Kriging‐based ionospheric TEC, ROTI and amplitude scintillation index ( S 4 ) maps for India
2020; Institution of Engineering and Technology; Volume: 14; Issue: 11 Linguagem: Inglês
10.1049/iet-rsn.2020.0202
ISSN1751-8792
AutoresP. Babu Sree Harsha, Venkata Ratnam Devanaboyina, Mutyala Lavanya Nagasri, M. Sridhar, Koppireddi Padma Raju,
Tópico(s)GNSS positioning and interference
ResumoIET Radar, Sonar & NavigationVolume 14, Issue 11 p. 1827-1836 Research ArticleFree Access Kriging-based ionospheric TEC, ROTI and amplitude scintillation index (S 4) maps for India Pasumarthi Babu Sree Harsha, Pasumarthi Babu Sree Harsha Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorDevanaboyina Venkata Ratnam, Corresponding Author Devanaboyina Venkata Ratnam dvratnam@kluniversity.in Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorMutyala Lavanya Nagasri, Mutyala Lavanya Nagasri Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorMiriyala Sridhar, Miriyala Sridhar Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorKoppireddi Padma Raju, Koppireddi Padma Raju Department of Electronics and Communication Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, IndiaSearch for more papers by this author Pasumarthi Babu Sree Harsha, Pasumarthi Babu Sree Harsha Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorDevanaboyina Venkata Ratnam, Corresponding Author Devanaboyina Venkata Ratnam dvratnam@kluniversity.in Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorMutyala Lavanya Nagasri, Mutyala Lavanya Nagasri Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorMiriyala Sridhar, Miriyala Sridhar Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, Green Fields, Vaddeswaram, Guntur, IndiaSearch for more papers by this authorKoppireddi Padma Raju, Koppireddi Padma Raju Department of Electronics and Communication Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, IndiaSearch for more papers by this author First published: 11 September 2020 https://doi.org/10.1049/iet-rsn.2020.0202Citations: 9AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract In this study, an attempt is made to develop regional ionospheric total electron content (TEC), rate of TEC index (ROTI) and amplitude scintillation index (S 4) maps using data from 26 dual-frequency global positioning system (GPS) receivers under GPS aided geo-stationary earth orbit augmented navigation programme over Indian region. One-minute spatial TEC maps are generated for the entire Indian region covering 65–100°E geographic longitudes, 5–40°N geographic latitudes with 2° × 2° grid separation using Kriging method. Also, the ROTI and amplitude scintillation index (S 4) maps for every 5 min are generated. The performance of the generated regional TEC maps is evaluated during a severe geomagnetic storm condition. Patterns of equatorial ionisation anomaly (EIA) TEC structures are investigated for St. Patrick's storm between the days 15 and 20 March 2015 with the generated TEC maps. The proposed TEC maps depict the northern crest of EIA features in latitudinal extent and TEC strength. The kriging method detects the suppressed and downshift of EIA features on 18 March 2015. The movement of plasma bubbles are analysed over Indian latitudes during the transition period of a geomagnetic storm on 17 March 2015. 1 Introduction Spatial ionospheric total electron content (TEC) maps derived from continuous operating global positioning systems (GPSs) stations are widely used for providing ionospheric corrections to single-frequency global navigation satellite system (GNSS) users [1]. TEC is directly proportional to ionospheric delay between GNSS L1 and L2 signals [2]. TEC map is an effective tool for monitoring the ionosphere. Grid-based ionospheric TEC maps would be useful for the satellite-based augmented systems (SBAS) like wide area augmentation system (WAAS) for the United States of America, GPS aided geo-stationary earth orbit (GEO) augmented navigation (GAGAN) for India, MTSAT (multi-functional transport satellite) satellite-based Augmentation System for Japan and European Geostationary Navigation Overlay Service (EGNOS) for Europe [3]. TEC maps can be useful for investigating ionospheric gradients due to geomagnetic storm effects [4, 5], Equatorial ionisation anomaly (EIA), plasma bubbles, travelling ionospheric disturbances, atmospheric gravity waves from lower atmosphere etc. [6]. A multiple number of GPS TEC stations over an area of interest provide an opportunity to generate TEC maps using geospatial interpolation methods, tomographic and data assimilation techniques. Several thousands of GPS receiver's data provide an opportunity for ionosphere monitoring in global scale [7]. Various global TEC maps are developed using the International GNSS Service (IGS) data [8-14]. To characterise the post-sunset ionospheric irregularities rate of TEC index (ROTI) maps are generated with the help of IGS datasets [15-20]. Centre for Orbit Determination in Europe Global Ionospheric Maps (GIMs) use higher-order spherical harmonic function model to generate global ionospheric TEC map with 2.5° × 5° grid resolution [8]. Regional ionospheric maps (RIMs) can represent local ionospheric behaviour with improved spatial and temporal resolution. Indian SBAS, GAGAN have 26 dual frequency GPS TEC stations data sets are available to monitor regional ionospheric phenomenon over Indian region [21]. Geographically India comes under low latitude region, a suitable geostatistical interpolation technique is necessary to model ionospheric grid TEC values. Using GAGAN data, Sunda et al. [3] generated TEC maps using linear interpolation by binning all the ionospheric pierce points for 5 min of time to investigate the dynamic behaviour of ionospheric space weather results. Yadav et al. [22] effectively explained the evolution (sun rise to post noon hours) and devolution (post noon to mid-night hours) of EIA during St. Patrick's Day of 2015 storm conditions. The correlation between spread F and ionospheric scintillations over the Indian region during the St. Patrick's storm is observed by Sripathi et al. [23]. Yang and Liu [24] also found a good correlation between ROTI and ionospheric scintillation indices (S 4 (amplitude scintillation index) and (phase scintillation index)) during disturbed ionospheric conditions [25]. The local observations impact on TEC estimation lead to the development of regional ionospheric TEC maps. Several signal processing methods are proposed to generate RIMs those include empirical orthogonal function analysis, universal kriging (UK), three-dimensional variational approaches, Kalman filter etc. are thoroughly studied [26-29]. Regional TEC maps for Chinese, Japanese and Turkish regions are well represented [2, 30-32]. Several researchers have developed ionospheric TEC grid estimation using various techniques such as kriging, planar-fit, inverse distance weighted, minimum mean square error for GPS and SBAS applications over the Indian region [33-36]. Rao et al. [37] suggested that IPP altitude of 350 km is valid with higher elevation angle (>40°) over Indian region for significant TEC estimation. Among all these models, kriging method is a robust method for modelling ionospheric time delays. Kriging methods are applicable for SBAS ionospheric differential corrections, generation of scintillation maps, estimation of differential code biases and to enhance global ionospheric models [38-40]. Blanch [41] successfully implemented ordinary kriging (OK) based TEC model for USA-WAAS especially in worst case of ionospheric disturbances conditions based on knowledge of underlying deterministic and random spatial ionospheric structures. OK method is successfully implemented by [39] to improve accuracy and smoothing of GIM TEC maps about 2.5%. Deviren and Arikan [27] have implemented an automatic spatial interpolation algorithm based on isotropic UK with using Matérn family to fit the experimental variogram for mid-latitude region. Stanislawska et al. [42] has generated kriging TEC maps over Europe with a time rate of 1 h using 14 GPS receivers. Vernon and Cander [43] utilised the kriging method to provide TEC maps with a 10-min time rate. Southern European ionospheric TEC maps are generated using the kriging technique with 1-min temporal resolution and 0.5° × 0.5° [44]. Most of the interpolation algorithms use linear unbiased predictor for mapping the data points. The success of the kriging method lies in the minimisation of error variances under multivariate normality condition that lies within the estimate. OK and UK are both minimum error variances estimators but differ in priori trend model. The OK method has constant trend values whereas the UK method chooses a polynomial trend from different data points in two-dimensional space. The major challenge is in the characterisation of the under-sampled regions and further improve the kriging-based methods under intense ionospheric disturbance conditions. Two types of F region irregularities were observed at low latitudes during geomagnetic storms. One type is due to the mapping of equatorial plasma bubbles (EPBs) generated through the RT instability at equatorial latitude [45-49]. Also, the F region irregularities could be generated locally through Perkins instability [50]. EPB's are major sources in the occurrence of ionospheric scintillations those cause disruption in the trans-ionospheric signal propagation [50-55]. Jacobsen [18] gave details of ROTI estimation with the different sampling rate and time intervals. To investigate small scale structures ROTI estimated time interval is predominant factor whereas, to draw large-scale structures information from the ROTI maps larger estimated time interval could be chosen. Pi et al. [15] used 5 min for ROTI estimation to measure the short-time ionospheric irregularities. Krankowski et al. [20] choose 30 min as ROTI time interval to cover large patch structures at high latitudes. Cherniak et al. [16] provided global ROTI maps binning ROTI value with 2° × 2° of latitudinal separation using all the IGS stations’ datasets. Amerian et al. also generated the TEC maps and four-dimensional electron density maps over Iran region using IRI-2007 and Iranian permanent GPS network using basis functions to improve empirical model accuracy [56, 57]. Pasumarthi and Devanaboyina [58] utilised GAGAN and COSMIC RO data to generate hourly TEC maps over the Indian region. Foster and Evans [59] utilised an adaptive normalised convolution technique to generate TEC maps over North American Sector. Li et al. [60] also utilised OK method to provide regional TEC maps over Chinese region using crustal movement observation network of China and the IGS datasets. Sunda et al. [61] also generated hourly TEC and amplitude scintillation index linear interpolated maps using GAGAN SBAS satellites data over the Indian region. Kumar et al. [62] provided hourly TEC variations over 17 GAGAN stations using Taylor series expansion method. In this research work, an attempt is made to generate regional ionospheric TEC, ROTI and amplitude scintillation index (S 4) maps are generated for Indian region using the OK method. Global scale and regional scale ionospheric response for analysing the effects of the St. Patrick's storm prevailed on 17 March 2015 are studied earlier [5, 22, 23, 63]. The dual-frequency GPS receivers data observables namely TEC, ROTI and S 4 obtained from the GAGAN network is incorporated as data estimates into kriging method to provide minute-wise TEC maps, 5 min ROTI and S 4 maps over entire Indian region with spatial extent ranging from 65 to 100°N longitudes and 5 to 40°E latitudes. The gridded ROTI maps with 5 min estimation intervals to capture ionospheric irregularities over Indian region to monitor ionospheric irregularities, i.e. EPBs during a geomagnetic storm event. The remaining part of this paper is organised as follows: Section 2 deals with GPS data processing, semivariogram estimation and kriging method equations for spatial TEC, ROTI and S 4 maps generation. Section 3 demonstrates the results and discussions including spatial and temporal EIA variations, comparison of the proposed TEC maps with global TEC maps, validation of with IGS stations, spatial ROTI and S 4 maps. Section 4 describes conclusions focussing on advantages of the generated RIMs. 2 Methodology To develop Indian SBAS, the Indian Space Research Organization (ISRO) and Airports Authority of India jointly initiated the GAGAN project primarily for civil aviation applications. GAGAN uses 26 GPS dual-frequency receivers (Novatel GSV 4004) to provide additional accuracy, integrity and availability over the Indian region. These 26 GPS receivers record GPS week, GPS time, PRN number, elevation angle, azimuth angle, slant TEC, S 4 measurements. 2.1 GPS TEC data analysis The slant TEC measurements are obtained as code combined with carrier phase pseudorange measurements with a one-min sampling rate. The differential biases are estimated for all GPS satellites and GPS receivers using Kalman filter methodology [3, 21]. The day-wise collected ionospheric slant TEC data from almost 26 GAGAN stations is fed into pre-processing steps. The elevation cut-off is chosen to be above 40° to ensure minimum multipath effect [37]. The biases corrected slant total electron content (STEC) measurements and respective information of station latitude and longitude, elevation and azimuth angle of each satellite are used to derive ionospheric pierce point latitude, ionospheric pierce point longitude and vertical total electron content (VTEC) assuming ionosphere as a thin-shell to 350 km for the entire day. The is the slant TEC for a particular satellite for that receiver that is converted into based on the elevation dependent mapping function in (1) as given below in (2), where is the radius of earth, i.e. 6371 km, is the assumed height of ionosphere, i.e. 350 km [37], for satellite for a receiver , respectively [64] (1) (2) 2.2 ROTI estimation The rate of change of TEC index (ROTI) is a measure of small-scale ionospheric irregularity those prevails frequently over low-latitude region [15]. The (slant TEC for a particular satellite corresponding to a receiver) is chosen and rate of change of TEC is estimated [15] (3) where is chosen as 1 min. The standard deviation of is estimated within 5 min of duration, is termed as rate of TEC index (ROTI) [15] (4) where represents the average value over 5 min sliding window time duration. The ROTI values are estimated for all the visible satellites available in a day. 2.3 S 4 data analysis Ionospheric scintillations cause sudden rapid fluctuations in the amplitude and phase of received GPS signals. The amplitude scintillation index (S 4) derived from signal intensity can define the scale of amplitude scintillation (weak (0.2 < S 4 < 0.4), moderate (0.4 < S 4 < 0.6), strong (S 4 > 0.6)) activity. Estimating S 4 index from received signal intensity could be seen elsewhere [25]. The S 4 index with noise contribution is termed as total S 4 and S 4 index after eliminating noise contribution is termed as corrected S 4 index. The S 4 values are filtered with code and carrier divergence observations to eliminate multipath and noise [25]. The dual-frequency GPS receiver's data used in this study has the capability to provide the corrected S 4 values. For further analysis, obtained corrected S 4 values >40° of elevation angle are considered. 2.4 Implementation of OK method for spatial TEC, ROTI and S 4 maps The foremost assumption is to consider data as a second-order stationary random variable , i.e. the data has a constant mean over the time (3) and the spatial correlation between the two points (h distance apart) of data depends on the distance alone (4), X represents the distance vector of IPP latitude and IPP longitude grid points (5) (6) The variance between two VTEC data points is given by (5) [27] (7) Substituting (6) in (7), the variance of VTEC data is given by (8) and is equal to twice the that is to be the fitted variogram based on geostatistical properties of the data (8) The crucial step is to find a good matching between theoretical semivariogram and experimental semivariogram for the data. Based on the geostatistical properties (sill, range and nugget) measured either the experimental semivariogram can be fitted to Gaussian, spherical, exponential [27] etc. Based on the hourly observations from GPS receivers, with their IPP latitude and IPP longitudes respective sill, range and nugget values are estimated based on least-squares fit. The spatial range defines the latitudinal and longitudinal distance before the model response becomes flattened and the value given by the model semivariogram is termed as a sill. The nugget effect could be attributed to the biases through the measurement errors, sampling time variations in the data sets. Upon observing the geostatistical properties for data, a three-step procedure is followed in choosing the theoretical semivariogram that fits the empirical data as follows. Step 1: The semivariogram should pass to the centre of the binned data points. Step 2: The semivariogram should be as close to the average of all the binned data points. Step 3: The semivariogram should pass at the middle of local polynomials. By observation, the spherical semivariogram has a good correlation with empirical data sets (for TEC, ROTI and S 4 data points). These theoretical semivariogram functions are estimated for each hour based on (9) with as the structure variance, d is the effective range, is the sill and is the nugget variance (9) The framework of theoretical semivariogram for each hour is used to estimate data points at unknown grid locations (latitude, longitude) based on known data point grid locations. A best linear unbiased estimator is to be modelled by maintaining proper weights calculated using theoretical semivariogram function. Kriging method is useful in maintaining specified spatial resolution. The kriging estimator is given by (10), standard OK equations are seen elsewhere [30] (10) The weights to be provided for a known VTEC location is assumed to follow (11) and are estimated accordingly to the spherical semivariogram estimates given in (7) (11) where is the known spatial measurement location vector that constitutes latitude and longitude , is the unknown spatial measurement location vector at latitude and longitude , is to be estimated VTEC on unknown location , is the known VTEC estimate at the location . There are different methods to provide optimal weight estimates by satisfying (9). The value of N is arbitrary, i.e. chosen accordingly to the required spatial resolution. For all the N spatial locations, the kriging method estimates the TEC value of that by providing different weight factors according to the semivariogram function is computed with all the available IPP's for that hour. Semivariogram is much useful to capture spatial correlations observed in data for chosen location in this manner. Similar procedure is carried out for ROTI and S 4 maps generation with 5 min of temporal resolution by maintaining proper spatial extent. 3 Results and discussions To understand the kriging TEC method response to St. Patrick's day geomagnetic storm of March 2015 conditions, a total six days (from 15 to 20 March 2015) GPS TEC data over Indian region are considered. The data collected from the GPS TEC network reference stations (red triangles) (Fig. 1) are utilised. Table 1 gives the geographic and geomagnetic coordinates for the GAGAN network stations. Ionospheric TEC, ROTI and S 4 maps are drawn for the entire Indian region with a latitudinal range of 5–40°N and longitudinal range of 65–100°E with 2° × 2° of grid separation. The temporal resolution is chosen as 1 min for TEC maps and 5 min for ROTI and S 4 maps. Fig. 1Open in figure viewerPowerPoint GAGAN network comprising 26 dual frequency GPS station locations Table 1. GAGAN Network Station Coordinates Station Name Geographic Geomagnetic Latitude °N Longitude °E Latitude °N Longitude °E Ahmedabad 23.02 72.51 14.84 146.98 Aizwal 23.84 92.67 14.31 166.11 Bgatti 10.83 72.17 2.81 145.44 Bagdogra 26.68 88.32 17.34 162.17 Bangalore 12.95 77.68 4.42 151.02 Bhopal 23.28 77.34 14.68 151.58 Delhi 28.56 77.22 19.93 151.95 Guwahati 26.12 91.59 16.62 165.2 Hyderabad 17.45 78.47 8.81 152.16 Jodhpur 26.26 73.05 18 147.82 Kolkata 22.64 88.44 13.31 162.05 Lucknow 26.76 80.88 17.87 155.21 Mumbai 19.09 72.85 10.92 146.91 Port Blair 11.65 92.73 2.2 165.64 Raipur 21.18 81.74 12.27 155.58 Shimla 31.08 77.06 22.44 152.05 Trivandrum 8.49 76.9 0.07 149.88 Vishakhapatnam 17.78 83.22 8.8 156.75 Nagpur 21.08 79.06 12.37 153.02 Khajuraho 24.82 79.92 16.01 154.15 Agra 27.16 77.97 18.48 152.52 Aurangabad 19.86 75.39 11.45 149.41 Hubli 15.36 75.08 7.02 148.7 Madurai 9.83 78.09 1.29 151.16 Bhubaneswar 20.25 85.8 11.09 159.39 Gaya 24.74 84.94 15.6 158.87 3.1 Variations in solar and geomagnetic indices The global solar indices like F10.7 and solar wind (SW) plasma speed variations are presented in Fig. 2. The storm sudden commencement time, peak storm time and different phases of the storm can be measured with the help of interplanetary magnetic field (IMF) variations along the Z -direction (IMF B z), planetary-K index (K p) and disturbed storm time (D st) index (Fig. 2). The above-mentioned solar and geomagnetic indices data can be freely downloaded (https://omniweb.gsfc.nasa.gov/form/dx1.html) elsewhere. All the observed solar and geomagnetic indices are in quiet ionospheric condition scales on the first two days, i.e. 15 and 16 March 2015. The B z index lies in −6.6 to 12.1 nT, plasma speed (298–430 km/s), K p index is below 4, D st index is almost positively varied in the range of (−1 to 30 nT) and daily averaged F10.7 index is 113.1 and 116 sfu on 15 and 16 March 2015, respectively. For 15 and 16 March 2015, K p < 4, positive D st index values indicate geomagnetic quiet (pre-storm days) condition (Fig. 2). Fig. 2Open in figure viewerPowerPoint Hourly solar and geomagnetic indices for days 15–20 March 2015 On 17 March 2015, at 05:00 UTC there is sudden turn of IMF B z towards north reaching 20.1 nT and on successive hour 06:00 UTC it pointed towards south with −5.9 nT and SW plasma speed travelled with 500 km/s (Fig. 2). The K p index is 5.7 and D st index gave a positive peak with 56 nT at 05:00 UTC. The daily averaged F10.7 on 17 March 2015 is noted as 113.2 sfu. The sudden storm commencement (SSC) time (initial phase of storm) is observed around 05:00 UTC. The interesting observation is B z turned towards south with −16.3 nT at 08:00 UTC. At 11:00 UTC plasma is moving with 568 km/s, K p index reached a maximum value of 7.7, D st index with −63 nT indicating main phase of geomagnetic storm conditions. The IMF B z remained south for the rest of the day (from noon to midnight) on 17 March 2015. The plasma speed (∼550 km/s) maintained and D st index is in decrement and reached a minimum of −223 nT at 23:00 UTC on 17 March 2015 (main phase completion). During the main phase, K p index value is 7.7. 3.2 Spatial EIA TEC structures The spatial movement of northern EIA is analysed during geomagnetic storm conditions (15–20 March 2015) during 05:00 UTC, 09:00 UTC and 16:00 UTC, respectively, (Fig. 3). Fig. 3Open in figure viewerPowerPoint Spatial TEC maps generated through proposed kriging method for days 15–20 March 2015 for 05:00 UTC, 09:00 UTC and 16:00 UTC It is evident that EIA entry on 15 March 2015 is observed at 05:00 UTC (Fig. 3). The regular EIA structure is developed during 09:00 UTC (14:30 LT) with the maximum TEC intensity of 110 TECU. The EIA exit is clearly noticed at 16:00 UTC (Fig. 3). TEC estimated from kriging method are indicating daily EIA features representing quiet ionospheric conditions on 15 March 2015. The formation of EIA is still under development at 05:00 UTC on 16 March 2015. The EIA crest was formed between 20 and 30°N latitudes during 09:00 UTC time. EIA exit conditions are observed at 16:00 UTC indicating quiet time ionospheric conditions for 16 March 2015. The SSC of geomagnetic storm is observed at 05:00 UTC on 17 March 2015. It is evident that EIA is not yet developed at 05:00 UTC. The complete EIA structure is developed at 09:00 UTC. It is noticed that B z turned southward at 08:00 UTC. The suppressed EIA TEC values are observed at 16:00 UTC. On 18 March 2015, the enhanced TEC values are observed at 05:00 UTC (10:30 LT) between the latitude range of 0–20°N due to north to south fluctuations of IMF B z. The absence of EIA is observed at 09:00 UTC due to negative storm conditions (Fig. 3). The EIA structure shifted towards geomagnetic equator and western longitudes with more TEC intensity values. The intensified storm time TEC fluctuations are mainly due to disturbance dynamo electric fields [22]. The depleted TEC values are noticed at 16:00 UTC. indicating a continuation of negative storm conditions. The long recovery phase of the geomagnetic storm continued for the days 19 and 20 March 2015. The formation of EIA structure is indicated at 05:00 UTC (Fig. 3). The EIA crest was formed at 20°N latitudes at 09:00 UTC. At 16:00 UTC, decrement in TEC values are noticed on 19 March 2015 whereas a clear EIA exit condition is observed on 20 March 2015. The daily EIA formations are well captured by kriging TEC method at 05:00 UTC, 09:00 UTC and 16:00 UTC, respectively. 3.3 Temporal EIA TEC structures To observe the temporal EIA structures, a meridional cross-section at 75°E longitude is considered for the kriging TEC method and the Universitat Politècnica de Catalunya Rapid GIM (UQRG) [11], for the days 15–20 March 2015 (Fig. 4). The x -axis provides the diurnal variation for VTEC and y -axis is the geographic latitude. The colour bar represents the VTEC value. Fig. 4Open in figure viewerPowerPoint Temporal EIA structures provided by proposed kriging method and UQRG TEC maps around 75°E longitude for days 15–20 March 2015 On 15 March 2015, absence of EIA structure is noticed till 03:00 UTC from both kriging and UQRG TEC maps (Fig. 4). The significant EIA temporal structure is developed between 04:00 and 16:00 UTC by the kriging method on 15 and 16 March 2015 and UQRG TEC map shows the EIA temporal structure between 04:00 and 18:00 UTC. The maximum TEC value is 107 TECU for the proposed method and whereas 91 TECU for UQRG TEC maps at 09:00 UTC on 15 March 2015. On 16 March 2015, the proposed kriging method gave maximum TEC of 97 TECU and UQRG TEC map gave peak TEC of 85 TECU. The TEC peak formation (EIA crest) is noticed at 23°N latitude by the proposed method whereas UQRG TEC map gave EIA crest at 22.5°N latitude for the days 15 and 16 March 2015 (Fig. 4). For 17 March 2015, peak TEC of 87 TECU (21°N latitude) and 92 TECU (20°N latitude) at 09:45 and 10:15 UTC is observed from the proposed kriging method and UQRG TEC maps, respectively. On the storm day (17 March 2015) around 10:00 UTC, there is TEC enhancement over 75°E longitude observed from kriging method and UQRG, could be due to prompt penetration electric field (PPEF). After 13:00 UTC, reduction in EIA strength is noticed from the proposed kriging method and UQRG [22]. The maximum TEC of 74 TECU around 09:45 UTC at 15°N latitude is estimated by the kriging method and with 84 TECU at 07:15 UTC at 12.5°N latitude by UQRG on 18 March 2015. There is a clear suppression of EIA crests noticed on 18 March 2015 by the kriging method and UQRG TEC maps. Almost, 6° of a latitudinal shift in EIA crest is estimated by the proposed kriging method between 17 and 18 March 2015. The long recovery phase of the geomagnetic storm continues for the days 19 and 20 March 2015. The maximum TEC is 79 TECU at 23°N latitude is observed from kriging method and 81 TECU at 17.5°N latitude for UQRG TEC maps at 09:00 UTC. On 20 March 2015 there is an enhancement of TEC noticed from the kriging method and UQRG of about 110 and 94 TECU at 09:15 and 10:00 UTC, respectively, (Fig. 4). There are significant errors between the proposed kriging method TEC and UQRG during 20 to 24 UTC hours. These could be due to sparse data sets into the global ionospheric model (only four IGS stations namely Lucknow, Bangalore, Hyderabad, and Port Blair are driven to UQRG model). Pasumarthi and Devanaboyina [58] also observed about ten TECU of deviation between global models (CORG, JPRG and EHRG) and IGS data over the Indian region. The mean TEC deviations are calculated over the entire Indian region between
Referência(s)