Stormenhanced plasma density features ... - Wiley Online Library

Click Here

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 115, A06305, doi:10.1029/2009JA014674, 2010

for

Full Article

Storm‐enhanced plasma density features investigated during the Bastille Day Superstorm Ildiko Horvath1,2 and Brian C. Lovell1,2 Received 23 July 2009; revised 29 December 2009; accepted 7 January 2010; published 15 June 2010.

[1] Field‐aligned passes track true profiles. Such Defense Meteorological Satellite Program

passes permitted investigating storm‐enhanced plasma density (SED) feature development during the Bastille Day Superstorm in a comprehensive way. We tracked equatorial ionization anomaly (EIA) and SED features and their underlying forward fountain circulation and downward SED plume plasma flows, respectively. Northward subauroral polarization stream E fields detaching plasma and producing SED plumes were also detected. We assessed the effects of South Atlantic Magnetic Anomaly and summer‐to‐ winter interhemispheric plasma flows on the EIA and found a southward dipping gradient in drift/flow when no storm/evening‐related fountain strengthening occurred. We investigated the relative importance of different plasma sources in SED development. An extremely large plasma enhancement seen over Florida at 2200 UT on 15 July 2000 was a SED feature that was tracked by many GPS total electron content (TEC) maps as a 200 TEC unit (TECU) enhancement. We tracked its equally large conjugate pair over Trelew (Argentine Patagonia) and unraveled their development. Their underlying SED plume supplied most of the plasma. Appearing between these two SED features, a small and highly asymmetrical EIA offered on each side a low baseline upon which the downward streaming SED plume plasma piled up. Contradicting a currently accepted explanation, there was no enhanced fountain action detected to contribute 150 TECU to the 200 TECU. Later (∼2400 UT), there was enhanced fountain action, but SED plume contribution still dominated. Proven by observational evidence, SED development is a complex process of SED plume plasma flows and equatorward wind effects that cannot be described by one single explanation. Citation: Horvath, I., and B. C. Lovell (2010), Storm‐enhanced plasma density features investigated during the Bastille Day Superstorm, J. Geophys. Res., 115, A06305, doi:10.1029/2009JA014674.

1. Introduction [2] In 2000, a coronal mass ejection (CME) emitted from the Sun at around midday on 14 July, a day celebrated by the French as Bastille Day. On the following day, 15 July, this CME hit the Earth’s magnetosphere and triggered the so‐called Bastille Day Superstorm [Basu et al., 2001; Kil et al., 2003] that is one of the largest superstorms of solar cycle 23. The Bastille Day Superstorm has attracted considerable interest due to the development of a well‐defined magnetospheric tail [Kelley et al., 2004] and its occurrence near solstice [Kil et al., 2003]. Its other notable features are the large plasma depletions appearing over the South Atlantic Magnetic Anomaly (SAMA) [Lin and Yeh, 2005] and the

1 Security and Surveillance, School of Information Technology and Electrical Engineering, University of Queensland, St. Lucia, Brisbane, Queensland, Australia. 2 National ICT Australia, Queensland Research Laboratory, St. Lucia, Brisbane, Queensland, Australia.

Copyright 2010 by the American Geophysical Union. 0148‐0227/10/2009JA014674

extremely large storm‐enhanced plasma density (SED) feature seen over Florida [Vlasov et al., 2003; Kelley et al., 2004]. Lin et al. [2007] associated these plasma density depletions with the steady increases of both the zonal (eastward) and the radial (outward) convective electric (E) field components. It was also suggested that these locally increased E fields could drive the plasma from the SAMA region to the SED features in the conjugate hemisphere, and thus this plasma transport could be the plasma source producing these SED features [Lin et al., 2007]. However, Kelley et al. [2004] explained the development of SED feature appearing over Florida with westward advection and enhanced fountain effects caused by prompt penetrating E fields. Vlasov et al. [2003] also noted the unusual behavior of the equatorial ionization anomaly (EIA) in the American longitude sector and over the Pacific but investigated the northern crest only. These above described studies clearly indicate that the development of SED features during the Bastille Day Superstorm is still not clear. Furthermore, it is also unclear what caused the unusual behavior of the EIA. [3] During magnetic storms, SED features regularly develop equatorward of the base of westward subauroral polarization stream (SAPS), on the equatorward edge of the midlatitude

A06305

1 of 13

A06305

A06305

HORVATH AND LOVELL: EXTREMELY LARGE SED FEATURES

Table 1. Ground‐Based Magnetometer Observatory Sites Providing Data for This Study and Their Geographic Locations and Geomagnetic Latitudes Station

Geographic Latitude (°N)

Geographic Longitude (°E)

Geomagnetic Latitude (°N)

Ascension Island Bangui Bay St Luis Chichijima Guam Hermanus Huancayo M’Bour San Juan San Pablo‐Toledo Trelew Tsumeb

−7.95 4.33 30.35 27.10 13.59 −34.42 −12.04 14.38 18.11 39.55 −43.25 −19.20

345.62 18.57 270.37 142.18 144.87 19.23 284.68 343.03 293.15 355.65 294.68 17.58

−2.36 4.20 40.05 18.47 5.30 −33.98 1.80 20.11 28.31 42.78 −33.05 −18.77

trough in the dusk sector [Foster, 1993; Foster and Rich, 1998], and are magnetically connected to the plasmaspheric tail or detached plasma regions [Chappell, 1974; Chen and Grebowsky, 1974; Carpenter et al., 1993]. Detachment processes are caused by northward SAPS E fields [Foster and Burke, 2002; Foster et al., 2007] and produce sunward streaming SED plumes in the plasmasphere [Foster et al., 2002]. SED plume downward plasma flows are fed by the peeled off (i.e., detached) plasmaspheric plasma [Carpenter et al., 1993]. Although Su et al. [2001] demonstrated that SED features are the low‐altitude signatures of SED plumes, the SED plume plasma’s downward movement has been regularly overlooked as the attention was focused on the westward movement. This fact was highlighted in our recent study [Horvath and Lovell, 2009a]. We presented observational evidence [Horvath and Lovell, 2009a] that the SED features detected during the November superstorms were created by large downward plasma flows, and that the downward streaming plasma was the combination of SED plume plasma generated by detachment processes [Foster et al., 2002] and plasma from enhanced fountain effects [Vlasov et al., 2003; Kelley et al., 2004]. Invoking the recent modeling results of Balan et al. [2008, 2009], we also discussed the significance of equatorward neutral winds in maintaining the SED features’ high plasma densities [Horvath and Lovell, 2009a]. It has been proven by recent studies utilizing various models [e.g., Lin et al., 2005; Vijaya Lekshmi et al., 2008; Lu et al., 2008; Balan et al., 2008, 2009, 2010] that prompt penetration E field can only increase low‐ latitude and midlatitude plasma densities in the presence of equatorward neutral winds (or less poleward winds) and that equatorward wind alone can produce stronger plasma density enhancements than together with a prompt penetration E field. [4] This study’s main aims are to further investigate the development of SED features and the unusual evening/ nighttime EIA behavior over both hemispheres, and to explore how the EIA was related to the SED features during the Bastille Day Superstorm. Results will provide additional information to our recent SED studies [Horvath and Lovell, 2009a] regarding the relative importance of neutral winds, enhanced fountain effects and SED plume plasma in the development of SED features. How the EIA and SED features, and their underlying plasma flows were influenced by prompt penetration E field, SAMA effects, and strong summer‐to‐winter

interhemispheric and equatorward/poleward hemispheric neutral winds will also be revealed.

2. Database and Methodology [5] To study the response of evening and nighttime topside ionosphere to the Bastille Day Superstorm, the period of 29 June–19 July 2000 was investigated utilizing multi‐ instrument in situ Defense Meteorological Satellite Program (DMSP) data detected at ∼840 km along the ascending (northbound) passes of spacecraft F12, F13, F14 and F15. Orbit characteristics are described in detail by Horvath [2006, 2007] and Horvath and Lovell [2009a]. Our DMSP database contains ion density (Ni; i+/cm3), electron temperature (Te; °K), vertical (Z) and horizontal (Y) drift (VZ and VY; m/s) measurements, and derived vertical plasma flow {FZ; in [i+/(cm2s)]} values. Altogether 305 passes were analyzed. Regional surface maps provided continuous surface coverage, and line plots revealed spatial and temporal variations. We also noted the passes’ alignment with respect to the magnetic field lines, and innovatively gave preference to field‐aligned passes and to detection made perpendicular to the magnetic field lines. Such observations provide scientifically significant information, as the EIA and SED features are created by field‐aligned plasma flows and the important plasma movements take place in the horizontal direction (perpendicular to the magnetic field lines) at high latitudes. [6] Interplanetary magnetic field (IMF) vectors (BX, BY, BZ; nT) measured by the Advanced Composition Explorer (ACE) [Stone et al., 1998] satellite and a small collection of geomagnetic indices [Kp, Dst (nT), AE (nT)] were employed to monitor the underlying magnetic activity. DH (Hequ − Hnon−equ; nT) values were computed for various longitude sectors. DH was considered as an index of the equatorial electrojet (EEJ) strength [Rastogi and Klobuchar, 1990], and provided an indication of zonal E field and equatorial E × B drift variations [Anderson et al., 2002; Fejer et al., 2007]. Since we could not estimate vertical plasma drift, the DH data were utilized to observe E × B drift variations. Four equatorial‐non‐equatorial magnetometer station pairs: (1) Tsumeb‐Bangui, (2) Guam‐Chichijima, (3) Huancayo‐San Juan, and (4) Ascension Island‐M’Bour were considered (see Table 1). These are situated in the (1) African (∼18°E; geographic), (2) Australian (∼145°E), (3) American (∼290°E) and (4) Atlantic (∼345°E) longitude

2 of 13

A06305


A06305

Figure 1. (top) The regional surface maps depict the distribution of plasma density (Ni), vertical plasma flow (FZ), and vertical and horizontal plasma drifts (VZ; VY) in the quiet time topside ionosphere at evening. Their most apparent characteristics are the signatures of summer–to–winter interhemispheric plasma flows (in FZ) and SAMA effects (in VZ and VY; see circled areas). (bottom) Same as above but with average field‐ aligned profiles constructed for 312°E (geomagnetic). The white dots indicate the locations of largest SED features detected during the Bastille Day Superstorm. These SED features are shown in Figures 3a, 4a, and 5a.

sector, respectively. We also computed total B field (F; nT) values to observe B field strength variations at Hermanus (South Africa), San Juan (Puerto Rico), Bay St. Luis (Mississippi), Trelew (Argentine Patagonia) and San Pablo‐ Toledo (Spain; see Table 1). Only limited simulations were offered by the Coupled Thermosphere‐Ionosphere Plasmasphere (CTIP) model [Fuller‐Rowell et al., 1996; Webb et al., 2009] for 15 July 2000. For observing global neutral wind

variations, we chose the running times of 1912:00 UT and 2136:00 UT.

3. Observational Results and Interpretations 3.1. Characteristics of the Evening Topside Ionosphere Before the Bastille Day Superstorm [7] In order to get a perspective on the state of the quiet time evening topside ionosphere before the Bastille Day

3 of 13

A06305


Superstorm, we investigated the period of 29 June to 8 July 2000 (Kp < 3+) with a series of maps and field‐aligned line plots. In Figure 1 (top), these maps show how the plasma density (Ni), vertical plasma flows (FZ), and vertical and horizontal plasma drifts (VZ and VY) varied over the two hemispheres. In Ni, the most outstanding features are the pronounced asymmetry about the geomagnetic equator and a small equatorial peak (>160·103 i+/cm3; red) appearing over the dip equator (∼1800 LT). Lee et al. [2002] noted this asymmetry in the longitude sector of 230–350°E only. This asymmetry was caused by some large plasma depletions (370 m/s; red) are related to the auroral two‐cell convection zone [Heelis and Mohapatra, 2009]. Shown by the maximum eastward (antisunward) drifts [−(230–175) m/s; dark brown], antisunward convections completed the two‐cell convection pattern [Heelis and Mohapatra, 2009] over the polar cap (60–240°E). From the averaged data of these maps, a series of field‐aligned average profiles are constructed for the American longitude sector (∼312°E) where the magnetic field lines are almost north‐ south directed. Plotted in the same scale as the storm time line plots, so they can be compared, these average field‐aligned line plots illustrate the true latitudinal cross sections of the above described quiet time features. 3.2. Description of the 14–16 July 2000 Events [8] Plotted for the period of 14–17 July 2000 in Figure 2a, the IMF vectors and various indices depict the nature of the Bastille Day Superstorm. There were several substorms before the sudden commencement (SSC) occurred at ∼1500 UT on 15 July [Basu et al., 2001]. During the initial phase (I), all three IMF components turned northward (+) and the AE index attained a sub peak (1450 nT). Marking the beginning of main phase (M) at ∼1930 UT, the IMF BZ component turned southward (−) and started its steep decrease toward −60 nT. Meanwhile, the AE index maximized (1700 nT) at ∼2000 UT. On the following day (16 July) at

A06305

∼0100 UT, the Dst index reached its minimum of −370 nT and thus registered a superstorm. A few hours later, at ∼0300 UT, the Kp index reached 9. The following sudden increase of BZ (from −16 nT to 16 nT) triggered the recovery phase (R) that covered also the next two days (17 and 18 July). [9] In Figure 2b, the DH (nT) plots show how the EEJ intensity varied in the four chosen longitude sectors (see details in section 2). Since EEJ variations reflect zonal E field variations, DH (nT) values also provide E × B drift estimations [Rastogi and Klobuchar, 1990; Anderson et al., 2002]. During the main phase, there was a large positive DH amplitude (∼170·103 nT) in the Australian daytime sector, and a smaller negative DH amplitude (∼−80·103 nT) in the nighttime African and Atlantic sectors. In the American evening sector, the maximum positive DH amplitude (∼325·103 nT) indicates an upward E × B drift surge due to the increased net E field and to the SAMA’s weak B field as the drift magnitude is E × B/B2 [Abdu et al., 2005]. Figure 2c depicts the geographic locations of station pairs and other observatories providing magnetometer data. Figures 2d and 2e illustrate how the total B field intensity varied in the American longitude sector and around Africa. During the main phase, there was a sudden drop in B field intensity in the SAMA region. However, the drop was less significant away from the SAMA region where the F component was stronger, like at Bay St. Luis. [10] As the DH variations indicate, there were two sudden increases in EEJ strength and consequently in equatorial upward E × B drift during the main phase: (1) between 2000 UT and 2100 UT, when the Dst index dropped down to −130 nT, and (2) between 2100 UT and 2200 UT, when the Dst index reached −295 nT [Basu et al., 2001]. There was a further increase, (3) between 2200 UT and 2400 UT, when Dst reached −370 nT [Kil et al., 2003]. We tracked EIA and SED features with their underlying plasma flows and drifts during the first and third periods in the evening sector at different local times. 3.3. First Period of Upward ExB Drift Increase: 2000–2100 UT on 15 July 2000 [11] In Figures 3a and 3b, two CTIP simulated wind vector maps show how the global wind varied at 300 km altitude just before and after this first period, at ∼1900 UT and ∼2130 UT, respectively. Two field‐aligned Ni line plots illustrate at these times the sudden redevelopment of the EIA over Africa and the Atlantic, respectively. This redevelopment was due to the storm conditions, during which the fountain and its ionospheric signature, the EIA, can reach DMSP altitude (∼840 km [Balan et al., 2009, 2010]) resulting in the detection of distinct EIA crests and well‐defined EIA trough (see line plots below). In these longitude sectors, the equatorward winds developed at auroral latitudes were strongest (up to 1127 m/s) and reached low latitudes in the north in all longitudes and only in some longitudes (0–160°E) in the south. Over the South Atlantic, the winds were mainly poleward directed at auroral latitudes and moderate. 3.3.1. Asymmetrical Forward Fountain Circulation Underlying a Symmetrical EIA [12] Figures 4b and 4c are constructed with data from the above described F12 and F13 passes, respectively. With another pass (see Figure 4d), the EIA and underlying forward

4 of 13

A06305


A06305

Figure 2. (a) The various line plots depict the nature of the Bastille Day Superstorm. (b) EEJ strength variations, reflecting E × B drift variations, are estimated with the DH data computed for four longitude sectors. (c) The global map depicts the locations of ground‐based magnetometer observatories providing data for this study. (d and e) The F line plots depict the variation of total B field intensity during the period investigated. fountain circulation will be investigated for the effects of prompt penetration E field, summer‐to‐winter interhemispheric plasma flows, SAMA and hemispheric neutral winds. [13] In Figure 4b, the latitudinal Ni profile tracked at ∼20.30 LT over Africa the EIA’s redevelopment producing symmetrical crests [∼40·104 (i+/cm3)]. Proven by observational evidence, presented and described below, a SED feature

(indicated as S) in each hemisphere appeared equatorward of both the plasmapause and the midlatitude trough [Foster, 1993]. Their coinciding locations are indicated by a subauroral Te peak [Prölss, 2006; Horvath and Lovell, 2009b]. Here in the African sector, the northern (summer) upward and southern (winter) downward plasma flows, and the SAMA effects were still strong (see details in section 3.1). Their

5 of 13

A06305


A06305

Figure 3. CTIP simulated wind vector maps show global wind variations. Two color‐coded field‐aligned Ni plots show the sudden redevelopment of the equatorial ionization anomaly over Africa and the American continents. combined effects created and maintained over low latitudes a southward dipping gradient in vertical plasma flows and drifts (see FZ and VZ plots), and distorted the forward plasma fountain’s usual flow pattern and drift pattern. Normal fountain circulation was demonstrated in our previous study [Horvath and Lovell, 2008] and was explained invoking SUPIM results [Balan and Bailey, 1995; Balan et al., 1997; Bailey et al., 1997]. In this case, due to this southward dipping gradient, the fountain’s upward plasma flow (see red arrow) is small. At the northern EIA crest, the fountain’s downward plasma flow (see blue arrow) became decreased by the interhemispheric upward (summer) flows. Oppositely, at the southern EIA crest, the fountain’s downward plasma flow (see blue arrow) became further increased by the interhemispheric downward (winter) flows. The VZ plot tracked a similar pattern. However, these asymmetrical flow patterns and drift patterns were underlying a symmetrical EIA showing a ∼2.5 times increase with respect to the quiet time maximum Ni of ∼15·104 (i+/cm3) (see Figure 1). These could be due to the converging equatorward neutral winds (see CTIP wind vector map in Figure 3a) and to the strengthening of the plasma fountain caused by the prompt penetration E field suddenly increasing the vertical upward E × B drift. In the local (Bangui‐Tsumeb) DH plot, the positive DH amplitude indicates that the shock impact increased suddenly the EEJ strength, eastward E field and vertical upward E × B drift, and thus caused the EIA’s sudden redevelopment. [14] There was an asymmetry in the midlatitude plasma densities due to differences in SED feature (S) development. The southern SED feature appeared in a significantly better developed form (∼45·104 i+/cm3) and at lower latitudes (∼30°S; geomagnetic) where the equatorward winds were

converging (see CTIP wind vector map in Figure 3a). Its northern counterpart was significantly less developed (∼13·104 i+/cm3) and appeared at higher latitudes (∼50°N) where the equatorward winds were strongest (see CTIP wind vector map in Figure 3a). We tracked increased northward SAPS E fields (see VY plots) that create plasma detachment events [Chappell, 1974; Chen and Grebowsky, 1974; Carpenter et al., 1993] that in turn produce downward SED plume plasma flows, which were detected by the FZ plot, building up SED features (see details in section 1). These VY and FZ line plots provide observational evidence that the plasma enhancements (S) appearing equatorward of the trough are SED features. The significant differences between the SED features tracked could be due to differences in the northward SAPS E fields arising from the different UTs of detection, and to the SAMA effects and hemispheric wind effects. Unaffected by the SAMA, the northern SED feature was detected earlier, at the end of the initial phase, when a weaker SAPS E field increase (see VY plots showing an increase from 800 to 1000 m/s) created a weaker plasmaspheric detachment event [Foster et al., 2002] that in turn produced a weak downward SED plume plasma flow [∼108 (i+/(cm2s)]. However, the strong equatorward winds (see CTIP wind vector map in Figure 3a) helped keeping the ionization at higher altitudes and lower recombination rates. Oppositely, at a latter UT, during the main phase, there was a significantly larger intensification at southern latitudes in northward SAPS E field magnitude (see VY plots showing an increase from 700 to 1400 m/s). This increase created a stronger plasma detachment event [Foster et al., 2002] in the south that in turn produced a stronger downward SED plume plasma flow that was quite large [∼60·108 (i+/(cm2s); ∼two times of the

6 of 13

A06305


Figure 4. (a) The global map shows the ground tracks of some DMSP‐F12, ‐F13, and ‐F15 passes with the modeled magnetic field lines and different equators. (b–d) The line plot series depict the plasma density, thermal and vertical flow structures, and vertical and horizontal drift patterns of the evening/nighttime topside ionosphere during the first period at the beginning of main phase. (e) The underlying magnetic conditions were probed with the IMF BZ component and AE index. Equatorial EEJ (E × B drift) estimates are illustrated by the local DH plots. The F line plots illustrate the variation of local total B field intensity. 7 of 13

A06305

A06305


quiet time values of −(38.2–23.6) ·108 (i+/(cm2s) in blue] and intensive as the scattered nature of the Ni detection indicates. These strong SED plume downward plasma flows were tracked over Hermanus (South Africa), in the SAMA region. There, the SAMA related downward drifts (see details in section 3.1) contributed to their acceleration and the strong converging equatorward winds (see CTIP wind vector map in Figure 3a) helped maintaining these high plasma densities. [15] We also tracked northward SAPS E field signatures and SAMA signatures of the local total B field (F) data. The F plot detected the low total B field intensity (∼26.46·103 nT) of the SAMA and a sudden drop in intensity (∼1 UT hour before this SED feature was detected) that remained low (∼26.30·103 nT) for 2 UT hours. Due to ring current increases that reduce B field strength [Brandt et al., 2001], this total B field intensity drop is a signature of increased northward SAPS E field. SAPS E fields are produced by ring currents [Foster and Burke, 2002; Foster et al., 2007], and thus a sudden B field decrease indicates increased northward SAPS E fields. 3.3.2. Symmetrical Forward Fountain Circulation Underlying a Symmetrical EIA [16] In Figure 4c, the Ni latitudinal profile shows the EIA at the western edge of the Atlantic, where both the summer‐to‐ winter interhemispheric plasma flows and the SAMA effects maximized during the magnetically quiet times investigated (see details in section 3.1). Its symmetrical peak plasma densities of ∼82·104 (i+/cm3) are ∼5.5 times larger than the maximum Ni detected during quiet conditions, while the Ni increase was ∼2 times over the equator. This pass was detected one UT hour later (at 20.65 UT) and ∼2.5 LT hours earlier (at ∼1800 LT) than the previous pass over Africa (see Figure 4b). Thus, these large plasma density increases could be due to the combined effects of maximum prompt penetration E field maximizing the vertical upward E × B drift and to the converging storm time equatorward winds (see CTIP wind vector map in Figure 3a). At this earlier time (∼1800 LT), when the F layer dynamo is normally active, the regular evening prereversal enhancement occurred in a more advanced stage of the main phase. Then, both the prompt penetration E field and the vertical upward E × B drift maximized because of the zonal E field enhancement [Abdu et al., 2008]. Supporting these explanations, the VZ plot tracked vertical upward net drift ∼200 m/s over the magnetic equator. While the upward E × B drift was maximum, the fountain‐related plasma flows and drifts remained symmetrical (see FZ and VZ line plots) as the interhemispheric plasma flows (see Figure 1) had no affect on this strengthened fountain [Balan and Bailey, 1995; Balan et al., 1997; Bailey et al., 1997]. Further evidence is provided by the local (Huancayo‐San Juan) DH plot showing that this detection was taken on the steep ascending phase of a large DH amplitude indicating an upward E × B surge that maximized later, during the main phase. [17] SED feature development taking place equatorward of the northern trough was evidenced by the various line plots tracking the increased northward SAPS E fields (see VY plot) that created detachment events [Foster et al., 2002], and the resultant downward SED plume plasma flows (see green arrows in FZ plot) and SED features (S1, S2 in Ni plot). S2 is a more prominent feature than S1, since this better developed S1 blended in the high plasma density poleward region of the EIA. Strong equatorward neutral winds (see CTIP wind vector map in Figure 3b) helped maintaining the high plasma

A06305

densities of these SED features. However, the southern SED feature (S) was least developed. This was due to a decreasing SAPS E field (see VY plots) that was combined with strong poleward directed winds (see CTIP wind vector map in Figure 3b) and SAMA effects (see VZ map in Figure 1), both moving the ionization down to lower altitudes of increased recombination. 3.3.3. Asymmetrical Forward Fountain Circulation Underlying an Asymmetrical EIA [18] During the main phase, the F15 spacecraft tracked a strongly asymmetrical plasma density profile (see Ni plot of Figure 4d) in the African sector. Based on the observational evidence presented in Figure 4d and described below, this asymmetry was mainly caused by asymmetrical EIA and SED feature development, and by interhemispheric and hemispheric wind and SAMA effects. We note that only one UT hour earlier, the F12 Ni line plot detected a symmetrical EIA over the same location (see Figure 4c). [19] Estimated by the local (Bangui‐Tsumeb) DH plot, the vertical upward E × B drift was decreasing indicating weakening forward fountain action that is extremely sensitive to wind effects [Balan and Bailey, 1995; Balan et al., 1997; Bailey et al., 1997]. In good agreement with the local DH data, at 21.36 LT (∼3.5 LT hours after the regular prereversal strengthening seen in Figure 4c), the FZ plot tracked a forward fountain (see red and blue arrows) and some SED plume plasma flows (see green arrows) that were highly asymmetrical and significantly weaker than seen in Figure 4b. This asymmetrical plasma flow pattern maintained a highly asymmetrical Ni profile across the two hemispheres with a stronger northern EIA crest and a larger northern SED feature. [20] In the north, these high plasma densities could be due to the strong equatorward winds (see CTIP wind vector map in Figure 3b) and interhemispheric upward (summer) plasma flows (see FZ map in Figure 1). A maximum Ni of ∼50·104 i+/cm3 is related to the EIA crest and shows a ∼3.5 times increase compared to the quiet time peak value seen in Figure 1. Meanwhile, a SED feature occurred (∼30·104 i+/cm3) over northern midlatitudes (over San Pablo‐Toledo; Spain) where the SAMA effects were absent. However, this SED feature (∼30°N; geomagnetic) is not an outstanding feature, as it blends in the high Ni region appearing poleward of the northern EIA crest. SED feature development was caused by the increased northward SAPS E field (VY ≈ 800 m/s) that triggered detachment events [Foster et al., 2002] and produced downward SED plume plasma flows (see green arrows) that were tracked by the FZ plot. A northward SAPS E field increase is also indicated by the local San Pablo‐Toledo total B field intensity (F) data of which magnitude was ∼2 times as large (∼44.11·103 nT) away from the SAMA region than in the SAMA region, at Hermanus. Indicating ring current and SAPS E field increases, there was a sudden drop in F at San Pablo‐Toledo that had occurred at around the time when the F12 spacecraft detected a SAPS E field increase (see VY plot), a SED plume plasma flow (see FZ plot) and a SED feature (see Ni plot). [21] In the south, there was a large plasma depletion (∼1·104 i+/cm3) around the larger area of Hermanus, between ∼15°S and the trough (∼60°S). There, the interhemispheric winter downward drifts combined with the SAMA‐related downward drifts and kept increasing (up to –450 m/s, which is less than the quiet time average of ∼–550 m/s) over low

8 of 13

A06305


latitudes and midlatitudes, and significantly decreased the southern EIA crest and its entire poleward region. Clearly, the mechanical effects of these combined downward plasma flows overruled those of strong equatorward winds occurring there (see CTIP wind vector maps in Figure 3b). Indeed, the process of southern plasma depletion seen in the quiet time data (see Figure 1) became further enhanced over low latitudes and midlatitudes during the main phase. Over Hermanus, the total absence of SED feature is obvious. Possibly, an increase in total B field intensity (at 20.69 UT; see Hermanus F plot) indicating ring current and SAPS E field decreases prevented any SED feature development. Supporting this explanation, the VY plots tracked a drop from 1100 to 900 m/s and thus show that there was a decrease in northward SAPS E field in the south. 3.4. Third Period of Upward ExB Drift Increase: 2200–2400 UT on 15 July 2000 [22] Shown in Figure 5, three line plot series provide observational evidence of the dramatic increase of SED features, up to ∼85·104 (i+/cm3) that is ∼5.5 times larger than the maximum Ni detected during quiet conditions (see details below). These increases were tracked in the American longitude sector and over the Eastern Pacific at ∼1800 LT during the third period (see details in section 3.2), covering the end of main phase and the beginning of recovery phase, and were maintained by storm time equatorward neutral winds. [23] At ∼2200 UT, when the Dst index reached its first minimum, the partially field‐aligned 12–13 passes of F13 were situated over the American continents (see map). The spacecraft made field‐aligned detection between southern and northern midlatitudes (see pink trace in Figure 5b). In Figure 5b, the line plot series tracked a large EIA‐like or double‐peak (∼ ± 30°N; geomagnetic) plasma density structure (see Ni plot), and their underlying plasma flows (see FZ plot) and drifts (see VZ plot). It becomes obvious from these Ni, FZ and VZ line plots that this double‐peak structure was the combination of a small asymmetrical EIA surrounded by two extremely large SED features situated at ∼ ± 30°N. Supporting evidence is provided by the FZ plot that tracked the underlying forward fountain circulation (see red and blue arrows) that was possibly created by a prompt penetration E field and downward SED plume plasma flows (see green arrow). Meanwhile, the VY plots detected in both hemispheres the signatures of some very strong northward SAPS E field intensifications that triggered detachment events that in turn produced SED plume plasma flows [Chappell, 1974; Chen and Grebowsky, 1974; Carpenter et al., 1993]. These high SED plasma densities were maintained by storm time equatorward neutral winds (see details in section 4). The northern SED feature was tracked over San Juan (close to Florida) where the total B field intensity (see F plots) was higher (∼39·103 nT) than over Trelew (Argentine Patagonia; ∼26.5·103 nT) where the southern SED feature appeared. A sudden drop in total B field intensity was registered by these local F plots at each of these SED features indicating also SAPS E field increases (see details in section 3.3). [24] Tracked by the Ni plot’s field‐aligned section shown in Figure 5b, we can see traces of the EIA crests (C) that developed at 15°N and 12.5°S (geomagnetic) or 4.1°N and 24.1°S (geographic), and that were embedded in the high background ionization of SED features. The EIA crest‐related

A06305

plasma densities are significantly lower than the SED‐related maximum plasma densities at ∼ ± 30°N (geomagnetic). The magnitude of northern EIA crest (∼45·104 i+/cm3) was only half of that of the northern SED feature (∼85·104 i+/cm3). The southern EIA crest was very small (∼22.5·104 i+/cm3), slightly larger than the minimum Ni detected at the EIA trough (∼20.0·104 i+/cm3) showing a ∼2.5 times increase compared to quiet conditions. Thus, the southern crest blended in the EIA trough region and created a broader low plasma density region. Although there is a southward dipping gradient in the underlying plasma flow structures and drift structures, similar to what we observed in Figures 4b and 4d), the FZ line plot of Figure 5b clearly shows that the SED‐ related plasma flows (see green arrow) and the fountain circulation (see blue arrows) are separate. [25] In Figure 5c, a non‐field‐aligned line plot set shows the forward superfountain’s behavior at around the same LT (1800 LT) as Figure 5b but 2 UT hours later (∼2400 UT), at the end of main phase when the Dst index was still rapidly decreasing. The Ni plot tracked a symmetrical EIA over the Eastern Pacific, where the SAMA and summer‐to‐winter interhemispheric plasma flow‐related effects were less intensive (see Figure 1). Both the symmetrical nature of this EIA and the underlying plasma circulation (see FZ plot) suggest that this forward superfountain was strengthening as the prompt penetration E field maximized due to zonal E field enhancement [Abdu et al., 2008]. Furthermore, it is also suggested that this strengthening took place simultaneously over a wider geographic area, so a non‐field‐aligned pass could also detect it. These suggestions are supported by the local (Huancayo‐San Juan) DH line plot indicating an upward E × B surge that peaked at ∼1.5 UT hours earlier. The time delay was possibly caused by the process of plasma buildup. Because the F13 13‐01 passes are not field‐aligned, the non‐field‐aligned FZ and VZ line plots do not show the typical plasma flow structures and drift structures over the coinciding geomagnetic and dip equators we expect to see. However, the northward SAPS E field intensifications causing detachment events and the resultant SED plume plasma flows were tracked by the VY and FZ line plots, respectively. In both hemispheres, the SAPS E fields moved to higher latitudes from their previous positions. Therefore, both hemispheres’ SED features (S1; S2) appeared at higher latitudes than in Figure 5b and thus became distinctly separated from the EIA crests. The northern (S2) and southern (S2) SED features were tracked over Bay St. Luis (Mississippi) and close to Trelew (Argentine Patagonia), respectively. At each station, there was a drop in the local B field data seen (see F lots) indicating a northward SAPS E field increase in an independent way. Meanwhile, the high EIA crest‐ and SED‐related plasma densities were maintained by storm time equatorward neutral winds (evidence is discussed in section 4). [26] In Figure 5d, the non‐field‐aligned line plot sets show how the evening EIA that was symmetrical at ∼2400 UT (see Figure 5c) lost its symmetry ∼2 UT hours later over the Eastern Pacific due to strong asymmetrical equatorward neutral winds. At the beginning of recovery phase (∼1.74 UT), when the Dst index started its slow increase, the prompt penetration E field was weakening and caused the decline of vertical upward E × B drift as is indicated by the local (Huancayo‐San Juan) DH line plot. This suggests that fountain strengthening ceased. As a result of these

9 of 13

A06305


Figure 5. Same as for Figure 4 but for the third period covering the second half of the main phase.

10 of 13

A06305

A06305


wind effects and drift effects, the Ni profile became highly asymmetrical indicating stronger equatorward winds in the south. There, a stronger EIA crest and SED feature appeared as separate formations with a large plasma depletion (∼26·104 i+/cm3) between them. The southern SED feature (∼59·104 i+/cm3) was larger than the southern EIA crest (∼52.5·104 i+/cm3). According to the FZ plot, the SED related downward plasma flow (see green arrow), triggered by a strong SAPS E field (see VY plots), was separate from the fountain circulation (see red and blue arrows). In the north, the SED features were significantly smaller (∼10·104 i+/cm3) due to the decrease of northward SAPS E fields (see VY plots) and to the possibly weaker equatorward winds.

4. Discussion [27] We have presented various scenarios (see Figures 4 and 5) showing the development and maintenance of EIA and SED features under the influence of prompt penetration E field, storm time equatorward winds, summer‐to‐winter interhemispheric plasma flows and SAMA effects. These examples demonstrate the relative importance of storm time equatorward neutral winds producing high plasma densities (i.e., positive storm effects) via their mechanical effects on EIA‐related downward plasma flows that were successfully modeled by SUPIM [Lin et al., 2005; Balan et al., 2010, and references therein] and on SED‐related downward plasma flows. For the first time the relative importance of interhemispheric and SAMA‐related plasma flows, distorting forward fountain circulation and accelerating/damping SED‐ related plasma flows, has also been revealed. These scenarios also demonstrate the overall importance of northward SAPS E fields, which create detachment processes [Chappell, 1974; Chen and Grebowsky, 1974] and SED plume plasma [Carpenter et al., 1993] building up SED features [Foster et al., 2002]. Since these underlying physical mechanisms were observed in the SAMA region, these scenarios do not support the suggestion of Lin et al. [2007] that the plasma could be driven from the SAMA region to the SED features in the conjugate hemisphere. The proposed plasma movement could produce upward plasma flows that are not in agreement with the downward plasma flows we observed regularly. [28] Kelley et al. [2004] utilized a GPS total electron content (TEC) map, tracking the North American region at 2200 UT on 15 July 2000, as an example to explain the development of a SED feature reaching 200 TEC units (TECU, 1 TECU = 1016 el m−2) over Florida during the Bastille Day Superstorm. Based on the estimation of Vlasov et al. [2003], 50 TECU were due to westward advection and 150 TECU were due to enhanced fountain effects. [29] In Figure 5b, we have provided a set of field‐ aligned line plots tracking true profiles of plasma density and underlying plasma flows of the topside evening/nighttime ionosphere over the American continents at that time (2200 UT on 15 July). Figure 5b confirmed the occurrence of this extremely large SED feature over Florida (∼30°N; geomagnetic), detected its conjugate pair (∼30°S) over Trelew (Argentine Patagonia) and revealed also the presence of a small EIA between them. Due to the summer–to–winter interhemispheric plasma flows and the SAMA effects, this small EIA was highly asymmetrical with a larger northern and a smaller southern crest. The TEC map of Foster appearing in

A06305

the article of Vlasov et al. [2003] shows not only the northern SED feature (200 TECU) over Florida but the lower TEC values (∼75 TECU) at 4.1°N, 294.4°E (geographic) where we suggest the northern EIA crest was located. This means, that the forward fountain at that time (2200 UT) was not strong enough to enhance fountain action. Thus, there was no enhanced fountain action and therefore there was no contribution from the forward fountain in building up neither the northern nor the southern SED feature. This contradicts the explanation of Kelley et al. [2004] and the estimation of Vlasov et al. [2003]. As the plasma could not flow from the low plasma density EIA crest (i.e., low‐pressure region; ∼75 TECU) to the high plasma density SED feature (i.e., high‐pressure region; ∼200 TECU), fountain effects could not contribute ∼150 TECU to the SED feature’s 200 TECU. Our explanation is further verified by Figure 5b’s FZ and VZ profiles revealing that the SED feature‐related downward plasma flows were separate from the fountain circulation. Thus, the SED plume plasma alone was the downward plasma flow (see green arrow) underlying each SED feature. These field‐aligned profiles provide observational evidence that the 200 TECU SED feature tracked at 2200 UT over Florida was due to detachment processes [Chappell, 1974; Chen and Grebowsky, 1974] producing SED plume plasma [Carpenter et al., 1993] and to the mechanical effects of equatorward neutral winds keeping the ionization at high altitude of low recombination [Balan et al., 2008, 2009, 2010]. The EIA’s poleward sides provided small baselines upon which the downward streaming SED plume plasma piled up. [30] Later on (∼2400 UT), when the forward superfountain action became enhanced, the northern SED features became smaller than the northern EIA crest (see Figure 5c). Thus, there was a direct contribution from the enhanced fountain action adding to the SED plume plasma created by detachment processes. This created a scenario that fits the explanation of Vlasov et al. [2003] and Kelley et al. [2004] partially. The relative importance of detachment processes was possibly higher. [31] Based on their own global GPS TEC map series only (possibly no TEC line plots were available at that early stage of their investigation), Vlasov et al. [2003] interpreted the appearance of a broad (∼40° in geographic latitudes) low TEC region at 2200 UT on 15 July as a broad equatorial trough. These authors explained its creation with the sunward convection produced by a poleward penetrating E field. In fact, as this study’s Figure 5b’s field‐aligned (Ni, FZ, VZ) profiles revealed, the smaller southern EIA crest of the above described small EIA blended in the EIA trough and created a broad equatorial trough‐like structure. Thus, the underlying physical processes can be explained with a normal forward fountain operating under the influence of strong summer‐ to‐winter interhemispheric plasma flow and SAMA effects. When this EIA strengthened (see Figure 5c), the plasma density of the entire EIA increased and became symmetrical. It did not deplete the equatorial trough region as Vlasov et al. [2003] suggested. There was an increase from 20·104 (i+/cm3) (detected in Figure 5b) to 32·104 (i+/cm3) (detected in Figure 5c). Furthermore, the GPS TEC maps of Vlasov et al. [2003] did not have the resolution to track the separation between the EIA crest and SED feature that we illustrated in Figures 5c and 5d. Thus, the overall structure was interpreted

11 of 13

A06305


by those authors as an EIA with a southern crest displaced way poleward. However, we mapped the southern crest locations and SED feature locations (see Figure 5a) clearly illustrating that the southern EIA crest was not placed way poleward of its usual position. In fact, the southern SED feature appeared more poleward at ∼2400 UT and became separated from the southern EIA crest. Then, the southern SED feature was significantly larger than the southern EIA crest. We tracked SED features at ∼45°S dip latitude (see Figure 5a) over the southeastern Pacific and southwestern Atlantic where the neutral winds are most efficient (see details given by Horvath [2007]). This modeled 45°S dip contour line provides further evidence of the significant role of storm time equatorward winds in maintaining high plasma densities by keeping the ionization at higher altitudes where the recombination rates are low.

5. Conclusion [32] We have investigated the development and maintenance of EIA and SED features during the Bastille Day Superstorm with field‐aligned line plots providing true profiles of the plasma density features detected and their underlying plasma circulation. Their analysis revealed the strong influence of summer‐to‐winter interhemispheric plasma flows and SAMA effects on the operation of forward plasma fountain during normal periods and strengthening, and on the development and maintenance of SED features. With some limited CTIP simulations, we demonstrated the relative importance of equatorward (poleward) winds’ mechanical effects in producing high (low) plasma densities. [33] Tracking the signatures of increased northward SAPS E fields triggering detachment events and the resultant downward SED plume plasma flows and SED features, we unraveled the development of the extremely large (200 TECU) SED feature appearing over Florida at 2200 UT on 15 July 2000. Contradicting current speculations, we provided observational evidence that there was no enhanced fountain action at that time to contribute 150 TECU for the SED feature’s 200 TECU. Field‐aligned vertical flux (FZ) and drift (VZ) line plots detected a weak forward fountain circulation occurring separately from the stronger downward SED plume plasma flows. Because of the small northern EIA crest (75 TECU), there was no plasma flow from the EIA crest’s low‐pressure region to the SED feature’s high‐pressure region. The EIA provided only a small baseline upon which the SED plume plasma piled up. We detected its equally large conjugate pair developing the same way over Trelew (Argentine Patagonia). Tracking downward SED plume plasma flows and the resultant SED features along 45°S dip latitude, where equatorward neutral winds are strongest, we also demonstrated that SED plume plasma flows and mechanical wind effects were responsible for the development and maintenance of these extremely large SED features. [34] This study’s results related to this 2200 UT scenario, and to another scenario at 2400 UT when there was enhanced fountain action observed but its importance was secondary compared to the dominating SED plume plasma flows, indicate that the phenomenon of SED cannot be described by one single explanation. SED development is a complex process, particularly when it occurs under the combined effects of the SAMA, summer‐to‐winter interhemispheric

A06305

plasma flows and strong equatorward winds like during the Bastille Day Superstorm. [35] Acknowledgments. NICTA is funded by the Australian Government’s Backing Australia’s Ability initiative, in part through the Australian Research Council. We are grateful to the ACE SWEPAM instrument team, to the ACE Science Center for providing the ACE data, and to M. Hairston for his advice. We gratefully acknowledge the Center for Space Sciences at the University of Texas at Dallas and the U.S. Air Force for providing the DMSP thermal plasma data. We are also grateful to the Community Coordinated Modeling Centre for the CTIP model simulations. Simulation results have been provided by the Community Coordinated Modeling Center at Goddard Space Flight Center through their public Runs on Request system (http:// ccmc.gsfc.nasa.gov). The CCMC is a multiagency partnership between NASA, AFMC, AFOSR, AFRL, AFWA, NOAA, NSF, and ONR. The CTIP Model was developed by T. Fuller‐Rowell at the Space Environment Center (NOAA SEC). We also thank the World Data Center for Geomagnetism at Kyoto (http://swdcwww.kugi.kyoto‐u.ac.jp/index.html) for providing the Kp, Dst, and AE indices and the ground‐based magnetometer data. [ 36 ] Wolfgang Baumjohann thanks Eduardo Araujo‐Pradere and another reviewer for their assistance in evaluating this paper.

References Abdu, M. A., I. S. Batista, A. J. Carrasco, and C. G. M. Brum (2005), South Atlantic magnetic anomaly ionization: A review and a new focus on electrodynamic effects in the equatorial ionosphere, J. Atmos. Sol. Terr. Phys., 67, 1643–1657, doi:10.1016/j.jastp.2005.01.014. Abdu, M. A., et al. (2008), Abnormal evening vertical plasma drift and effects on ESF and EIA over Brazil‐South Atlantic sector during the 30 October 2003 superstorm, J. Geophys. Res., 113, A07313, doi:10.1029/2007JA012844. Anderson, D., K. Yumoto, M. Ishitsuka, and E. Kudeki (2002), Estimating daytime vertical ExB drift velocities in the equatorial F‐region using ground‐based magnetometer observations, Geophys. Res. Lett., 29(12), 1596, doi:10.1029/2001GL014562. Bailey, G. J., N. Balan, and Y. Z. Su (1997), The Sheffield University plasmasphere‐ionosphere model‐A review, J. Atmos. Terr. Phys., 59, 1541–1552, doi:10.1016/S1364-6826(96)00155-1. Balan, N., and G. J. Bailey (1995), Equatorial plasma fountain and its effects: Possibility of an additional layer, J. Geophys. Res., 100, 21,421–21,432, doi:10.1029/95JA01555. Balan, N., G. J. Bailey, M. A. Abdu, K. I. Oyama, P. G. Richard, J. MacDougall, and I. S. Batista (1997), Equatorial plasma fountain and its effects over three locations: Evidence for an additional layer, the F3 layer, J. Geophys. Res., 102, 2047–2056, doi:10.1029/95JA02639. Balan, N., H. Alleyne, Y. Otsuka, D. V. Lekshmi, B. G. Fejer, and I. McCrea (2008), Relative effects of electric field and neutral wind on positive ionospheric storms, Earth Planets Space, 61, 439–445. Balan, N., K. Shiokawa, Y. Otsuka, S. Watanabe, and G. J. Bailey (2009), Super plasma fountain and equatorial ionization anomaly during penetration electric field, J. Geophys. Res., 114, A03310, doi:10.1029/ 2008JA013768. Balan, N., K. Shiokawa, Y. Otsuka, T. Kikuchi, D. Vijaya Lekshmi, S. Kawamura, M. Yamamoto, and G. J. Bailey (2010), A physical mechanism of positive ionospheric storms at low latitudes and midlatitudes, J. Geophys. Res., 115, A02304, doi:10.1029/2009JA014515. Basu, S., S. Basu, K. M. Groves, H.‐C. Yeh, S.‐Y. Su, F. J. Rich, P. J. Sultan, and M. J. Keskinen (2001), Response of the equatorial ionosphere in the South Atlantic region to the Great Magnetic Storm of July 15, 2000, Geophys. Res. Lett., 28, 3577–3580, doi:10.1029/2001GL013259. Brandt, P. C., D. G. Mitchell, E. C. Roelof, and J. L. Burch (2001), Bastille Day Storm: Global response of the terrestrial ring current, Sol. Phys., 204, 377–386, doi:10.1023/A:1014286728053. Carpenter, D. L., B. L. Giles, C. R. Chappell, P. M. E. Decreau, R. R. Anderson, A. M. Persoon, A. J. Smith, Y. Corcuff, and P. Canu (1993), Plasmaspheric dynamics in the duskside bulge region: A new look at an old topic, J. Geophys. Res., 98, 19,243–19,271, doi:10.1029/ 93JA00922. Chappell, C. R. (1974), Detached plasma regions in the magnetosphere, J. Geophys. Res., 79, 1861–1870, doi:10.1029/JA079i013p01861. Chen, A. J., and J. M. Grebowsky (1974), Plasma tail interpretations of pronounced detached plasma regions measured by OGO 5, J. Geophys. Res., 79, 3851–3855, doi:10.1029/JA079i025p03851. Fejer, B. G., J. W. Jensen, T. Kikuchi, M. A. Abdu, and J. L. Chau (2007), Equatorial ionospheric electric fields during the November 2004 magnetic storm, J. Geophys. Res., 112, A10304, doi:10.1029/2007JA012376.

12 of 13

A06305


Foster, J. C. (1993), Storm Time Plasma Transport at Middle and High Latitudes, J. Geophys. Res., 98, 1675–1689, doi:10.1029/92JA02032. Foster, J. C., and W. J. Burke (2002), SAPS: A new characterization for sub‐auroral electric fields, Eos Trans. AGU, 83, 393, doi:10.1029/ 2002EO000289. Foster, J. C., and F. J. Rich (1998), Prompt mid‐latitude electric field effects during severe geomagnetic storms, J. Geophys. Res., 103, 26,367–26,372, doi:10.1029/97JA03057. Foster, J. C., P. J. Erickson, A. J. Coster, J. Goldstein, and F. J. Rich (2002), Ionospheric signatures of plasmaspheric tails, Geophys. Res. Lett., 29(13), 1623, doi:10.1029/2002GL015067. Foster, J. C., W. Rideout, B. Sandel, W. T. Forrester, and F. J. Rich (2007), On the relationship of SAPS to storm‐enhanced density, J. Atmos. Sol. Terr. Phys., 69, 303–313, doi:10.1016/j.jastp.2006.07.021. Fuller‐Rowell, T. J., D. Rees, S. Quegan, R. J. Moffett, M. V. Codrescu, and G. H. Millward (1996), A coupled thermosphere‐ionosphere model (CTIM), in STEP Handbook of Ionospheric Models, edited by R. W. Schunk, pp. 217–238, Utah State Univ., Logan. Heelis, R. A., and S. Mohapatra (2009), Storm time signatures of the ionospheric zonal ion drift at middle latitudes, J. Geophys. Res., 114, A02305, doi:10.1029/2008JA013620. Horvath, I. (2006), A total electron content space weather study of the nighttime Weddell Sea Anomaly of 1996/1997 southern summer with TOPEX/Poseidon radar altimetry, J. Geophys. Res., 111, A12317, doi:10.1029/2006JA011679. Horvath, I. (2007), Impact of 10 January 1997 geomagnetic storm on the nighttime Weddell Sea Anomaly: A study utilizing data provided by the TOPEX/Poseidon mission and the Defense Meteorological Satellite Program, and simulations generated by the Coupled Thermosphere/ Ionosphere Plasmasphere model, J. Geophys. Res., 112, A06329, doi:10.1029/2006JA012153. Horvath, I., and B. C. Lovell (2008), Formation and evolution of the ionospheric plasma density shoulder and its relationship to the superfountain effects investigated during the 6 November 2001 great storm, J. Geophys. Res., 113, A12315, doi:10.1029/2008JA013153. Horvath, I., and B. C. Lovell (2009a), Storm‐enhanced plasma density features, daytime polar cap plasma enhancements, and their underlying plasma flows investigated during superstorms, J. Geophys. Res., 114, A11304, doi:10.1029/2009JA014465. Horvath, I., and B. C. Lovell (2009b), Distinctive plasma density features of the topside ionosphere and their electrodynamics investigated during southern winter, J. Geophys. Res., 114, A01304, doi:10.1029/ 2008JA013683. Kelley, M. C., M. N. Vlasov, J. C. Foster, and A. J. Coster (2004), A quantitative explanation for the phenomenon known as storm‐enhanced density, Geophys. Res. Lett., 31, L19809, doi:10.1029/2004GL020875. Kil, H., L. J. Paxton, X. Pi, M. R. Hairston, and Y. Zhang (2003), Case study of the 15 July 2000 magnetic storm effects on the ionosphere‐ driver of the positive ionospheric storm in the winter hemisphere, J. Geophys. Res., 108(A11), 1391, doi:10.1029/2002JA009782.

A06305

Lee, J. J., K. W. Min, V. P. Kim, V. V. Hegai, K.‐I. Oyama, F. J. Rich, and J. Kim (2002), Large density depletions in the nighttime upper ionosphere during the magnetic storm of July 15, 2000, Geophys. Res. Lett., 29(3), 1032, doi:10.1029/2001GL013991. Lin, C. H., A. D. Richmond, R. A. Heelis, G. J. Bailey, G. Lu, J. Y. Liu, H. C. Yeh, and S.‐Y. Su (2005), Theoretical study of the low‐ and midlatitude ionospheric electron density enhancement during the October 2003 superstorm: Relative importance of the neutral wind and the electric field, J. Geophys. Res., 110, A12312, doi:10.1029/2005JA011304. Lin, C. S., and H. C. Yeh (2005), Satellite observations of electric fields in the South Atlantic anomaly region during the July 2000 magnetic storm, J. Geophys. Res., 110, A03305, doi:10.1029/2003JA010215. Lin, C. S., H.‐C. Yeh, and C. K. Chao (2007), On a possible relationship between density depletions in the SAA region and storm‐enhanced densities in the conjugate hemisphere, J. Atmos. Sol. Terr. Phys., 69, 151–158, doi:10.1016/j.jastp.2006.07.013. Lu, G., L. P. Goncharenko, A. D. Richmond, R. G. Roble, and N. Aponte (2008), A dayside ionospheric positive storm phase driven by neutral winds, J. Geophys. Res., 113, A08304, doi:10.1029/2007JA012895. Prölss, G. W. (2006), Subauroral electron temperature enhancement in the nighttime ionosphere, Ann. Geophys., 24, 1871–1885. Rastogi, R. G., and J. A. Klobuchar (1990), Ionospheric electron content within the equatorial F 2 layer anomaly belt, J. Geophys. Res., 95, 19,045–19,052, doi:10.1029/JA095iA11p19045. Stone, E. C., A. M. Frandsen, R. A. Mewaldt, E. R. Christian, D. Margolies, J. F. Ormes, and F. Snow (1998), The Advanced Composition Explorer, Space Sci. Rev., 86, 1–22,, doi:10.1023/A:1005082526237. Su, Y.‐J., M. F. Thomsen, J. E. Borovsky, and J. C. Foster (2001), A linkage between polar patches and plasmaspheric drainage plumes, Geophys. Res. Lett., 28, 111–113, doi:10.1029/2000GL012042. Venkatraman, S., and R. Heelis (2000), Interhemispheric plasma flows in the equatorial topside ionosphere, J. Geophys. Res., 105, 18,457– 18,464, doi:10.1029/2000JA000012. Vijaya Lekshmi, D., N. Balan, V. K. Vaidyan, H. Alleyne, and G. J. Bailey (2008), Response of the ionosphere to super geomagnetic storms: Observations and modeling, Adv. Space Res., 41, 548–555, doi:10.1016/j. asr.2007.08.029. Vlasov, M., M. C. Kelley, and H. Kil (2003), Analysis of ground‐based and satellite observations of F region behavior during the great magnetic storm of July 15, 2000, J. Atmos. Sol. Terr. Phys., 65, 1223–1234, doi:10.1016/j.jastp.2003.08.012. Webb, P. A., M. M. Kuznetsova, M. Hesse, L. Rastaetter, and A. Chulaki (2009), Ionosphere‐thermosphere models at the Community Coordinated Modeling Center, Radio Sci., 44, RS0A34, doi:10.1029/2008RS004108. I. Horvath and B. C. Lovell, Security and Surveillance, School of Information Technology and Electrical Engineering, University of Queensland, St. Lucia, Brisbane, Qld 4072, Australia. ([email protected]. edu.au)

13 of 13