Intensity variations of the equivalent Sq current ... - Wiley Online Library

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 116, A10308, doi:10.1029/2011JA016632, 2011

Intensity variations of the equivalent Sq current system along the 210° magnetic meridian Y. Yamazaki,1 K. Yumoto,1,2 T. Uozumi,2 and M. G. Cardinal1 Received 8 March 2011; revised 15 July 2011; accepted 18 July 2011; published 15 October 2011.

[1] The total current intensity (Jtotal) of the equivalent Sq current system along the

210° magnetic meridian shows the following variations: (1) solar activity variations, (2) seasonal variations, and (3) day‐to‐day variations. These variations arise from different physical mechanisms. The main objective of the present paper is to determine the relative amount contributed by each variation to the Jtotal. First, the empirical Sq field model by Yamazaki et al. (2011) is analyzed to examine the impact of solar activity and seasonal variations. The results show that Jtotal changes by ±33% with the change of solar radiation activity in one solar cycle and by ±17% with the change of season. Next, observation data are analyzed to examine day‐to‐day variations. The daily values of Jtotal from 2000 to 2002 are derived after removal of the solar activity and seasonal contributions. The results show that Jtotal changes by ±14% from day to day. Therefore, we conclude that variations in Jtotal are mainly controlled by solar radiation activity, while the impact of seasonal effects is about half of the solar activity contribution and the impact of day‐to‐day effects is also about half of the solar activity contribution. Citation: Yamazaki, Y., K. Yumoto, T. Uozumi, and M. G. Cardinal (2011), Intensity variations of the equivalent Sq current system along the 210° magnetic meridian, J. Geophys. Res., 116, A10308, doi:10.1029/2011JA016632.

1. Introduction [2] The electric current is one of the key parameters in the study of the ionosphere. According to Ohm’s law, the ionospheric current density (J) is expressed as a product of the ionospheric conductivity (s) and ionospheric‐dynamo electric field (Eid), i.e., J = sEid. The ionospheric conductivity is basically a function of the electron number density Ne and hence controlled by solar radiation in the extreme ultraviolet (EUV) wavelength that creates ions and electrons in the ionosphere. On the other hand, the midlatitude ionospheric‐dynamo electric field arises from the neutral wind (U) driven by the local solar heating and the waves that propagate from below the ionosphere [Richmond and Roble, 1987]. The existence of the ionospheric current, flowing at an altitude of 100 km, has been confirmed by rocket measurements [e.g., Burrows and Hall, 1965]. The ionospheric current produces daily regular variations of the geomagnetic field on the Earth’s surface. The regular magnetic variation on each day is called SR variation [Mayaud, 1965] while the averaged SR variation over geomagnetically quiet days is called Sq variation [Chapman and Bartels, 1940]. The equivalent ionospheric current system can be derived from SR (or Sq) variations. The spatial pattern 1

Department of Earth and Planetary Sciences, Kyushu University, Fukuoka, Japan. 2 Space Environment Research Center, Kyushu University, Fukuoka, Japan. Copyright 2011 by the American Geophysical Union. 0148‐0227/11/2011JA016632

of the equivalent Sq current system is characterized by a dayside vortex in the low‐to‐middle latitude (within approximately ±60° magnetic latitude) of each hemisphere, counter‐clockwise in the Northern Hemisphere and clockwise in the Southern Hemisphere. It is known that the intensity and shape of the equivalent Sq current system change on various timescales. [3] There is a solar activity variation in the Sq current intensity, which has been found by comparing the equivalent Sq current system of the solar maximum and solar minimum [Chapman, 1919; Matsushita, 1968; Campbell and Matsushita, 1982; Takeda, 1999, 2002a]. The amplitude of Sq variation increases with increasing sunspot number, showing almost linear relationship [Chapman and Bartels, 1940; Yacob and Prabhavalkar, 1965; Yacob and Radhakrishna Rao, 1966; Yamazaki and Yumoto, 2011]. The dependence of the amplitude of Sq variation on solar radiation activity is explained by the change in the ionospheric conductivity [Takeda, 2002b; Takeda et al., 2003]. [4] The pattern and intensity of the current system slowly changes on a seasonal timescale (a few months to a year) [Matsushita and Maeda, 1965; Suzuki, 1973; Campbell and Schiffmacher, 1985, 1988; Campbell et al., 1993; Takeda, 1999, 2002a; Yamazaki et al., 2009]. Takeda [1999, 2002a] found that the total current intensity of equivalent Sq current system shows a semi‐annual variation with equinoctial maxima. He also pointed out that the semi‐annual variation of the total current intensity becomes more distinct during solar maximum than solar minimum. Both the semi‐annual variation and solar activity effects on the semi‐annual variation have been found in long‐term variations of the amplitude of

A10308

1 of 8

A10308

YAMAZAKI ET AL.: Sq CURRENT VARIATIONS

Sq variation [Rastogi and Iyer, 1976; Rastogi et al., 1994; Stening, 1995]. The semi‐annual variation of the amplitude of Sq variation is much greater than the semi‐annual variation of ionospheric conductivity [e.g., Wagner et al., 1980]. Yamazaki et al. [2009] suggested that the semi‐annual variation of equivalent Sq current system arises from the semi‐annual variation of the diurnal wind field, which shows amplitude maxima near the equinoxes. [5] The equivalent Sq current system also shows temporal and spatial variations less than the seasonal timescale, which are recognized as day‐to‐day variations [Hasegawa, 1960; Suzuki, 1978; Takeda, 1984; Takeda and Araki, 1984; Stening et al., 2005a, 2005b; Chen et al., 2007]. The source of the day‐ to‐day variation of equivalent Sq current system is not well understood. According to observational results by Takeda and Araki [1984], the day‐to‐day variation is partly explained by the additional current system of polar origin that is controlled by interplanetary magnetic field (IMF). Simulation results using general circulation model (GCM) by Miyahara and Ooishi [1997] and Kawano‐Sasaki and Miyahara [2008] indicated that day‐to‐day variability of the atmosphere below the ionosphere has a significant effect on the day‐to‐day variation of the ionospheric current system. Moreover, Olson [1989], based on numerical calculations, estimated the direct effects of magnetospheric currents (magnetopause currents, tail currents and ring currents) on the Sq variation and suggested that day‐to‐day variations of magnetospheric currents significantly contribute to day‐to‐day variability of the amplitude of Sq variation. [6] The variations of the equivalent Sq current system on different timescales arise from different factors. To understand the variability of the ionosphere and the physical processes behind it, it is important to evaluate the relative importance of each variation and identify the sources of each variation. In the present paper, we aim to determine the relative contribution of (1) solar activity variations, (2) seasonal variations, and (3) day‐to‐day variations, to the total current intensity Jtotal of equivalent Sq current system, and discuss the sources of each variation. We focus on the equivalent Sq current system along the 210° magnetic meridian (210 MM) [Yumoto et al., 1996].

2. Solar Activity Variations and Seasonal Variations [7] An empirical Sq field model was constructed by Yamazaki et al. [2011] based on the functional fitting of the Sq variations (Kp ≤ 2+) observed along the 210 MM. Magnetic data from 21 stations of the Circum‐pan Pacific Magnetometer Network (CPMN) [Yumoto et al., 2001], covering both the Northern Hemisphere and the Southern Hemisphere, were used in the construction of this empirical model. The analysis time span (1996–2007) covers an entire solar cycle. This empirical model describes the Sq variation in H (geomagnetic northward), D (geomagnetic eastward) and Z (vertical downward) components at each station as a product of functions of solar activity (SA), day of year (DOY), lunar age (LA) and local time (LT). That is, M ðSA; DOY ; LA; LT Þ ¼ f ðSAÞ g ð DOY Þ hð LAÞ ið LT Þ

ð1Þ

A10308

where M denotes the Sq variation in H, D or Z for a given station, and functions f(SA), g(DOY), h(LA) and i(LT) are given as follows: f ðSAÞ ¼ a1 þ a2 SA

g ð DOY Þ ¼ b1 þ

2 X

ð2Þ

½b2k cosð2k DOY =365Þ

k¼1

þ b2kþ1 sinð2k DOY =365Þ hð LAÞ ¼ c1 þ c2 cosð2 2LA=24Þ þ c3 sinð2 2LA=24Þ

ið LT Þ ¼ d1 þ

4 X

ð3Þ ð4Þ

½d2k cosð2k LT =24Þ þ d2kþ1 sinð2k LT =24Þ

k¼1

ð5Þ

where the coefficients ai, bi, ci and di can be determined by a least mean square fit of equation (1) to the observed Sq variations. The solar activity parameter SA is derived from the daily F10.7. Yamazaki et al. [2011] divided the empirical Sq variation into S variation and L variation: the S variation is a component that is independent of the lunar‐age parameter LA and the L variation is a component that is dependent on LA. After smoothing latitudinal distributions, they applied the spherical harmonic analysis (SHA) on the S and L fields under the assumption that the longitude and local time (or lunar time) are equivalent, and examined fundamental features of the equivalent ionospheric current system. [8] Using the same method as Yamazaki et al. [2011], we have derived the S variation from the empirical Sq field model. In this study, we use only the S variation that is independent of the lunar‐age parameter LA. This is practically same as we did not use equation (4) in fitting. The SHA for orders m = 0 to m = 4 and degrees n = m to n = m + 13 are applied to the S field in order to obtain the external current function. The total current intensity Jtotal is derived as the difference between the maximum and minimum values of the external current function. We examine the solar activity and seasonal variations in the total current intensity Jtotal of the equivalent current system. Although actual observations are always affected by both the solar activity and seasonal effects, we can independently evaluate solar activity effects and seasonal effects by analyzing the empirical model that calculates Jtotal using SA and DOY as independent variables. For example, Figure 1, which is originally given by Yamazaki et al. [2011], shows month‐to‐month variations of Jtotal(SA, DOY) at different levels of solar activity. A prominent semi‐annual variation in Jtotal with equinoctial maxima can be found at any level of solar activity SA. At a fixed month (DOY), Jtotal increases with increasing SA, showing dependence of Jtotal on solar activity. Note that the results obtained from the empirical model include errors of less than 20% although the error bars are not shown in the figure. [9] To evaluate the solar activity variation of Jtotal (SA, DOY), we calculated variability of Jtotal (SA, DOY) from its mean value using all the following SA values, SA = 70, 80, .., 230 for each month (DOY = 15, 46, 74, 105, 135, 166,196, 227, 258,

2 of 8

A10308

A10308


and 349 at each solar activity level (SA = 70, 80, .., 230). The results are summarized in Table 2. From the table, it can be seen that the DOY effect is about ±17% of mean Jtotal (SA, DOY). Also, it is found that the DOY effect on the Jtotal (SA, DOY) variation does not depend on SA. Possible explanations will be discussed later.

3. Day‐to‐Day Variations

Figure 1. Month‐to‐month variations of the total current intensity Jtotal at different solar activity levels, as derived from the empirical model of Yamazaki et al. [2011]. Figure 1 is originally given by Yamazaki et al. [2011]. Thick, middle and thin lines indicate the results for SA = 220, 150 and 80, respectively. 288, 319 and 349 are used as DOY values for each month from January to December, respectively). The results are summarized in Table 1. SA = 70 and SA = 230 represent the lowest and highest solar activity level during 1996–2007, respectively. The lowest and highest values of the monthly mean SA during 1996–2007 were observed in October 2007 (monthly mean SA = 67.8) and December 2001 (monthly mean SA = 226.6), respectively. From Table 1, it can be seen that the SA effect is about ±33% of the mean Jtotal (SA, DOY). It is also interesting to note that the SA effect on the Jtotal (SA, DOY) variation is different in each month. The SA effect is most significant in February (±35.9%) and least significant in June (±30.4%). [10] The seasonal variation of Jtotal (SA, DOY) is evaluated in a similar way. We calculate variability of Jtotal (SA, DOY) from its mean value using all the following DOY values, DOY = 15, 46, 74, 105, 135, 166, 196, 227, 258, 288, 319

Table 1. Quantitative Evaluation of Solar Activity Variation

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Mean

Mean Jtotal (kA)

SA Effect (kA)

Variability (%)

271.1 320.1 358.5 360.4 333.1 323.3 340.6 363.5 370.7 345.9 294.5 259.5 328.4

±93.0 ±115.1 ±127.8 ±121.7 ±104.6 ±98.4 ±105.9 ±116.7 ±119.4 ±109.6 ±93.0 ±83.4 ±107.4

±34.3 ±35.9 ±35.6 ±33.8 ±31.4 ±30.4 ±31.1 ±32.1 ±32.2 ±31.7 ±31.6 ±32.1 ±32.7

[11] We have evaluated the solar activity and seasonal effects on the Jtotal variation by using the empirical Sq field model constructed by Yamazaki et al. [2011]. However, the Sq field of the empirical model does not reproduce day‐to‐ day variations because this empirical model uses a function of solar activity (SA), day of year (DOY), lunar age (LA) and local time (LT), and does not include any variation on shorter timescale. In this section, daily values of Jtotal from 2000 to 2002 are derived from observation data to examine the impact of day‐to‐day variations. [12] Geomagnetic data are obtained from 21 stations along the 210 MM: 12 stations from CPMN, six stations from International Real‐time Magnetic Observatory Network (INTERMAGNET) [Kerridge, 2001], and three stations from World Data Center for Geomagnetism, Kyoto (WDC). The name and location of each station are listed in Table 3 (see also Figure 2). The days when the maximum Kp index does not exceed 5 are selected from the year 2000 to 2002. The selected days cover about 80% of the total day number but does not include the main phase of the magnetic storm. Since the data set covers moderately disturbed periods, the electric field from the high‐latitude ionosphere can play an important role. Such possible effects will be discussed later in section 4.3. Magnetic data in H, D and Z components at each station are obtained after the component transformation from geographic coordinates to geomagnetic coordinates. The disturbance fields in H that arise from the magnetospheric ring current are corrected by subtracting values of Dst index from H‐component magnetic field. (Hourly values of Dst index, provided by World Data Center for Geomagnetism, Kyoto, were interpolated into 1‐min values by using spline curves.) Hourly mean data in H, D

Table 2. Quantitative Evaluation of Seasonal Variation

SA = 70 SA = 80 SA = 90 SA = 100 SA = 110 SA = 120 SA = 130 SA = 140 SA = 150 SA = 160 SA = 170 SA = 180 SA = 190 SA = 200 SA = 210 SA = 220 SA = 230 Mean

3 of 8

Mean Jtotal (kA)

DOY Effect (kA)

Variability (%)

221.7 234.8 248.0 261.2 274.5 287.8 301.2 314.6 328.0 341.5 355.0 368.6 382.1 395.7 409.3 422.9 436.5 328.4

±37.7 ±40.1 ±42.3 ±44.4 ±46.6 ±48.8 ±51.0 ±53.2 ±55.4 ±57.7 ±59.9 ±62.2 ±64.5 ±66.8 ±69.1 ±71.4 ±73.7 ±55.6

±17.0 ±17.1 ±17.1 ±17.0 ±17.0 ±17.0 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9 ±16.9

A10308

A10308

YAMAZAKI ET AL.: Sq CURRENT VARIATIONS Table 3. Geographic and Geomagnetic Coordinates of Stations Used in the Study Station

Code

GGLat. (N°)

Rikubetsu Onagawa Kagoshima Okinawa Lunping Muntinlupa Guam Biak Darwin Weipa Birdsville Dalby

RIK ONW KAG OKN LNP MUT GAM BIK DAW WEP BSV DAL

43.48 38.43 31.48 26.75 25.00 14.37 13.58 −1.08 −12.40 −12.68 −25.54 −27.18

Memembetsu Kakioka Kakudu Charters Towers Alice Springs Canberra Esashi Mizusawa Kanoya

GGLon. (E°)

GMLat. (N°)

GMLon. (E°)

CPMN Database 143.76 141.47 130.72 128.22 121.17 121.02 144.87 136.05 130.90 141.88 139.21 151.20

36.32 31.15 24.37 19.57 17.89 6.26 5.61 −9.73 −22.06 −21.93 −36.1 −36.64

214.67 212.63 202.36 199.75 192.71 192.22 215.55 207.39 202.78 214.44 213.08 226.94

MMB KAK KDU CTA ASP CNB

INTERMAGNET Database 43.91 144.19 36.23 140.19 −12.69 132.47 −20.09 146.26 −23.77 133.88 −35.3 149.4

35.35 27.37 −22.00 −28.01 −32.92 −42.69

211.26 208.75 205.61 220.96 208.18 226.98

ESA MIZ KNY

39.24 39.11 31.42

WDC Database 141.35 141.20 130.88

30.08 29.94 21.89

150.81 150.92 200.75

and Z components are produced in local time, and daily variations DH(LT), DD(LT) and DZ(LT) are calculated as deviations from the nighttime level that is defined as the average of the six hourly values for local times 0000, 0200, 0300, 2200, 2300 and 2400 of each day. For each day, the SHA for orders m = 0 to m = 4 and degrees n = m to n = m + 13 is applied to the daily variations DH(LT), DD(LT) and DZ(LT) in the same way as section 2, and daily values of Jtotal are derived. It is assumed that characteristics of day‐ to‐day variations in Jtotal are statistically significant although, in reality, Jtotal may change on a timescale shorter than 24 hours (see Suzuki [1978] and Takeda [1984] for such variations in Jtotal). [13] In Figure 3, the obtained daily values of Jtotal from 2000 to 2002 were plotted. The solid line represents a 90‐day moving average, including solar activity variations and seasonal variations. As was mentioned in the former section, there is a prominent semi‐annual variation in Jtotal with equinoctial maxima. To obtain day‐to‐day variations, we have removed solar activity variations and seasonal variations in the following way. We first used linear regression method to adjust the daily Jtotal value to a solar flux index of 180 units (the seasonal difference in solar activity dependence shown in Table 2 was not taken into account in this adjustment). Next, the data were arranged as a function of day of year, and 30‐day moving average, defined as e J total, were subtracted from each value. The residuals are day‐to‐day variation DJtotal. The standard deviation of DJtotal is used to evaluate the day‐to‐day variation. We plotted in Figure 4 the size of the 30‐day moving standard deviation of DJtotal in relation to the seasonal average (e J total). It can be seen that the size of the day‐to‐day variation is ±14% of the seasonal average but different in different months: largest during the northern winter (±16% in January) and smallest during the northern

summer (±12% in May). A possible reason for the seasonal dependence of the day‐to‐day variability in Jtotal is seasonal changes in the atmospheric waves that propagate from below the ionosphere. For example, the sudden stratospheric warming tends to occur during the northern winter, and can affect the thermosphere/ionosphere [e.g., Goncharenko and Zhang, 2008; Liu et al., 2011a, 2011b]. Another possible reason is seasonal changes of the electric field in the high‐latitude ionosphere. Variability of the high‐latitude electric field shows seasonal changes and most significant during winter months [e.g., Matsuo et al., 2003]. The variability of the high‐latitude electric field causes the Joule heating to increase, and thus has an impact on the thermosphere/ionosphere [Deng et al., 2009].

4. Summary and Discussion [14] We have examined the solar activity variations, seasonal variations and day‐to‐day variations of the equivalent Sq current system observed along the 210 MM. The analysis results of the empirical Sq field model have shown that Jtotal changes by ±33% from its mean value with the change of solar radiation activity and also shown that Jtotal changes by ±17% with the change of season. Moreover, day‐to‐day variations were evaluated by analyzing observation data from 2000 to 2002. The results have shown that Jtotal changes by ±14% from day to day. In this section, we discuss sources of solar activity variations, seasonal variations and day‐to‐day variations in Jtotal. 4.1. Sources of Solar Activity Variations [15] The solar activity dependence of Jtotal can be explained by the change in the ionospheric conductivity. An increase of the solar radiation activity results in an increase of ionospheric conductivity. The neutral wind of the lower thermosphere (U), the driver of the current, does not seem to

4 of 8

A10308


A10308

Figure 3. Daily values of Jtotal from 2000 to 2002. The solid line is 90‐day moving average.

solar radiation activity. These features are consistent with the results by Takeda [1999, 2002a]. The semi‐annual variation of Jtotal is considered to be due to the semi‐annual variation of the diurnal wind field [Yamazaki et al., 2009]. Jtotal also shows the solstitial asymmetry: Jtotal (SA, DOY) during the June solstice is greater than that of the December solstice. Campbell [1990] pointed out that the Northern Hemisphere produces stronger current system than the Southern Hemisphere, which is consistent with our results. Probably, hemispheric asymmetry is caused by the asymmetric distribution of the main geomagnetic field. The ionospheric‐dynamo electric field is expressed as a product of the neutral wind U and the Earth’s magnetic field B. Thus, even if the neutral wind is distributed symmetrically, the wind‐driven current can be asymmetric. Moreover, the north‐south asymmetry in the main geomagnetic field causes the asymmetry in the ionospheric conductivity. It is

Figure 2. A map of the stations used in the analysis of section 3. depend on solar radiation activity [Bremer et al., 1997]. Also, the variation in the ionospheric electric field is not correlated with that of the Sq amplitude on the solar‐cycle timescale [Takeda et al., 2003]. An interesting feature shown in this study is that the dependence of Jtotal on the level of solar activity is different in different months (see Table 2). The solar activity dependence of Jtotal is most significant in February and least significant in June. This suggests that Jtotal is larger in June than February for the same level of the ionospheric conductivity, and may be explained by the seasonal change of the neutral wind. Further research is required to determine the physical mechanism. 4.2. Sources of Seasonal Variations [16] As shown in Figure 1, Jtotal shows a prominent semi‐ annual variation and its amplitude increases with increasing

Figure 4. The size of the day‐to‐day variability of Jtotal as a function of day of year. The day‐to‐day variability is defined as the ratio of the standard deviation of DJtotal to the seasonal average. The gray curve is the Fourier series representation using only five expansion coefficients.

5 of 8

A10308

A10308


Figure 5. Scatterplot of DJtotal (daily values) versus the effective interplanetary electric field Ee (the average over 1800(−) − 0600 LT). The error bars represent half of the standard deviation of Ee. The best‐fit regression line is also indicated.

also noted that the Sun‐Earth distance is changing with season. However, since the Sun‐Earth distance is largest around July and smallest around January, this effect cannot explain the solstitial asymmetry of Jtotal. From Table 2, it can be found that the relative amount of the seasonal variation in Jtotal is almost constant (±17%) at any level of solar radiation activity. This can be interpreted that although seasonal changes in the neutral wind U and the north‐south asymmetry of the main geomagnetic field lead to seasonal variations in Jtotal, the amplitude of seasonal variations in Jtotal is controlled by the magnitude of the ionospheric conductivity. 4.3. Sources of Day‐to‐Day Variations [17] In section 3, intensive day‐to‐day variations are found in Jtotal. We discuss here possible effects of the electric field from the high‐latitude ionosphere. The potential electric fields in the Earth’s polar caps (often characterized by the cross polar cap potential) are generated through the magnetosphere‐ionosphere (M‐I) system and hence significantly affected by changes in solar wind parameters. For example, Shepherd et al. [2002] demonstrated that the polar cap potential in the Northern Hemisphere increases as interplanetary electric field increases. The effective interplanetary electric field Ee is expressed as follows: Ee ¼ VBT sin2 ð=2Þ

ð6Þ

where V is the antisunward component of the solar wind velocity, BT is the transverse component of the interplanetary magnetic field (IMF); BT = (B2y + B2z ) , and is the clock angle of the IMF; = cos−1 (Bz/BT). These solar wind parameters time‐shifted to Earth are extracted from NASA/GSFC’s OMNI data set through OMNIWeb (see King and Papitashvili [2005] for details of the OMNI data). In Figure 5, we compare Ee with day‐to‐day variations of the total current intensity DJtotal. For Ee, the values averaged from 1800(−) LT to 0600 LT are used, where LT indicates the local time of the 210 MM sector and 1 2

the superscript (−) on the value of LT denotes the local time in the preceding day. The results reveal the tendency that DJtotal is smaller when Ee is larger. Although the trend is distinct, the data points in the figure are very scattered (correlation coefficient C = −0.144). Thus, the effects of Ee cannot fully explain day‐to‐day variability of Jtotal. Table 4 shows the correlation coefficient C between Ee and DJtotal with different time shifts in Ee. In the table, the values of Ee averaged from LTs to LTe are used. The interval from Cup to Clow indicates the 95% confidence interval of each C, which is estimated by 2000 times bootstrap resampling. The negative trend is statistically significant when both Cup and Clow are negative. In Table 4, it is interesting to point out that the dependence of DJtotal on Ee is most significant when the values of Ee are averaged over the nighttime hours of the preceding day (2100(−) − 0900 LT). This indicates that the changes in Ee do not immediately affect the midlatitude ionospheric current system but it takes approximately 9 hours to be effective. It has been previously suggested that the thermospheric winds generated by Joule heating in the polar ionosphere can decrease the intensity of the midlatitude ionospheric current system (the so‐called disturbance dynamo effects) (see Blanc and Richmond [1980] for numerical simulations; see also Le Huy and Amory‐Mazaudier [2008] for case studies). In future work, it needs to be clarified whether the disturbance dynamo effects can explain this 9‐hour time lag.

5. Conclusion [18] The total current intensity Jtotal of the equivalent Sq current system observed along the 210 MM shows these variations: (1) solar activity variations, (2) seasonal variations, and (3) day‐to‐day variations. We have shown that the relative amount that each variation contributes to variations of Jtotal is ±33%, ±17% and ±14%, respectively. It is concluded that variations in Jtotal are mainly controlled by solar radiation activity, while the impact of seasonal effects is about half of the solar activity contribution and the impact of day‐to‐day effects is also about half of the solar activity contribution. [19] The sources of variations in Jtotal, especially day‐to‐ day variations, remain to be understood. We have shown in this study that the day‐to‐day variability is most significant during the northern winter, which can be related to the variations in the atmospheric waves that propagate from below the ionosphere. Also, we have examined the effects of the electric field from the high‐latitude ionosphere. This is the first time that the impact of the changes in solar wind parameters on midlatitude ionospheric current system is Table 4. Correlation Coefficients Between Ee and DJtotal LTsa (LT)

LTea (LT)

C

Cup

Clow

0600 0300 0000 2100(−) 1800(−) 1500(−) 1200(−) 0900(−) 0600(−)

1800 1500 1200 0900 0600 0300 0000 2100(−) 1800(−)

−0.0518 −0.1035 −0.1312 −0.1483 −0.1441 −0.1337 −0.1244 −0.1197 −0.0808

0.0214 −0.0353 −0.0617 −0.0842 −0.0797 −0.0700 −0.0569 −0.0515 −0.0023

−0.1272 −0.1740 −0.1972 −0.2072 −0.2058 −0.1960 −0.1890 −0.1893 −0.1594

6 of 8

a

Local time of the 210 MM sector.

A10308


statistically evaluated. It has been found that DJtotal tends to be smaller when the effective interplanetary electric field Ee is larger, but it has also been found that the effects of Ee are too small to explain day‐to‐day variability of Jtotal. In the future it will be important to compare the day‐to‐day variability of atmospheric parameters and Jtotal for a better understanding of the sources of day‐to‐day variations in Jtotal. [20] Acknowledgments. The authors thank all the members of the CPMN project for their ceaseless support. The CPMN project is financially supported by the Ministry of Education, Science and Culture of Japan (and Japan Society for the Promotion of Science) as the Grant‐in Aid for Overseas Scientific Survey (05041060, 0841105, 10041122, 12373003, 15253005, 18253005). Also, the results presented in this paper rely on data collected at magnetic observatories that collaborate with INTERMAGNET and World Data Center for Geomagnetism, Kyoto. We thank the institutes that support the magnetic observatories and we also thank INTERMAGNET for promoting high standards of magnetic observatory practice (http://www.intermagnet.org). We are grateful to the Herzberg Institute of Astrophysics for providing F10.7 solar radiation data. The planetary geomagnetic disturbance index Kp can be downloaded at http://www‐app3.gfz‐potsdam.de/kp_index/index.html. The geomagnetic equatorial Dst index can be downloaded at http://wdc.kugi.kyoto‐u. ac.jp/dstdir/. We acknowledge use of NASA/GSFC’s Space Physics Data Facility’s OMNIWeb service (http://omniweb.gsfc.nasa.gov/hw.html), and OMNI data. The first author is supported by a grant from Research Fellowship of the Japan Society for the Promotion of Science (JSPS) for Young Scientists. [21] Robert Lysak thanks the reviewers for their assistance in evaluating this paper.

References Blanc, M., and A. D. Richmond (1980), The ionospheric disturbance dynamo, J. Geophys. Res., 85, 1669–1686. Bremer, J., R. Schminder, K. M. Greisiger, P. Hoffmann, D. Kürschner, and W. Singer (1997), Solar cycle dependence and long‐term trends in the wind field of the mesosphere/lower thermosphere, J. Atmos. Sol. Terr. Phys., 61, 765–774. Burrows, K., and S. H. Hall (1965), Rocket measurements of the geomagnetic field above Woomera, South Australia, J. Geophys. Res., 70, 2149–2158, doi:10.1029/JZ070i009p02149. Campbell, W. H. (1990), Differences in geomagnetic Sq field representations due to variations in spherical harmonic analysis techniques, J. Geophys. Res., 95, 20,923–20,936. Campbell, W. H., and S. Matsushita (1982), Sq currents: A comparison of quiet and active year behavior, J. Geophys. Res., 87, 5305–5308. Campbell, W. H., and E. R. Schiffmacher (1985), Quiet ionospheric currents of the Northern Hemisphere derived from geomagnetic field records, J. Geophys. Res., 90, 6475–6486. Campbell, W. H., and E. R. Schiffmacher (1988), Quiet ionospheric currents of the Southern Hemisphere derived from geomagnetic field records, J. Geophys. Res., 93, 933–944. Campbell, W. H., B. R. Arora, and E. R. Schiffmacher (1993), External Sq currents in the India‐Siberia region, J. Geophys. Res., 98, 3741–3752. Chapman, S. (1919), The solar and lunar diurnal variations of terrestrial magnetism, Philos. Trans. R. Soc. London A, 218, 1–118. Chapman, S., and J. Bartels (1940), Geomagnetism, Oxford Univ. Press, London. Chen, G.‐X., W.‐Y. Xu, A.‐M. Du, Y.‐Y. Wu, B. Chen, and X.‐C. Liu (2007), Statistical characteristics of the day‐to‐day variability in the geomagnetic Sq field, J. Geophys. Res., 112, A06320, doi:10.1029/ 2006JA012059. Deng, Y., A. Maute, A. D. Richmond, and R. G. Roble (2009), Impact of electric field variability on Joule heating and thermospheric temperature and density, Geophys. Res. Lett., 36, L08105, doi:10.1029/2008GL036916. Goncharenko, L., and S.‐R. Zhang (2008), Ionospheric signatures of sudden stratospheric warming, Geophys. Res. Lett., 35, L21103, doi:10.1029/ 2008GL035684. Hasegawa, M. (1960), On the position of the focus of the geomagnetic Sq current system, J. Geophys. Res., 65, 1437–1447. Kawano‐Sasaki, K., and S. Miyahara (2008), A study on three‐dimensional structures of the ionospheric dynamo currents induced by the neutral winds simulated by the Kyushu‐GCM, J. Atmos. Sol. Terr. Phys., 70, 1549–1562. Kerridge, D. (2001), INTERMAGNET: Worldwide near‐real‐time geomagnetic observatory data, paper presented at Space Weather Workshop:

A10308

Looking Towards a European Space Weather, Eur. Space Res. and Technol. Cent., Noordwijk, Netherlands. King, J. H., and N. E. Papitashvili (2005), Solar wind spatial scales in and comparisons of hourly Wind and ACE plasma and magnetic field data, J. Geophys. Res., 110, A02104, doi:10.1029/2004JA010649. Le Huy, M., and C. Amory‐Mazaudier (2008), Planetary magnetic signature of the storm wind disturbance dynamo currents: Ddyn, J. Geophys. Res., 113, A02312, doi:10.1029/2007JA012686. Liu, H., E. Doornbos, M. Yamamoto, and S. Tulasi Ram (2011a), Strong thermospheric cooling during the 2009 major stratosphere warming, Geophys. Res. Lett., 38, L12102, doi:10.1029/2011GL047898. Liu, H., M. Yamamoto, S. T. Ram, T. Tsugawa, Y. Otsuka, C. Stolle, E. Doornbos, K. Yumoto, and T. Nagatsuma (2011b), Equatorial electrodynamics and neutral background in the Asian sector during the 2009 stratospheric sudden warming, J. Geophys. Res., 116, A08308, doi:10.1029/ 2011JA016607. Matsuo, T., A. D. Richmond, and K. Hensel (2003), High‐latitude ionospheric electric field variability and electric potential derived from DE‐2 plasma drift measurements: Dependence on IMF and dipole tilt, J. Geophys. Res., 108(A1), 1005, doi:10.1029/2002JA009429. Matsushita, S. (1968), Sq and L current systems in the ionosphere, Geophys. J. R. Astron. Soc., 15, 109–125. Matsushita, S., and H. Maeda (1965), On the geomagnetic solar quiet daily variation field during the IGY, J. Geophys. Res., 70, 2535–2558. Mayaud, P. N. (1965), Analyse morphologique de la variabilité jour à jour de la variation “régulière” Sr du champ magnétique terrestre, II, Le système de courants Cm (régions non polaires), Ann. Geophys., 21, 515–544. Miyahara, S., and M. Ooishi (1997), Variation of Sq induced by atmospheric tides simulated by a middle atmosphere general circulation model, J. Geomagn. Geoelectr., 49, 77–87. Olson, W. P. (1989), The contribution of magnetospheric currents to Sq, Pure Appl. Geophys., 131, 447–462. Rastogi, R. G., and K. N. Iyer (1976), Quiet day variation of geomagnetic H‐field at low latitudes, J. Geomagn. Geoelectr., 28, 461–479. Rastogi, R. G., S. Alex, and A. Patil (1994), Seasonal variations of geomagnetic D, H and Z fields at low latitudes, J. Geomagn. Geoelectr., 46, 115–126. Richmond, A. D., and R. G. Roble (1987), Electrodynamic effects of the thermospheric winds from the NCAR thermospheric general circulation model, J. Geophys. Res., 92, 12,365–12,376. Shepherd, S. G., R. A. Greenwald, and J. M. Ruohoniemi (2002), Cross polar cap potentials measured with Super Dual Auroral Radar Network during quasi‐steady solar wind and interplanetary magnetic field conditions, J. Geophys. Res., 107(A7), 1094, doi:10.1029/2001JA000152. Stening, R. J. (1995), Variations in the strength of the Sq current system, Ann. Geophys., 13, 627–632. Stening, R., T. Reztsova, and L. H. Minh (2005a), 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. Stening, R., T. Reztsova, D. Ivers, J. Turner, and D. Winch (2005b), Morning quiet‐time ionospheric current reversal at mid to high latitudes, Ann. Geophys., 23, 385–391. Suzuki, A. (1973), New analysis of geomagnetic Sq‐field, J. Geomagn. Geoelectr., 25, 259–280. Suzuki, A. (1978), Geomagnetic Sq field at successive universal time, J. Atmos. Terr. Phys., 40, 449–463. Takeda, M. (1984), Day‐to‐day variation of equivalent Sq current system during March 11–26, 1970, J. Geomagn. Geoelectr., 36, 215–228. Takeda, M. (1999), Time variation of global geomagnetic Sq field in 1964 and 1980, J. Atmos. Sol. Terr. Phys., 61, 765–774. Takeda, M. (2002a), Features of global geomagnetic Sq field from 1980 to 1990, J. Geophys. Res., 107(A9), 1252, doi:10.1029/2001JA009210. Takeda, M. (2002b), The correlation between the variation in ionospheric conductivity and that of the geomagnetic Sq field, J. Atmos. Sol. Terr. Phys., 64, 1617–1621. Takeda, M., and T. Araki (1984), Time variation of instantaneous equivalent Sq current system, J. Atmos. Terr. Phys., 46, 911–915. Takeda, M., T. Iyemori, and A. Saito (2003), Relationship between electric field and currents in the ionosphere and the geomagnetic Sq field, J. Geophys. Res., 108(A5), 1183, doi:10.1029/2002JA009659. Wagner, C. U., D. Möhlmann, and K. Schäfer (1980), Large‐scale electric fields and currents and related geomagnetic variations in the quiet plasmasphere, Space Sci. Rev., 26, 391–446. Yacob, A., and A. S. Prabhavalkar (1965), Solar control of the amplitude and phase of the yearly mean Sq(H) at Alibag for the period 1905 to 1960, J. Atmos. Terr. Phys., 27, 73–80. Yacob, A., and D. Radhakrishna Rao (1966), Solar cycle and annual variations of Sq(H) at Alibag, J. Atmos. Terr. Phys., 28, 351–360.

7 of 8

A10308


Yamazaki, Y., and K. Yumoto (2011), Long‐term behavior of annual and semi‐annual Sq variations, Earth Planets Space, in press. Yamazaki, Y., K. Yumoto, T. Uozumi, A. Yoshikawa, and M. G. Cardinal (2009), Equivalent current systems for the annual and semiannual Sq variations observed along the 210‐MM CPMN stations, J. Geophys. Res., 114, A12320, doi:10.1029/2009JA014638. Yamazaki, Y., et al. (2011), An empirical model of the quiet daily geomagnetic field variation derived from CPMN 210 MM data from 1996 to 2007, J. Geophys. Res., doi:10.1029/2011JA016487, in press. Yumoto, K., et al. (1996), The STEP 210 magnetic meridian network project, J. Geomagn. Geoelectr., 48, 1297–1310.

A10308

Yumoto, K., et al. (2001), Characteristics of Pi 2 magnetic pulsations observed at the CPMN stations: A review of the STEP results, Earth Planets Space, 53, 981–992. M. G. Cardinal, Y. Yamazaki, and K. Yumoto, Department of Earth and Planetary Sciences, Kyushu University, 6‐10‐1 Hakozaki, Higashi‐ku, Fukuoka 812‐8581, Japan. ([email protected]) T. Uozumi, Space Environment Research Center, Kyushu University, Fukuoka 812‐8581, Japan.

8 of 8