GPS-TEC Variation during Low to High Solar Activity Period ( 2010-2014 ) under the Northern Crest of Indian Equatorial Ionization Anomaly Region

The measurements of ionospheric TEC (total electron content) are conducted at a low latitude Indian station Surat (21.16 ̊N, 72.78 ̊E Geog.), which lies under the northern crest of the equatorial anomaly in Indian region. The data obtained are for a period of five years from low to high solar activity (20102014) using GPS (Global Positioning System) receiver. In this study, we report the diurnal and seasonal variation of GPS-TEC, dependence of GPS-TEC with solar activity, geomagnetic condition and EEJ strength. From the seasonal analysis, it is found that greater values of the GPS-TEC are observed during equinox season followed by winter and summer. The appearance (in the year 2011 and 2014) and disappearance (in the year 2010 and 2012) of “winter anomaly” have been observed at the station. From the correlation of GPS-TEC with different solar indices, i.e. solar EUV flux, F10.7 cm solar radio flux and Zurich sunspot number (SSN), it is concluded that the solar index EUV flux is a better controller of GPS-TEC, compared to F10.7 cm and SSN. Further, it is observed that there is no effect of rising solar activity on correlation. Moreover, the percentage variability of GPS-TEC and the standard deviation of GPS-TEC obtained for quiet and disturbed days show that dependence of GPS-TEC on geomagnetic condition is seasonal. Also, there is a positive correlation observed between GPS-TEC and EEJ strength.


Introduction
The partially ionized region of the upper atmosphere called the earth's ionos-phere, reaches altitudes ranging above 60 km from the surface of the earth.The ionospheric plasma density in the ionosphere exhibits significant variations with time (sunspot cycle, seasonal, diurnal, latitude, longitude, solar and geomagnetic activities), which can result in change of special vital ionospheric parameters such as total electron content (TEC).TEC is expressed in units of TECU (1TECU = 10 16 electrons per square meter), which is the altitude integral parameter of electron density, can be defined as of the total numeral of electrons from satellite to receiver path in a single square meter cross-section area.The main contribution to the maximum TEC in the ionosphere is at the F-region density that mainly affects the radio wave propagation [1] [2].In the equatorial low latitude ionosphere at F-region, the ionization density distribution is characterized by a trough at the equator and dual crests on either side of the equator, nearly at about ±17˚ magnetic latitudes are called as the crests of EIA.Due to the complexity in dynamic processes involved in ionization of the equatorial low latitude ionosphere at F-region, the morphology of this region becomes significant to study.The Global positioning system (GPS) signal propagating through the ionosphere is advanced in phase and delayed in time.This time delay being a function of electron density when measured by using dual frequency receiver can compute the TEC.So, the GPS receiver is one of the most practicable tools for TEC study [3] [4].The ionospheric delay constitutes the major source of error for a single frequency GNSS operation.The sensitivity of the ionospheric range delay to TEC for the primary GPS signal is 0.162 m/TECU [5].The TEC in the upper atmosphere plays a crucial role in the determination of the range delays by the electromagnetic signals while traversing through the ionosphere [6].According to Chowdhary et al. [7] the development and the discrepancy of the EIA in electron density are reflected in TEC as the F layer forms the largest part of TEC.So, it is important to study of TEC.
The GPS-TEC data for the Surat station (located under the northern crest of the equatorial ionization anomaly in Indian sector) have been analyzed earlier in the year 2009 at low solar activity period [27].However, the analysis of high solar activity has never been performed for this station.Further, we obtained correlation coefficient between GPS-TEC and solar indices, especially to the intensity of the ascending phase of the 24 th solar cycle being subdued.

Data Analysis
Since 2008, the dual frequency NovAtel GSV4004B GPS ionospheric scintillation and TEC monitor (GISTM) receiver continuously records the GPS data at the Department of Applied Physics (APD), SardarVallabhbhai National Institute of Technology (S.V.N.I.T), at Surat station (21.16˚N, 72.78˚E, Geog.) in India.The GSV4004B GPS receiver tracks up to 11 satellites at a time, which are at different elevation angles [29].The data is collected at every minute, which do not include the 50-Hz sampled raw data, but the reduced TEC and S4 index and other parameters are included.To extract TEC from the GPS observation file, the pseudorange measurements on L1 (1575.42MHz) and L2 (1227.60MHz) frequencies have been considered.The ionospheric delay can be expressed in terms of the signal carrier frequency as follows: where,

( )
ion c f δ is the signal propagation delay at a given carrier frequency.
In the above Equation ( 2), there is a bias error correction, which is different for different satellite-receiver pairs.[30] [31] [32] [33] reported that if accurate estimates of the ionospheric TEC are to be obtained, the differential instrumental biases must be removed.In the present study, the receiver part of the bias is corrected by taking the value supplied by the manufacturer by calibrating the receiver against Wide Area augmentation system (WAAS).The satellite bias error corrections are different for different satellite-receiver pairs.The magnitudes of these biases per PRN, in nanoseconds were used in TEC calculations.The biases are calculated by the International GNSS Service (IGS).Calculation details, analysis, and results are available at (http://aiuws.unibe.ch/spec/dcb.php).These values are frequently updated.The detailed explanation is given in [34] [35] [36] [37].The slant TEC measurements made are the sum of the real slant TEC, the GPS satellite differential delay b S (satellite bias) and the receiver differential delay, b R (receiver bias).Except when at the zenith, the GPS satellite transmits the signal to the reference station through the ionosphere at some oblique angle.To remove the effect of the increased path length due to obliqueness, the STEC needs to be multiplied by a slant factor, S(E).
Therefore, the vertical TEC can be expressed as: The ionosphere puncture point (IPP) is where the GPS signals travel through the maximum electron density layer of the ionosphere.The mean ionospheric height of 350 Km is used for the determination of IPP locations, which is found to be valid for elevation > 50˚ in a low latitude sector [39].In the present study, TEC data collected from the low solar activity to high solar activity periods between January-2010 to February-2015 at the observing station are being used.
The data for the months of August-2013, November-2013, December-2013, January-2014 and February-2014 were unavailable due to some technical issues.
The analysis is limited to geomagnetic quiet conditions.
Rastogi and Klobuchar [40] showed that the difference between the corres-

Results and Discussion
To study the GPS-TEC variation with local time, season, solar activity and geomagnetic condition, we have used a data base from January 2010 to December 2014 situated under the northern crest of the equatorial anomaly region, Surat.
The results are discussed below.

Diurnal Variation of GPS-TEC
For Surat station the diurnal variation of hourly mean GPS-TEC is obtained, such that there is a gradual decrease in GPS-TEC between 00:00 to 05:00 IST (IST = local time (LT) = UT + 5:30 hrs) pre-dawn to sun up period, reaching a minimum between 05:00 to 06:00 IST over the Indian sector and steadily increase with the time of day attaining a maximum in the noon to afternoon and a gradual decrease after sunset.The observations of diurnal variation in GPS-TEC show that the time at which GPS-TEC reaches the diurnal peak vary from month to month and day to day.In general, the diurnal GPS-TEC attained their peak values mostly between 13:00 and 16:00 IST and the minimum in the GPS-TEC flat during most of the night time hours; i.e. from 22:00 to 06:00 during both low and high solar activity phase.
One very important noticeable feature that is the day-to-day uncertainty in the variation of GPS-TEC, particularly during the middle of the day to pre-dawn hours, which is of serious worry in forecasting, as well as in navigation system.
This uncertainty in the day-to-day variation, seasonal variation in TEC [6] [39] [46] [47] may be attributed to: 1) the changes in the activity of the Sun itself and to the connected changes in the intensity of the incoming radiations; 2) the zenith angle (χ) at which they impinge on the Earth's upper atmosphere, in addition to the changes which take place in the Earth's magnetic field; 3) the equatorial electrojet (EEJ) strength, added to the effects due to the dynamics of the EIA and 4) the meridional neutral winds.The diurnal variation of GPS-TEC may also be due to the motion of the ionized particles across the geomagnetic fields by tidal winds [48].Recently, Jonah et al. [49] studied GPS-TEC variation during high and low solar activity phase over the South American sector.They reported that a day-to-day uncertainty in the variation of GPS-TEC contains a component driven by planetary waves improved by tides as they propagate upward.A strong vertical coupling through the increasing propagating waves can also give rise to day-to-day oscillation in GPS-TEC.Their study also illustrates that apart from solar radiation effect, variations in the meridional wind or zonal winds also play an important role corresponding to the changes in GPS-TEC.

Seasonal Variation of GPS-TEC
To study the seasonal variation of GPS-TEC for different years (2010-2014), each year has been classified into three seasons, i.e. equinoxes (March, April, September and October), summer (May, June, July and August) and winter (November, December of current year) and (January, February of successive year).
Hourly mean of diurnal GPS-TEC for four months at a particular hour were obtained to derive the seasonal mean for all these three seasons, Figure 1 From Figure 1 we can see that the forenoon rate of ionization in equinox season is faster than that in winter and summer season.The plot in Figure 1 and  ratio is higher in winter season [70].Thus, the disappearance of winter anomaly in EIA crest may be attributed to a mechanical effect of the equator-ward neutral winds [66].
March 2014 to February 2015 is a high solar activity period.For the high solar activity period the nighttime resurgence of the anomaly (occasioning a secondary peak in GPS-TEC) associated to the evening prereversal enhancement is observed at low latitude station Surat as shown in Figure 1(e).This type of similar results observed for high solar activity period 2001 over South American Sector by Jonah et al. [49].
In order to check the reliability of data of one GPS station (Surat), we analyzed the GPS-TEC data obtained from three different stations (Bangalore, Hyderabad and Lucknow) available from the IGS network in RINEX format.The results are presented in Figure 3.It can be seen that the seasonal trends of TEC observed at these three stations are similar to those at Surat.Also the disappearance of winter anomaly is observed at all three stations for the year 2012 as observed at Surat station.

GPS-TEC Variation with Solar Activity
The main source of insolation on the earth is solar energy, in the form of radiation and charged particles.The earth's upper atmosphere takes in those solar radiations derived from heating, ionization and dissociation.The earth's ionosphere is constructed by the ionizing effect of solar extreme ultraviolet rays.In general, the upper atmosphere of the earth, solar radiations is measured in expression of solar indices, i.e.In an effort to know which solar index represents the ionospheric GPS-TEC well at Surat, we obtain the correlation between TEC obtain at Surat and all the solar indices is shown as Figure 5.The left column shows the correlation between daily average GPS-TEC with EUV flux values (photons cm −2 •s −1 ) and the correlation between daily average GPS-TEC with respective F10.7 cm (s.f.u.) values is shows in the middle column for the year 2010 to 2014.The GPS-TEC correlation coefficient (r) with EUV flux value for the year 2010 to be (r = 0.405) followed by the year 2011 (r = 0.  the African sector using GPS-TEC data over hrao station and reported a high positive coefficient r of (r = 0.75) for the year 2011.They found that the September-October equinox registered larger TEC values of 85 TECU with a high daily TEC variability compared to the March -April equinox for the year 2011.They attributed to the sun's radiation intensity in the two equinoxes calculated by the solar flux, whereby in the year 2011 the solar flux F 10.7 cm registered a value of about 170 sfu in spring equinox than fall equinox with a value of about 130 sfu.Our results for the year 2011 are in agreement with them.
Mostly researchers [6] [13] [28] [58] [63] [73]- [81] have focused on the correlation between TEC and F 10.7 solar flux.Only a few researchers have focused on the correlation between TEC and Solar EUV flux [6] [28] [73] [77].According to Prasad et al. [58] the F10.7 index originates mainly at high-temperature transition region of the solar atmosphere, but the solar EUV flux originates in the irregular layer above the Photosphere ("Chromospheres") and to some minor extent in the transition region and corona.At higher solar activity periods the F10.7 cm flux undergoes stronger fluctuations.During these periods the EUV emissions from higher excited atoms in the solar atmosphere correlate with GPS-TEC.
Table 2 represents the correlation of mean GPS-TEC with all three solar in-  pends, on the ionization due to solar radiation and growth of electron density due to transport [58].For the day-time period, a major part of the electron density at equator is caused by the solar radiation whereas under the northern crest of the EIA location the contribution of plasma is transported because of the fountain effect adds to that created due to solar radiations which are less powerful compared to those at the equator as a result of the solar zenith angle changes.This possibly would be one of the reason for the experimental less correlations between the GPS-TEC and solar indices parameters at a low latitude station under the northern crest of the equatorial ionization anomaly (EIA) region in the Indian sector.
It can be seen that value of correlation coefficient (r) for EUV flux and F10.7 cm flux is positive for all season except for the summer 2013.The correlation of GPS-TEC and SSN is found to be low as compared to those of EUV flux and F10.7 cm in all seasons.In general correlation coefficient of GPS-TEC with EUV is observed to be better for all season compare to F10.7 and SSN.The daytime maximum GPS-TEC for all day observation are plotted against the corresponding EUV flux which shows the linear positive correlation (r = 0.63) during the period 2010-2014 (see Figure 6).Recently, Bhuyan and Hazarika [64] have studied the ascending half of the solar cycle 24 from 2009 to 2012 over Dibrugarh (27.5˚N, 94.9˚E Geog.) and found a linear positive correlation (r = 0.56) between diurnal maximum GPS-TEC against corresponding solar EUV flux.A research of Chen et al. [82] shows that the relationship between the solar EUV flux and  the previous solar minimum.The main reason that controls TEC variations and the variability of thermospheric electron density, composition and temperature of neutral atmosphere is the solar EUV radiation [83] [84].To obtain the correlation between the decomposed TEC A6 and measured solar index EUV flux and F10.7 cm flux, Yadav et al. [85] found good correlation between decomposed TEC A6 and EUV flux.In general, for the seasonal and variation of GPS-TEC with different solar indices at EIA crest region over Surat; India, the solar index EUV flux is a better indicator of GPS-TEC, compare to F10.7 cm and SSN for low to high solar activity period (2010-2014).

Dependence of GPS-TEC with Geomagnetic Condition
To study the effect of geomagnetic activity on the GPS-TEC, we derived variability of GPS-TEC, for five quiet and five disturbed days in each month for year 2010-2014 taken from the website http://www.ga.gov.au/oracle/geomag/iqd_form.jsp.There is considerable dayto-day variability about the monthly mean, characterized by the standard deviation σ (GPS-TEC).The absolute standard deviation is important in practical applications [86], for our purposes, we prefer to express it as a percentage of the mean value of GPS-TEC, as described [86].Percentage variability of GPS-TEC (∑ GPS-TEC), during the disturbed days and quiet days is found to be highest in an equinox season in comparison to that in the winter and summer season (except for the year 2010, which is associated with the low solar activity year).Moreover, for both quiet and disturbed conditions we plotted the standard deviation of GPS-TEC against the years 2010-2014 as shown in Figure 7.It is observed that the standard deviation of GPS-TEC is high and it increases with an increase in solar activity (2010-2014) as seen in Figure 7.This could be attributed to the increase in input of energy to the ionosphere from the Sun in high solar activity.

Dependence of GPS-TEC on EEJ Strength
The characteristic of the equatorial electrojet (EEJ) strength plays a very important role in the electron density distribution over equatorial and low latitudes [26] [87] [88].The EEJ states to an enhanced eastward electric current in the E region because of a strong vertical polarization electric field developed into a latitude band of ±3˚ about the dip equator [89].Rastogi and Klobuchar [40] found dependence of TEC on the equatorial Electrojet by using ATS-6 TEC measurements from the Indian region.Figure 8 in left column illustrates a comparison of the measured monthly peak values of the GPS-TEC for 2010-2014 with equatorial electrojet (EEJ) strength.The overall trends of TEC obtained from GPS and equatorial electrojet (EEJ) strength are similar in different months of the years of study.It is observed that starting from the year 2010 peak value of GPS-TEC and the peak value of EEJ was maximum during equinox months and minimum during the summer months.Similar results were also found for other years 2011 to 2014.Rama Rao et al. [6] [39] and Venkatesh et al. [26] reported that the EEJ shows a clear seasonal behavior with stronger electrojet strength during equinoctial months over Indian equatorial and low latitude sectors.Kumar et al. [81] found the EIA crest in TEC is maximum in equinox months constitute due to EEJ strength.The research of Rama Rao et al. [39] reported that the positions of the EIA crest and its   EEJ strength, Venkatesh et al. [26] found (r = 0.53) over the Brazilian region; Rama Rao et al. [39] found (r = 0.62), Bagiya et al. [44] found (r = 0.71) and Venkatesh et al. [26] found (r = 0.51) in the Indian region stations.Our results agree with this study showing the positive relation between GPS-TEC and EEJ strength.However, it is to be noted that, Venkatesh et al. [26] have found a correlation coefficient of (r = 0.5) for both Brazilian and Indian sector stations spread over the entire EIA region.The averaging effect might have dampened the correlation whereas we have focused only on the anomaly crest region where we found better correlation.

Summary
In this study ionospheric GPS-TEC is investigated during the ascending phase of = 10 22 Wm −2 •Hz −1 ).However, for the year (2012) its disappearance may due to higher EEJ strength value during summer months compare to winter months (see in Figure 8). From the correlation of seasonal and annual variation of GPS-TEC with different solar indices at EIA region over the given location, it can be concluded that the solar index EUV flux is a better controller of GPS-TEC, compared to F10.7 cm and SSN for low to high solar activity period (2010-2014). The overall trends of TEC obtained from GPS and equatorial electrojet (EEJ) strength are similar in all the months of the years 2010-2014.The correlation between GPS-TEC and EEJ strength, showed a positive correlation for the year 2010 (r = 0.73), 2011 (r = 0.60), 2012 (r = 0.76) 2013 (r = 0.80) and 2014 (r = 0.84) (see in Figure 8).

R x =
Mean earth radius, 6371 km, h m = altitude of the IPP, χ = elevation angle and

Figure 1 .
Figure 1(d) 2013 and Figure 1(e) 2014 respectively.In general it is seen from Figures 1(a)-(e) that the greater values of the GPS-TEC were observed in equinoctial months for all five years, followed by winter in the year 2011 and 2014 also in summer for the year 2010 and 2012.The highest value of GPS-TEC is recorded for equinoctial months in the year 2014 (~130 TECU), 2012 (~91 TECU),

Figure 2
Figure2states that the amplitude of the diurnal maximum is higher in the equinoctial months, which may be due to higher thermospheric [O/N 2 ] ratio and EEJ strength over the Surat station.It is normally accepted that the thermospheric neutral composition has a direct control on the seasonal variation of GPS-TEC[44] [50].Oryema et al.[51] and the references therein reported that, during the daytime in an equinoctial season, the sun is overhead the equator and temperature level at the equator are hotter than at the poles, therefore from the equator meridional wind flows towards the pole.This flow of meridional wind

Figure 2 .
Figure 2. Plots of seasonal maximum GPS-TEC for the years 2010-2014 from normal quiet period under the northern crest of EIA at Surat.

1 )
The solar EUV flux (0.1 nm -50 nm): The daily average of full solar disk EUV flux recorded in the 0.1 -50 nm spectral band and measured at 1˚AU (astronomical unit).The solar EUV flux data, represented by units of 10 10 photons cm −2 •s −1 , has been used from the Solar EUV Monitor (SEM) on-board the Solar and Heliospheric Observatory (SOHO) satellite.2) F10.7 cm solar radio flux (2800 MHz): The Solar flux is a compute of the intensity of solar radio emission at the reciprocal of the period of 2800 MHz also known as 10.7 cm flux (the wavelength of the radio signals at 2800 MHz) commonly measured in solar flux unit (1 s.f.u.= 10 22 W•m −2 •Hz −1 ) and is proportional to sunspot activity, and 3) Zurich sunspot number (SSN).

Figure 4
shows yearly mean of different solar indices, i.e. a) solar EUV flux, b) F10.7 cm solar radio flux and c) Zurich sunspot number (SSN) with respect to GPS-TEC.Pattern of all three solar indices with GPS-TEC in low and moderate in the solar activity year 2010 and 2011, also high solar activity year 2013 and 2014 trend matches well except for the year 2012.

Figure 3 .
Figure 3. Plots of seasonal variation of GPS-TEC for the different Indian stations (Bangalore, Hyderabad, Surat and Lucknow) for the year 2012.
dices during the three seasons, i.e. winter, equinox and summer for the period of March 2010 to February 2011.In Table2bold fonts represents a greater value of a correlation with respective season of the year, similarly for the year 2011, 2012, 2013 and 2014 respectively.For each season, the data of four months are grouped together.The electron density in the equatorial and low latitudes de-

F10. 7
cm flux during the deep solar minimum years (2007-2009) is different from the previous solar minimum years, the lessening in two parameters being 5% in F10.7 cm flux and 15% in 0.1 -50 nm EUV irradiation compared to

Figure 6 .
Figure 6.Mass plot for the daily GPS-TEC against corresponding daily EUV flux from January 2010 to December 2014.The solid line is the linear regression trough the data points.
of percentage variability of GPS-TEC (∑ GPS-TEC) for low to high solar activity year 2010 to 2014 situated under the northern crest of the equatorial anomaly region, over Surat station, are shown in Table3.The differences between two equinox months March and April, and between September and October, are mostly due to the transitions between summer and winter, which was earlier reported by Rishbeth and Mendillo[86].

Figure 7 .
Figure 7. Standard deviation of GPS-TEC during geomagnetically quiet and disturbed days for each month of the year 2010, 2011, 2012, 2013 and 2014 (The data for the months of August 2013, November 2013 to February 2014 were not available due to technical issues).

Figure 8 .
Figure 8. Plot of comparison of the measured monthly peak values of the GPS-TEC with equatorial electrojet (EEJ) strength in left column and the statistical relation of monthly peak values of GPS-TEC with equatorial electrojet (EEJ) strength in right column for 2010-2014.

Figure 8
in the right column, represents the statistical relation of monthly peak values of TEC with equatorial electrojet (EEJ) strength.For each plot in Figure 8 in the right column the correlation coefficients (r) were determined and the corresponding fitting is shown by the straight line.The scattered points around the straight line may be due to other factors (local electric fields, neutral winds) contributing to the control of EEJ strength on the day-to-day variation of GPS-TEC.We found a positive correlation coefficient for the year 2010 (r = 0.73), 2011 (r = 0.60), 2012 (r = 0.76), 2013 (r = 0.80) and 2014 (r = 0.84).The correlation between EIA crest TEC and the 24 th solar cycle (2010-2014) for a low latitude station under the northern crest of the equatorial ionization anomaly region in the Indian station Surat (21.16˚N, 72.78˚E Geog.).The major findings of the study are as follows:  The diurnal variation of GPS-TEC demonstrates a gradual increase of total electron content with sunrise around 05:00 and 06:00 IST (IST = local time = UT + 05:30 hrs); peaks up around local afternoon between 13:00 and 16:00 IST, and a gradual decrease after sunset.The diurnal variation amplitude of the GPS-TEC curve varies with evident positions of the Sun over a station. While investigating the seasonal variation of GPS-TEC over the region illustrates grater GPS-TEC is highest during the equinox season for the all the years 2010-2014, followed by summer and winter for the year 2010 and 2012.However, winter anomaly was observed for the year 2011 and 2014 (see Figure 1 and Figure 2) this could be attributed to the fact that atomic and molecular ratio [O/N 2 ] which is greater during high solar activity periods in the northern hemisphere.The disappearance of the winter anomaly for the year (2010) may be attributed to low solar activity when the F10.7 solar radio flux is a smaller amount than the threshold level of 90 -100 solar flux units (1 s.f.u.

Table 1 .
Literature review of correlation studies of TEC with different solar parameters.

Table 3 .
The mean values of percentage variability of GPS-TEC (∑ GPS-TEC %) for low to high solar activity year 2010 to 2014 during geomagnetically quiet and disturbed days.