F. N. Okeke and Y. Hamano
Department of Earth and Planetary Physics, Faculty of Science, University of Tokyo, Tokyo, Japan
The abnormal quiet days (AQDs) events have been studied for many years by various workers. The form and causes of these AQDs events have been studied by Brown  and Butcher . Sastri  and Sastri and Murthy  found that occurrence of an abnormal Sq(H) phase confined to the EEJ belt and was closely associated with the incidence of complete or partial counter-electrojet (CEJ) condition. He also investigated the caustic mechanism of the Sq(H) phase variability on normal quiet days (NQDs). Mayaud , described the equatorial CEJ on the observed AQDs.
Several works on cause of AQDs are based on the reversed equatorial electrojet or CEJ. These include works of Gouin and Mayaud  which pointed out that the horizontal geomagnetic variation dH on occasions fell below its nighttime value on magnetically quiet days at Addis Ababa. Stening  suggested that on occasion the reversed jet seems to be associated with the additional imposition of a current system generated by a semidiurnal tidal mode. Earlier work of Butcher and Brown  suggested that an AQD results from combination of the occurrence of a day of low range in Sq(H) and a small magnetospheric substorm or mini-bay which forms the minimum in H. Stening et al. , raised up the issue that "The reversal of the equatorial electrojet (REJ) from east to westward on magnetically quiet days have challenged researchers since it was identified by Gouin and Mayaud , as a counter electrojet." They also noted that one of the important unanswered questions concerning the reversed REJ is whether it is generated by changes in local winds in the equatorial region or by global changes in mean winds or tides in the ionospheric dynamo region. They finally suggested that the change in background winds will lead to changes in the amplitudes and phases of tides reaching the dynamo region, but that this needs to be confirmed in particular cases. Last et al.  suggested the origin of AQDs to be different in the two latitude regions, because of the difference in the characteristics of the phase variability of Sq(H) in their local time variation and spatial extent. Fambitakoye and Mayaud , in their conclusion asked a pertinent question "What is the origin of counter-electrojet?" Also Sastri  showed that on days with abnormal Sq(H) phase limited to the electrojet belt, conditions characteristic of complete or partial CEJ prevail around the usual time interval of the diurnal maximum of the H field.
Rastogi  showed that the storm-time variation of geomagnetic field at an equatorial station has two components, one due to the equatorial ring current and another due to changes of the electric field of magnetospheric origin penetrating from high latitudes. Sizova  showed that CEJ caused by interplanetary magnetic field (IMF) occurs under magnetoperturbed conditions when components Bz of the IMF turns from the south direction to the north. Earlier work of Schlapp and Mann  showed that single current vortex acts not only on AQDs, but more generally on days of small amplitude of Sq(H) at stations like Hartland which are poleward of Sq focus. A striking abnormal features was found by Alex et al.  in both H and Z components, which showed maximum values around the midday hours. It is a rare abnormal feature and they suggested as absence of Sq(H) enhancement. Okeke  found that the obvious phase shift of Sq(H) on these AQDs were due to CEJ.
From literature reviewed, it is evident that there still exist controversy as of the cause of AQDs. Again, no work has really tried sorting out various forms of AQDs, as this will help give explanations as to the causes. Brown and Williams  and Sastri and Murthy  has defined AQDs as days on which the diurnal maximum of H occurred outside the time interval 0930-1230 LT. It is the purpose of this paper to sort out different forms of AQDs, attempt explanation as to the cause of each form of AQDs. This paper will adopt the definition of AQDs as given by Brown and Williams  and Sastri and Murthy , but with slight modification, and at the same time look at the days with abnormal features even when the diurnal variation of H seems normal. For example study the features where dH amplitude has broad constant maximum for some hours, where both H and Z components show abnormal noon maximum and some other features to be identified. This will help to easily identify AQDs and their possible causes.
The data set is the published minute values of all the three components of geomagnetic fields. The H and Z components minute interval records were subjected to hourly interval values and employed in the analysis. The three EEJ regions used in this study are Huancayo of -75.20o, -12.06o, Kiritimati Island -157.50o, 2.05o, and Pohnpei 158.33o, 7.00o of geographic longitude and geographic latitude, respectively. While their geomagnetic longitude and geomagnetic latitude are 356.12o, 1.40o, 273.49o, 3.09o, and 229.19o, 0.99o, respectively. The five International Quiet Days (IQDs) of each month of the year 1998 was chosen in this present study.
For indicating the amplitude of the Sq(H) and Sq(Z) variation at these regions, the hourly departure on these IQDs from the midnight base value were used. The midnight base value was defined as the average values of H or Z at hours preceding and succeeding local midnights on each of the IQDs of each month. The results are values of dH and dZ accordingly for every IQDs and for all the months of the year. The correction for the non-cyclic variation removes the disturbance effects due to some local sources. This paper looks at AQDs as days on which diurnal maximum of H occurred outside the time interval 1030-1330 LT. To really ascertain the days when the amplitude was reduced, the mean monthly amplitude of dH was calculated for each month and where the amplitude is reduced such that it is less than its monthly mean it is regarded as AQDs. This study for the first time adopts this method. Also where there is a fairly constant dH maximum which extends from morning hours at about 1000 LT to evening hour at about 1700 LT, it is regarded as AQD. Morning peak of dH, post-noon peak, afternoon minimum and dH and dZ showing noon maximum are all classified as AQDs.
The results of the analysis are presented in Figures 1, 2, and 3. The results as seen in Figure 1 show the NQDs with the observed diurnal variation of dH peaking around the normal noon hour local time. The diurnal maximum of dH occurs within the time interval 1030-1330 LT as had earlier been defined in this work. The selected NQDs of months and dates are indicative in the first row of Figure 1. The second row of Figure 1 shows different forms of AQDs in respective months and days of the dH diurnal variation, while the last row shows the dZ diurnal variation on the same day of dH AQD variation. From this figure, it could be observed that in Pohnpei, on January 26, the diurnal variation pattern seemingly assumes a regular pattern and as such could be classified as NQDs. But analysis shows that there is an indication of Sq(H) reduction shown in the dH instead of enhancement. This is a typical AQD that can easily be regarded as a NQD. Also on February 6, there is similar case where there is a noticeable reduction in dH and with maximum occurring outside the defined time interval. On May 28 there was an early noon (morning) occurrence of dH maximum around 0800 LT and rather reaching its minimum at local noon and again maximizing around 2000 LT. On the other hand, on June 17, late maximum was observed at about 1600 LT. While on September 16 there was pre-noon maximum around 1000 LT, minimum at 1330 LT and again reaching maximum at about 1800 LT. On December 27, there was early noon maximum at about 1000 LT which gradually decayed to zero during midnight LT.
The analysis of results of Kiritimati Island are illustrated in Figure 2. The NQDs presented in this figure are as seen in the first row, the days are August 17, September 28, and December 13. The AQDs of dH diurnal variation are shown in second and fourth row, while the corresponding dZ component diurnal variations are seen on the third and fifth row, respectively. In the sixth row, the first column contains the dH diurnal variation while the second column contains the dZ diurnal variation of December 27 at Kiritimati. It is important to note the AQD as observed on August 9 in this region. The dH started increasing from early morning at about 0600 LT, reaching its maximum at about 1000 LT and continued at a fairly constant amplitude till about 1600 LT and then decayed to minimum. This AQD has a broad maximum as seen in that figure. On August 15, there was a maximum occurrence of dH at about 0830 LT and sudden fall to about 1100 LT, then increases from 1400 LT to a maximum at 1600 LT and then fell gradually to midnight level. On August 18, it is clearly seen that dH amplitude almost remained constant from midnight till about 1300 LT and gradually peaked at about 1600 LT. While on September 16 there was pre-noon maximum at about 0900 LT and early minimum occurring between 1200 noon and 2400 LT. On December 17, 18, and 27 early noon maximum was observed.
At Huancayo, as seen in Figure 3, the NQDs are presented in the first row and in respective months and days as indicated in the figure. The AQDs in dH and dZ are also indicated in the same figure. A striking feature was observed in this region on January 23, where both dH and dZ attain their maximum around local noon, this is a very rare feature. This is a new feature or new aspect of AQD. On January 26 the dH variation seemingly looked normal but analysis showed a reduction in dH amplitude which fell below the mean monthly amplitude of 101 nT. On February 24, there was an early morning peak and noon decrease. On February 26 there was maximum before local noon, and sudden decrease from noon period. On March 7, broad maximum was observed lasting for barely 8 hours, from 1000 LT to 1800 LT, reduction in dH was also observed. There was also reduction in dH amplitude on March 9 as each fall below their monthly mean amplitude of 105 nT.
Several forms of AQDs have been found in this study. The noon minimum (depression) of dH amplitude on May 28 at Pohnpei is a typical of effect due to counter or reversed electrojet. Gouin  was the first to observe this type of abnormality in dH amplitude at Addis Ababa. Also Gouin and Mayaud  pointed out that the dH occasionally fall below nighttime value at same place. Most noon and afternoon depression so observed had been explained by several workers as effect due to counter electrojet, e.g., Onwumechili and Akasofu , Rastogi , Stening , and some others. The AQD in Pohnpei on March 28 was observed a few hours after dawn and a few hours before dusk which is typical of CEJ. It is also clearly indicative from this figure that dH amplitude which measures the magnitude of Sq(H) got depressed below its nighttime level on this day. This is a clear indication that current above this region has reversed direction [Onwumechili, 1997]. Further confirming the existence of counter electrojet, the resulting value of dH of the difference between the dH in these EEJ regions and non-EEJ was seen to be less than zero, which indicates the existence of CEJ.
The observed morning dH maximum, noon minimum or decrease in dH amplitude and late noon maximum which were all as a result of CEJ had been associated with late morning reversal and early evening reversal of vertical electric field [Ezema et al., 1996]. While Takeda and Maeda [1980, 1981] and Marriot et al.  suggested that both diurnal and semi-diurnal tidal winds can occasionally vary in such a way as to contribute to the features of a counter electrojet even though the semi-diurnal tidal winds are normally weaker, but may be enhanced on some days to cause abnormal Sq current like CEJ. The new feature of AQD was identified on January 23 at Huancayo, where both the horizontal and vertical components of the geomagnetic field showed maximum values around the local noon. This is not a common phenomenon. The reduction of dH amplitude on this day was striking, and was suggested to be due to local irregularities in Sq current system. But it has long been established that the EEJ is enhanced by localized ionospheric currents flowing at the dip equator with higher current intensities during the daytime. Here the reverse is the case, therefore we tend to suggest that this must be due to current which flows in opposite direction that inhibits the equatorial enhancement of dH over this region. Alex et al.  noted that this is a characteristic of a westward CEJ current which completely inhibits the equatorial enhancement of dH over the dip equator.
Another peculiar feature was identified on August 9 at Kiritimati and on March 7 at Huancayo, where the dH amplitude were found to be fairly constant between 0800 LT and 1700 LT, indicating no Sq(H) enhancement. This suggests absence of equatorial enhancement of dH in EEJ regions. Alex et al.  found similar variation and suggested the absence or cancellation of the EEJ to be caused by a westward flowing current system much wider than the conventional EEJ. This was attributed to the excitation of certain tidal modes at low latitudes on such AQDs.
The review of many studies on AQDs showed that types or various forms of AQDs had not been studied separately. This study has identified new forms of AQDs and attempted explanations as to the cause of these AQDs. Some of the identified features require very careful study before they could be identified. We have suggested that if these features are such that the amplitudes fall well below the nighttime level, it is referred to as CEJ, which appears during the morning or afternoon hour local time [Gouin, 1962]. On the other hand if these amplitudes show irregular fluctuations that have comparable amplitudes with the EEJ, it is associated with substorm [Onwumechili et al., 1973]. The new feature of both dH and dZ reaching maximum at noon with reduction in dH was attributed to absence or cancellation of EEJ.
It is important that further studied be carried out on such similar AQDs as had been identified in this paper for detail understanding of these complexities associated with AQDs. It is being suggested that a combination of data from VHF back scatter radar, the HF doppler radar ground based magnetometer near these regions would give more insight to forms and caused of AQDs.
Alex, S., B. D. Kadam, and R. G. Rastogi, A new aspect of daily variations of the geomagnetic field at low latitude, J. Atmos. Terr. Phys., 54 (7/8), 861, 1992.
Brown, G. M., On the nature of abnormal quiet days in Sq(H), Geophys. J. Roy. Astron. Soc., 64, 513, 1981.
Brown, G. M., and W. R. Williams, Some properties of day to day variability of Sq(H), Planet. Space Sci., 17, 455, 1969.
Butcher, E. C., An investigation on the causes of abnormal quiet days in Sq(H), Geophys. J. Roy. Astron. Soc., 69, 101, 1982.
Butcher, E. C., and G. M. Brown, On the nature of abnormal quiet days in Sq(H), Geophys. J. Roy. Astron. Soc., 64, 513, 1981.
Ezema, P. O., C. A. Onwumechili, and S. O. Oko, Geomagnetically quiet day ionospheric currents over the Indian sector, 3, Counter electrojet currents, J. Atmos. Terr. Phys., 58, 567, 1996.
Fambitakoye, O., and P. N. Mayaud, Equatorial electrojet and regular daily variation SR, iv, Special features in particular days, J. Atmos. Terr. Phys., 38, 123, 1975.
Gouin, P., Reversal of the magnetic daily variation Sq, J. Atmos. Terr. Phys., 38, 113, 1962.
Gouin, P., and P. N. Mayaud, A proposde l'existence possible d'um countre-electrojet aux latitudes magetiques equatoriales, Ann. Geophys., 23, 41, 1967.
Last, B. J., D. A. Emilia, and A. K. Outhred, Abnormal quiet day occurrence at Addis Ababa, Trivandrum, and Alibag, Planet. Space Sci., 24, 567, 1976.
Marriot, R. T., A. D. Richmond, and S. V. Venkateswaram, The quiet time equatorial electrojet and counter electrojet, J. Geomagn. Geoelectr., 31, 311, 1979.
Mayaud, P. N., The equatorial counter-electrojet - a review of its geomagnetic aspect, J. Atmos. Terr. Phys., 39, 1055, 1977.
Okeke, F. N., The Sq(H) abnormal quiet day variation in the low latitude region of Indian, Nig. J. Phys., 9, 78, 1997.
Onwumechili, C. A., The Equatorial Electrojet, 371 pp., Gordon and Breach Sci. Publ., UK, 1997.
Onwumechili, C. A., and S. I. Akasofu, On the abnormal depression of Sq(H) under the equatorial electrojet in the afternoon, J. Geomagn. Geoelectr., 24, 161, 1972.
Onwumechili, C. A., K. Kawasaki, and S. I. Akasofu, Relationship between the equatorial electrojet and polar magnetic variations, Planet. Space Sci., 21, 1, 1973.
Rastogi, R. G., Westward equatorial electrojet during daytime hours, J. Geophys. Res., 79, 1503, 1974.
Rastogi, R. G., Midday reversal of equatorial ionospheric electric field, Ann. Geophys., 15, 1309, 1997.
Sastri, J. H., Occurrence of abnormal Sq(H) phase in the equatorial electrojet region, Planet. Space Sci., 29 (7), 741, 1981.
Sastri, J. H., Phase variability of Sq(H) on normal quiet days in the equatorial electrojet region, Geophys. J. Roy. Astron. Soc., 71, 187, 1982.
Sastri, J. H., and B. S. Murthy, On the occurrence of abnormal quiet days at Kodaikanal, Indian J. Radio Space Phys., 7, 62, 1978.
Schlapp, D. M., and R. J. Mann, On the relation between days of small amplitude of Sq(H) and abnormal quiet days, Geophys. J., 95, 633, 1988.
Sizova, L. Z., Dynamics of equatorial counter electrojet caused by interplanetary magnetic field, Geomagn. Aeron., 35 (1), 141, 1995 (in Russian).
Stening, R. G., Magnetic variations at other latitudes during reverse equatorial electrojet, J. Atmos. Terr. Phys., 39, 1071, 1977.
Stening, R. G., C. E. Meek, and A. H. Manson, Upper atmosphere wind systems during reverse equatorial electrojet events, Geophys. Res. Lett., 23 (22), 3243, 1996.
Takeda, M., and H. Maeda, Three dimensional structure of ionospheric currents, 1, Currents caused by diurnal tides, J. Geophys. Res., 85, 6895, 1980.
Takeda, M., and H. Maeda, Three dimensional structure of ionospheric currents, 2, Currents caused by semidiurnal tidal winds, J. Geophys. Res., 86, 5861, 1981.