Critical frequencies <Emphasis Type="Italic">foF</Emphasis>2 as an indicator of trends in thermospheric dynamics

The time variations in three parameters during the last decades are considered. R(foF2) is the correlation coefficient between the nighttime and daytime values of foF2 for the same day. Stable trends are found for the minimum (R(foF2)(max)) and maximum (R(foF2)(min)) values of R(foF2) during a year. The foF2(night)/foF2(day) ratio demonstrates both, negative and positive trends, and the trend sign depends on the inclination I and declination D of the magnetic field. The correlation coefficient r(h, fo) between foF2 and the 100 hP level in the stratosphere demonstrates a decrease (in the years of maximum and minimum solar activity) from the 1980s to the 1990s. The trends in all three groups of data are considered under the assumption of long-term changes in the circulation in the upper atmosphere.     

Long-term trends in the ratio of the daytime and nighttime values of <Emphasis Type="Italic">foF</Emphasis>2

The consideration of the relation between the daytime and nighttime values of the critical frequency F2, foF2 of the ionospheric F2 layer, started in the previous publication of the authors, is continued. The main regularities in variations in the correlation coefficient R(foF2) characterizing this relation are confirmed using larger statistical material (more ionospheric stations and longer observational series). Long-term trends in the R(foF2) value are found: at all stations the negative value of R(foF2) increases with time after 1980.     

Regular changes in the critical frequency of the <Emphasis Type="Italic">F</Emphasis>2 layer of the quiet midlatitude ionosphere

A method for constructing the empirical model of the F2 layer critical frequency (foF2) under magnetically quiet conditions, aimed at analyzing disturbances of any nature, is proposed. This method has been analyzed, and typical features of regular changes in foF2 of the quiet ionosphere and day-to-day foF2 variability are analyzed using the data from Irkutsk and Slough stations as an example. In particular, it has been obtained that this model differs from the international IRI model, and this difference is mainly caused by the fact that the foF2 values in the IRI model do not correspond to quiet conditions. Therefore, this model gives a larger amplitude of the annual and semiannual variations in foF2 than the IRI model. In addition, this model more accurately reproduces the rate of foF2 annual variations at a fixed local time, especially in equinoxes, when foF2 variations can exceed 1 MHz within one month.     

Spatial and time variations in critical frequencies of the ionospheric <Emphasis Type="Italic">F</Emphasis> region above the zone of equatorial earthquake preparation

Spatial and time variations in the critical frequencies foF2 before the strong earthquake of August 15, 1963, with the magnitude M = 7.75 are analyzed. The epicenter of the earthquake was located in the vicinity of the magnetic equator in the American longitudinal sector. The data of the topside ionosonde on board Alouette-1 and of the series of ground-based ionosondes has been used for this purpose. The ground-based ionosondes made it possible to detect an insignificant anomalous decrease in foF2 within the zone of earthquake preparation a few days prior to the earthquake. This result confirms the conclusion drawn earlier on the basis of the satellite data. The modification of the ionosphere at the F-region level is more evident in the satellite than in the ground-based data. It is also noted that the character of the time variations in foF2 a day before the earthquake is similar to the so called “quiet time” Q-disturbances in the ionosphere, when the electron concentration at the F region maximum differs from the median values by more than 20% under undisturbed geophysical conditions.     

Spatial inhomogeneity of variations in the <Emphasis Type="Italic">F</Emphasis><Subscript>2</Subscript> layer critical frequencies under the conditions of winter 1995

The daily samples of the hourly measurements of the foF2 critical frequency, obtained on January 5–21, 1995, at the midlatitude and high-latitude automated ionospheric stations (geographic latitude higher than 60°), are considered. The {fo} sets are transformed into the {δfoF2} sets of relative variations, for which asymmetry (A) and excess (E) are calculated. The selected stations are grouped into 20 pairs of automated ionospheric stations (AISs) located at distances of 200–10 000 km from one another. Sign estimates of the cross-correlation coefficients between the sets of 16 A and E values for different pairs of stations are used. Two types of structures of the statistical invariant spatial distribution are established: the structures with a scale of about 300 km, invariant with respect to latitude, and with a scale of about 6000 km (for only high latitudes).     

Properties of the <Emphasis Type="Italic">F</Emphasis>2-layer critical frequency median in the nocturnal subauroral ionosphere during low and moderate solar activity

Based on an analysis of data from the European ionospheric stations at subauroral latitudes, it has been found that the main ionospheric trough (MIT) is not characteristic for the monthly median of the F2-layer critical frequency (foF2), at least for low and moderate solar activity. In order to explain this effect, the properties of foF2 in the nocturnal subauroral ionosphere have been additionally studied for low geomagnetic activity, when the MIT localization is known quite reliably. It has been found that at low and moderate solar activity during night hours in winter, the foF2 data from ionospheric stations are often absent in the MIT area. For this reason, a model of the foF2 monthly median, which was constructed from the remaining data of these stations, contains no MIT or a very weakly pronounced MIT.     

Geomagnetic activity that corresponds to the median of the <Emphasis Type="Italic">F</Emphasis>2-layer critical frequency at various latitudes

On the basis of the F2-layer critical frequency foF2 for the noon at some European stations for 1958–2005, it is found that the geomagnetic activity corresponding to the foF2 median is systematically lower than that averaged over the month; the difference increases with an increase in latitude. Moreover, the dispersion of geomagnetic activity for the foF2 median at relatively high latitudes is lower than at middle latitudes. These regularities are related to the fact that high geomagnetic activity usually leads to a distinct deviation of foF2 from the typical average value, i.e., from the foF2 median, and such deviation is more substantial at relatively high latitudes. That is why the geomagnetic activity for the foF2 median is lower at relatively high latitudes than at middle latitudes.     

Long-term changes in the relation between <Emphasis Type="Italic">foF</Emphasis>2 and <Emphasis Type="Italic">hmF</Emphasis>2

The change in the dependence of the F2-layer critical frequency on its height hmF2 is considered based on two sources of initial data used earlier by the authors. It is found that the slope k of the foF2 dependence on hmF2 systematically decreases from the earlier (“etalon”) period, 1958–1980, to the later periods of 1988–2010, 1998–2010, and 1998–2014. Since the foF2 value depends on the atomic oxygen concentration in the F region much more strongly than hmF2, the found decrease in k confirms the concept of a decrease in the atomic oxygen concentration in the thermosphere with time previously formulated by the authors.     

Variations in the fractal dimension of fluctuations of the ionospheric <Emphasis Type="Italic">F</Emphasis>2 layer critical frequency

The 40-year period of observations of short-term variations (with characteristic times of up to 1–2 days) in the critical frequency of the ionospheric F2 layer (foF2) is analyzed. The continuous (with a step of 1 h) series of fluctuations (F) of the foF2 critical frequency (with eliminated daily variations) has been calculated using the hourly variations in foF2 at Moscow stations. The fractal dimension (FRH) of the fluctuations, characterizing short-term variations in foF2, has been determined and analyzed on a 30-day interval, using the Higuchi method. It has been established that FRH estimates substantially change in time. The 11-year cycle, which is in antiphase with the solar cycle, and the total annual and semiannual variations, similar to the variations observed in the normalized critical frequency of the E region and in the electron density of the D region, are clearly defined in these changes. Thus, the parameters of fast variations in the ionospheric F2 layer are affected by the phase of the 11-year solar cycle and by the position of the Earth in the orbit or seasonal variations in the atmosphere.     

Global empirical model of critical frequency of the ionospheric <Emphasis Type="Italic">F</Emphasis>2-layer for quiet geomagnetic conditions

Using the foF2 database obtained from satellites and ground-based ionospheric stations, we have constructed a global empirical model of the critical frequency of the ionospheric F2-layer (SDMF2—Satellite and Digisonde Data Model of the F2 layer) for quiet geomagnetic conditions (Kp < 3). The input parameters of this model are the geographical coordinates, UT, day, month, year, and the integral index F10.7 (day, τ = 0.96) of solar activity for a given day. The SDMF2 model was based on the Legendre method for the spatial expansion of foF2 monthly medians to 12 in latitude and 8 in longitude of spherical harmonics. The resulting spatial coefficients have been expanded by the Fourier method in three spherical harmonics with respect to UT. The effect of the saturation of critical frequency of the ionospheric F2-layer at high solar activity was described in the SDMF2 model by foF2 as a logarithmic function of F10.7 (day, τ = 0.96). The difference between the SDMF2 and IRI models is a maximum at low solar activity as well as in the Southern Hemisphere and in the oceans. The testing on the basis of ground-based and satellite data has indicated that the SDMF2 model is more accurate than the IRI model.     

Increase in the critical frequency of the ionospheric <Emphasis Type="Italic">F</Emphasis> region prior to the substorm expansion phase

Specific variations in the critical frequency of the ionospheric F ₂ layer during magnetospheric substorms have been found based on the data of vertical sounding stations in Europe and North America. Maximal attention has been paid to the positive peaks of ΔfoF2 with a duration of 6–8 h before the beginning of the substorm expansion phase (T ₀). The possible physical mechanisms by which these peaks are formed (related to the impact of fast particles in the foreshock region of the solar wind on the Earth’s magnetosphere and different for middle and high latitudes) have been considered. The positive peaks of ΔfoF2 can be used in a short-term prediction of the ionospheric disturbance onset and space weather on the whole.     

Relation of the <Emphasis Type="Italic">F</Emphasis>2 region to stratospheric parameters: Dependence on magnetic activity in two solar cycles

The dependence of the correlation coefficient r(h, fo) between the stratospheric parameter h(100) and critical frequency foF2 revealed in the data of two solar cycles (1979–1989 and 1990–2000) on geomagnetic activity is analyzed. It is shown that the character of the r(h, fo) dependence on limitation on the Ap geomagnetic index is the same in both cycles but depends on the time of day and solar activity level for the given year. It is also found that there is a considerable difference in the absolute values of r(h, fo) between two cycles.     

Long-Term Trends in the Critical Frequency of the E-layer

A search for trends k(foE) in the critical frequency of the ionospheric E layer at Juliusruh and Slough stations is performed by the method often used by the authors to analyze trends in the F2-layer parameters. It is found that k(foE) could differ in both magnitude and even sign within different time intervals. However, the k(foE) trends have been stably negative over the last two decades for both stations and all months of the year. The k(foE) values averaged over a year are ?0.012 and ?0.005 MHz per year for Juliusruh and Slough stations, respectively. The method used in the recent paper by La?tovi?ka et al. (2016) to determine foE trends is analyzed, and it is shown that the difference in linear approximation of the dependence of the observed foE values on F10.7 within different time intervals could be interpreted not as the presence of a different foE dependence on the F10.7 index within these intervals but as the presence within them of foE trends that change the slope of the linear approximation.     

Diurnal and Seasonal Variations in Trends in the <Emphasis Type="Italic">E</Emphasis>-Layer Critical Frequency

Long-term changes in the E-layer critical frequency foE at three stations in the European region (Juliusruh, Slough, and Rome) are analyzed by the method described in detail in the previous paper by the authors. It is found that two former stations demonstrate a well-pronounced change in foE (a trend) during the two previous decades. At the same time, the same features of the behavior of the aforementioned trend k(foE) are obtained for both stations. The trend is positive and negative in the morning and evening hours, respectively. It is minimal near the local noon. That explains the small value of k(foE) obtained in the previous paper for 1200 LT. A well-pronounced seasonal behavior of k(foE) is detected: the trend is minimal and maximal in the summer period and at the end of fall—beginning of winter, respectively. The trend maximal amplitude in the morning hours reaches +0.04 MHz per year, whereas the minimal amplitude in evening hours is–0.06 MHz per year. No systematic changes in foE exceeding 0.01 MHz in magnitude per year are found for Rome station.     

Spatial and time variations in the <Emphasis Type="Italic">foF</Emphasis>2 (night)/<Emphasis Type="Italic">foF</Emphasis>2 (day) ratio: Specification of some effects

A series of important aspects of the time and spatial variations in the critical frequency ratio fo(night)/fo(day) considered by one of the authors [Danilov, 2007, 2008] is specified. The list of the Eastern-Hemisphere stations, for which an analysis of the above indicated ratio was performed, is completed. The available stations of the Western Hemisphere are considered. It is shown that the character of the variations in the fo(night)/fo(day) ratio is independent of limitations imposed on the Ap index of magnetic activity for the analyzed days. The consideration of the long-term variations in geomagnetic activity using the annual mean value of the Ap index does not influence the principal conclusions of the study, although in some cases changes statistical characteristics of the obtained behavior of the fo(night)/fo(day) ratio after 1980. A comparison of the results, based on the interpretation of the trends of the fo(night)/fo(day) value with the changes in the stratosphere-ionosphere coupling from the 1980s to the 1990s obtained by the authors earlier, confirms the assumption that there occurs a systematic change in the zonal wind in the upper atmosphere.     

Comparison of Ionospheric Parameters during Similar Geomagnetic Storms

The degree of closeness of ionospheric parameters during one magnetic storm and of the same parameters during another, similar, storm is estimated. Overall, four storms—two pairs of storms close in structure and appearance according to recording of the magnetic field Х-component—were analyzed. The examination was based on data from Sodankyla observatory (Finland). The f-graphs of the ionospheric vertical sounding, magnetometer data, and riometer data on absorption were used. The main results are as follows. The values of the critical frequencies foF2, foF1, and foE for different but similar magnetic storms differ insignificantly. In the daytime, the difference is on average 6% (from 0 to 11.1%) for all ionospheric layers. In the nighttime conditions, the difference for foF2 is 4%. The nighttime values of foEs differ on average by 20%. These estimates potentially make it possible to forecast ionospheric parameters for a particular storm.     

Relation of the <Emphasis Type="Italic">F</Emphasis> region to stratospheric parameters: Possible relation to the atmospheric dynamics

Time variations in the correlation coefficient r(h, fo) between the h(100) stratospheric parameter and the foF2 ionospheric parameter for more than two solar activity cycles (1979–2004) are considered. It is obtained that, for daytime values of the correlation coefficient r(h, fo) when all available years are considered, the absolute value of r(h, fo) decreases in time; i.e., the correlation weakening from the 1980s to the 1990s. This weakening is interpreted as a weakening of the eastward winds in the MLT region of the atmosphere, where presumably a filtration of internal waves providing interrelation of various atmospheric layers occurs.