic points. Friedrich and Torkar [1992] actually described in their model only the dependence of the electron concentration on the solar zenith angle. This was due to the very limited database of the initial data (72 profiles obtained by the rocket method). The model is created for moderate solar activity ( W =60 ), middle latitudes and undisturbed geomagnetic conditions. McNamara [1979] collected and analyzed about 700 profiles of ne obtained mainly by rocket methods and PR methods. This publication is the only one in which an attempt is made to take into account the longitudinal effects in the lower ionosphere mentioned earlier by George [1971] and Mechtly et al. [1969]. However, the longitudinal effects in the McNamara [1979] model are taken into account in the latitudinal dependence of n_e due to the difference in the measurement points of geographic ( j ) and geomagnetic ( q ) latitudes as well as of the modified dipole latitude X = arctan (I/ cos j ). The expression for the electron concentration at the given height is presented by the linear function: where M is the number of the month.