Structure of the High-Latitude Noon Ionosphere of the Southern Hemisphere

Key points:

• The structure of the southern winter noon ionosphere was studied using Ne measurements onboard the CHAMP satellite.
• Using the auroral precipitation model, the trough associated with the cusp, the main trough, and the Ne peak were identified.
• It is shown that the Ne peak is pronounced at sunlit longitudes, and the main ionization trough is formed only in the shadow.

Abstract

The structure of the winter noon ionosphere of the southern hemisphere was studied. This structure includes the dayside cusp, associated high-latitude ionospheric trough (HLT), main ionospheric trough (MIT), electron density (Ne) peak at latitudes about 70°, mid-latitude ring ionospheric trough (RIT), and low-latitude quasi-trough. Data from the CHAMP satellite in the southern hemisphere for quiet geomagnetic conditions under high solar activity were selected for analysis. The DMSP satellite data and a model of auroral diffuse precipitation were also used. This model represents two zones of auroral diffuse precipitation on the equatorward and poleward edges of the auroral oval. It is shown that the situation in the winter noon ionosphere of the southern hemisphere depends cardinally on longitude. At sunlit longitudes, only the HLT is observed, and MIT is formed in the shadow region. At intermediate longitudes, both troughs can be observed and, therefore, there is a problem of their separation. The positions of all structures of the ionosphere depend on the longitude; in particular, the positions of the daytime MIT are changed by 6°−7°. At latitudes of the dayside cusp, both the peak and the minimum of Ne can be observed. A high and narrow peak of Ne is regularly recorded in the CHAMP data at latitudes of the equatorward zone of diffuse precipitation (68°−72°). In the shadow region, this peak forms the MIT poleward wall, and at sunlit longitudes a quasi-trough equatorward of this peak is sometimes observed. The RIT is rarely formed during the day, only at the American and Atlantic longitudes.

1. Introduction

This article continues a series of studies on the irregular structure of the winter high-latitude ionosphere [1,2,3]. The main purpose of these studies is the separation and classification of ionospheric troughs. First of all, this applies to the identification of the high-latitude trough (HLT), which is located inside the auroral oval [4], and the main ionospheric trough (MIT), which is a subauroral trough because it is located 2°–5° equatorward of the auroral oval [5]. The problem of separating the MIT and HLT arises because the position of the equatorward edge of the auroral oval is usually unknown. At mid-latitudes, the MIT is separated from the ring ionospheric trough (RIT), which is regularly formed during even weak geomagnetic disturbances at latitudes of the residual magnetospheric ring current [6,7]. In addition, there are various types of mid-latitude quasi-troughs. Within the framework of this task, the structures of the midnight ionosphere [1], the early morning and evening ionosphere [2], and the late morning ionosphere [3] were studied in detail. The study was performed using an advanced method based on a simple visual model of the auroral oval [8]. This model was developed according to the data from the DMSP series satellites at the Polar Geophysical Institute, Murmansk, and is available on the website http://apm.pgia.ru (accessed on 18 August 2019). The model represents zone I of auroral diffuse precipitation on the equatorward edge of the auroral oval and zone II of diffuse precipitation on its poleward edge. Taylor [9] described observations of the trough poleward wall made using incoherent scatter radar at Malvern in the UK. He found that the poleward wall was tilted along the direction of the local geomagnetic field. Afterwards, Jones et al. [10], from the EISCAT radar data, confirmed that the steep gradient in electron density that forms the poleward edge of the trough accurately aligned with the geomagnetic field. These results demonstrated clearly that the poleward wall is produced by the effects of the precipitation of soft electrons into the ionospheric F-layer. As a rule, zone I forms the poleward wall of the MIT and, similarly, zone II forms the poleward wall of the HLT. This is the key point for separating the MIT and HLT.

To separate the MIT and RIT, a previously developed method of analyzing all geomagnetic disturbances, even weak substorms, was used [6,7]. In particularly complex and confusing cases, separation of the MIT/HLT and MIT/RIT was performed on the basis of a thorough analysis of the dynamics of all troughs over time.

Another specific feature of the developed method is the analysis within the framework of the longitudinal effect because the positions of all ionospheric structures depend on the longitude. This applies both to the positions of the troughs and boundaries of the auroral oval. Unfortunately, longitudinal variations in the position of the auroral oval boundaries were earlier determined only for pre-midnight conditions [11,12]. Therefore, the longitudinal variations of the auroral oval were specified according to the DMSP satellite data in the previous study [3]. As expected, they are determined by the geomagnetic dipole tilt angle. This paper is the latest in the series of studies devoted to the separation and classification of the troughs, which examines the most difficult situation for analysis of the noon ionosphere in the southern hemisphere. This difficulty is due, first, to the presence of the dayside cusp. Second, the satellite, even with high orbit inclination, does not reach the cusp and its associated structures at most of the longitudes. Third, the terminator divides the winter ionosphere in the southern hemisphere into sunlit and unlit parts, with a completely different ionospheric background.

The dayside cusp is the main structure of the noon ionosphere. The cusp is a funnel in the geomagnetic field, where particles of the solar wind fall and then precipitate into the ionosphere [13]. The position of the cusp depends on the dipole tilt angle, which changes with the season [14]. During magnetically quiet periods in winter, the cusp is located at latitudes of 77°–78° [14,15]. Slightly equatorward of the cusp, the precipitation from the low-latitude boundary layer is observed, and poleward of the cusp, the precipitation from the mantle occurs. On the morning and evening side, they are complemented by precipitation from the boundary plasma sheet. As a result, a high-latitude band of precipitation is formed, which occupies a local time interval of 09–15 LT or even wider [16]. Therefore, this combined structure is sometimes defined as a cleft. The width of the cleft is 1°–3° [15]. The results of long-term observations of the cusp/cleft are summarized in [17].

The low-energy fraction of precipitation (<1 keV) in the cleft forms a plasma peak at ionospheric heights. The plasma peak in the daytime combined with the peak at night forms a plasma ring, which was first observed from the Alouette 1 satellite data [18]. Sato and Colin [19] and Chan and Colin [20], also according to Alouette 1, distinguished two fractions of the plasma ring: high-latitude and low-latitude fractions. On the dayside, under quiet geomagnetic conditions, the high-latitude fraction corresponds to 75°–80° cusp latitudes and the low-latitude fraction occupies 68°–72° geomagnetic latitudes. However, only the high-latitude peak was usually highlighted and considered [21,22,23,24].

Equatorward of the cusp by 2°–4°, i.e., at latitudes ≥ 70° under quiet geomagnetic conditions, a high-latitude ionization trough has been observed [24,25,26,27,28,29,30]. In most studies, this trough is presented as a single branch combined with the evening/night MIT [5,31,32,33,34,35,36,37]. Therefore, the HLT was usually defined as a main or mid-latitude trough. No one has paid attention to this obvious contradiction. During observations using the Millstone Hill radar in the 1971–1973 winter, daytime troughs were recorded at latitudes below 68° GMLat, but only under high geomagnetic activity up to Kp = 7+ [38]; thus, it is difficult to determine what type they are. All the above-mentioned observations were made in the northern hemisphere. Only in the winter southern hemisphere, according to Ariel 3 data, was the trough in the afternoon observed not only at high latitudes but also at latitudes of 62°–68°, which correspond to the MIT [29,39]. Furthermore, the observations of daytime winter trough according to DMSP data in the southern hemisphere were directly interpreted as the formation of the MIT because of plasma stagnation in the shadow region [40]. However, these observations were performed at 13.5–16.4 LT [39] and ~09 LT [40]; midday conditions in the winter southern hemisphere were never considered.

Thus, in the winter southern hemisphere, in contrast to in the northern hemisphere, there are undoubtedly both troughs: the daytime MIT and HLT associated with the cusp. However, the question of separating these troughs has never been raised. This is what piqued our interest in the winter southern hemisphere. The main objective of this study is the clear separation of the MIT and HLT in the noon ionosphere of the southern hemisphere. This problem is solved through an analysis of the latitudinal/longitudinal distribution of the following structures of the midday winter ionosphere: the peak or minimum of ionization at the dayside cusp latitudes (i.e., at latitudes of zone II precipitation), the HLT associated with the cusp, the MIT, the RIT, the ionization peak at latitudes of zone I, and the quasi-trough equatorward of this peak.

2. Materials and Methods

2.1. CHAMP Satellite Data

The CHAMP satellite conducted in situ measurements of electron density (Ne). Variations in Ne are presented below in terms of plasma frequency fp (Ne[cm⁻³] = 1.24·10⁴fp²[MHz]). The CHAMP was placed on a nearly polar orbit with the inclination of 87°. The orbit altitude changed in the period of 2000–2002 from ~465 km to ~405 km, which is close to the height of the F2 layer maximum. High inclination makes it possible to record the dayside cusp and the daytime trough, although not at all longitudes, as will be seen below. The CHAMP data time resolution of 15 s is less than 1° of latitude, which allows determining the fine structure of the ionosphere. To analyze the structure of the noon (12 ± 1 LT) ionosphere, the satellite data in the southern hemisphere for high solar activity for the period 28−31 July and 2−25 August 2000, as well as 1−29 June 2002, were selected. Data were used only for relatively quiet geomagnetic conditions Kp ≤ 4. The CHAMP data are available on the website http://op.gfz-potsdam.de/champ (accessed on 15 January 2009).

2.2. DMSP Satellite Data

Analysis of the structure of the noon ionosphere is considerably simplified with the presence of simultaneous data on electron density and soft particle precipitation. In principle, such data are supplied by DMSP series satellites. The inclination of 97° of these satellites allows high-latitude troughs to be recorded, although not at all longitudes. Unfortunately, all satellites of this series are oriented along the meridians of 06–18 LT or 09–21 LT and are generally not suitable for cusp monitoring. During the orbit transition from sector 06 LT (09 LT) to sector 18 LT (21 LT), the satellite appears in the noon sector. However, this occurs when the satellite moves along the orbit at extremely high latitudes and when the local time (and longitude) changes very quickly. Moreover, the dayside cusp latitudes are reached by the satellite only at longitudes of the eastern hemisphere. Therefore, observations of the structure of the dayside cusp from the DMSP data are very limited. Nevertheless, at certain longitudes, it was possible to select several clear examples of observations in the cusp region. Figure 1a shows a classic example recorded onboard the DMSP F15 satellite on 5 June 2001. The satellite data were copied from the website https://dmsp.bc.edu/ (accessed on 15 October 2022). Figure 1a shows that the precipitation of soft particles from the cusp at latitudes of 76°–79° formed a plasma peak, equatorward of which there is a well-expressed trough. The observations belong to ~09 LT and were recorded in the longitudinal sector ~170°E, i.e., in the sunlit region of the ionosphere. The upper panel in Figure 1a shows the latitudinal fp cross-section recorded onboard the CHAMP on 11 June 2002 in the longitudinal sector 70°E. This plot depicts the peak and trough ionization similarly to DMSP data. Both latitudinal plasma density profiles have the classical form described in many studies; however, according to CHAMP data, a steeper equatorward wall is much more typical. Plasma irregularities observed on the equatorward wall of the trough are most likely associated with diffuse precipitation of zone I.

In almost all cited studies, the cusp is associated with the ionization peak. Only in one study [41], according to Alouette 1 data for the 1963/64 winter for 11–13 LT at altitude of 1000 km and in the longitudinal sector 30°–180°W, electron density minima were recorded along with peaks. As a result, when averaging the data, a smooth latitudinal Ne profile was obtained. According to the DMSP satellite, both the ionization peak and minimum are also observed at the cusp latitudes. Figure 1b shows the DMSP F15 measurements for 26 June 2001, which show a deep Ne minimum in the dayside cusp region. The data were obtained in the sunlit longitudinal sector ~150°E. The Ne minimum is probably associated with the effect of both horizontal and vertical drift caused by electric field action. The upper panel in Figure 1b shows a similar Ne profile obtained from CHAMP data on 3 August 2002 at longitude of 135°E. Its interpretation caused great difficulties but turned out to be simple when compared with DMSP data. There have been many such cases; thus, the data from the DMSP satellites considerably assist in interpretation of complex cases observed from the CHAMP satellite data.

3. Location of Winter Noon Ionosphere Structures

Figure 2 shows the location of the structures of the noon ionosphere: diffuse auroral precipitation, the Ne minimum at latitudes of the cusp, HLT, MIT, Ne peak, and low-latitude trough. The CHAMP data were used for 11–13 LT, Kp ≤ 4. The localization is presented in terms of the longitudinal effect because the pattern of the daytime ionosphere greatly varies with longitude. A solid thin curve shows the satellite inclination in terms of geomagnetic latitude. Even with a large inclination of the orbit, observations at high geomagnetic latitudes are severely limited at most longitudes. The terminator divides the daytime winter ionosphere of the southern hemisphere into illuminated and unlit parts.

Zones I and II of auroral diffuse precipitation of low-energy particles were taken for 12 LT and Kp = 2 from the model of the Polar Geophysical Institute [8]. The positions of both zones change with longitude by 2.5°. The longitudinal variations in the position of the precipitation zones in the southern hemisphere were identified from the TIMED data for pre-midnight hours [10] and from DMSP data for morning hours [3]. These variations are consistent with changes in the dipole tilt angle; therefore, they are used in Figure 2 for the analysis of daytime conditions as well. Because the precipitation model was created by taking into account the previous results, it agrees well with them. In particular, the average value of the equatorward boundary of zone I of auroral diffuse precipitation (68°) is consistent with the systematics of precipitation in [42]. Similarly, the position of the equatorward boundary of zone II (76°) corresponds to the average position of the equatorward boundary of the dayside cusp, as analysis of the above-mentioned references shows. Of note, in this case we mean a cleft because the model does not separate precipitation in the cusp from precipitation from the mantle and low-latitude boundary layer.

In order to eliminate the dependence on geomagnetic activity, so that all structures correspond to the position of the auroral oval, they were also reduced to Kp = 2 according to simple formulas:

Λ_cusp = −79.3° + 0.4°Kp for the Ne minimum in the cusp,

Λ_HLT = −76.4° + 0.9°Kp for HLT,

Λ_peak = −74.1° + 1.7°Kp for Ne peak,

Λ_MIT = −70.2° + 0.6°Kp for MIT.

The HLT at 12 LT according to ISIS 1 and Injun 5 data for Kp ≤ 3 was at latitude of 73° [5], according to OGO 6 data for Kp ≤ 3 at latitude of 77.5° [4], according to DE 2 data at ~78° (Kp value was not indicated) [35], according to tomography data for Kp ≤ 2 at latitude of ~75° [36], and according to NNSS satellite data (TEC) for 2_- ≤ Kp ≤ 3_- at latitude of 70.5° [33]. As can be seen, the spread of values is quite large, which requires explanation. Note also that the values of the coefficients for the Kp in the formulas are much smaller than those in other experiments. For example, from the Alouette 1 data for the cusp, Titheridge derived the formula 80°–2°Kp [24], and according to AEROS-B for the daytime trough, Neske deduced the formula 78°–2.4°Kp [27]. The reason for such a large difference is unclear; this problem has existed for a very long time and requires detailed analysis.

Figure 2 shows that the MIT is observed mainly in the shadow area. This is practically an expected result, but it has never been as obvious as in Figure 2. It is also seen that the MIT poleward wall is formed by diffuse precipitation of zone I. The inclination of the satellite does not interfere with the recording of the MIT; therefore, an approximating curve (a polynomial of the 5th degree) is plotted for all longitudes, although there is no MIT at sunlit longitudes. The amplitude of the longitudinal effect in the MIT position according to the approximating curve is ~7°. Because the MIT position strongly depends on the longitude, the average latitude of −70.2° for Kp = 2 can be only used for a rough comparison with other observations. However, currently, there are no such observations, e.g., in the only study of the daytime MIT [39], the Ariel 3 data were considered only for the afternoon hours of 13.5–16.4 LT.

An example of a classical MIT is shown in Figure 3a. The MIT was determined by the presence of a fairly well-defined Ne minimum equatorward of the poleward wall, which was located at latitudes of zone I of diffuse precipitation. At the longitudes of America, under low geomagnetic activity, the trough shifted to the pole, and the poleward wall was not reached by the satellite. Such structures were not considered.

The inclination of the satellite does not allow recording of high-latitude troughs associated with the cusp at a large interval of longitudes. Therefore, the approximating curve (a 3rd degree polynomial) for the HLT is given only in the longitude interval of 60°E–210°E. The average position of −76.4° for Kp = 2 corresponds well to the equatorward boundary of the dayside cusp. There were no large problems with determining the HLT at sunlit longitudes. The electron density at these longitudes usually sharply drops toward high latitudes, and if the formed Ne minimum was accompanied by at least a small Ne peak at the cusp latitudes, such a structure was defined as the HLT. A pronounced example of the HLT is shown in Figure 3b. It was recorded on 5 June 2002 in the longitudinal sector 135°E. Thus, the number of HLTs in Figure 2 reflects the number of Ne peaks at the cusp latitudes.

Figure 3b shows a well-expressed but atypical example of an HLT. More often, according to CHAMP data, the structure shown in Figure 3c is observed at the sunlit longitudes. It is characterized by a high and narrow Ne peak at about 70° latitude. The electron density sharply decreases toward high latitudes, forming the HLT. The HLT poleward wall at the latitudes of the cusp is more weakly expressed than that in Figure 3b. Figure 3d shows a structure that looks similar to the structure in Figure 3c. However, the Ne minimum is now reached at the latitudes of the cusp. If the Ne peak in the cusp was not formed and a minimum was recorded, then such a structure was defined as the trough in the cusp, as well as the structure in Figure 1b. It is indicated in Figure 2 by a red triangle. The ratio of the number of occurrences of peaks to the number of trough occurrences in the cusp in the longitudinal interval of 75°–180° was estimated to be 55:45; thus, the trough in the cusp is observed not significantly less frequently than the peak. The same situation was described earlier according to the Alouette 1 data [41]. However, until now, only the Ne peak has usually been associated with the cusp (see, for example, [43]).

The red asterisks in Figure 2 indicate the ring ionospheric troughs. The RIT is associated with the precipitation of hot particles from the magnetospheric ring current [6,7]. Figure 3e shows an example of the simultaneous existence of the MIT and RIT. The latitudinal fp profile was recorded on 10 June 2002 in the longitudinal sector 326°E. The RIT is often formed even after a weak increase in geomagnetic activity; in this case after Kp = 3₊. However, it is rarely observed during the day and only at American and Atlantic longitudes because the magnetic field is weak here, hot ion precipitation from the magnetospheric ring current is intense, and a deep trough is formed. The RIT is located far away from the auroral oval and the MIT. Figure 2 shows that the RIT during the day at latitudes of 61–64° is observed. In the recent study [44], these latitudes at 10–14 MLT in the southern hemisphere correspond to the MIT, while in the northern hemisphere a more adequate estimate of ~65° was obtained.

Clearly, the separation of the troughs at brightly sunlit longitudes and in the deep shadow region does not cause many problems because only the HLT is observed at the sunlit longitudes, and the MIT is mainly observed in the shade. Questions arise for intermediate longitudes where both troughs can exist. Figure 3f shows two examples of simultaneous observation of the MIT and HLT. In the first case, which was recorded on 14 June 2002, the troughs are clearly separated according to the developed method: the poleward wall of the MIT is formed by diffuse precipitation of zone I, and the HLT poleward wall is produced by precipitation in the cusp. The second example was recorded at 9:36 UT on 14 June 2002 in almost the same longitudinal sector. In a brief analysis, the Ne minimum at latitude of ~74° was assigned to the MIT, and Ne minimum at ~77° was assigned to the HLT. However, after careful analysis, the weakly expressed trough was found to be located at the equatorward border of zone I precipitation at the latitude of −68°, which is more typical for the MIT, and the minimum at the latitude of −74° refers to the HLT. This example once again underlines the complexity of the troughs’ separation problem.

4. Discussion

Figure 3 shows a maximum and four peaks of Ne at moderately high latitudes near 70°. The two peaks in Figure 3a,e refer to the region of a deep shadow, and the two peaks in Figure 3c,d and the maximum in Figure 3b refer to the sunlit longitudes. The Ne peak in the shadow area forms the poleward wall of the MIT. The MIT as a structure is determined by the poleward wall and a rather deep minimum. Clearly, this minimum is formed as a result of the decay of ionization in the absence of solar radiation. This was shown previously in [40]. Thus, the reasons for the formation of the daytime MIT are the same as for the night MIT.

The mechanism of HLT formation has not been fully investigated. It is usually assumed that it is associated with the westward plasma drift. If the drift is strong, then the plasma heats up due to collisions, recombination increases, and the trough is formed (see, for example, [45]). If the drift is weak at the edge of the evening convection cell, then it carries ionospheric plasma with a reduced density from the night side (see, for example, [34,36]). The influence of field-aligned currents is also possible. The incoming field-aligned current was observed in the minimum of a high-latitude daytime trough according to data from the HILAT satellite [45]. Such a current causes a decrease in the electron density [46,47].

The Ne peak at the sunlit longitudes is formed on the high background ionization level as a separate structure. In the absence of this peak, the background density would fall monotonically towards high latitudes, as shown by the dashed curve in Figure 3c. Therefore, the Ne minimum created by this peak is a fictive structure. It can be called a “quasi-trough”. All cases of well-defined quasi-troughs are shown in Figure 2. This is done because they can be confused with MITs during a cursory analysis, as experience has shown.

All Ne peaks are similar in that they are located at latitudes of zone I of diffuse precipitation. Noticeable Ne peaks at these latitudes were observed in December 1963 onboard the Alouette 1 satellite [20]. This peak is more often observed at night and forms a single low-latitude plasma ring at latitudes near 70°. In the CHAMP data, pronounced Ne peaks are observed in a wide range of noon and afternoon hours. Thus, there is not one but two plasma rings: a high-latitude one, which is observed during the day at the latitudes of zone II (75°–80°), and a low-latitude one at the latitudes of zone I (68°–73°). Clearly, the Ne peaks are associated with the precipitation of low-energy auroral particles. However, although everything is clear regarding the plasma peak at the latitudes of the dayside cusp, the connection of the second peak with precipitation is not as obvious. For proof, we turn again to DMSP data. The most pronounced example of a low-latitude peak was recorded onboard the DMSP F18 satellite in the winter northern hemisphere on 3 January 2022. These data are compared with CHAMP data (from Figure 3c) in Figure 4. The Ne peak at the moderately high latitudes is formed by the diffuse precipitation of zone I, which was to be proved. A weak plasma drift does not participate in the formation of this peak in this case; however, we can assume that the drift operates in the case of a wide Ne maximum, as in Figure 3b.

The data analysis shows that the latitudinal profiles of the electron density in the daytime winter ionosphere of the southern hemisphere vary greatly with longitude. A characteristic Ne profile can be chosen for each longitude sector. The schematic pattern of the longitudinal effect is depicted in Figure 5, which shows the most characteristic latitudinal fp profiles selected for August 2000 and June 2002.

The upper latitudinal fp profile was recorded on 18 June 2002 in the longitudinal sector 25°E. It represents a typical example of the MIT for the shadow region, with the poleward wall formed by the precipitation of zone I. The next latitudinal profile was recorded on 13 June 2002 at a longitude of 30°E, i.e., also in the shadow. Under these conditions, the MIT and HLT can be simultaneously observed. Latitudinal profiles in the shadow region are characterized by a monotonic decrease in electron density to high latitudes. The next latitudinal profile was recorded on 2 June 2002 in the longitudinal sector 105°E. At the sunlit longitudes, solar radiation creates a high level of background ionization up to 8–9 MHz. Diffuse auroral precipitation is superimposed on this background, so that the maximum of electron density is formed at latitudes of zone I. Then, the density drops sharply toward high latitudes, forming a steep HLT equatorward wall. The poleward wall of the HLT is formed by the precipitation in the dayside cusp. A typical latitudinal profile of the electron density is formed, which is regularly observed in the CHAMP data at the sunlit longitudes. The Ne maximum at the latitudes of zone I has often been observed as a narrow and high peak, as shown by the fp profiles at the longitudes of 133°E and 184°E. The profile at the longitude of 133°E was recorded on 9 June 2002. The electron density in this case decreases toward high latitudes, forming a minimum at the cusp latitudes. These cases, which are also quite common, are indicated in Figure 2 by triangles. The latitudinal Ne profile in the longitudinal sector 184°E was recorded on 9 August 2000. On the contrary, it is characterized by the presence of the classical Ne peak at the cusp latitudes. In addition, at the latitude of ~58°, there is a Ne minimum, which is sometimes expressed as a rather deep quasi-trough (as on Figure 3c). When moving from the sunlit longitudes to the shadow region, the latitudinal profile with the maximum at latitudes of zone I is formed again. Such a profile was observed on 15 August 2000 in the longitudinal sector 221°E. When moving further into the depths of the shadow, the electron density at mid-latitudes decays, and the MIT is formed, as at the longitude of 241°E. The MIT is most pronounced and most often observed in the region of the deepest shadow, i.e., at the longitudes of America. At these longitudes, the poleward wall of the MIT often manifests as a high and narrow Ne peak. This is also reflected in Figure 2. At the longitudes of America, the satellite inclination does not allow recording a high-latitude trough; however, clearly, the HLT exists at all longitudes. It is schematically depicted on the lowest latitudinal fp profile in Figure 5.

5. Conclusions

The data from the CHAMP and DMSP satellites allowed us to construct a typical pattern of the midday winter ionosphere of the southern hemisphere. It is schematically shown in Figure 5. The specifics of this pattern are the large difference between the geomagnetic and geographical poles. This determines the dramatic difference between the sunlit and shaded winter ionosphere in the southern hemisphere. In the western hemisphere, especially in the region of deep shadow at American longitudes, the background electron density is low and monotonously decreases toward high latitudes. At this low background level, the MIT is formed, often with a well-defined poleward wall in the form of a pronounced Ne peak. Neither the cusp nor the HLT are recorded at these longitudes, because even a polar orbit does not reach high geomagnetic latitudes. These are the latitudes of Antarctica.

At the sunlit longitudes of the eastern hemisphere, the background density is high due to solar radiation. At this high background density, a maximum is formed, often as a narrow and high Ne peak at the latitudes of zone I of diffuse precipitation. Then, the electron density drops sharply toward high latitudes, and the HLT is formed with the poleward wall created by precipitation in zone II, i.e., precipitation in the dayside cusp. However, at the latitudes of the cusp, the Ne minimum is frequently observed. Thus, at the sunlit longitudes, the Ne maximum, or a pronounced peak, the HLT or a minimum in the cusp are observed. The MIT is not formed at sunlit latitudes. It is worth reminding that we are discussing conditions of high solar activity.

At the intermediate longitudes, both troughs can be observed, and there can be a problem of their identification, i.e., the problem of separation. This problem was solved by means of the previously developed method: the poleward wall of the MIT should be formed by the precipitation of zone I, and the HLT poleward wall should correspond to the latitudes of zone II.

The pattern of polar plasma peaks presented in the studies [19,20] is fully confirmed. According to this pattern, there are two plasma rings: low-latitude and high-latitude. These rings are formed by the precipitation of zone I on the equatorward edge of the auroral oval and precipitation of zone II on its poleward edge, as determined by the model [8]. During the day, under quiet geomagnetic conditions, zone I occupies the latitudes of 68°–73°, and zone II is located at latitudes of 75°–80°, i.e., at the cusp latitudes. However, as mentioned above, at the cusp latitudes, not a peak, but a minimum, of Ne can be observed.

Some minor features of the ionosphere were also revealed. These include the formation of a quasi-trough equatorward of the Ne peak at the sunlit latitudes. With a cursory analysis, it can be confused with the MIT. At the longitudes of the deep shadow (America and Atlantic), the RIT is sometimes recorded together with the MIT. The RIT is formed most often in the morning, but sometimes it can be observed in the daytime.

Thus, the specifics of the structure of the daytime winter ionosphere in the southern hemisphere are determined by the large difference between the geographic and geomagnetic poles. This difference is much smaller in the northern hemisphere; therefore, the pattern of the daytime ionosphere in the northern hemisphere will obviously be very different. The pattern will also be different under low solar activity.