Comparing daytime , equatorial E × B drift velocities and TOPEX / TEC observations associated with the 4-cell , non-migrating tidal structure

We investigate the seasonal and longitude dependence of the daytime, vertical E×B drift velocities, on a dayto-day basis, using a recently-developed technique for inferring realisticE×B drifts from ground-based magnetometer observations. We have chosen only quiet days, Ap<10, from January 2001 through December 2002, so that the main contribution to the variability is due to the variability in the tidal forcing from below. In order to study the longitude dependence in daytime E×B drift velocities, we use appropriately-placed magnetometers in the Peruvian, Philippine, Indonesian and Indian longitude sectors. Since we are particularly interested in quantifying the E×B drift velocities associated with the 4-cell, non-migrating tidal structure, we compare the seasonal and longitude E×B drift structure with TOPEX satellite observations of Total Electron Content (TEC). We outline a plan to establish the magnitude of the longitude gradients that exist in the daytime, vertical E×B drift velocities at the boundaries of the observed 4-cell patterns and to theoretically identify the physical mechanisms that account for these sharp gradients. The paper demonstrates that sharp gradients in E×B drift velocities exist at one of the 4-cell boundaries and outlines how the C/NOFS IVM and VEFI sensor observations could be used to establish theE×B drift longitude gradients at the boundaries of each of the 4 cells. In addition, the paper identifies one of the theoretical, atmosphere/ionosphere models that could be employed to identify the physical mechanisms that might explain these observations.


Introduction
In the Earth's ionospheric F-region, between 200 and 800 km altitude, the daytime distribution of electrons and ions as a function of altitude, latitude, longitude and local time are determined by ionospheric production, loss and transport mechanisms.Production is primarily through photo-ionization of atomic oxygen by solar EUV radiation, λ<91.1 nm and loss is through charge exchange of O + ions with N 2 and O 2 , to give NO + and O + 2 , followed by recombination with electrons.Transport of ionization perpendicular to B is due to E×B drifts and transport parallel to B is due to ambipolar diffusion and the component of the neutral wind parallel to B. At low latitudes, the primary transport mechanism is via E×B drifts in the vertical and meridional plane.At the magnetic equator, these E×B drifts are upward in the daytime and primarily downward at night.The daytime upward drifts are responsible for producing crests in the Fregion peak electron density, Nmax, at ±15 to 18 • magnetic latitude, known as the equatorial anomaly (Hanson and Moffett, 1966;Anderson, 1973).
It is well known that diurnal and semi-diurnal tides in the E-region cause positive ions to move relative to the electrons (which are fixed to geomagnetic field lines) through collisions with neutrals, thus setting up a current that must be divergence free.To do this, polarization electric fields are set up and this process is known as the "E-region wind dynamo".Primarily, in the equatorial, daytime E-and F-region these polarization electric fields are eastward giving rise to upward, vertical E×B drifts that produce the F-region equatorial anomaly mentioned in the previous paragraph.
Recently, several observational studies have identified the existence of a 4-cell pattern in low latitude, ionospheric parameters with longitude that are primarily associated with Published by Copernicus Publications on behalf of the European Geosciences Union.
the equinoctial season.The first evidence emerged from IMAGE satellite FUV (135.6 nm) radiance observations after sunset (20:00 LT) that clearly showed enhancements in airglow-inferred Nmax values at the crests of the equatorial anomaly in 4 specific longitude zones during March-April 2001 (Sagawa et al., 2005;Immel et al., 2006).The latter study attributed the 4-cell pattern to the effects of a 4cell pattern in daytime, vertical E×B drift velocities associated with the diurnal, eastward propagating, non-migrating, wave number 3 (DE3) tidal mode (Hagan and Forbes, 2002) that have their origin due to deep tropical convection and latent heat release at tropospheric heights.A recent paper by Forbes et al. (2008) examines in detail the tidal variability in the dynamo region, the migrating and non-migrating diurnal (DE3) tides and gives a succinct explanation for the relation between ground-based and space-based perspectives of atmospheric tides.
Since the IMAGE observations were at night, Immel et al. (2006) could not rule out a 4-cell pattern in the prereversal enhancement in E×B drift that occurs after sunset.A subsequent paper by England et al. (2006), however, established that the 4-cell pattern was observed in CHAMP satellite in-situ electron densities at 12:00 LT.Further, the 4-cell pattern has also been observed in ROCSAT-1, daytime electron densities and E×B drift velocities at 600 km (Kil et al., 2007;Fejer et al., 2008) and COSMIC occultation observations (Lin et al., 2007).While Kil et al. (2007) find a 4-cell structure in daytime E×B drift velocities, this study averaged over the years from 1999 to 2004 and over season and did not determine seasonal variability or the longitude gradients in E×B drift that define the edges of the individual cells.However, Fejer et al. (2008) also analyzing ROCSAT-1 observations of daytime and nighttime E×B drift velocities at 600 km find daytime E×B drift velocities with a strong wave-number 4 signature in the Equinox and June solstice periods.They present the longitude gradients in these daytime E×B drift velocity signatures.
This current study is important and unique in that it is the first investigation to specifically 1.) Relate observed, ground-based, daytime E×B drift velocities with TOPEXobserved 4-cell TEC structures, 2.) Determine the sharp longitude gradients in E×B drift velocities that are responsible for the sharp gradients in observed TEC values, 3.) Study, for the first time, the seasonal dependence in observed E×B drifts at 4 longitudes and how they can explain the seasonal/longitude dependence in the TOPEX-observed TEC values, and 4.) Outline a plan to incorporate both observations and theoretical models to further our understanding of the physical mechanisms that are responsible for the 4-cell structure.
In this paper, we incorporate recently-developed techniques that use ground-based magnetometer observations to infer daytime, vertical E×B drift velocities and relate these E×B drifts to TOPEX TEC observations (Scherliess et al., 2008) in 4 longitude sectors under Equinoctial, June and De-cember solstice conditions.The next "Data" section briefly describes 1.)The ground-based magnetometer technique for inferring daytime, vertical E×B drifts velocities and 2.) The TOPEX observational database (Scherliess et al., 2008).This is followed by a "Results" section in which we compare the TOPEX observations with the "average" E×B drift velocity vs local time curves for the 3 seasons in 2001 and 2002.We conclude with a "Conclusions and Future Work" section.

Data
A recent technique has been developed to infer the daytime, vertical E×B drift velocity from ground-based magnetometer observations (Anderson et al., 2002).Utilizing a magnetometer located on the magnetic equator (Jicamarca, Peru) and one off the magnetic equator at 6 • N mag.lat.(Piura, Peru), Anderson et al. (2004) developed various relationships between the observed difference in the H component, H (H Jic -H Piura ), and the vertical E×B drift velocity observed by the JULIA (Jicamarca Unattended Long-term Ionosphere Atmosphere) coherent scatter radar measuring the Doppler shift of 150 km echo returns.These 150 km E×B drifts have been shown to be essentially equivalent to F-region E×B drift velocities by comparing them with the Jicamarca ISR (Incoherent Scatter Radar) E×B drifts.Anderson et al. (2004) developed a neural network technique that gave realistic, daytime E×B drift velocities.The neural network was trained with over 450 quiet and disturbed days between 2001 and 2004, using 5 min observations of H and JULIA E×B drift velocities between 09:00 and 16:00 LT.A subsequent paper by Anderson et al. (2006), demonstrated that realistic E×B drift velocities could be obtained with the Peruvian sector-trained neural network, when applied to other longitude sectors where appropriately-placed magnetometers existed, such as in the Philippine and Indian sectors.
Figure 1 displays the magnetometer-inferred, daytime E×B drift velocities for quiet days (A p <10) during the equinoctial period in the Peruvian (top figure), the Philippine (right figure) and the Indian (bottom figure) sectors.For the Peruvian sector, all of the 165 days are displayed as thin colored lines, where the thick red line is the average curve for all 165 days.This is compared to the thick blue line which is the Scherliess-Fejer (1999) climatological E×B drift curve.Similarly, for the Philippine sector, all of the 129 quiet days are displayed along with the average curve and for the Indian sector, all of the 92 days are displayed along with the average curve and each is compared to the Scherliess-Fejer climatological curve.The excellent comparisons give us confidence that realistic E×B drifts can be obtained from the H technique.
We have compared the "average" magnetometer-inferred, daytime E×B drift velocities with the Scherliess-Fejer (1999)   the TEC data to a common baseline in order to circumvent this problem and the same normalization has been applied in the current paper.In a nutshell, the normalization is accomplished by first finding the maximum TEC value for each ascending and descending pass between ±30 • geomagnetic latitude.Next, the peak values are longitudinally averaged to give daily values, again for ascending and descending passes, separately.These daily values were used as normalization factors, where each 18 s TEC data point was divided by its corresponding normalization factor.More detail concerning this applied normalization is given in Scherliess et al. (2008).The figures in this paper refer to "relative" TEC values that have been "normalized" using these factors.3a.Referring to Fig. 3b, there also appear to be sharp longitude gradients at 220 • E and 320 • E which would imply sharp gradients in the daytime, vertical E×B drift velocities at these longitudes, although these measurements have yet to be made.

In
In the June solstice period, Fig. 4b shows that the 4-cell pattern still persists, but to a much lesser degree.The maxima of the TEC values occur at the same longitudes and the minima also occur at the same longitudes as displayed in Fig. 3b for the Equinox period.With the exception of the 100 • E long.sector, what is striking about the June solstice compared with the Equinox period is that the relative TEC values at the magnetic equator and at 15 • mag.lat.are much closer to each other.This means that the equatorial anomaly crests in TEC are closer to the magnetic equator during the June solstice period which implies that the daytime, upward E×B drift velocities are smaller during June solstice than during Equinox.This is exactly what is observed when Fig. 4a is compared with Fig. 3a -a   In the June solstice period, Figure 4b shows that the 4-cell pattern still persists, but to a much lesser degree.The maxima of the TEC values occur at the same longitudes and the 6  Comparisons for the December solstice are displayed in Fig. 5.The 4-cell pattern no longer exists and has been replaced by a 3-cell pattern.While Figs. 3 and 4 clearly show that the magnetometer-inferred E×B drift velocities are capable of explaining the TOPEX/TEC crest separation in the four longitude sectors for Equinox and June solstice conditions, the comparison for the December solstice season displayed in Fig. 5 is somewhat ambiguous.Clearly the E×B drifts in the Philippine sector are greater than in the Peruvian sector and this accounts for the greater crest separation seen in the TOPEX/TEC observations between these two sectors.What needs further investigation is why the E×B drift velocities in the Indonesian and Indian sectors seem to be lower than in the Philippine sector, while the TEC crest separation appears to be equivalent to the Philippine sector.Interestingly, there now exists a gradient in TEC between Peruvian solstice than during Equinox.This is exactly what is compared with Figure 3a -a substantial decrease in E longitudes.and the East Brazilian/Atlantic sector.Since appropriatelyplaced magnetometers in the Eastern Brazilian sector were not available, daytime E×B drift velocities could not be compared with the Peruvian values.

Conclusions and future work
The results presented in Figs.Question 2.) To date, all of the multi-technique observations of the 4-cell pattern have been "averages" -typically a month-long average.Establishing if the 4-cell pattern exists on a day-to-day basis will help to determine whether the identified mechanism, the latent heat release in the tropospheric regions, is acting on a day-to-day basis.
Question 3.) If the state-of-the-art theoretical models are capable to capture the gradients, then the model results can be analyzed to determine the causes of the sharp boundaries and their longitude dependence.
ecember Figure 5b.Same as Figure 3b for December solstice.
and 5 are unique and the comparisons between locities and the satellite TEC observations that relate and longitudinal dependence have not previously unity to research three, fundamentally-important and w sharp are the longitude gradients in daytime, ine the boundaries of each of the 4 cells?2.) Is the 4n a-day-to-day basis?and 3.) Are current, state-off reproducing the sharpness of the boundaries in s? previous section (Figure 3a), the observed difference locity in the Philippine sector (23 m/sec) and the 15) or 8 m/sec.This is across 15 o longitude or about uestion from an observational standpoint will set the e theoretical modelers to understand the physical results with observations.ti-technique observations of the 4-cell pattern have -long average.Establishing if the 4-cell pattern exists of reproducing the sharpness of the boundaries in daytime, vertical E×B drift velocities?Question 1.)As demonstrated in the previous section (Fig. 3a), the observed difference between the maximum E×B drift velocity in the Philippine sector (23 m/s) and the Indonesian sector (15 m/s) is (23-15) or 8 m/s.This is across 15 • longitude or about 0.5 m/s/degree.Answering this question from an observational standpoint will set the "benchmarks" that are needed by the theoretical modelers to understand the physical mechanisms and to compare model results with observations.Question 2.) To date, all of the multi-technique observations of the 4-cell pattern have been "averages" -typically a month-long average.Establishing if the 4-cell pattern exists on a day-to-day basis will help to determine whether the identified mechanism, the latent heat release in the tropospheric regions, is acting on a day-to-day basis.Question 3.) If the state-of-the-art theoretical models are capable to capture the gradients, then the model results can be analyzed to determine the causes of the sharp boundaries and their longitude dependence.
C/NOFS' Ion Velocity Meter (IVM) and Vector Electric Field Instrument (VEFI) sensors could be used to obtain the daytime, vertical E×B drift velocities at the magnetic equator as a function of longitude, local time and season.The IVM sensor measures the E×B drift velocity perpendicular to B, and can be used to obtain the vertical E×B drifts at the magnetic equator by mapping along the geomagnetic field line.The VEFI sensor measures the electric field component perpendicular to B and can be similarly mapped to the magnetic equator.
The theoretical model that could be used to address these specific science questions has been described in a paper by Fuller-Rowell et al. (2008), which was developed to demonstrate the impact of terrestrial weather on the upper atmosphere.The Integrated Dynamics through Earth's Atmosphere (IDEA) model consists of the Whole Atmosphere Model (WAM) and a Global Ionosphere Plasmasphere (GIP) model.
climatological E×B drift model, because this analytic model is the accepted climatological model that has velocity technique.

Figure 1 .Fig. 1 .
Figure 1.Average ∆H-inferred ExB drifts (red curve) compared with the Sche climatological model (blue curve) in the Peruvian (top), Philippine (right figur Indian (bottom) longitude sectors.The thin line colored curves in each figure individual, quiet day curves for each sector -Peruvian (165 days), Philippine and Indian (92 days) (see text for details).
this section we present the TOPEX TEC quiet-day observations binned by season and longitude for the years 2001 and 2002 and compare these observations with the quiet-day H -inferred, daytime E×B drift velocities in 4 longitude www.ann-geophys.net/27/2861/2009/Ann.Geophys., 27, 2861-2867, 2009 comparisons for the Equinoctial period while Figures 4 and 5 present the comparisons for June and December solstice periods, respectively.

Figure 2 .
Figure 2. The geomagnetic coordinates for the Peruvian, Philippine, Indonesian and Indian magnetometer locations.

FigureFig. 2 .
Figure 3a presents the average, daytime vertical ExB drift velocities as a function of local time in 4 longitude sectors for Equinox periods in 2001 and 2002.Note the large difference in maximum ExB drift velocity between the Philippine sector and the Indonesian sector, 23 m/sec vs 15 m/sec.These two locations are only 15 degrees apart in longitude.In Figure 3b, the TOPEX/TEC values are plotted for Equinox, 2001 and 2002, between 12 and 16 LT.The top portion of Figure 3b displays the relative TEC values vs longitude at the geomagnetic equator (red curve) and at 15 o magnetic latitude (black curve).The values correspond to an average over of the northern and southern hemisphere relative TEC values.The bottom portion of Figure 3b is simply the average

Figure
Figure 3a presents the average, daytime vertical E×B drift velocities as a function of local time in 4 longitude sectors for Equinox periods in 2001 and 2002.Note the large difference in maximum E×B drift velocity between the Philippine sector and the Indonesian sector, 23 m/s vs. 15 m/s.These two locations are only 15 degrees apart in longitude.In Fig. 3b, the TOPEX/TEC values are plotted for Equinox, 2001 and 2002, between 12:00 and 16:00 LT.The top portion of Fig. 3b displays the relative TEC values vs. longitude at the geomagnetic equator (red curve) and at 15 • magnetic latitude (black curve).The values correspond to an average over of the Northern and Southern Hemisphere relative TEC values.The bottom portion of Fig. 3b is simply the average of the Northern and Southern Hemisphere relative TEC values plotted as a function of geographic longitude and absolute value of the geomagnetic latitude, to emphasize the location of the longitude gradients and the latitude crest separations.The longitude locations of the Peruvian (blue), Philippine (red), Indonesian (purple) and Indian (green) are indicated as vertical lines in the figure.The color of the vertical lines corresponds to the colored curves in Fig. 3a.Referring to the top portion of Fig. 3b, there clearly exist 4 maxima in relative TEC at ∼0 • , 100 • , 190 • , and 260 • E geog.long.as Scherliess et al. (2008) have pointed out.The edges substantial decrease in E×B drift velocities at all 4 longitudes.

Figure 3a .
Figure 3a.Average ExB drifts for Equinox Figure 3b.TOPEX relative TEC as a 2001-2002 in the Peruvian, Philippine, function of geographic Indonesian and Indian sectors (see text for and magnetic latitude for Equinox details).2001-2002 and 12-16 LT (see text for details).

Fig. 3a .
Fig. 3a.Average E×B drifts for Equinox 2001-2002 in the Peruvian, Philippine, Indonesian and Indian sectors (see text for details).

Figure 4a .Fig. 4a .
Figure 4a.Same as Figure 3a for June solstice Fig So

Figure 4a .Fig. 4b .
Figure 4a.Same as Figure 3a for June solstice Figure 4b.Same as Figure 3b for June Solstice.

Figure 5a .
Figure 5a.Same as Figure 3a for December Figure 5b.Same as Figure 3b for solstice.December solstice.