PUBLICATIONS
Journal of Geophysical Research: Atmospheres
RESEARCH ARTICLE
10.1002/2013JD020798
Key Points:
• Use two mechanisms to explain the
activity of Es layer
• The ID index is used for the
comparison
• The formation of Es layer is explained
more completely
Correspondence to:
J.-Y. Liu,
[email protected]
Citation:
Yeh, W.-H., J.-Y. Liu, C.-Y. Huang, and
S.-P. Chen (2014), Explanation of the
sporadic-E layer formation by comparing
FORMOSAT-3/COSMIC data with meteor
and wind shear information, J. Geophys.
Res. Atmos., 119, 4568–4579, doi:10.1002/
2013JD020798.
Received 26 AUG 2013
Accepted 2 APR 2014
Accepted article online 4 APR 2014
Published online 24 APR 2014
Explanation of the sporadic-E layer formation
by comparing FORMOSAT-3/COSMIC data
with meteor and wind shear information
Wen-Hao Yeh1,2, Jann-Yenq Liu2,3,4, Cheng-Yung Huang1, and Shih-Ping Chen2
1
GPS Science and Application Research Center, National Central University, Chung-Li, Taiwan, 2Institute of Space Science,
National Central University, Chung-Li, Taiwan, 3Center for Space and Remote Sensing Research, National Central University,
Chung-Li, Taiwan, 4National Space Organization, Hsinchu, Taiwan
Abstract The formation of the sporadic E (Es) layer can be interpreted in several different ways, with wind shear
theory and the meteor ionization mechanism being the most commonly used explanations. Nevertheless, neither
the wind shear theory nor the meteor ionization mechanism alone can completely explain the formation of
the Es layer. The meteor ionization mechanism cannot interpret the different activity in this layer between the
Northern and Southern Hemispheres, while the wind shear theory cannot explain the source of the large amount
of ionized particles in the Es layer. In this study, the activity in the Es layer is compared with information about
meteors and the global vertical speed of ionized particles. The information about meteors is obtained from
International Meteor Organization and Radio Meteor Observing Bulletin. The global vertical speed information for
ionized particles is calculated using the International Geomagnetic Reference Field model, Horizontal Wind Model
(HWM07), and Mass Spectrometer-Incoherent Scatter model. The activity in the Es layer is based on the value
of the irregular degree index, which is derived from the signal-to-noise ratio obtained from Formosa Satellite
Mission-3/Constellation Observing System for Meteorology, Ionosphere, and Climate (FORMOSAT-3/COSMIC)
Global Positioning System radio occultation mission. Taking both wind shear theory and the meteor ionization
mechanism together, the source of the ionized particles in the Es layer and the difference in the activity in the
Es layer between Northern and Southern Hemispheres can thus be explained more completely.
1. Introduction
The ionosphere of the Earth is an envelope containing partially ionized gases ranging from about 60 to a
thousand kilometers in altitude. The sporadic E (Es) layer is a layer with enhanced electron density which
appears sporadically in the lower E region from 90 to 120 km in altitude. The Es layer has been the subject of
research for many decades [e.g., Whitehead, 1970, 1989; Mathews, 1998]. Although there are many
mechanisms to explain the formation of the Es layer, including thunderstorms [Whitehead, 1970] and solar
radiation [Pavelyev et al., 2007], it is most often explained by the wind shear theory [Whitehead, 1961;
Mathews, 1998; Carrasco et al., 2007; Haldoupis, 2011] and the meteor ionization mechanism [Mathews, 1998].
In wind shear theory, the ionized particles are affected by wind and the Earth’s magnetic field, where the
convergence and divergence mechanisms determine whether the ionized particles form a layer. Otherwise,
as meteoroids enter the Earth’s atmosphere, the meteor count rate has a near-Gaussian height distribution
with a centroid altitude near 90 km [Hocking et al., 2001; Stober et al., 2008], which is close to the altitude of
the appearance of the Es layer. The dependency of the meteor count rate on the altitude is observed through
radar echo reflected by meteor trail plasma.
The radio occultation (RO) technique has been used to explore the atmosphere of other planets in the solar
system [Fjeldbo and Eshleman, 1969]. With the advent of the Global Positioning System (GPS), the ionosphere and
atmosphere can be observed globally using the RO technique. The RO technique is based on the use of a low
Earth orbit (LEO) satellite to receive the GPS signals which propagate through the atmosphere and ionosphere. At
the present time there are many satellites that can be used for this purpose, such as GPS/Meteorology (GPS/MET)
experiment, German Challenging Minisatellite Payload (CHAMP), and Formosa Satellite Mission-3/Constellation
Observing System for Meteorology, Ionosphere, and Climate (FORMOSAT-3/COSMIC or F3/C) [Liou et al., 2007].
Previous studies of the layer structures in the atmosphere and ionosphere using RO phase and amplitude profiles
have been conducted [Pavelyev, 2002; Pavelyev et al., 2003, 2008, 2010; Wickert et al., 2004; Liou and Pavelyev,
2006; Yeh et al., 2012]. On the other hand, Hocke et al. [2001] found that the irregularity in the Es layer occurs at
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4568
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 1. Four time intervals from Lon-Lat-Alt analysis in January 2010. (a–d) The time intervals are 0 to 6, 6 to 12, 12 to 18,
and 18 to 24 LT, respectively.
altitudes of 90–110 km using GPS/MET data. Wu et al. [2005] established a seasonal map of the Es layer by
using data from CHAMP. Arras et al. [2009] studied the occurrence frequency of the Es layer by using the wind
shear information from 80 to 100 km in altitude obtained from the Collm Observatory. Dou et al. [2010]
compared F3/C data with data from meteor radar and sodium fluorescent lidar to study the possible relations
between incoming meteors, the Es layer, and the sporadic sodium atom layers. Yeh et al. [2012] used the
Horizontal Wind Model (HWM07) to compare information about the Es layer with wind shear from 90 to
120 km at the latitude where the maximum Es layer occurs. Jacobi et al. [2013] compared F3/C data with the
data from the Collm VHF meteor radar in study of the correlation between Es rates and meteor flux variation.
The above studies show that wind shear and meteor ionization separately play important roles in formation of
the Es layer. Nevertheless, the comparison between Es activity and wind shear shows lack of correspondence
in spring and autumn, and meteor ionization cannot explain the difference in Es layer activity between the
Northern and Southern Hemispheres. In this study, we combine the wind shear and the meteor ionization
mechanisms together to explain the formation of the Es layer. These two mechanisms are found to be
complementary and not needed in additional explanation.
In this study, we not only compare the Es layer activity with meteor observations data but also with the global
annual variation in meteor number. The data for meteor observations are obtained from the Radio Meteor
Observing Bulletin (RMOB), and information about the global annual variation in the number of meteors is
obtained from the International Meteor Organization (IMO). Additionally, the information related to vertical
speed is calculated using the International Geomagnetic Reference Field (IGRF) model, HWM07, and the
Mass Spectrometer-Incoherent Scatter (MSISE-90) model. The irregular degree (ID) index, which was
developed by Yeh et al. [2012] using the signal-to-noise ratio (SNR) of the P code for the L1 channel, is
used to investigate the activity in the Es layer. The method for analysis of the F3/C data is described in
section 2. The method for analyzing meteor information and the calculation of the vertical speed of
ionized particles are described in section 3. The comparison of the results and the discussions are given in
section 4, followed by a summary.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4569
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 2. (left) Two-dimensional maximum ID index distribution showing four time intervals in January 2010. (right) The
location of data points in the area in the 0.4 to 0.5 ID range interval in January 2010.
2. Data Analysis of the FORMOSAT-3/COSMIC Data
The SNR of the L1/P code of F3/C from 2008 to 2011 is used for the present investigation. In order to study the
irregularity of the Es layer and compare this with meteor information, one needs to analyze the annual
variation of the ID index, which indicates the activity in the Es layer. The first step is to transfer the SNR to the
ID index by using the ID index derivation described in Yeh et al. [2012], the height of the perigee of the
straight line connecting GPS and LEO satellites is adopted as the altitude. After transferring the SNR from
2008 to 2011, the data for each year are separated into 12 months, and for each month, the data are
separated into four local time (LT) intervals, which are 0 to 6, 6 to 12, 12 to 18, and 18 to 24. The data for each
LT interval in each month are then used for Longitude-Latitude-Altitude (Lon-Lat-Alt) analysis. An example is
shown in Figure 1 where the four panels show the four time intervals for January 2010. Due to the bias caused
by traditional determination of the tangent points of the signal and the influence of ionospheric inclined
layers and irregularities, the essential values of the ID index are seen in the 50–80 km altitude interval [Wickert
et al., 2004; Pavelyev et al., 2012]. So a more robust analytical model is needed to militate the influence if using
the Abel inversion [Pavelyev et al., 2012], although the bias does not have much influence on the analytical
results in this study. Time intervals 1 to 4 are 0 to 6, 6 to 12, 12 to 18, and 18 to 24 LT, respectively. The
horizontal spatial resolution in Figure 1 is 1° in longitude and latitude, and the vertical resolution is 1 km. The
color in Figure 1 indicates the median value of all the ID index data located in each grid.
The results of Lon-Lat-Alt analysis are used to determine the two-dimensional maximum ID index (maxID)
distribution. An example of maxID distribution is shown in Figure 2 (left). The four panels in Figure 2 (left)
show the two-dimensional maxID distribution for the four time intervals in January 2010. The resolution in
longitude and latitude is 1°. The maxID in each grid of two-dimensional maximum ID index distribution is
the value of maximum ID index in all vertical grids with similar Lon and Lat in Lon-Lat-Alt analysis. For the
Lon-Lat-Alt and two-dimensional maxID distribution, the individual ID index values over 7° in longitude and
latitude and 1 km in altitude, while the coordinates of the center are shifted every 1° along the longitude and
latitude, respectively. The maxID distribution is separated into six areas based on the different ranges of the
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4570
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 3. The annual variation of ID values in six range intervals in 2010. The black lines indicate the original profiles, and the red lines the profiles after smoothing.
ID index. The ID range intervals for these six areas are 0.0 to 0.1, 0.1 to 0.2, 0.2 to 0.3, 0.3 to 0.4, 0.4 to 0.5, and
larger than 0.5. Next, we gather the data which are located in each area in each time interval. The four panels
in Figure 2 (right) show examples of data located in the area for 0.4 to 0.5 ID range interval in January 2010.
Then the data located in each area and time interval are used for analysis of annual variation. The black lines
in Figure 3 indicate the annual variation of ID values in the six range interval areas in 2010. The values in
Figure 3 are the median value of the ID index in each time interval in each day of the year (DOY). We then
smooth the annual variation of the ID index. The smoothing process is used to make the variation more clear
for the purpose of determination. The smoothing method is formulated as fs = [I + STΓS] 1fu [Twomey, 1996],
where fs, fu, I, S, and Γ are the smoothed result, raw data, unitary matrix, constraint smoothing matrix, and
diagonal matrix with degrees of control element, respectively. The value of the diagonal elements in Γ is
1000. The smoothed results are the red lines in Figure 3.
The plots in Figures 3a–3d show that although the ID values are different, similar values are maintained
throughout the whole year. In Figures 3e and 3f, the ID values show obvious seasonal variations, especially in
Figure 3f, which indicates that the activity of Es layer in the corresponding area is violent. The annual
variations of the ID index in the area where the ID range is larger than 0.5 in 2010 to 2011 are shown in
Figures 5 and 6 (top).
3. Data Analysis of Meteor and Vertical Speed Information
3.1. Global Annual Variation in Meteor Information
The meteor information in this study comes from the meteor shower calendar provided by IMO [McBeath,
2009, 2010]. The duration, maximum date, and zenithal hourly rate (ZHR) of the meteor shower are used in
the calculation. Based on this information, the variation of meteor numbers during the duration of the shower
is described by using the cosine functions. A diagram of the meteor number variation in one meteor shower
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4571
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
is shown in Figure 4 as an example. In this figure,
the length of the horizontal axis indicates the
duration of the meteor shower and the blue
arrow points to the date of the shower peak. The
length of the vertical axis indicates the value of
the ZHR, which represents the maximum number
of meteors that an ideal observer would see in
perfectly clear skies with the shower’s radiant
overhead. The left side of the shower
maximum at the beginning and the right side
of the shower maximum at the end of a shower
are fitted by cosine functions from π to 2π
and 0 to π, respectively, to conform to the
situation of continuity.
Figure 4. A diagram of the variation in the number of meteors
in one meteor shower. The length of the horizontal axis indicates the duration of the meteor shower, the blue arrow points
to the date of shower peak, and the length of the vertical axis
indicates the ZHR value.
We considered all the meteor showers in the IMO
meteor shower calendar in 2010 and 2011. The
annual variations of the amount of meteors from
2010 and 2011 are shown in Figures 5 and 6
(bottom). The variations in the ID index in the
area where the ID range is larger than 0.5 are shown in Figures 5 and 6 (top).
3.2. Local Annual Variation of Meteor Information
The meteor observation data provided by the RMOB website are used for local comparison with RO data. Five
set of meteor observations from 2008 and 2011, and three from 2009 to 2010, which have the most complete
meteor observation of the whole year, are used in this study. The observer and the location of the meteor
observations are shown in Table 1, and more information regarding the meteor observations can be found on
the RMOB website. RMOB provides hourly meteor number observations and daily variations in numbers of
meteors from 2008 to 2011, as shown in Figures 7–10 (bottom). The value of meteor number is the
summation of the number over 24 h in a day. If the information for meteor number is not complete for a day,
the data point for that day is not plotted in the panels.
The annual variation of ID index corresponding to each set of meteor observation is shown in Figures 7–10
(top and middle). The position of each ID index profile located in an area of 10° in longitude and latitude is
Figure 5. (top) Annual variation in the ID in the area where the ID range is larger than 0.5. The black line indicates the raw
variation, and the red line is the smoothed result. (bottom) Annual variation of the number of meteors in 2010, as obtained
from IMO. The red lines indicate the meteor number variation in each meteor shower, and the black crosses indicate the
summation of the meteor number.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4572
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 6. (top) Annual variation in the ID in the area where the ID range is larger than 0.5. The black line indicates the raw
variation, and the red line is the smoothed result. (bottom) Annual variation of the number of meteors in 2011, as obtained
from IMO. The red lines indicate the meteor number variation in each meteor shower, and the black crosses indicate the
summation of the meteor number.
considered with the observational location located at the center. Figures 7–10 (top) show the variation in the
median value of the ID index in each DOY and each altitude. Figures 7–10 (middle) show the mean ID value
from 75 to 125 km in altitude in Figures 7–10 (top). As described in section 2, the essential values of the ID
index seen in the 50–80 km altitude interval of the ray perigee are caused by influence of ionospheric inclined
layers and irregularities [Wickert et al., 2004; Pavelyev et al., 2012].
3.3. Vertical Speed of Ionized Particles
The ion’s vertical speed w at a steady state, as associated with wind shear theory, can be calculated with the
following equation [Mathews, 1998]
U cos I sin I þ ωvii V cos I
w¼
(1)
2
1 þ ωv ii
where vi and ωi are the ion-neutral collision frequency and ion gyrofrequency, respectively, U and V are
neutral velocity in the magnetic south and east, respectively, and I is the magnetic dip angle. In this study, the
IGRF model is used to calculate the magnetic dip angle, declination angle, and magnetic field. The neutral
velocity in geodetic coordinate is calculated by HWM07, and the magnetic declination angle is then used to
transform the neutral velocity from geodetic coordinate to geomagnetic coordinate.
The ion-neutral collision frequency is calculated by using the equation [Kelley, 2009]
1
v i ¼ 2:6 109 ðnn þ ni ÞA2
(2)
where nn and ni are the number density of neutral particles and ions, respectively; and A is the mean
Table 1. The Observers and the Location of the Observations Obtained From RMOB That are Used in This Study
Observer
Location
Ed Majden
Gaspard De Wilde
Mike Otte
Enric Fraile Algeciras
Andy Smith
Willy Camps
Jeff Brower
Kurt Fisher
Steve Roush
YEH ET AL.
Courtenay, Canada (49°41′N, 125°01′W)
Dessel, Belgium (51°14′N, 5°06′E)
Pearl City, IL, USA (42°20′N, 90°W)
Centre Meteorologic de l’Alt Camp (Tarragona) (41°17′N, 01°15′E)
Tavistock, Devon, UK (50°33′N, 4°08′W)
Tessenderlo, Belgium (51°05′N, 5°06′E)
Kelowna, British Columbia, Canada (49°51′N, 119°34′W)
Salt Lake City, Utah, USA (40°46′N, 111°53′W)
Apache Junction, AZ, USA (33°23′N 111°32′W)
©2014. American Geophysical Union. All Rights Reserved.
4573
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 7. (top) Yearly distributions of the ID index, (middle) yearly mean ID value variations from 75 to 125 km in altitude as shown in Figure 7 (top), and (bottom) the
daily resolution for meteor number variations in 2008.
Figure 8. (top) Yearly distributions of the ID index, (middle) yearly mean ID value variations from 75 to 125 km in altitude as shown in Figure 8 (top), and (bottom) the
daily resolution for meteor number variations in 2009.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4574
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 9. (top) Yearly distributions of the ID index, (middle) yearly mean ID value variations from 75 to 125 km in altitude as shown in Figure 9 (top), and (bottom) the
daily resolution for meteor number variations in 2010.
Figure 10. (top) Yearly distributions of the ID index, (middle) yearly mean ID value variations from 75 to 125 km in altitude as shown in Figure 10 (top), and (bottom)
the daily resolution for meteor number variations in 2011.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4575
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Figure 11. (top) The global mean ID index distribution and (bottom) global distributions of the mean gradient of vertical speed obtained using RO data and the
calculation results from 90 to 115 km in altitude for four periods from 2008 to 2011.
neutral molecular mass in atomic mass units. In (2), ni is neglected due to the low ratio of ni to nn. The
information for nn and A is calculated using the MSISE-90 model. The number density of the helium
atoms (He), oxygen atoms (O), nitrogen molecules (N2), oxygen molecules (O2), argon atoms (Ar),
hydrogen atoms (H), and nitrogen atoms (N), which are provided by the MSISE-90 model, is used in
this study. The summation of the number density of each particle is nn, and A is calculated following
7
X
Sj NUMj, where NUMtotal, NUMj, and Sj are total number density, number density
the equation NUM1 total
j¼1
of particle j, and the atomic/molecular mass of particle j in atomic mass units, respectively; and j = 1
to 7 indicates He, O, N2, O2, Ar, H, and N, respectively. The gyrofrequency in (1) is calculated by qB
m
[Kivelson and Russell, 1995], where m and q are the particle mass and charge, respectively, and B is
the magnetic field. In this study, the value of m is the mean particle mass, which is considered to be
composed of the seven particles calculated by MSISE-90. The calculation results for the global mean
gradient of vertical speed (dw/dz, where z is the altitude) from December to February (D-months),
March to May (M-months), June to August (J-months), and September to November (S-months) from
90 to 115 km in altitude for 2008 to 2011 are shown in Figure 11 (bottom).
4. Analysis Results and Conclusions
4.1. Ionized Particle Source of Es Layer
In Figures 5 and 6, the global annual variation of the number of meteors can be separated into four portions,
two high meteor number portions (HMNPs) and two low meteor number portions (LMNPs). The two HMNPs
are from about 110 to 235 day of year (DOY) (first HMNP) and about 340 to 10 DOY (second HMNP). The two
LMNPs are from about 10 to 110 DOY (first LMNP) and about 235 to 340 DOY (second LMNP). The
first HMNP contains a group of meteor showers with three large peaks, and the second HMNP contains
two. Comparison of the annual variation of meteor number and the annual variation of ID value shown
in Figures 5 and 6 (top) indicates that the HMNP/LMNP correspond to high/low ID value periods.
Furthermore, the local maximum LMNP corresponds to the local maximum ID value variation. The
obvious local maximum LMNP is in about 30, 72, 270, and 294 DOY and corresponds to the local
maximums for about 35, 76, 260, and 294 DOY in ID value variations.
The observational locations given in Table 1 are from the Northern Hemisphere. In Figures 7–10, the ID value
is larger in the J-months than in the other periods for each observational location, which indicates that the
activity in the Es layer is larger in the J-months than in other periods [Arras et al., 2008, 2009; Yeh et al., 2012].
As can be seen in Figures 7–10, the local maximums of most of the annual meteor number variations are
between 150 and 200 DOY, which correspond to the maximum annual ID value variation.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4576
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Conclusively, the comparisons of the global and
local meteor information with Es layer activity
show that meteors make a nonnegligible
contribution to the amount ionized particles to
the Es layer. Nevertheless, in the middle of
December and early January, the variation in the
local annual meteor number variations in the
Northern Hemisphere shows two peaks with large
numbers of meteors that do not reflect large Es
layer activity. Furthermore, for global annual
meteor number variations, the high ID value
corresponding to the first HMNP only occurs in
the Northern Hemisphere while the high ID value
corresponding to second HMNP only occurs in
the Southern Hemisphere. These phenomena
should be associated with the generation
environment of the Es layer, which is discussed in
the subsections below.
4.2. Generation Environment of the Es Layer
As mentioned in the last subsection, in addition to
the source of ionized particles, which comes from
meteors, a generation environment is another
requirement for formation of the Es layer which
can be explained by wind shear theory. When the
direction of the wind is to the magnetic west/east
Figure 12. Illustration of the wind shear convergence mechanism for ions. (top) The geomagnetic zonal wind shear convergence mechanism for ions. (bottom) The geomagnetic
meridional wind shear convergence mechanism for ions in the
Northern Hemisphere. The arrows on the right-hand side of
the panels indicate the upward direction Z, and V ðu; v; w Þ is the
motion vector.
at above/below, the term
vi
ωi
1þ
V cos I
2 in (1) signifies
vi
ωi
the air drag and Lorentz force, which is qV × B,
causing the ions to drift down/up. When the
direction of the wind in the geomagnetic Northern/Southern Hemispheres is to the magnetic north/south
I sin
2I in (1) signifies that ions are driven
above and to the magnetic south/north below, the term U cos
1þ
vi
ωi
horizontally by the air drag while at the same time they are constrained to gyrate around the inclined
magnetic field lines. The ions then converge and a layer forms at the wind shear null where geomagnetic
zonal or meridional wind velocity is 0. The illustrations of wind shear theory can be seen in Figure 12. The
collision frequency between electron and neutral is much smaller than the gyrofrequency. Therefore, in order
to maintain plasma neutrality, the electrons are Coulomb-forced to follow the ions [Haldoupis, 2011].
The global mean ID index distributions from 90 to 115 km in altitude obtained using RO data from 2008 to
2011 for the four periods are displayed in Figure 11 (top). In the D-months and J-months, the summer
hemisphere has a high ID index; in the M-months and S-months, the ID index is low and the difference
between the Northern and Southern Hemispheres is not obvious. The comparison of the distribution of the
ID index with dw/dz in the D-months and J-months shows that the ions in the hemispheres with high/low ID
index correspond to negative/positive dw/dz. Furthermore, the negative and positive dw/dz indicate the
convergence and divergence of the ions, respectively. Thus, a large meteor number does not reflect increased
Es layer activity in the Northern Hemisphere in the D-months. In the M-months and S-months, the global
distribution of dw/dz is similar to the D-months and J-months, respectively, although the activity in the Es layer is
much smaller. Consider the annual variations of the meteor number as discussed in the last subsection. Two
HMNP correspond to the D-months and J-months, and two LMNP correspond to the M-months and S-months. As
there are fewer ionized particles contributed by meteors in the M-months and S-months than in the D-months
and J-months, there is less activity in the Es layer showing an obscure difference between the two hemispheres.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4577
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Furthermore, the ID index along the magnetic equator is low and shows a gap dividing both hemispheres in
Figure 11 (top). Compared to the values in Figure 11 (bottom), dw/dz is close to zero near the magnetic
equator. This indicates that neither of the two convergence mechanisms works well near the magnetic
equator, where the magnetic dip angle is extremely close to 0° and the magnetic field is fairly horizontal. For
I sin
2I in (1) is 0, which signifies that the
the case with the geomagnetic meridional wind direction, the term U cos
1þ
vi
ωi
wind moves the ions to gyrate around the horizontal magnetic field lines rather than moving vertically. For
the case with the geomagnetic zonal wind direction, the term
vi
ωi
V cos I
2 in (1) becomes
1þ
vi
ωi
V
1þ
vi
ωi
2, which signifies
vi
ωi
that the ions are strongly Lorentz forced to move vertically. However, they cannot converge easily into a layer
because they are kept near a fixed magnetic field line by the strongly magnetized electrons since the plasma
must remain neutral [Haldoupis, 2011]. It should be noted, as shown in Figure 11 (top) that there is a hole with
no Es layer activity southwest of South Africa in all periods, especially the D-months. This phenomenon is also
evident in the F3/C global S4 index distributions [Chen et al., 2013]. Comparison with Figure 11 (bottom)
shows that the value of dw/dz at the location of the hole is close to zero which indicates that the convergence
mechanism is not operating in this region.
5. Summary
In this study, we compared the annual variation of Es layer activity with global and local annual variations of
meteor numbers and the global distribution of the vertical speed of ionized particles. Based on the
comparison results, it can be seen that both wind shear theory and meteor ionization mechanisms are
essential factors in Es layer generation. The meteor ionization mechanism contributes ionized particles to the
Es layer. Examination of annual variations in meteor numbers shows that the amount of meteors in the
D-months and J-months is larger than in the M-months and S-months, and this corresponds to the activity in
the Es layer in the D-months and J-months which is larger than in the M-months and S-months. Nevertheless,
the meteor ionization mechanism cannot explain the difference between the Es layer activity in the Northern
and Southern Hemispheres, and the wind shear theory can be used to make up this defect. In D-months and
J-months, the distribution of Es layer activity is correlated with the distribution of the convergence
mechanism, which is associated with wind shear theory, but in M-months and S-months, the correlations are
not obvious. The activity of the Es layer in the M-months and S-months cannot be explained using wind shear
theory along but can be explained using the meteor ionization mechanism. The activity in the Es layer in
M-months and S-months occurs because there are fewer ionized particles in these months than in the
D-months and J-months. In other words, the wind shear theory and the meteor ionization mechanism are
complementary mechanisms that can be used to explain the formation of the Es layer more completely.
Acknowledgments
The authors would like to thank the
support from National Space
Organization (NSPO) and Ministry of
Science and Technology. The research
is supported by grant NSC-102-2811M-008-048, NSC-102-2119-M-008-015,
and NSPO-S-102100.
YEH ET AL.
References
Arras, C., J. Wickert, G. Beyerle, S. Heise, T. Schmidt, and C. Jacobi (2008), A global climatology of ionospheric irregularities derived from GPS
radio occultation, Geophys. Res. Lett., 35, L14809, doi:10.1029/2008GL034158.
Arras, C., C. Jacobi, and J. Wickert (2009), Semidiurnal tidal signature in sporadic E occurrence rates derived from GPS radio occultation
measurements at higher midlatitudes, Ann. Geophys., 27, 2555–2563, doi:10.5194/angeo-27-2555-2009.
Carrasco, A. J., I. S. Batista, and M. A. Abdu (2007), Simulation of the sporadic E layer response to prereversal associated evening vertical
electric field enhancement near dip equator, J. Geophys. Res., 112, A06324, doi:10.1029/2006JA012143.
Chen, S. P., J. Y. Liu, and P. K. Rajesh (2013), Scintillation hole observed by FORMOSAT-3/COSMIC, EGU General Assembly, Vienna, Austria.
Dou, X.-K., X.-H. Xue, T. Li, T.-D. Chen, C. Chen, and S.-C. Qiu (2010), Possible relations between meteors, enhanced electron density layers, and
sporadic sodium layers, J. Geophys. Res., 115, A06311, doi:10.1029/2009JA014575.
Fjeldbo, G., and V. R. Eshleman (1969), Atmosphere of Venus as studied with the Mariner 5 dual radio-frequency occultation experiment,
Radio Sci., 4, 879–897, doi:10.1029/RS004i010p00879.
Haldoupis, C. (2011), A tutorial review on sporadic E layers, in Aeronomy of the Earth’s Atmosphere and Ionosphere, IAGA Spec. Sopron Book Ser.,
pp. 381–394, Springer, Berlin, doi:10.1007/978-94-007-0326-1-29.
Hocke, K., K. Igarashi, M. Nakamura, P. Wilkinson, J. Wu, A. Pavelyev, and J. Wickert (2001), Global sounding of sporadic E layers by the GPS/MET
radio occultation experiment, J. Atmos. Sol. Terr. Phys., 63, 1973–1980, doi:10.1016/S1364-6826(01)00063-3.
Hocking, W. K., B. Fuller, and B. Vandepeer (2001), Real-time determination of meteor-related parameters utilizing modern digital technology,
J. Atmos. Sol. Terr. Phys., 63(2–3), 155–169, doi:10.1016/S1364-6826(00)00138-3.
Jacobi, C., C. Arras, and J. Wickert (2013), Enhanced sporadic E occurrence rates during the Geminid meteor showers 2006–2010, Adv. Radio Sci.,
11, 313–318, doi:10.5194/ars-11-313-2013.
Kelley, M. C. (2009), The Earth’s Ionosphere: Plasma Physics and Electrodynamics, 2nd ed., Academic, San Diego, Calif.
©2014. American Geophysical Union. All Rights Reserved.
4578
Journal of Geophysical Research: Atmospheres
10.1002/2013JD020798
Kivelson, M. G., and C. T. Russell (1995), Introduction to Space Physics, Cambridge Univ. Press, New York.
Liou, Y.-A., and A. G. Pavelyev (2006), Simultaneous observations of radio wave phase and intensity variations for locating the plasma layers
in the ionosphere, Geophys. Res. Lett., 33, L23102, doi:10.1029/2006GL027112.
Liou, Y.-A., A. G. Pavelyev, S. F. Liu, A. A. Pavelyev, N. Yen, C. Y. Huang, and C. J. Fong (2007), FORMOSAT-3 GPS radio occultation mission:
Preliminary results, IEEE Trans. Geosci. Remote Sens., 45(10), 3813–3826, doi:10.1109/TGRS.2007.903365.
Mathews, J. D. (1998), Sporadic E: Current views and recent progress, J. Atmos. Sol. Terr. Phys., 60(4), 413–435, doi:10.1016/S1364-6826(97)00043-6.
McBeath, A. (2009), International Meteor Organization 2010 Meteor Shower Calendar, IMO_INFO(2-09).
McBeath, A. (2010), International Meteor Organization 2011 meteor shower calendar, IMO_INFO(2-10).
Pavelyev, A. G. (2002), Radio vision of the vertical structure of the layers and a study of radio wave propagation conditions in the atmosphere
using high-stability satellite signals, Radiophys. Quantum Electron., 52(5–6), 363–369, doi:10.1007/s11141-009-9151-x.
Pavelyev, A. G., T. Tsuda, K. Igarashi, Y. A. Liou, and K. Hocke (2003), Wave structures in the electron density profile in the ionospheric
D and E-layers observed by radio holography analysis of the GPS/MET radio occultation data, J. Atmos. Sol. Terr. Phys., 65(1), 59–70,
doi:10.1016/s1364-6826(02)00226-2.
Pavelyev, A. G., Y. A. Liou, J. Wickert, T. Schmidt, A. A. Pavelyev, and S. F. Liu (2007), Effects of the ionosphere and solar activity on radio occultation
signals: Application to CHAllenging Minisatellite Payload satellite observation, J. Geophys. Res., 112, A06326, doi:10.1029/2006JA011625.
Pavelyev, A. G., Y. A. Liou, J. Wickert, A. A. Pavelyev, T. Schmidt, K. Igarashi, and S. S. Matyugov (2008), Location of layered structure in the
ionosphere and atmosphere by use of GPS occultation data, Adv. Space Res., 42(1), 224–228, doi:10.1016/j.asr.2008.02.006.
Pavelyev, A. G., Y. A. Liou, J. Wickert, T. Schmidt, and A. A. Pavelyev (2010), Phase acceleration: A new important parameter in GPS occultation
technology, GPS Solutions, 14, 3–11, doi:10.1007/s10291-009-0128-1.
Pavelyev, A. G., Y. A. Liou, K. Zhang, C. S. Wang, J. Wickert, T. Schmidt, V. N. Gubenko, A. A. Pavelyev, and Y. Kuleshov (2012), Identification and
localization of layers in the ionosphere using the eikonal and amplitude of radio occultation signals, Atmos. Meas. Tech., 5, 1–16,
doi:10.5194/amt-5-1-2012.
Stober, G., C. Jacobi, K. Frohlich, and J. Oberheide (2008), Meteor radar temperatures over Collm (51.3°N, 13°E), Adv. Space Res., 42(7),
1253–1258, doi:10.1016/S1364-6826(00)00138-3.
Twomey, S. (1996), Introduction to the Mathematics of Inversion in Remote Sensing and Indirect Measurement, Dover Publications, Mineola, N. Y.
Whitehead, J. D. (1970), Production and prediction of sporadic E, Rev. Geophys., 8, 65–144, doi:10.1029/RG008i001p00065.
Whitehead, J. D. (1989), Recent work on mid-latitude and equatorial sporadic-E, J. Atmos. Terr. Phys., 51(5), 401–424, doi:10.1016/
0021-9169(89)90122-0.
Whitehead, J. K. (1961), The formation of the sporadic-E layer in the temperate zones, J. Atmos. Terr. Phys., 20, 49–58, doi:10.1016/
0021-9169(61)90097-6.
Wickert, J., A. G. Pavelyev, Y. A. Liou, T. Schmidt, C. Reigber, K. Igarashi, A. A. Pavelyev, and S. Matyugov (2004), Amplitude variations in GPS
signals as a possible indicator of ionospheric structures, Geophys. Res. Lett., 31, L24801, doi:10.1029/2004GL020607.
Wu, D. L., C. O. Ao, G. A. Hajj, M. de la Torre Juarez, and A. J. Mannucci (2005), Sporadic E morphology from GPS-CHAMP radio occultation,
J. Geophys. Res., 110, A01306, doi:10.1029/2004JA010701.
Yeh, W.-H., C.-Y. Huang, T.-Y. Hsiao, T.-C. Chiu, C.-H. Lin, and Y.-A. Liou (2012), Amplitude morphology of GPS radio occultation data for
sporadic-E layers, J. Geophys. Res., 117, A11304, doi:10.1029/2012JA017875.
YEH ET AL.
©2014. American Geophysical Union. All Rights Reserved.
4579