Features of the F 3 layer in the low latitude ionosphere at sunset

Transcription

1 JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 116,, doi: /2010ja016111, 2011 Features of the F 3 layer in the low latitude ionosphere at sunset Biqiang Zhao, 1,2 Weixing Wan, 1 Bodo Reinisch, 3,4 Xinan Yue, 5 Huijun Le, 1 Jing Liu, 1,6 and Bo Xiong 1,6 Received 10 September 2010; revised 28 October 2010; accepted 16 November 2010; published 25 January [1] The F 3 layer is a common feature within ±10 of the magnetic equatorial ionosphere in the daytime. According to Balan et al. (1998) the F 3 layer occurs mainly during the morning noon period due to the combined effect of the upward E B drift and the neutral wind that provides upward plasma drifts at and above the F 2 layer. The F 3 layer occurrence rate is higher in summer and decreases with increasing solar activity. In this study, the characteristic of the sunset F 3 layer is first investigated using a solar cycle of ionosonde data ( ) from the magnetic equatorial station at Jicamarca, and compared with the features derived from the four subtropical stations at Sao Luis, Fortaleza, Kwajalein, and Vanimo. Evidence shows that the local time distribution of the occurrence of the F 3 layer can extend to the postsunset time ( local time). The sunset F 3 layer has a strong seasonal dependence occurring mainly during the summertime. Unlike the daytime F 3 layer, the occurrence of the sunset F 3 layer clearly increases and the virtual height of the bottom side of the F 3 layer statistically increases from 620 to 1000 km with increasing solar activity. In addition, the occurrence of the sunset F 3 layer at the other stations is much less than that at Jicamarca. These features of the dependence on the season, solar activity, and latitude are clearly related to the geomagnetic control of the evening prereversal enhancement of the equatorial zonal electric field and geomagnetic configuration. Citation: Zhao, B., W. Wan, B. Reinisch, X. Yue, H. Le, J. Liu, and B. Xiong (2011), Features of the F 3 layer in the low latitude ionosphere at sunset, J. Geophys. Res., 116,, doi: /2010ja Introduction [2] The revelation of multilayer structure in the F 2 region over the equatorial ionosphere can be traced back to the middle of the last century. Ground based ionosonde measurements show that the high frequency end of the virtual height frequency (h f ) ionogram records a spur which moves to higher frequencies during the morning noon sector [e.g., Sen, 1949; Ratcliffe, 1951; Skinner et al., 1954]. Sayers et al. [1963] detected the topside ledges in the equatorial topside ionosphere using a Langmuir probe onboard the Ariel I satellite. Such ledges were later detected by the topside sounding technique and revealed as cusps in the topside ionograms at low latitudes after the launch of the Alouette ISIS program [e.g., Lockwood and Nelms, 1964; Raghavarao and Sivaraman, 1974; Sharma and Raghavarao, 1989]. 1 Beijing National Observatory of Space Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China. 2 State Key Laboratory of Space Weather, Chinese Academy of Sciences, Beijing, China. 3 Center for Atmospheric Research, University of Massachusetts Lowell, Lowell, Massachusetts, USA. 4 Lowell Digisonde International, LLC, Lowell, Massachusetts, USA. 5 COSMIC Program Office, University Corporation for Atmospheric Research, Boulder, Colorado, USA. 6 Graduate University of Chinese Academy of Sciences, Beijing, China. Copyright 2011 by the American Geophysical Union /11/2010JA [3] Studies of the stratification of the F 2 layer gradually subsided until recent theoretical work based on the Sheffield University Plasmasphere Ionosphere Model (SUPIM) suggested that the equatorial plasma fountain could produce a temporary additional layer ( G layer) above the normal F 2 when the plasma flow is distorted by a meridional neutral wind [Balan and Bailey, 1995]. This additional layer was later renamed the F 3 layer because it has the same chemical composition as the F region [Balan et al., 1997]. After that, the subject of the F 3 layer received extensive investigations at low and equatorial latitudes concerning its spatial and temporal distribution [Jenkins et al., 1997; Balan et al., 1998; Hsiao et al., 2001; Batista et al., 2002; Rama Rao et al., 2005; Zain et al., 2008;Uemoto et al., 2007; Sreeja et al., 2010; Zhao et al., 2011], and also its storm time characteristics [Zhao et al., 2005; Paznukhov et al., 2007; Balan et al., 2008; Sreeja et al., 2009; Zong et al., 2010]. The mechanismofthestormtimef 3 layer resembles that in quiet periods and was caused by a rapid upward E B drift resulting from an eastward penetration electric field but sometimes can be caused by strong field aligned diffusion due to the storm time meridional wind surges as proposed by Lin et al. [2009a, 2009b]. [4] Lynn et al. [2000] discussed the problem of the nomenclature to describe F 2 stratification. They proposed that the additional layer be referred to as F 3 or F 1.5 depending on whether the transitory layer moved above or stayed below the F 2 layer peak which maintained continuity with the prestratification and poststratification F 2 layer. Their oblique 1of9

2 Table 1. Location of the Ionosondes Used in the Study Station Latitude (deg) Longitude (deg) Magnetic Dip Latitude (deg) Magnetic Dip (deg) Data Coverage Jicamarca Sao Luis , Kwajalein Fortaleza , Vanimo ionosonde measurements in Southeast Asia show that the transitory layer was seen as an F 3 layer close to the magnetic equator but invariably as an F 1.5 layer farther from the magnetic equator. These observations suggest that the distortion in the equatorial electron density profile, associated with a movement toward the base of the F 2 layer as magnetic field lines, descended with increasing latitude. Fagundes et al. [2007] reported stratification of the F 2 layer over an equatorial anomaly crest location in Brazil, which was mainly considered as the result of the propagation of atmospheric gravity waves (AGWs) often observed in the middle latitudes [Heisler, 1962]. [5] The study on the stratification of F 2 layer at the topside ionosphere was also in progress. Uemoto et al. [2004, 2006] performed a statistical analysis of the topside ledges in the equatorial ionosphere using the Ohzora (EXOS C) and ISIS II satellite. They concluded that the characteristics of the F 3 layer and topside ledges are similar, except for the seasonal dependence of the occurrence probabilities. Their results also show that the ionization ledge is observable at almost all local time (LT) sectors except the period from 0300 to 0800 LT. However, Depuev and Pulinets [2001], using the data from the Intercosmos 19 satellite, reported the first observations of ionization ledges at midnight and early morning hours (0400 LT) at 165 longitude. Thampi et al. [2005, 2007] also discussed the possibility of a link between the F 3 layer and topside ledges through the total electron content (TEC) measured by low Earth orbiting satellites. [6] From these investigations, it is known that the F 3 layer can be observed under geomagnetically quiet conditions. The critical frequency becomes greater than that of the F 2 layer during daytime ( LT) within about ±10 magnetic latitudes. The F 3 layer appears with higher occurrence and lasts longer on the summer side of the geomagnetic equator during low solar activity periods. Rama Rao et al. [2005] point out that the altitude of the F 3 layer is very high at the magnetic equator ( km) compared with that at subtropical stations Waltair and SHAR (around 500 km). It can also be detected by topside sounders as topside ledges although the two concepts are not completely equivalent. [7] In this paper we report a type of the F 3 layer produced by the E B prereversal enhancement (PRE), which was not fully discussed in previous work [e.g., Balan et al., 1998]. A statistical result with respect to the seasonal and solar activity dependence of the sunset F 3 layer is given on the basis of more than a solar cycle ( ) of ionogram data recorded at Jicamarca (dip latitude 0.5 N). A comparative study has also been carried out with ionograms recorded at four subtropical stations, Sao Luis (2.0 S), Kwajalein (3.8 N), Fortaleza (6.6 S), and Vanimo (11.2 S). The results are expected to enhance our knowledge about the characteristics of the F layer stratification phenomenon on equatorial aeronomy, although it has been investigated for years. 2. Data Sources [8] In this study, ionogram data recorded at five subtropical and equatorial stations (Jicamarca, Sao Luis, Fortaleza, Kwajalein, and Vanimo) are used. The geographic locations, magnetic inclinations, declinations, and data coverage of the stations are listed in Table 1. Geomagnetic information is calculated by International Geomagnetic Reference Field (IGRF) Measurements at Jicamarca, Sao Luis, Fortaleza, and Kwajalein are taken from the University of Massachusetts Lowell digital ionosonde database [Reinisch et al., 2004; Khmyrov et al., 2008] with ionogram temporal resolution of 15, 10, and 5 min. Note that the ionosonde at Jicamarca during was a DPS 1 [Reinisch et al., 2009], running most of the time with a 30 min ionogram cadence. Measurements at Vanimo were performed every 15 min and were obtained from the Ionospheric Prediction Service (IPS) clean ionogram data available at ips.gov.au/world_data_centre. 3. Results 3.1. Features of the Sunset F 3 Layer at Jicamarca [9] The typical F 3 layer is a cusp structure occurring at the high frequency end of the trace recorded in the ionogram during daytime periods. The stratification of the F 2 layer on the days of occurrence started around 0800 LT and occasionally extended as late as 1600 LT [Balan et al., 1998; Rama Rao et al., 2005]. However, there is usually an evening PRE of the eastward electric field in the dusk sector which occurs simultaneously in the equatorial E and F regions [Heelis et al., 1974]. This upward E B drift raises the ionization peak and plasma converges in the topside ionosphere and then forms the F 3 layer which could be detected by ground based ionosondes for a short period when its peak density is greater than that of the F 2 layer. One such set of observations is presented in Figure 1, which displays some sample ionograms recorded at Jicamarca on 21 November 1995, 9 February 1999, and 14 January 2002 corresponding to different solar activity. The geomagnetic conditions during the three events were quiet with maximum Kp index not exceeding 2. The ionograms recorded at 30 min intervals show some distortions at the high frequency end starting at around 1700 LT. The distortions developed into a clear cusp, and an additional trace appears at the high frequency end at around 1900 LT. The trace remained distorted until the whole F layer started to descend at 2100 LT. Note that on 9 February 1999 there was a strong spread F after 2030 LT which made the F 3 trace difficult to identify. 2of9

3 total of 536 cases during , of which 247 cases were for Kp < 3. The virtual heights of the F 3 layer (VH d ) at sunset ( 1800 LT), when the best stratification with distinct additional layers was seen on the ionograms, were recorded as illustrated in Figure 3. The maximum of magnetic activity index Kp during which the F 3 layer appeared is also shown in Figure 3. Figure 3 shows that the F 3 layer appears more frequently during the summer months (November February), much less frequently during the equinoctial months (March April and September October), and with no appearance during the winter months (June August). The amount of the occurrence of the sunset F 3 layer increases as the solar activity increases from low solar activity years ( and ) to middle high solar activity years ( ). The VH d was shown clearly to increase with solar activity. It varies from km during the years and to about km during the years Figure 4 plots the VH d versus the corresponding F10.7 at every sunset F 3 layer case, and a robust regression method was applied to derive the relationship between the two parameters. It was shown that VH d generally grows linearly with solar flux. The fact is in agreement with the feature of the evening PRE of the F region vertical drifts Figure 1. Sequence of ionograms recorded at Jicamarca on 21 November 1995, 9 February 1999, and 14 January 2002 during different solar activity from LT showing the development and disappearance of the F 3 layer. [10] Figure 2 shows the monthly occurrence probability of the F 3 layer versus local time at Jicamarca during the low, middle, and high solar activity years 2006, 1999, and 2002, respectively. The monthly occurrence probability refers to the percentage of days of available data when stratification of the F 2 layer appears. As illustrated in Figure 2 the higher occurrence probability of the F 3 layer appears at two local time bands, the morning noon period LT and the afternoon sunset period LT, which have a clearly seasonal dependence. In summer, the F 3 layer occurrence in the afternoon sunset band is higher than that of the morningnoon band, while at equinox, the situation is the reverse. In winter, occurrence probabilities at both local time bands are very low. Another noteworthy feature is that the occurrence probability of the F 3 layer near sunset local time has clear solar activity dependence. In the following section, statistical results are given according to the Jicamarca data during Statistical Results of the Sunset F 3 Layer at Jicamarca [11] Unlike the high occurrence of the daytime F 3 layer the amount of sunset F 3 layer days is limited. We selected the days when F 3 appeared during the sunset sector and recorded the local time when F 3 disappeared or became fuzzy due to strong nighttime spread F in Jicamarca station. There are a Figure 2. Histograms showing the percent occurrence of the F 3 layer during LT over Jicamarca in 1999, 2002, and 2006, years of different solar activity. 3of9

4 Figure 3. Day to day variations of the occurrence of the F 3 layer recorded at Jicamarca during The virtual height of the F 3 layer for Kp < 3 (green circles) and Kp 3 (red circles) when best seen is shown, as is the maximum Kp (squares) during which the F 3 layer appeared. The data coverage is denoted as black horizontal bar. which display a linear dependence on the solar flux as revealed by Fejer et al. [1991]. Actually, the solar activity dependence can also be found in the altitude of the daytime F 3 layer with a model simulation at the June solstice corresponding to 1100 LT at 4 N magnetic latitude at the longitude of Fortaleza [Balan et al., 1998]. The visible virtual height of the F 3 peak increases from 500 to 600 km as F10.7 changes from 95 to 194. In comparison, the virtual height of the F 3 peak during the daytime varies from about 430 km in June to about 625 km in January at subtropical regions [Balan et al., 1998], which should be attributed to the seasonal variation of the wind lifting effect. [12] There is no one to one correspondence between the height of the sunset F 3 layer and the Kp index. However, sufficiently strong magnetic activity can induce a distinct additional layer during the sunset period such as was the case on 23 September 1999 and also on 30 October The mechanism responsible for the storm time F 3 layer is similar to that in quiet periods but with a much faster formation due to the rapid uplift of the F layer by an upward E B drift resulting from an eastward penetrating electric field [e.g., Zhao et al., 2005; Paznukhov et al., 2007; Balan et al., 2008]. Such an example from a theoretical model is presented in detail by Lin et al. [2009a, 2009b] Occurrences of the F 3 Layer at Five Stations During Low Solar Activity [13] The morning noon ionosphere becomes broad and intense with increasing solar activity, while the corresponding driving mechanism such as E B drift and neutral wind for F 3 layer formation remains more or less constant. Thus, the upward force arising from drift and wind becomes less 4of9

5 poleward wind, which was confirmed by the SUPIM model. Figure 5 shows that the characteristics of the F 3 layer appearing at Jicamarca near the magnetic equator are different from those at subtropical stations Fortaleza, Kwajalein, and Vanimo. The F 3 layer of Jicamarca has lower occurrence probability (<30%) during the daytime and higher occurrence probability during the sunset period. The occurrence probability, as well as its duration, decreases with the increase in dip latitude, which implies a geomagnetic control of the sunset F 3 layer. Figure 4. Relationship between the virtual height of the F 3 layer when best seen and solar flux during for Kp < 3. The solid line shows the linear fitting line of the data. efficient to raise the morning F 2 plasma to the topside altitude to form an F 3 layer during solar maximum [Balan et al., 1998]. Then the occurrence of the F 3 layer is higher during periods of low solar activity than under periods of high solar activity as revealed by long term observational ionogram data [Batista et al., 2002; Rama Rao et al., 2005]. Considering the availability, integrity, and quality of the ionograms, data from July 2006 to June 2007 at Jicamarca, Kwajalein, and Vanimo, data from May to December 2006 and January to April 2010 at Sao Luis, and data from March to September 2007 and October 2009 to February 2010 at Fortaleza were chosen. The average F10.7 cm solar flux indices during the above three periods are 77.6, 80.0, and 75.8 solar flux units (sfu) (1 sfu = Wm 2 Hz 1 ). [14] Figure 5 shows the monthly occurrence probability of the F 3 layer versus local time at five stations. The occurrence probability refers to the percentage of days of available data when stratification of the F 2 layer appears. According to Lynn et al. [2000], the stratification was recognized in terms of a kink in the profile of the F 2 layer, which could rise above the peak of the background F 2 layer (F 3 layer) or remain below the peak depending on the latitude of observation (F 1.5 layer). In this paper the F l.5 /F 3 phenomenon is generally designated as an F 3 layer for want of a concise description. As illustrated in Figure 5, the F 3 layer occurs most frequently (>75%) in summer, that is December February at Fortaleza and Vanimo, and June July at Kwajalein, during LT. Furthermore, the layer occurs earlier in summer than in other periods; the layer also lasts longest in summer from morning to presunset. The F 3 layer occurs least frequently (<20%) in winter except at Fortaleza in July and at Kwajalein in January, when the occurrence can exceed 40% at some intervals during daytime. Balan et al. [1998] discussed the possibility of the higher occurrence of the F 3 layer in winter and attributed it to the strong upward E B drift in winter associated with less 4. Discussion [15] The model result of Balan and Bailey [1995] shows that the F 3 layer starts to form during the morning hours when the upward E B drift increases. Once formed, the peak concentration of the F 3 layer remains greater than that of the F 2 layer for a short period of time before noon when the drift is large. The layer then weakens but continues to exist for more than 10 hours at the equator. However, the possibility of detection for the F 3 layer depends on the magnitude of the upward E B drift as the plasma at the topside ionosphere will diffuse downward along the geomagnetic field lines to higher latitudes due to the field aligned pressure gradient and gravity forces. Without a sustaining upward plasma flux, the F 3 layer usually becomes weak and invisible because of the shielding effect of the F 2 layer (NmF 2 NmF 3 ) and disappears on the ionograms in the morning noon period. On the other hand, interhemispheric wind blowing from the summer hemisphere to the winter hemisphere acts to raise the plasma at the upwind latitudes near the magnetic equator helping to maintain the diffused plasma at high altitude. As a result, the peak value of the F 3 layer exceeds that of the F 2 layer and hence enhances the possibility of the F 3 layer in the ionogram. This mechanism adequately explains the high occurrence of the F 3 layer observations at Fortaleza, Kwajalein, and Vanimo in summer, and the low occurrence at Jicamarca. Similar observational results were also obtained at other magnetic equatorial stations [Jenkins et al., 1997; Rama Rao et al., 2005; Uemoto et al., 2007]. [16] Although the neutral wind controls where and when the F 3 layer appears with higher probability, the main driving force for the formation and maintenance of the F 3 layer is the upward E B drift velocity. The strength and direction of the equatorial zonal electric field undergoes large day to day variability. Thus, the appearance of the F 3 layer may not be presented as described in the model studies. Fejer et al. [1991] give the average pattern of the equatorial F region drift, which shows that the upward E B drift increases from morning hours to noon, during which the F 3 layer can be best seen. Another period during which the F region undergoes a quick ascent is around dusk when PRE of the eastward electric field develops. Balan et al. [1998] suggested that although the driving force undergoes a large and sudden upward strengthening during the evening hours, formation of another layer at lower altitudes is unlikely to occur during when the F 2 layer drifts upward, because of the absence of the production of ionization. Our results fail to support the above assumption. As shown in Figures 2 and 3, the amount of the sunset F 3 layer during the high solar activity and summer is significant. The sunset F 3 layer may appear when the magnitude of the PRE of the eastward electric field is strong 5of9

6 Figure 5. Occurrence probability of the F 3 layer at Vanimo, Fortaleza, Kwajalein, Sao Luis, and Jicamarca during low solar activity. The dip latitude and the solar terminator around sunset (thick line) at 300 km are shown. enough. Note that the dusk upward drift develops from 1600 LT and peaks at 1900 LT while the production of ionization still exists at the high altitudes of the F 2 region. [17] The occurrence of the sunset F 3 layer was shown to be location dependent, as illustrated in Figure 5. At Kwajalein and Vanimo, sunset F 3 was hardly observed at middle and low solar activity (data absent during high solar activity). One possible reason is that the PRE has a longitude variation. Kil et al. [2009] derived a global PRE map (vertical ion velocity within ±5 dip latitudes at LT during ) using the drift data measured by ROCSAT 1 at 600 km [Su et al., 2006]. As shown by Su et al. [2006, Figure 6], strong upward drift (>30 m s 1 ) was measured at longitude ranges 45 W 75 W in summer months. In comparison, at longitude 120 E 180 E, where Kwajalein and Vanimo are located, the drift is small (8 24 m s 1 ). The PRE is weak at these longitudes unable to produce the sunset F 3 layer. At Fortaleza and Sao Luis, 40 east of Jicamarca, a distinctive sunset F 3 layer was also seldom observed at dusk in summer. [18] We propose that this regional difference may be related to the configuration of the geomagnetic field. Figure 6 (top) shows the average virtual height changes Dh F at Sao Luis, Jicamarca, and Kwajalein, superimposed within a summer month s data during 2002 (black dots, solid line) and 2004 (shaded dots, dashed line). At Sao Luis we added data in November because of less data in December. There were no data during 2002 for Kwajalein. Figure 6 (bottom) displays the drift derived from Dh F/Dt at Sao Luis, Jicamarca, and Kwajalein. The Dh F/Dt reflects the vertical movement of the base of the layer and can be used to estimate ionosonde vertical plasma drifts [Oyekola, 2006; Oyekola et al., 2008]. In an earlier study comparing coincident incoherent scatter and ionosonde F region vertical plasma drifts, Batista et al. [1986] showed that the reversal times after sunset derived 6of9

7 Figure 6. (top) Monthly mean value of ionospheric virtual height, h F, and (bottom) its time derivative, Dh F/Dt, at Sao Luis, Jicamarca, and Kwajalein during the summer months of 2002 (black dots, solid lines) and 2004 (shaded dots, dashed lines). by the two techniques are in good agreement, although the ionosonde generally underestimates the PRE peak amplitude. As illustrated in Figure 6, the PRE drift is most significant at Sao Luis with peak value reaching 30 m s 1, and 18 m s 1 at Jicamarca during The values are a little smaller than the drift data of ROCSAT 1 as reported by Kil et al. [2009]. During solar minimum 2004 the PRE drifts are around m s 1 at Sao Luis and Jicamarca, and no PRE was observed at Kwajalein. [19] We also calculated the equatorward neutral wind along the magnetic meridian U eq = U cos (D) ±U F sin (D) atan altitude 500 km for Sao Luis and Jicamarca from Horizontal Wind Model 07 [Drob et al., 2008]. The input parameters are appropriate to magnetically quiet (Ap = 4) solstical conditions (day 355) at middle solar activity (F10.7 = 120, F10.7A = 120); where U (positive equatorward) and U F (positive eastward) are the meridional and zonal wind velocities, respectively, and D (positive eastward) is the magnetic declination angle; the plus and minus signs apply to Southern and Northern hemispheres, respectively. Figure 7 gives daily U eq variation at Sao Luis and Jicamarca. Combining Figures 6 and 7 shows that the PRE value is large at Sao Luis during LT; however, the U eq of this period is poleward which is unfavorable for the plasma to be maintained at high altitude. The appearance of PRE at Jicamarca is LT, about an hour behind that at Sao Luis while the U eq turns equatorward. Thus, the combined effects of the PRE electric field and the equatorward wind allow the F 3 layer to be observed more frequently at Jicamarca than at the Sao Luis longitude, which has a large magnetic declination of 18. [20] The occurrence of sunset F 3 layers also shows a large day to day variability. Part of the variability of PRE can be attributed to forcing from lower altitudes. Evidence has shown that the thermospheric and electrodynamic conditions of the ionosphere in the afternoon may affect the intensity of the postsunset E B drift velocity. Large, sudden increases of km in the h F were observed in the postsunset sector ( LT) on eight geomagnetically quiet days (Ap 5) during The perturbations were characterized by postnoon enhancements of the equatorial electrojet (EEJ) and an accompanying intensification of the equatorial ionization anomaly (EIA) before evening. Sastri [1998] attributed the phenomenon to the modifications of the equatorial thermospheric zonal winds and a flux tube integrated Pedersen conductivity distribution favorable to a very effective F region dynamo. A similar case was investigated by Zhao et al. [2008] for the East Asian region, which suggests that planetary waves modulating the semidiurnal tide might be responsible for the corresponding anomaly enhancement of EIA in the postsunset period. Basu et al. [2009] revealed the fact that a counter electrojet (CEJ) event in the afternoon may contribute to the inhibition of EIA. The cause of the day today variability of the E B drift corresponding to the PRE remains unresolved as the information regarding altitude, latitude, longitude, and local time variations of the ion drifts, the neutral winds, and the electric fields are only starting to become available. Future observational initiatives from the ground and space are designed to specifically address these issues. [21] Thampi et al. [2007] have discussed the relationship between the F 3 layer and topside ledges. The simultaneous occurrence of the ledges in latitude TEC variation and F 3 layer in the ground ionograms suggests that topside ledges are upward propagating F 3 layers. A potential relationship between the topside ledge and bottomside F 3 layer was also pointed out by Uemoto et al. [2006] who analyzed the topside ionograms of 19 and 430 passages obtained from the Ohzora and ISIS II satellites. They also found that the topside ledges can appear from the evening to slightly after midnight and occur mostly in equinox. This nighttime ionization ledge is possibly caused by the large daytime upward vertical drifts, or probably attributed to the large PRE drift around sunset during moderate and high solar activity according to our investigation. Although the local time dependence of the ionization ledge, as well as the observed dip latitude range, is consistent with those of the F 3 layer reported by Balan et al. [1998], this does not prove that the topside ledges can form independently of an F 3 layer. If the topside ionization ledge is caused by the upward movement of the ionospheric plasma driven by the zonal electric fields, a diurnal variation should be observed in the altitude parameter. The topside ledge should undergo a downward drift at night. However, the results of Uemoto et al. [2006] show that the altitude variations of the ionization ledge were largely scattered. The study by Yizengaw et al. [2009] may provide a clue in understanding the above inconsistency. They found that very strong EIAs in the postmidnight ( LT) sector during magnetically quiet periods are common which should 7of9

8 Figure 7. Effective neutral wind velocity (equatorward positive) at 500 km altitude at Jicamarca (solid line) and Sao Luis (dashed line). be caused by a reversed vertically upward drift observed from the ROCSAT 1 measurement. 5. Summary [22] In this study we have shown, from five stations at subtropical and equatorial areas, the seasonal, local time, and geomagnetic latitude variations of the F 3 layer occurrence. The main conclusions are as follows. High occurrence of the sunset F 3 layer should be distinguished from the traditional morning noon F 3 layer feature. The occurrence of sunset F 3 layers, longitude, which appear mainly at Jicamarca, is clearly dependent on the magnetic latitude as well as longitude. It also has a strong seasonal dependence which occurs mainly during November February. Unlike the daytime F 3 layer, occurrences of the sunset F 3 layer clearly increase with increasing solar activity. It is suggested that PRE of a zonal electric field associated with an equatorward wind contributes mostly to the F 3 layer formation. Statistical analyses for other areas are needed in future studies to clarify longitude and magnetic latitude dependence of the occurrence probability of the sunset F 3 layer. [23] Acknowledgments. This research was supported by the National Natural Science Foundation of China ( ), the National Important Basic Research Project of China (2011CB811405), and the Open Research Program of the State Key Laboratory of Space Weather, Chinese Academy of Sciences. This work was also supported by AF contract FA C 0072, and by NSF grant ATM Xinan Yue is supported by the National Science Foundation under grant AGS Special thanks should be given to the Ionospheric Prediction Service of Australia for providing data in Australia and to the Digisonde Global Ionospheric Observatory (GIRO). [24] Robert Lysak thanks Kenneth Lynn and another reviewer for their assistance in evaluating this paper. References 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: /95ja Balan, N., G. J. Bailey, M. A. Abdu, K. I. Oyama, P. G. Richards, J. Macdougall, and I. S. Batista (1997), Equatorial plasma fountain and its effects over three locations: Evidence for an additional layer, the F 3 layer, J. Geophys. Res., 102, , doi: /95ja Balan,N.,I.S.Batista,M.A.Abdu,J.MacDougall,andG.J.Bailey (1998), Physical mechanism and statistics of occurrence of an additional layer in the equatorial ionosphere, J. Geophys. Res., 103, 29,169 29,181, doi: /98ja Balan, N., S. V. Thampi, K. Lynn, Y. Otsuka, H. Alleyne, S. Watanabe, M. A. Abdu, and B. G. Fejer (2008), F 3 layer during penetration electric field, J. Geophys. Res., 113, A00A07, doi: /2008ja Basu, Su., S. Basu, J. Huba, J. Krall, S. E. McDonald, J. J. Makela, E. S. Miller, S. Ray, and K. Groves (2009), Day to day variability of the equatorial ionization anomaly and scintillations at dusk observed by GUVI and modeling by SAMI3, J. Geophys. Res., 114, A04302, doi: / 2008JA Batista, I. S., M. A. Abdu, and J. A. Bittencourt (1986), Equatorial F region vertical plasma drifts: Seasonal and longitudinal asymmetries in the American sector, J. Geophys. Res., 91, 12,055 12,064. Batista, I. S., M. A. Abdu, J. McDougall, and J. R. Souza (2002), Long term trends in the frequency of occurrence of the F 3 layer over Fortaleza, J. Atmos. Sol. Terr. Phys., 64, Depuev, V. H., and S. A. Pulinets (2001), Intercosmos 19 observations of an additional topside ionization layer: The F3 layer, Adv. Space Res., 27, , doi: /s (01) Drob, D. P., et al. (2008), An empirical model of the Earth s horizontal wind fields: HWM07, J. Geophys. Res., 113, A12304, doi: / 2008JA Fagundes, P. R., V. Klausner, Y. Sahai, V. G. Pillat, F. Becker Guedes, F. C. P. Bertoni, M. J. A. Bolzan, and J. R. Abalde (2007), Observations of daytime F2 layer stratification under the southern crest of the equatorial ionization anomaly region, J. Geophys. Res., 112, A04302, doi: /2006ja Fejer, B. G., E. R. de Paula, S. A. Gonzales, and R. F. Woodman (1991), Average vertical and zonal F region plasma drifts over Jicamarca, J. Geophys. Res., 96, 13,901 13,906, doi: /91ja Heelis, R. A., P. C. Kendall, R. J. Moffet, D. W. Windie, and H. Rishbeth (1974), Electric coupling of the E and F regions and its effect on F region drifts and winds, Planet. Space Sci., 22, Heisler, L. H. (1962), The anomalous ionospheric stratification F1.5, J. Atmos. Sol. Terr. Phys., 24, , doi: / (62)90212-x. Hsiao, C. C., J. Y. Liu, R. T. Tsunoda, S. Fukao, S. Saroso, K. Nozaki, V. L. Badillo, F. T. Berkey, S. W. Chen, and M. Yamamoto (2001), Evidence for the geographic control of additional layer formation in the low latitude ionosphere, Adv. Space Res., 27, Jenkins, B., G. J. Bailey, M. A. Abdu, I. S. Batista, and N. Balan (1997), Observations and model calculations of an additional layer in the topside ionosphere above Fortaleza, Brazil, Ann. Geophys., 15, , doi: /s Khmyrov, G. M., I. A. Galkin, A. V. Kozlov, B. W. Reinisch, J. McElroy, and C. Dozois (2008), Exploring digisonde ionogram data with SAO X and DIDBase, AIP Conf. Proc., 974, Kil, H., L. J. Paxton, and S. J. Oh (2009), Global bubble distribution seen from ROCSAT 1 and its association with the evening prereversal enhancement, J. Geophys. Res., 114, A06307, doi: / 2008JA Lin, C. H., A. D. Richmond, J. Y. Liu, G. J. Bailey, and B. W. Reinisch (2009a), Theoretical study of new plasma structures in the low latitude ionosphere during a major magnetic storm, J. Geophys. Res., 114, A05303, doi: /2008ja Lin, C. H., A. D. Richmond, G. J. Bailey, J. Y. Liu, G. Lu, and R. A. Heelis (2009b), Neutral wind effect in producing a storm time ionospheric additional layer in the equatorial ionization anomaly region, J. Geophys. Res., 114, A09306, doi: /2009ja Lockwood, G. E. K., and G. L. Nelms (1964), Topside sounder observations of the equatorial anomaly in the 75W longitude zone, J. Atmos. Terr. Phys., 26, Lynn, K. J. W., T. J. Harris, and M. Sjarifudin (2000), Stratification of the F 2 layer observed over Southeast Asia, J. Geophys. Res., 105, 27,147 27,156, doi: /2000ja Oyekola, O. S. (2006), Comparison between nighttime ionosonde, incoherent scatter radar, AE E satellite, and HF Doppler observations of F region vertical electrodynamic plasma drifts in the vicinity of the magnetic equator, J. Geophys. Res., 111, A11318, doi: /2006ja Oyekola, O. S., A. Ojo, and J. Akinrimisi (2008), A comparison of ground and satellite observations of F region vertical velocity near the dip equator, Radio Sci., 43, RS1005, doi: /2007rs Paznukhov, V. V., B. W. Reinisch, P. Song, X. Huang, T. W. Bullett, and O. Veliz (2007), Formation of an F3 layer in the equatorial ionosphere: A 8of9

9 result from strong IMF changes, J. Atmos. Sol. Terr. Phys., 69, , doi: /j.jastp Raghavarao, R., and M. R. Sivaraman (1974), Ionization ledges in the equatorial ionosphere, Nature, 249, , doi: /249331a0. Rama Rao, P. V. S., K. Niranjan, D. S. V. V. D. Prasad, P. S. Brahmanandam, and S. Gopikrishna (2005), Features of additional stratification in ionospheric F 2 layer observed for half a solar cycle over Indian low latitudes, J. Geophys. Res., 110, A04307, doi: /2004ja Ratcliffe, J. A. (1951), Some regularities in the F2 region of the ionosphere, J. Geophys. Res., 56, Reinisch, B. W., I. A. Galkin, G. Khmyrov, A. Kozlov, and D. F. Kitrosser (2004), Automated collection and dissemination of ionospheric data from the digisonde network, Adv. Radio Sci., 2, Reinisch, B. W., et al. (2009), New Digisonde for research and monitoring applications, Radio Sci., 44, RS0A24, doi: /2008rs Sastri, J. H. (1998), On the development of abnormally large postsunset upward drift of equatorial F region under quiet geomagnetic conditions, J. Geophys. Res., 103, , doi: /97ja Sayers, J., P. Rothwell, and J. H. Wager (1963), Field aligned strata in the ionization above the ionospheric F 2 layer, Nature, 198, , doi: /198230a0. Sen, H. Y. (1949), Stratification of the F2 layer of the ionosphere over Singapore, J. Geophys. Res., 54, Sharma, P., and R. Raghavarao (1989), Simultaneous occurrence of ionization ledge and counter electrojet in the equatorial ionosphere: Observational evidence and its implications, Can. J. Phys., 67, Skinner, N. J., R. A. Brown, and R. W. Wright (1954), Multiple stratification of the F layer at Ibadan, J. Atmos. Terr. Phys., 5, Sreeja, V., N. Balan, S. Ravindran, T. K. Pant, R. Sridharan, and G. J. Bailey (2009), Additional stratifications in the equatorial F region at dawn and dusk during geomagnetic storms: Role of electrodynamics, J. Geophys. Res., 114, A08309, doi: /2009ja Sreeja, V., S. Ravindran, and T. K. Pant (2010), Features of the F3 layer occurrence over the equatorial location of Trivandrum, Ann. Geophys., 28, , doi: /angeo Su, S. Y., C. H. Liu, H. H. Ho, and C. K. Chao (2006), Distribution characteristics of topside ionospheric density irregularities: Equatorial versus midlatitude regions, J. Geophys. Res., 111, A06305, doi: /2005ja Thampi, S. V., S. Ravindran, C. V. Devasia, T. K. Pant, P. Sreelatha, and R. Sridharan (2005), First observation of topside ionization ledges using radio beacon measurements from low Earth orbiting satellites, Geophys. Res. Lett., 32, L11104, doi: /2005gl Thampi, S. V., N. Balan, S. Ravindran, T. K. Pant, C. V. Devasia, P. Sreelatha, R. Sridharan, and G. J. Bailey (2007), An additional layer in the low latitude ionosphere in Indian longitudes: Total electron content observations and modeling, J. Geophys. Res., 112, A06301, doi: /2006ja Uemoto, J., T. Ono, A. Kumamoto, and M. Iizima (2004), Ionization ledge structures observed in the equatorial anomaly region by using PPS system on board the Ohzora (EXOS C) satellite, Earth Planets Space, 56, e21 e24. Uemoto, J., T. Ono, A. Kumamoto, and M. Iizima (2006), Statistical analysis of the ionization ledge in the equatorial ionosphere observed from topside sounder satellites, J. Atmos. Sol. Terr. Phys., 68, , doi: /j.jastp Uemoto, J., T. Ono, T. Maruyama, S. Saito, M. Iizima, and A. Kumamoto (2007), Magnetic conjugate observation of the F 3 layer using the SEALION ionosonde network, Geophys. Res. Lett., 34, L02110, doi: /2006gl Yizengaw, E., M. B. Moldwin, Y. Sahai, and R. de Jesus (2009), Strong postmidnight equatorial ionospheric anomaly observations during magnetically quiet periods, J. Geophys. Res., 114, A12308, doi: / 2009JA Zain, A. F. M., S. Abdullah, M. J. Homam, F. C. Seman, M. Abdullah, and Y. H. Ho (2008), Observations of the F3 layer at equatorial region during 2005, J. Atmos. Sol. Terr. Phys., 70, , doi: /j.jastp Zhao, B., W. Wan, and L. Liu (2005), Responses of equatorial anomaly to the October November 2003 superstorms, Ann. Geophys., 23, Zhao, B., W. Wan, L. Liu, K. Igarashi, M. Nakamura, L. J. Paxton, S. Y. Su, G. Li, and Z. Ren (2008), Anomalous enhancement of ionospheric electron content in the Asian Australian region during a geomagnetically quiet day, J. Geophys. Res., 113, A11302, doi: /2007ja Zhao, B., W. Wan, X. Yue, L. Liu, Z. Ren, M. He, and J. Liu (2011), Global characteristics of occurrence of an additional layer in the ionosphere observed by COSMIC/FORMOSAT 3, Geophys. Res. Lett., doi: /2010gl045744, in press. Zong, Q. G., B. W. Reinisch, P. Song, Y. Wei, and I. Galkin (2010), Dayside ionospheric response to the intense interplanetary shocks/solar wind discontinuities: Observations from the Digisonde Global Ionospheric Radio Observatory, J. Geophys. Res., 115, A06304, doi: / 2009JA H.Le,J.Liu,W.Wan,B.Xiong,andB.Zhao,BeijingNational Observatory of Space Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing , China. (zbqjz@mail. igcas.ac.cn) B. Reinisch, Center for Atmospheric Research, University of Massachusetts Lowell, 600 Suffolk St., Lowell, MA 01854, USA. X. Yue, COSMIC Program Office, University Corporation for Atmospheric Research, PO Box 3000, Boulder, CO 80307, USA. 9of9