The latitudes of the foci of the Sq current systems over Australia and Japan are determined on a daily basis during the period from December 1989 to June 1990. The availability of a dense network of observatories in Australia during that time enabled a better determination in that region. Latitudinal movements of the foci are compared with the strength of the equatorial electrojet, and generally an increase in electrojet strength is accompanied with a poleward movement of the focus, especially in Japan. Some examples are noted where the foci in the two hemispheres move poleward or equatorward together from one day to the next, but this relationship was not found to be statistically significant. It is hard to disentangle effects due to other current systems such as Sqp from changes related to atmospheric tides. If some of the observed effects are due to tides, then the (2,3) and (2,4) semidiurnal modes are more likely contributors than the (2,2) mode
JOURNAL OF GEOPHYSICAL RESEARCH, VOL 110, A10308, doi:10.1029/2005JA011219, 2005 Day-to-day changes in the latitudes of the foci of the Sq current system and their relation to equatorial electrojet strength Robert Stening and Tamara Reztsova School of Physics, University of New South Wales, Sydney, New South Wales, Australia Le Huy Minh Institute of Geophysics, Vietnamese Academy of Science and Technology, Hanoi, Vietnam Received May 2005; revised 11 July 2005; accepted 15 July 2005; published 28 October 2005 [1] The latitudes of the foci of the Sq current systems over Australia and Japan are determined on a daily basis during the period from December 1989 to June 1990 The availability of a dense network of observatories in Australia during that time enabled a better determination in that region Latitudinal movements of the foci are compared with the strength of the equatorial electrojet, and generally an increase in electrojet strength is accompanied with a poleward movement of the focus, especially in Japan Some examples are noted where the foci in the two hemispheres move poleward or equatorward together from one day to the next, but this relationship was not found to be statistically significant It is hard to disentangle effects due to other current systems such as Sqp from changes related to atmospheric tides If some of the observed effects are due to tides, then the (2,3) and (2,4) semidiurnal modes are more likely contributors than the (2,2) mode Citation: Stening, R., T Reztsova, and L H Minh (2005), Day-to-day changes in the latitudes of the foci of the Sq current system and their relation to equatorial electrojet strength, J Geophys Res., 110, A10308, doi:10.1029/2005JA011219 Introduction [2] During times when the level of magnetic disturbance is low, the main current system flowing in the ionosphere has come to be known as the Sq system (or ‘‘solar quiet’’) The main driver of this system is thought to be the (1, 2) atmospheric tidal mode, generated in situ in the ionospheric E region by absorption of solar radiation [Tarpley, 1970; Stening, 1971] The Sq system has two current whorls, one in each of the northern and southern hemispheres, which together feed into an equatorial electrojet flowing eastward along the magnetic equator There are conspicuous seasonal changes in the current system [Matsushita and Maeda, 1965; Stening, 1971] and, while some features are similar between one quiet day and the next, noticeable changes may also be observed on similarly quiet days Some of the most remarkable changes seen are variations in the strength of the equatorial electrojet with occasional reversals in direction of the electrojet, commonly known as a ‘‘counterelectrojet.’’ Another remarkable change occurs in the position of the center or ‘‘focus’’ of the current whorl in each hemisphere It is thought that such changes on magnetically quiet days must be due to changes in the (tidal) winds systems in the ionosphere It seems unlikely that changes in the diurnal (1, 2) mode could produce the changes observed in the currents or rather in the magnetic fields generated by the currents It is more likely that Copyright 2005 by the American Geophysical Union 0148-0227/05/2005JA011219 semidiurnal tides, known to also be present in the E region, will be responsible for these day-to-day changes [Stening, 1991] Knowledge of how the focus positions relate in each hemisphere may enable us to at least deduce whether these tidal changes responsible are predominately symmetric or asymmetric about the equator [3] During 1989 – 1990 a dense network of magnetic observatories were set up on the Australian mainland [Chamalaun and Barton, 1993] This network, known as AWAGS (Australia-Wide Array of Geomagnetic Stations), enabled a much more accurate pinpointing of the Sq focus position over Australia than is usually available Background [4] One of the more thorough examinations of this problem was performed by Schlapp [1976] He concluded that the latitudinal positions of the northern and southern foci were only weakly related but their tendency was to move poleward and equatorward together The correlation with electrojet strength was also weak but the tendency was for a stronger electrojet to correlate with foci more poleward Schlapp used values of DH at an hour near noon, which he suggests is nearly the same as using DH at the time when DY is zero He omitted all days for which the magnetic disturbance index Cp > 0.5 He used data from the IQSY and from the IGY Correlations from IGY data were rather higher than for the IQSY data since, as Schlapp suggests, ‘‘disturbance current systems tend to fluctuate synchronously over wide areas.’’ His A10308 of A10308 STENING ET AL.: CHANGES IN LATITUDES OF Sq SYSTEM FOCI A10308 Figure Current vectors over Australia at h UT on May 1990 most significant correlations came from Spanish and African stations and from a Japanese and an Australian station with Koror (7.3°N, 134.5°E) as the electrojet station [5] Takeda and Araki [1984] followed the form of the Sq current system through 18 consecutive days in March 1980 when magnetic disturbance was low They noted day-to-day changes Some of these they attributed to currents flowing outside of the ionosphere The others represented increases or decreases in the overall current amplitude or the addition of an apparent semidiurnal effect None clearly demonstrated a change in focus latitude Further similar studies were performed by Takeda [1984] using data from March 1970 Here changes in shape rather than intensity were noted but these were attributed to disturbances, as measured by the AE index, or, again, semidiurnal tides [6] Kane [1974] examined data in the Indian region during 1964 and found that the Sq focus latitude shifted equatorward when the overall Sq current strength was larger and also when the equatorial electrojet strength was larger This is opposite to the relation found by Schlapp [1976] (Kane estimated the Sq strength from values of DY at Indian and Russian stations near the focus and used Trivandrum DH data to give the electrojet amplitude) His conclusions were reached from qualitative inspection of average curves during equinox in 1964 [7] So why did Kane and Schlapp reach different conclusions? First, it is not clear that Kane would have obtained a negative correlation if he had evaluated it, though his data seem to indicate that this would be likely Second, the Sq current system over India/Russia has some peculiarities It almost disappears in winter [Rastogi et al., 1996], though Kane’s results were restricted to equinox Kane [1990] also discusses the variability of the focus position in South America in 1958, as evidenced by changes in DH at Trelew (43.3°S, 65.3°W) Method [8] Different methods used in determining the latitude of the Sq system focus were discussed by Stening et al [2005] The preferred method was that in which the time when the declination variation DD changed sign was found first This was the time when DD changed from negative to positive in the southern hemisphere and from positive to negative in the northern hemisphere For the Australian data the eastern magnetic element DY was used The horizontal element DH or northward element DX was then evaluated at the time when DD = [9] We calculated the focus positions in both north and south hemispheres each day for all months from December 1989 to June 1990, except February 1990 The latter month had so many disturbed days that it was not possible to obtain a useful plot [10] The southern hemisphere focus was determined from an array of Australian stations which were operating during that period In some cases we checked the focus position over Australia by drawing a full map of the current vectors from the AWAGS network as in Figure The current vectors were obtained by rotating the magnetic field vectors clockwise through 90° while the magnetic vectors were derived from magnetometer data at the individual observatory sites The strength of the electrojet at Baclieu (9.3°N, 105.7°E, geographic) is estimated from the horizontal magnetic field variation DH The mean of the preceding and succeeding midnight values are subtracted from the maximum value to determine the DH value plotted There are occasional gaps in the Baclieu data of STENING ET AL.: CHANGES IN LATITUDES OF Sq SYSTEM FOCI A10308 Table Observatory Geographic Coordinates Observatory Code Latitude Longitude Guam Lunping Kanoya Kakioka Memambetsu Weipa Cooktown Robinson River Mount Isa Winton Alpha Birdsville Quilpie Etadunna Bourke Menindee Port Augusta Condobolin Portland GUA LNP KNY KAK MMB WEI CKT ROB ISA WTN ALP BIR QUI ETA BUK MEN PTA CDN POL 14N 25N 31N 36N 44N 13S 15S 17S 21S 22S 24S 26S 27S 29S 30S 32S 32S 33S 38S 145E 121E 131E 140E 144E 142E 145E 137E 139E 143E 147E 139E 144E 139E 146E 142E 138E 147E 141E [11] For the northern hemisphere focus, usually determined from data from Japanese observatories, we were only able to determine the local time of the focus from changes in DD (we did not have a magnetometer array like that in Australia) This causes some uncertainty In addition we found that during northern winter, the DD variation at the stations nearest to the equator sometimes did not exhibit the usual daily variation with a morning maximum followed by a later minimum Instead a southern hemisphere type of variation appeared with a morning minimum Such occurrences of this ‘‘invasion phenomenon’’ [Mayaud, 1965] led to obviously incorrect values If the program yielded a time for DD = outside of the range of to 14 h LT, we replaced it with the zero time most commonly seen for that month at that station Results [12] Table gives a listing of the geographic coordinates of the observatories used Those used vary a little from month to month on account of breaks in the availability of data We choose the ‘‘best’’ set of observatories for each month In fact the same northern hemisphere observatories were used for all months but January, namely LNP, KNY, KAK, and MMB In January we added GUA (Guam) for a better result [13] It is difficult to decide at what level of magnetic disturbance we should start to reject data Unfortunately, 1990 is near the maximum of the solar cycle, so a fairly high level of disturbance frequently occurs On some disturbed days there is an additional westward current flow at high latitudes which extends on to the Australian mainland, or at least its magnetic effect extends there This will result in pushing the observed focus to a lower latitude than normal It is also well known [Onwumechili et al., 1973; Reddy et al., 1979] that the amplitude of the equatorial electrojet is often diminished on disturbed days, so a disturbance will lead to a positive correlation between focus latitude and electrojet strength Days with a Kp disturbance index greater than 3+ in the hours around local noon are marked with an asterisk on the bottom of the diagram In February 1990, 18 A10308 out of the 28 days are disturbed according to this criterion, so we omit consideration of that month [14] We look first at the plot for January 1990 in Figure where we show the latitudes of the two foci and the strength of the electrojet at Baclieu, measured as described above The Australian observatories used were CKT, QUI, BUK, CDN, and POL Several days in this plot can be seen where all three plotted parameters move together, but clearly this does not happen all the time [15] The day of 25 January 1990 may be an example of the disturbance influence mentioned above as both foci move equatorward together The equatorward focus movements on and 10 January we would also attribute to magnetic disturbance; Kp values are 4+ and 40 during local noon hours in eastern Australia Another factor that arises on10 January is that the D variation at Lunping is of the southern hemisphere type with a morning minimum followed by a maximum, the ‘‘invasion’’ phenomenon mentioned above [16] In March the observatories used were WEI, ISA, BIR, ETA, and PTA, a chain slightly to the west of those used in January, selected because they gave the best data coverage There are a lot of disturbed periods Dips in the Baclieu DH record usually occur at these times as can be seen in Figure on March (Kp = 50), 13 March (Kp = 6+), and 26 March (Kp = 60) On 13 March the foci appear to move to very high latitudes and on 26 March they move to very low latitudes [17] On March the magnetic records not show clear signs of disturbance and so, even though the equatorial electrojet amplitude is diminished, the focus positions are probably not much affected by the disturbance On 13 and 26 March there are strong westward currents all over continental Australia The declination D variation is also irregular and this invalidates the method we use to find the focus [18] Two other notable changes might be noted in March From to March the focus over Japan moves poleward by 12° and the focus over Australia moves poleward by 6° (The day of March has a Kp of 50 but there is little clear evidence of disturbance on the records) From 17 to 18 Figure Variations with the day of the month, in January 1990, of the latitude of the northern hemisphere focus (full lines and open squares), the latitude of the southern hemisphere focus (dotted lines and filled squares), and of the strength of the equatorial electrojet (dashed lines and crosses) at Baclieu The actual value of DH at Baclieu (in nT) is obtained by multiplying the ordinate value by of A10308 STENING ET AL.: CHANGES IN LATITUDES OF Sq SYSTEM FOCI A10308 Figure As for Figure but for March 1990 Figure As for Figure but for June 1990 March the Australian focus moves poleward by 12° while that over Japan moves poleward by about 4° On these days the Kp values are and 10 [19] On and 18 March there are clear afternoon reversed electrojets at Baclieu accompanied by poleward movements in both foci These were the only clear examples of this phenomenon identifiable during quiet times within the period under examination [20] The variations of focus positions and electrojet strength for May are shown in Figure The Australian observatories used were CKT, ALP, BUK, and CDN In a cursory inspection one again may see that all three parameters move up and down together If the Baclieu electrojet plot is moved back (left) by one day, the correspondence may seem even more remarkable Yet there are other times when no correspondence can be noted The days of 11, 26, and 27 May had significantly large disturbances which would have influenced the results [21] On 12 May 1990 the equivalent currents over Australia look quite strange If there is an identifiable focus in Eastern Australia, it is certainly at quite a low latitude, less than 15°S and north of Cooktown DH is positive at Port Moresby (9.4°S) so a focus exists between 15°S and 9°S The Kp value is 3+ but the disturbance does not look severe The day of 20 May is similar (Kp is ) with a focus near 15°S [22] On 19 and 22 May, magnetic disturbance obliterates the Sq system On 20 May there is some evidence of a focus near the latitude of Cooktown but this is too far north to show a complete whorl [23] The days of 13 to 17 May are a group of five quiet days where Kp