Estimating the solar wind conditions during an extreme geomagnetic storm: a case study of the event that occurred on March 13–14, 1989
© Nagatsuma et al. 2015
Received: 31 October 2014
Accepted: 15 May 2015
Published: 27 May 2015
The solar wind conditions of an extreme geomagnetic storm were examined using magnetic field observations obtained from geosynchronous satellites and the disturbance storm-time (Dst) index. During geosynchronous magnetopause crossings (GMCs), magnetic field variations at the magnetosheath, which is the modulated interplanetary magnetic field (IMF), were observed by geosynchronous satellite. The dawn to dusk solar wind electric field (VBS) was estimated from the Dst index by using an empirical formula for Dst prediction; these data were then used to estimate the IMF and solar wind speed. This method was applied in the analysis of an extreme geomagnetic storm event that occurred on March 13–14, 1989, for which no direct solar wind information was available. A long duration of the GMC was observed after the second storm sudden commencement (SSC) of this event. The solar flare possibly associated with the second SSC of this storm event was identified as the March 12 M7.3/2B flare. The IMF B z was estimated to be about −50 nT with a solar wind speed of about 960 km/s during the 5 h in which the main phase of the storm rapidly developed, assuming an Alfvén Mach number (M A) during this period of more than 2.
KeywordsExtreme geomagnetic storm Geosynchronous magnetopause crossing Geomagnetic indices Solar wind parameters
Geomagnetic disturbances are driven by the effect of the solar wind–magnetosphere–ionosphere compound system. A number of studies have been conducted to obtain empirical relationships for the compound system and to develop a numerical simulation code for understanding this interaction (e.g., Tanaka 2007 and references therein). However, there is little knowledge of this compound system under extreme conditions because of the infrequent occurrence of extreme events.
To examine geomagnetic disturbances under extreme conditions, the following two aspects are important. The first is the condition of the solar wind, which is the main driver of geomagnetic disturbances, and the second is the behavior of the solar wind–magnetosphere–ionosphere coupling processes under extreme conditions of solar wind. Because of the complex nature of this system, it can be difficult to assess whether the response of the system during a normal space environment can be applied to extreme conditions of solar wind. Thus, the study of events occurring under extreme conditions is important for understanding the behavior of the compound system.
Coverage of solar wind observation was limited prior to operation of the Advanced Composition Explorer (ACE) at the L1 point in 1997. During the maximum period of solar cycle 22, almost the only solar wind monitoring platform was Interplanetary Monitoring Platform (IMP) 8, which had a 35 Earth radii circular orbit, and was used to measure Earth’s magnetotail, magnetosheath, and solar wind. The coverage of solar wind observation by IMP 8 was only for 7 to 8 days of every 12.5 days, which was the orbital period of IMP 8. Such circumstances led to many severe geomagnetic disturbances for which there was no solar wind observation in the maximum period of solar cycle 22. Many severe geomagnetic storm events have thus not been analyzed in detail owing to a lack of solar wind information. For example, an extreme storm that occurred on March 13–14, 1989, which is the focus of this paper, had no solar wind coverage during its onset, main phase, and most of the recovery phase. Therefore, McKenna-Lawlor et al. (2005) attempted to reconstruct the interplanetary shocks from five major flares hitting both Earth and Mars during the March 9–23, 1989, by using the Hakamada–Akasofu–Fry version 2 (HAFv.2) solar wind model (Fry et al. 2001, 2003). Other papers have reconstructed solar wind parameters during large storms by using geomagnetic variations or indices (e.g., Li et al. 2006; Cliver et al. 2009).
The three major solar wind monitoring platforms currently around Earth’s orbit include ACE and Solar Terrestrial Relations Observatories (STEREO) Ahead (A) and Behind (B). STEREO-A and STEREO-B are drifting in heliocentric orbits away from Earth; the former is leading, and the latter is lagging. Baker et al. (2013) examined the extreme solar wind conditions obtained from STEREO-A on July 23–24, 2012. They also used the WSA-ENLIL model to estimate the proton density of the solar wind for this event because the quality of the low-energy particle measurements might have been poor owing to contamination from high-energy particles. On the basis of their data analysis, they suggested that an extreme solar wind with their estimated conditions could produce an extreme geomagnetic storm if it reached the geospace environment; the lowest possible disturbance storm-time (Dst) index was −1182 nT.
In this paper, we estimate the solar wind conditions for the extreme storm event that occurred on March 13–14, 1989, by using the geostationary magnetopause crossing (GMC) period obtained from the Geostationary Operational Environmental Satellites (GOES) magnetic field data and the Dst index. The solar origin of this event is also discussed.
Overview of the March 13–14, 1989, storm event
The geomagnetic storm that occurred on March 13–14, 1989, was the most extreme geospace disturbance since 1957. The minimum value of the Dst index during this event was −589 nT at 02 UT on March 14. From a probabilistic viewpoint, an event of this scale may occur once every 60 years (Kataoka 2013). Two storm sudden commencements (SSCs) have been identified for this storm event at 0128 and 0747 UT, both on March 13 (e.g., Fujii et al. 1992), with amplitudes of 43 and 76 nT at Kakioka Observatory, respectively. The main phase of the storm rapidly developed after 20 UT on March 13. The magnitude of the Dst index decreased by 330 nT in 5 h and reached a minimum of -589 nT.
This event is also well known for its practical implications. The Hydro–Quebec power system collapsed because of the strong geomagnetically induced current (GIC) that flowed in power lines in North America. As a result, six million residents in the province of Quebec were left without electrical power for more than 9 h. Furthermore, a low-latitude aurora was visible across the southern United States during the night of March 13 and early hours of March 14 (Allen et al. 1989). Although the solar wind information is not available for this period, geospace disturbances during this extreme storm have been studied by using satellite data (e.g., Fujii et al. 1992; Greenspan et al. 1991; Okada et al. 1993; Rasmussen and Greenspan 1993; Rich and Denig 1992; Shinbori et al. 2005; Sojka et al. 1994) and ground-based ionospheric data (e.g., Batista et al. 1991; Hajkowicz 1991; Lakshmi et al. 1991; Walker and Wong 1993; Yeh et al. 1992).
List of optical and X-ray solar flares with X-ray intensities of more than M5.0
Type II, IV
The shock transit speed of this event was also not extreme. According to Cliver et al. (1990), the solar flare associated with the first SSC of this storm event was identified as the March 10 X4.5/3B flare, which is in the top row of Table 1. McKenna-Lawlor et al. (2005) also calculated an interplanetary shock propagation by using the HAFv2 model with the solar origin of the X4 flare on March 10 and the first SSC of the storm on March 13–14. The transit time from the occurrence of the solar flare to the first SSC in this event has been estimated to be 54.8 h, which is 3.8 times longer than the event on that occurred on August 4, 1972, with a transit time of 14.6 h (Cliver et al. 1990). The maximum solar wind speed of the interplanetary shock that produced the first SSC of this event was estimated to be about 550 km/s in their paper. However, they did not discuss the second SSC of this event and its solar origin. At the time of the second SSC, the Dst had dropped to ~ −130 nT. The rapid enhancement of the ring current started at about 20 UT on March 13, approximately 12 h after the second SSC. The existence of the two SSCs strongly suggests that this great geomagnetic storm was caused by the combination of two CMEs (solar flares).
Estimating solar wind conditions during the March 13–14, 1989, event
In this study, we focus on the GMC by GOES. Because of the compression due to the high dynamic pressure of the solar wind and the erosion due to the enhanced dayside reconnection based on the intense southward interplanetary magnetic field (IMF), the location of the dayside magnetopause shrank and sometimes reached inside the geosynchronous orbit (6.6 Earth radii).
During the period from 14 UT on March 13 to 01 UT on March 14, several negative excursions of the GSM-Z magnetic field component were recorded for both GOES 06 and 07, which is a manifestation of the GMC. The extreme duration of the GMC in this event should be noted here. Excluding the time period from 1735 to 1945 UT, the GMC is continued for almost 11 h, which means that the GMC occurred during almost the entire daytime period of the GOES satellites.
Another important point to note here is that both the GSM-Y and GSM-Z components of the magnetic field variations from GOES 06 and 07 were almost coherent during the GMC. This characteristic suggests that both GOES 06 and 07 simultaneously observed the magnetosheath field during the GMC, which was the IMF modulated by the bow shock. Therefore, it should be possible to estimate the variations of the IMF from the magnetic field observations of GOES 06 and 07 during the GMC in the same manner as that of a solar wind monitoring satellite.
By analyzing the data of the storm that occurred on November 20–21, 2003, which is the fifth largest storm since 1957, Nagatsuma et al. (2007) showed that the GSM-Y and GSM-Z components of the magnetic field variations from GOES during the GMC were four times larger than the IMF assuming that the Alfvén Mach number (M A) of the solar wind was more than 2. Thus, the GSM-Y and GSM-Z components of the IMF were estimated to be a quarter of the magnetic field variations observed by GOES 06 and 07 during the GMC when M A for the solar wind was more than 2.
where r 0, α, D p, and B z represent the standoff distance at the subsolar point, the level of tail flaring, the solar wind dynamic pressure, and north–south component of the IMF, respectively. To estimate D p from these equations, the IMF B z estimated from the GOES observations is used as an input parameter. The estimation of D p from this empirical model is limited to the time at which the solar zenith angle of the satellite is less than 60° because this model should be applied only around the noon sector.
Another way to estimate the solar wind parameters is to use the geomagnetic index, which is a manifestation of the solar wind–magnetosphere–ionosphere coupling. By using an empirical model of geomagnetic indices, Nagatsuma et al. (2007) estimated the VBs and the merging electric field (E m) from the Dst and the PC index for the Northern Hemisphere (PCN) during the storm that occurred on November 20–21, 2003, respectively. They also showed that these estimations are in good agreement with values estimated from satellite observations of the solar wind. We applied the same method of VBs estimation to this event.
Next, we estimated the solar wind speed (V sw) from the estimated IMF B z and VBs averaged over 1 h. The estimated solar wind speed is highly scattered because this estimation is rather sensitive to the magnitude of the IMF B z. The average solar wind flow speed during this period was 960 km/s.
Results and discussion
List of X-ray solar flares of more than M5.0 and estimated shock speed and maximum solar wind speed
Previous studies have suggested that solar flares of modest intensity and duration can produce severe geomagnetic storms. In the case of the solar sources of the Bastille geomagnetic storm that occurred on July 15–17, 2000, with a Dst minimum of −301 nT, the duration of the solar flare was 40 min (Andrews 2001). In case of the solar sources of the Halloween geomagnetic storm that occurred November 20–21, 2003, with a Dst minimum of −422 nT, three long-durational event (LDE) flares with intensities less than M5.0 were reported (e.g., Gopalswamy et al. 2005). This demonstrates that a strong geoeffective solar wind driver can be produced from a solar flare modest of intensity and duration.
The dynamic pressure of the solar wind in this event may have been extreme because the GMC occurred around the noon sector and in the dawn and dusk sectors. This indicates that extreme dynamic pressure enhancement may have compressed the magnetosphere. However, it is unclear whether we can apply the empirical formula for the magnetopause location to such an extreme condition, especially for the dawn and dusk sectors, because the static gas pressure may have some effect on the magnetopause location, which is not considered in this empirical formula. By simply applying the empirical formula for the dawn and dusk sectors of the GMC on March 13–14, 1989, the solar wind dynamic pressure was estimated to be more than 100 nPa. However, there is insufficient information to confirm such an extreme dynamic pressure of the solar wind. This point should be examined in detail for future extreme storm events by considering a variety of data.
Our method of data analysis can be applied to severe geomagnetic storm events with no solar wind information. This type of reconstruction of solar wind information will be important for studying extreme geomagnetic storms in the past.
By using GOES magnetic field data and the Dst index, we have examined the solar wind conditions during the extreme geomagnetic storm that occurred on March 13–14, 1989. The solar flare possibly associated with the second SSC of this storm event was identified as the March 12 M7.3/2B flare. During the 5 h in which the main phase of the storm rapidly developed, IMF B z was estimated to be about −50 nT with a solar wind speed of about 960 km/s, assuming an M A during this period of more than 2. The number density of the solar wind during the GMC was at least 8–60/cm−3. Although these solar wind conditions are similar to those for the storm that occurred on November 20–21, 2003, the duration of the southward IMF was longer and the solar wind speed was 1.5 times faster in the latter event. This means that events such as the geomagnetic storm that occurred on March 13–14 could have occurred more frequently than previously thought.
advanced composition explorer
- M A :
Alfvén Mach number
central meridian distance
coronal mass ejection
- VBs :
dawn to dusk solar wind electric field
geocentric solar magnetosphere
geomagnetically induced current
Geostationary Operational Environmental Satellite
geosynchronous magnetopause crossing
Hakamada–Akasofu–Fry version 2
interplanetary magnetic field
interplanetary monitoring platform
- E m :
merging electric field
Solar Maximum Mission
Solar Terrestrial Relations Observatory
storm sudden commencement
We thank the National Oceanic and Atmospheric Administration and the World Data Center for Geomagentism, Kyoto, for providing GOES magnetometer data and the Dst index, respectively. We also thank the ACE instrument teams (SWEPAM and MAG) and the ACE Science Center for providing the ACE data. The OMNI data were obtained from the GSFC/SPDF OMNIWeb interface at http://omniweb.gsfc.nasa.gov.
- Allen J, Sauer H, Frank L, Reiff P (1989) Effects of the March 1989 solar activity. Eos Trans AGU 70(46):1479–1488. doi:10.1029/89EO00409View ArticleGoogle Scholar
- Andrews MD (2001) LASCO and EIT observations of the Bastille Day 2000 solar storm. Solar Phys 204:179–196View ArticleGoogle Scholar
- Baker DN, Li X, Pulkkinen A, Ngwira CM, Mays ML, Galvin AB, Simunac KDC (2013) A major solar eruptive event in July 2012: defining extreme space weather scenarios. Space Weather 11:585–591. doi:10.1002/swe.20097View ArticleGoogle Scholar
- Batista IS, De Paula ER, Abdu MA, Trivedi NB, Greenspan ME (1991) Ionospheric effects of the March 13, 1989, magnetic storm at low and equatorial latitudes. J Geophys Res 96(A8):13943–13952. doi:10.1029/91JA01263View ArticleGoogle Scholar
- Belov AV, Eroshenko EA, Oleneva VA, Yanke VG (2008) Connection of Forbush effects to the X-ray flares. J Atomos Sol Terr Phys 70:342–350. doi:10.1016/j.jastp.2007.08.021View ArticleGoogle Scholar
- Burkepile JT, St. Cyr OC (1993) A revised and expanded catalogue of mass ejections observed by the solar maximum mission coronagraph. NCAR, TN-369 + STR
- Cliver EW, Feynman J, Garrett HB (1990) An estimate of the maximum speed of the solar wind, 1938–1989. J Geophys Res 95(A10):17103–17112. doi:10.1029/JA095iA10p17103View ArticleGoogle Scholar
- Cliver EW, Balasubramaniam KS, Nitta NY, Li X (2009) Great geomagnetic storm of 9 November 1991: association with a disappearing solar filament. J Geophys Res 114:A00A20. doi:10.1092/2008JA013232
- Feynman J, Hundhausen AJ (1994) Coronal mass ejections and major solar flares: the great active center of March 1989. J Geophys Res 99(A5):8451–8464. doi:10.1029/94JA00202View ArticleGoogle Scholar
- Fry CD, Sun W, Deehr CS, Dryer M, Smith Z, Akasofu S-I, Tokumaru M, Kojima M (2001) Improvements to the HAF solar wind model for space weather predictions. J Geophys Res 106(A10):20985–21001. doi:10.1029/2000JA000220View ArticleGoogle Scholar
- Fry CD, Dryer M, Smith Z, Sun W, Deehr CS, Akasofu S-I (2003) Forecasting solar wind structures and shock arrival times using an ensemble of models. J Geophys Res 108(A2):1070. doi:10.1029/2002JA009474View ArticleGoogle Scholar
- Fujii R, Fukunishi H, Kokubun S, Sugiura M, Tohyama F, Hayakawa H, Tsuruda K, Okada T (1992) Field-aligned current signatures during the March 13–14, 1989, great magnetic storm. J Geophys Res 97(A7):10703–10715. doi:10.1029/92JA00171View ArticleGoogle Scholar
- Gopalswamy N, Yashiro S, Michalek G, Xie H, Lepping RP, Howard RA (2005) Solar source of the largest geomagnetic storm of cycle 23. Geophys Res Lett 32:L12S09. doi:10.1029/2004GL021639
- Greenspan ME, Rasmussen CE, Burke WJ, Abdu MA (1991) Equatorial density depletions observed at 840 km during the great magnetic storm of March 1989. J Geophys Res 96(A8):13931–13942. doi:10.1029/91JA01264View ArticleGoogle Scholar
- Hajkowicz LA (1991) Global onset and propagation of large-scale travelling ionospheric disturbances as a result of the great storm of 13 March 1989. Planet Space Sci 39(4):583–593. doi:10.1016/0032-0633(91)90053-DView ArticleGoogle Scholar
- Kataoka R (2013) Probability of occurrence of extreme magnetic storms. Space Weather 11:214–218. doi:10.1002/swe.20044View ArticleGoogle Scholar
- Lakshmi DR, Rao BCN, Jain AR, Goel MK, Reddy BM (1991) Response of equatorial and low latitude F-region to the great magnetic storm of 13 March 1989 (1991). Ann Geophys 9:286–290Google Scholar
- Li X, Temerin M, Tsurutani BT, Alex S (2006) Modeling of 1–2 September 1859 super magnetic storm. Adv Space Res 38:273–279View ArticleGoogle Scholar
- McKenna-Lawlor SMP, Dryer M, Fry CD, Sun W, Lario D, Deehr CS, Sanahuja B, Afonin VA, Verigin MI, Kotova GA (2005) Predictions of energetic particle radiation in the close Martian environment. J Geophys Res 110(A3):A03102. doi:10.1029/2004JA010587Google Scholar
- Nagatsuma T, Asai KT, Kataoka R, Hori T, Miyoshi Y (2007) S–M–I coupling during the super storm on 20–12 November 2003. Adv Geosci 14:237–244Google Scholar
- O’Brien TP, McPherron RL (2000) An empirical phase space analysis of ring current dynamics: solar wind control of injection and decay. J Geophys Res 105(A4):7707–7719. doi:10.1029/1998JA000437View ArticleGoogle Scholar
- Okada T, Hayakawa H, Tsuruda K, Nishida A, Matsuoka A (1993) EXOS D observations of enhanced electric fields during the giant magnetic storm in March 1989. J Geophys Res 98(A9):15417–15424. doi:10.1029/93JA01128View ArticleGoogle Scholar
- Rasmussen CE, Greenspan ME (1993) Plasma transport in the equatorial ionosphere during the great magnetic storm of March 1989. J Geophys Res 98(A1):285–292. doi:10.1029/92JA02199View ArticleGoogle Scholar
- Rich FJ, Denig WF (1992) The major magnetic storm of March 13–14, 1989, and associated ionosphere effects. Can J Phys 70(7):510–525. doi:10.1139/p92-086View ArticleGoogle Scholar
- Rufanach CL, Martin RF Jr, Sauer HH (1989) A study of geosynchronous magnetopause crossings. J Geophys Res 94(A11):15125–15134. doi:10.1029/JA094iA11p151254View ArticleGoogle Scholar
- Shinbori A, Nishimura Y, Ono T, Iizima M, Kumamoto A, Oya H (2005) Electrodynamics in the duskside inner magnetosphere and plasmasphere during a super magnetic storm on March 13–15, 1989. Earth Planets Space 57:643–659View ArticleGoogle Scholar
- Shue JH, Song P, Russell CT, Steinberg JT, Chao JK, Zastenker G, Vaisberg OL, Kokubun S, Singer HJ, Detman TR, Kawano H (1998) Magnetopause location under extreme solar wind conditions. J Geophys Res 103(A8):17691–17700. doi:10.1029/98JA01103View ArticleGoogle Scholar
- Shue JH, Song P, Russell CT, Chao JK, Yang YH (2000) Toward predicting the position of the magnetopause within geosynchronous orbit. J Geophys Res 105(A2):2641–2656. doi:10.1029/1999JA900467View ArticleGoogle Scholar
- Sojka JJ, Schunk RW, Denig WF (1994) Ionospheric response to the sustained high geomagnetic activity during the March ‘89 great storm. J Geophys Res 99(A11):21341–21352. doi:10.1029/94JA01765View ArticleGoogle Scholar
- Tanaka T (2007) Magnetosphere–ionosphere convection as a compound system. Space Sci Rev. doi: 10.1007/s11214-007-9168-4
- Walker GO, Wong YW (1993) Ionospheric effects observed throughout East Asia of the large magnetic storm of 13–15 March 1989. J Atmos Terr Phys 55(7):995–1008. doi:10.1016/0021-9169(93)90093-EView ArticleGoogle Scholar
- Yeh KC, Lin KH, Conkright RO (1992) The global behavior of the March 1989 ionospheric storm. Can J Phys 70(7):532–543. doi:10.1139/p92-088View ArticleGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.