Day-to-Day Variability of H and Z Components of the Geomagnetic Field at the African Longitudes

The Day-to-day variability of the geomagnetic field elements at the African longitudes has been studied for the year 1987 using geomagnetic data obtained from four different African observatories. The analysis was carried out on solar quiet days using hourly values of the Horizontal, H, and vertical, Z, geomagnetic field values. The results of this study confirm that Sq is a very changeable phenomenon, with a strong day-to-day variation. This day-to-day variation is seen to be superimposed on magnetic disturbances of a magnetospheric origin.


Introduction
Changes in the magnetic environment of the Earth are of interest to those studying space weather and climate change, particularly in the upper atmosphere.The upper atmosphere is ionized by the Sun's ultraviolet and X-radiation to create the ionosphere, and the free ions and electrons are moved by winds arising from the heating effects of the Sun.The currents in the ionosphere have magnetic effect on the ground and are monitored using magnetometers on the Earth surface.The records of any observatory show that on some days there are regular variations on the magnetic record while on other days the variation is irregular.The daily variations of the geomagnetic field when solar-terrestrial disturbances are absent are called solar quiet (Sq) variations [1].These Sq variations are caused mainly by electrical currents in the upper atmosphere, at altitudes of about 110 km above the Earth surface [2].Studies on solar quiet daily variation of the Earth's magnetic field show that Sq on one day can be different from Sq of the next day in amplitude, phase, and focal latitude [3][4][5].This change in Sq between two adjacent days is the day-to-day variability in Sq between the two days.This day-to-day variability has been highly attributed to changes in the ionospheric dynamo currents, which depend on the ionospheric conductivity and tidal winds, varying with solar radiation and ionospheric conditions [6,7].
Hasegawa [8] examined the day-to-day changes of the quiet day variation and the ionospheric current systems for the second polar year and suggested that some or all the dayto-day variabilities in solar quiet daily variation (Sq) are due to variability in the positions of the foci of the ionospheric current systems rather than changes in the distribution of ionization and conductivity.Studies conducted by [9][10][11][12][13][14][15][16] clearly showed that the variability of Sq occurred at all hours of the day.
Rabiu et al. [17] from their comprehensive study of Sq day-to-day variability at Addis Ababa, an equatorial electrojet station, found out that the daytime (0700-2000 hours) magnitudes of Sq  (horizontal component of the earth's magnetic field) and Z (vertical component of the earth's magnetic field) were greater than the nighttime (2000-0700 hours) for all the months they studied.They also found that the day-to-day variability peaks during the daytime mostly around the local noon within the range of 1000-1400 hours for all the months in the two Sq elements,  and .Their findings are therefore in agreement with the diurnal variation pattern of Sq in the earlier works of [18,19] which showed that the maximum intensity of Sq occurs around local noon.The diurnal variation of day-today variability, which followed the variation pattern of Sq, can be attributed to the variability of the ionospheric process and physical structures such as conductivity and wind structures,   which are responsible for Sq variation.Studying the Sq variability in Indian equatorial electrojet sector, Okeke et al. [13] noted that changes in the electric field control the phase and randomness of the variabilities, while the magnitude of the ionospheric conductivity controls the magnitude of the variabilities.Campbell [20], Rabiu [21], and Obiekezie and Obiadazie [22] found consistent nighttime variation in the horizontal magnetic field component at midlatitudes and attributed same to distant magnetospheric sources.
The similarities between the diurnal variation patterns of day-to-day variability and Sq in the earlier works of many researchers such as Onwumechili [18], Matsushita [19], and Fambitakoye and Mayaud [23] suggested that the root cause of Sq may also explains the day-to-day variability effect.Several causes of Sq have been identified which can thus explain the day-to-day variability.Onwumechili and Ezema [24] concluded that the diurnal variation of Sq of  arises in daytime which is consistent with atmospheric dynamo theory of the geomagnetic daily variation.
Accurate determination of Sq variability has found applications in the determination of the Earth's electrical conductivity, in determining the baselines for quantifying of magnetospheric disturbances, and in improving the satellite main-field modeling.Although so much work has been done on the Sq variability, Hibberd [25] noted that the phenomenon of day-to-day variability is poorly understood.Thus, monitoring the day-to-day variability could provide very important contributions to the understanding of the atmospheric dynamics affecting the Sq variability.The aim of this paper therefore is to analyze the day-to-day variability in the  and  geomagnetic field elements at four different locations along the African longitudes and also deduce the mechanisms responsible for the observed variations and variabilities.The result of this research is expected to add to the few existing ones in the African region thus bridging the gap between Africa and the other parts of the world where much work had been done.

Data and Method of Analysis
A set of observatories located in the East, West, Central, and Southern Africa supplied the dataset used in this work.The geographic and geomagnetic locations of the stations can be seen in Table 1.
Magnetically quiet days from ten internationally quiet days (IQDs) in each month for the year 1987 were selected.The ten International Quiet Days (IQD) are the ten quietest days of the month according to the classification of planetary magnetic index, Kp.The local time (LT) for all the four stations was employed throughout the analysis.The baseline values (  and   ) were calculated as the average of the values of the hours flanking the midnight plus the midnight values where  00 ,  00 ,  01 ,  01 , where  = 0 to 23 hours.The hourly departures were then corrected for noncyclic variation (Δ  ).This noncyclic variation (Δ  ) defined as a phenomenon in which the value at 00LT is different from the value at 23LT, [26,27] The linearly adjusted values at the hours are In other words, where  = 0 to 23 hrs and  can be either  or .
The hourly departures corrected for noncyclic variation on quiet days give the solar quiet daily variation in  and  denoted as Sq() and Sq().
The variabilities of these hourly amplitudes for the hour  from the day  to the next day  + 1 for all hours of the day are

Results and Discussion
Figures 1 and 2 show the monthly variations in the Sq() and Sq(), respectively, for the four stations.It is evident from Figure 1 that the amplitude of Sq() in AAB is higher than that in all other stations.The amplitude curves present the same shape for AAB, BANG, and MBR but different from HMN.The morphology of the curves for AAB, BANG, and MBR shows a regular increase in amplitude in the morning hours which picks around 8.00 hrs for AAB, 10.00 hrs for BANG, and 12.00 hrs for MBR and a gradual decrease from the peak value down to the night value.A kind of phase difference is observed between AAB, BANG, and MBR.This phase shift may be attributed to the differences in their latitudinal locations.AAB, BANG, and MBR stations are located in the equatorial region, AAB (0.18 ∘ dip latitude) is located within the equatorial electrojet (EEJ) zone; thus, from Figure 1 the amplitudes of AAB is seen to be higher than Bang, and MBR Sq() amplitudes.These high amplitudes could be as a result of influence of the Equatorial electrojet current.The EEJ current is an east-west current which is seen flowing positive in the morning thus causing an enhancement in the Sq() values of stations within the EEJ region.
The morphology of the amplitude curve for HMN which is seen different from the other stations is seen to be having a minimum when the others are having a maximum.H amplitude in HMN decreases from dawn to about noon, when a minimum is achieved; it later increases up to dusk.This variance could be attributed to the hemispherical difference between HMN and the other three stations.HMN is in the southern hemisphere while the other three stations are on the northern hemisphere.It has been established that the regular daily variations are mainly caused by electric currents flowing at approximately 100 km altitude in the ionosphere (external source current).The ionospheric currents typically form two global horizontal current vortices at the sunlit side of the Earth, one flowing clockwise in the southern hemisphere and the other flowing counterclockwise in the northern hemisphere.
The amplitude curves for Sq() are seen to be conspicuously opposite that of Sq() for MBR and HMN stations.The  variation at Mbour decreases from morning hours at about sunrise, a minimum at about local noon, and a gentle rise towards sunset period. variations for HMN show a maximum at about noon, given the clockwise sense of the southern hemisphere current circulation.The variations in  at the AAB station show a maximum between the periods from 0800 to 1200 hours LT within December and Feb-August.This result therefore is suggesting that there is a presence of counterelectrojet during these hours.Alex et al. [28] found "" variation to be in phase with the  variation and suggested that it could be the cancellation of EEJ.Maximum values were observed almost in all the months in BANG which suggest an anomaly in the current pattern.This is actually expected to be minimal as it is in line with the counterclockwise sense of the northern hemisphere current circulation.This observed anomaly could be said to be the reversal of the atmospheric dynamo electric field as it is found to occur after sunrise with a peak around local noon when the significant increase in E-region ionization must have been formed.Another possible explanation for this anomalous behavior in Bangui can be found in a combination of factors affecting the induced currentsignals.Obiekezie and Okeke [29] found that, at ground level, the induced currents signals have an opposite sign compared with the external signal.A minimum nocturnal variation is observed in the two components Sq() and Sq() for all the stations.This nighttime variation could be attributed to distant current of nonionospheric origin.This night variation is in line with the earlier works of [14-16, 21, 22, 30-32].Obiekezie and Okeke [16] attributed the observed nighttime variations to currents flowing in the magnetosphere (such as the ring currents) in which most of it filter into the ionosphere at night even during magnetic quiet periods.
The variabilities of Sq hourly amplitudes for the hour,  from the day  to the next day  + 1 for all hours of the day called the day-to-day variability, were investigated only for consecutive IQDs.The day-to-day variabilities of Sq() and Sq() are as shown in Figures 3 and 4. For the four stations, seven consecutive IQDs in the month of September were chosen.The variability occurrence was a dawn to dusk phenomena, although more noticeable in the daytime but turns very mild during the night in both Sq() and Sq() in all the stations.It could be seen from Figures 3 and 4 that the variabilities between two paired consecutive days are quite different from any other two paired subsequent consecutive days.For example, on the 3/4 September, 5/6 September, and 6/7 September, the variations are seen to be remarkably different from one another.Observing this figure on the 3/4 September at AAB, the day-to-day variability    in  was seen to have maximum amplitude of about 10 nT and a minimum amplitude up to −30 nT, while on the 4/5, at the same station, the maximum amplitude was about 40 nT and the minimum was about −37 nT while on the 6/7 the maximum was 37 nT and the minimum was about −40 nT.The maxima and minima were seen to be occurring at different times; thus, there exists phase variations.Significant differences in amplitude as well as in phase can be seen in the other three stations.These amplitude and phase variations are seen not to have a definite pattern; they are seen to be random.This result is in line with Okeke et al. [13] who noted that changes in the electric field control the phase and randomness of the variabilities, while the magnitude of the ionospheric conductivity controls the magnitude of the variabilities.It could be seen from Figures 3 and 4 also that the variabilities were maximum during daylight hours and mild at night suggesting that the root cause of Sq could be responsible for its day-to-day variability.
The observed day-to-day variability in Sq() on 18/19 September followed the exact pattern of variation of Sq() observed in Figure 1.These two days may be said to be very quiet suggesting that the observed randomness on the other consecutive days may actually be due to magnetic disturbances of a magnetospheric origin.Following Okeke et al. [13] who noted that changes in electric field control the phase and randomness, we are suggesting that it is the magnetic disturbances that give rise to the change in the electric field.These magnetic disturbances are actually found to affect the determination of a true Sq variation.
The day-day amplitudes in Sq() at Addis Ababa are found to be small compared to all other stations in all the consecutive days except on 6/7 September.this fact can be related to the observed counterelectrojet in Sq() at Addis Ababa because the widths of both the electrojet and the counterelectrojet are nearly equivalent, thus causing the amplitude to be greatly reduced.This finding is in agreement with the works of [33,34].

Conclusion
The results of this study confirm that Sq is a very changeable phenomenon, with a strong day-to-day variation, and that it is superimposed on magnetic disturbances of a magnetospheric origin.It could be seen from the results that the root cause of Sq could be responsible for its day-to-day variability.It is found that, for the Equatorial Electrojet stations, when the widths of both the electrojet and the counter-electrojet are nearly equivalent, the amplitude of variation is greatly reduced.Since the magnetic disturbances are of magnetospheric origin, monitoring the day-to-day variability could provide very important contributions to the knowledge of the ionospheric dynamics as it could be the key to investigate the magnetospheric/ionospheric interaction which actually affects the determination of the true Sq variation.

Figure 3 :
Figure 3: Day-to-day variations in Sq() for the month of September.

Figure 4 :
Figure 4: Day to day variations in Sq() for the month of September.

Table
: The stations and their locations.
22 ,  22 , and  23 ,  23 , represent the hourly values of  and  at 00, 01, 22, and 23 hours LT, respectively.The Sq amplitude Δ  , Δ  for any hour  is the difference between hourly values   ,   and the baseline value,  0 ,  0 .Thus,