GNSS data and ionospheric studies. Prof. Andrzej Krankowski

Transcription

1

2 GNSSdataandionosphericstudies Prof. Andrzej Krankowski University of Warmia and Mazury in Olsztyn, Poland Space Radio-Diagnostics Research Centre (SRRC/UWM) kand@uwm.edu.pl Outline Agenda Introduction. GNSS signals propagation Introduction. Ionosphere GNSS networks as data for the ionosphere monitoring GNSS data generation and RINEX data format Total Electron Content (TEC) TEC mapping approach IGS/UWM Ionosphre Combination Centre VTEC maps and the IONEX format COSMIC radio occultation Monitoring of the TEC fluctuations using GNSS data Regional monitoring of the ionosphere Conclusions IRI215 Workshop 2

3 Introduction. GNSS signals propagation The ionosphere medium where GNSS signals pass a long distance. The ionosphere delay is the significant error source for satellite navigation systems, but it can be directly measured and mitigated with using dual frequency GNSS receivers. Dual frequency GPS measurements can effectively provide integral information on the electron density along the ray path by computing differential phases of code and carrier phase measurements. The integral of the electron density along the ray path (TEC) between the transmitting GNSS satellite and the receiver. Introduction. Ionosphere Ionosphere is the part of the Earth s atmosphere, consisting of several ionized layers and extending from about 5 km up to 1, km. Plasma density distribution in the ionosphere varies with: altitude day/night seasons latitude/longitude solar activity geomagnetic conditions Global ionospheric maps of total electron content (TEC) IGS GIMs Image credit: UWM Equatorial Region: strongest effects; highest; strongest TEC gradients; Irregularities not correlated with magnetic activity Mid-Latitude Region: normally quiescent but with strong gradients during extreme levels of geomagnetic activity Auroral Region: aurora and structures. Phase scintillations. Ionosphere-plasmasphere system (courtesy of the Windows to the Universe) Electron density distribution with altitude

4 Introduction. Ionospheric radio waves propagation Study of the ionosphere in order to provide the broad understanding necessary to support space and groundbased radio systems. Global morphology and modeling of the ionosphere; Ionospheric space-time variations; Development of tools and networks needed to measure ionospheric properties and trends; Theory and practice of radio propagation via the ionosphere; Application of ionospheric information to radio systems. GNSS networks as data for the ionosphere monitoring The IGS and other permanent GNSS networks collects, archives, and distributes GPS observation data sets of sufficient accuracy to satisfy the objectives of a wide range of applications and experimentation. The GNSS observations provided by IGS and other permanent station networks, with a 3 s sampling RINEX data. International GNSS Service - IGS IGS polar stations PBO Network Plate Boundary Observatory IGS/EPN POLENET - The Polar Earth (EUREF Permanent Tracking Network) Observing Network

5 GNSS data generation RINEX data format IncludesASCIIfileformatsfor: observation( o ) navigation( n ) meteorological( m ) ionosphericdata( i ) Definedathttp:// Eachfiletypeconsistsofaheadersectionandadatasection Headersectioncontainsglobalinformationfortheentirefileandisplaced atthebeginningofthefile. Containsheaderlabelsincolumns618foreachlinecontainedinthe headersection Theselabelsaremandatoryandmustappearexactlyasperformat description RINEXfilenameconvention: ForsiteSSSS,ondayofyearDDD,sessionTandyearYY: SSSSDDDT.YYo(RINEXobservationfileiethesite sgpsdata) SSSSDDDT.YYn(RINEXnavigationfileiethebroadcastephem) E.g.,hers127.3oisobservationdataforHerstmonceux,day127, session,year23. AllthedatesandtimesinGPST

6 RINEX observation data. Header + Body. 2 OBSERVATION DATA RINEX VERSION / TYPE National GPS Network Ordnance Survey Oct 3 1:25:41 22PGM / RUN BY / DATE Active Station at Ordnance Survey Office Taunton COMMENT TAUN MARKER NAME TAUN MARKER NUMBER National GPS Network Ordnance Survey OBSERVER / AGENCY 8148 LEICA RS REC # / TYPE / VERS 348 LEIAT54 LEIS ANT # / TYPE The following coordinates are NOT APPROXIMATE COMMENT Approx coords replaced by official precise ETRS89 values COMMENT APPROX POSITION XYZ... ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 4 L1 C1 L2 P2 # / TYPES OF OBSERV TIME OF FIRST OBS TIME OF LAST OBS END OF HEADER PRN PRN PRN PRN PRN PRN PRN RINEX observation data. Header + Body. 2 OBSERVATION DATA RINEX VERSION / TYPE National GPS Network Ordnance Survey Oct 3 1:25:41 22PGM / RUN BY / DATE Active Station at Ordnance Survey Office Taunton COMMENT TAUN MARKER NAME TAUN MARKER NUMBER National GPS Network Ordnance Survey OBSERVER / AGENCY 8148 LEICA RS REC # / TYPE / VERS 348 LEIAT54 LEIS ANT # / TYPE The following coordinates are NOT APPROXIMATE COMMENT Approx coords replaced by official precise ETRS89 values COMMENT APPROX POSITION XYZ... ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 4 L1 C1 L2 P2 # / TYPES OF OBSERV TIME OF FIRST OBS TIME OF LAST OBS END OF HEADER PRN PRN PRN PRN PRN PRN PRN

7 Total Electron Content (TEC) Estimation of absolute Total Electron Content (TEC) using GPS involves two steps: - Levelling the phases to the pseudorange gives the relative TEC - Estimation/removal of instrumental biases (calibration) gives absolute TEC Absolute TEC: TEC = TEC R -( b R + b S ) Relative TEC: TEC R = TEC + <TEC P TEC > ARC b R - Receiver/station bias (TECU) b S - Satellite bias (TECU) TEC P - Differential pseudorange (TECU) TEC - Differential carrier phase (TECU) < > ARC - Average over phase-connected arc TECU TEC P TEC UT (hours) Vertical total electron content estimation RINEX files contains a measurements of dual-frequency signals delays at the frequencies L1=1.6 GHz and L2=1.2 GHz with 3 sec resolution. Dual-frequency radio signals, propagated through the ionosphere, are subject to a differential phase change due to the dispersive nature of the plasma. GPS receiver provides simultaneous measurements of pseudo-range (code) P1 and P2, and carrier phase delays of signals L1(1) and L2(2), which can be written as follows: I c ( t t ) P1 1 s _1 r _ 1 P2 I2 c ( ts _ 2 t r _ 2) L I L 1 1 1N1 2 I2 2N 2 Differential delay of both signals is proportional to the slant TEC: 2 f f f f 2 I TEC M TEC STEC can be transformed into VTEC by use of a mapping function M(): VTEC( rp, t) STEC( PS, PR, t) M ( ) where the zenith angle of the signal path at the piercing point P(r P ) Assuming that the inter-frequency differential delays R and S are known, the geometryfree linear combination can be transformed into VTEC estimation

8 Vertical total electron content estimation, DCBs In order to retrieve an absolute values of the slant TEC we should take into account the instrumental offset for a receiver and a GPS satellite so called, DCBs (differential code biases). calibrated STEC = STEC + Br + Bs where Br a receiver bias, Bs a GPS satellite bias Offsetisthedifferencebetweenrelative stecandabsolutestec Illustrationofthebiasapplicationfora zenithtecestimates Horvath and Crozier, RS, 27 Rideout and Coster, GPS Sol, 26 Vertical total electron content estimation, DCBs 1) For all GPS and GLONASS satellites, as well as for a selected set of receivers DCBs are estimated during the GIMs calculation as set of unknowns in the overestimated system of equations (e.g. Schaer et al., 1999). 2) To estimate DCB for a single receiver independently there can be used some suitable mathematical representation of the spatial and temporal variability of TEC over a station. For example, - polinomial expansion of all GPS TEC data derived at this GPS station over 1 or 3 days interval (Cirraolo et al., 27) - bi-quadratic expansion (Brunini and Azpilicueta, 29) For the case of a dense local GNSS network DCB for a single receiver can be determined (Ma and Maruyama, AnGeo, 23): either - from mathematical model of vtec expansion over some region (vtec mapping) or from the assumption of the equality of vtec data at the same bin/mesh or - from comparison of the concurrent stec data for 2 co-located GNSS receivers

9 TEC mapping approach For the representation of TEC estimates a spherical harmonics expansion with different degree and order was carried out, as shown in Eq. (1). TEC(, ) 15 n n m P nm (sin)( a nm cos( m) b nm sin( m)) where, are geographic latitude and longitude, Pnm are the normalized associated Legendre functions of degree n and order m; anm and bnm are the unknown SHE coefficients which were derived using GPS TEC observations. TEC mapping approach SHE (spherical harmonics expansion) is one of the most commonly used and approved techniques for TEC mapping. A number of regional TEC maps are based on SHE approach. 1) CODE s Global Ionospheric Maps (GIMs) 2) Australian (IPS) regional TEC maps 3) South American regional ionosphere maps (SAIMs) 4) South African regional TEC maps 5) Regional ionospheric maps over Japan (GEOTEC)

10 IGS/UWM Ionosphre Combination Centre International GNSS Service - IGS IGS directly manages ~4 permanent GNSS stations observing 4-12 satellites at 3 s rate: more than 25, STEC observations/hour worldwide, but there is lack of stations at some areas (e.g., over the oceans) IGS/UWM Ionosphre Combination Centre The IGS Ionosphere Working group started its activities in June 1998 with the main goal of a routinely producing IGS Global TEC maps. This is being done now with a latency of 11 days (final product) and with a latency of less than 24 hours (rapid product). This has been done under the direct responsibility of the Iono-WG chairmans: 1. Dr Joachim Feltens, ESA , 2. Prof.. Manuel Hernández- Pajares, UPC, Prof. Andrzej Krankowski, UWM, 28- CODE ESA JPL UPC IGS Ionosphere Maps UWM The IGS ionosphere product is a result of the combination of TEC maps derived by different Analysis Centers by using weights computed by Validation Center, in order to get a more accurate product. IGS Ionosphere Analysis Centers IGS Ionosphere Combination Center IGS GNSS data IGS Ionosphere Validation Center JASON TEC data UWM

11 Determining VTEC in a global network: main problem: lack of data - South and Oceans It can be seen that the typical holes appearing at the first stage of the global maps computation (each 2 hours). This requires an optimum spatial-temporal interpolation technique to cover all the Ionosphere. Lack of data over the equatorial Africa and Atlantic, and in part over equatorial and southern Pacific, hamper the detection of the equatorial anomalies (June 13, 24). 19 TheIONEXformatbody The IONEX (IONosphere interexchange) format allows to store the VTEC and its error estimates in a grid format, in consecutive values at different longitudes- for each latitude grid point. 1 START OF TEC MAP EPOCH OF CURRENT MAP LAT/LON1/LON2/DLON/H LAT/LON1/LON2/DLON/H LAT/LON1/LON2/DLON/H END OF TEC MAP 2 START OF TEC MAP END OF TEC MAP 1 START OF RMS MAP EPOCH OF CURRENT MAP LAT/LON1/LON2/DLON/H END OF RMS MAP END OF FILE

12 VTECmaps:theIONEXformat The IONEX (IONosphere interexchange) format allows to store the VTEC and its error estimates in a grid format, in consecutive values at different longitudes- for each latitude grid point. IONEX header 1. IONOSPHERE MAPS MIX IONEX VERSION / TYPE cmpcmb v1.2 gage/upc 11-may-4 13:1 PGM / RUN BY / DATE ionex file containing IGS COMBINED Ionosphere maps COMMENT global ionosphere maps for day 118, 24 DESCRIPTION IONEX file containing the COMBINED IGS TEC MAPS and DCBs DESCRIPTION EPOCH OF FIRST MAP EPOCH OF LAST MAP 72 INTERVAL 13 # OF MAPS IN FILE COSZ MAPPING FUNCTION. ELEVATION CUTOFF combined TEC calculated as weighted mean of input TEC valuesobservables USED 29 # OF STATIONS 28 # OF SATELLITES BASE RADIUS 2 MAP DIMENSION HGT1 / HGT2 / DHGT LAT1 / LAT2 / DLAT LON1 / LON2 / DLON -1 EXPONENT TEC values in.1 tec units; 9999, if no value available COMMENT DCB values in nanoseconds, reference is Sum_of_SatDCBs = COMMENT DIFFERENTIAL CODE BIASES START OF AUX DATA PRN / BIAS / RMS PRN / BIAS / RMS PRN / BIAS / RMS acor STATION / BIAS / RMS acu STATION / BIAS / RMS... zwen 1233M STATION / BIAS / RMS DIFFERENTIAL CODE BIASES END OF AUX DATA END OF HEADER Overall validation of VTEC maps during more than 1 years of IGS Iono WG operations Example of comparison of IGS vs JASON: UT 7-9UT Units: 1 TECUs Red: Jason-1 TEC Green: IGS final TEC JASON dual frequency altimeter provides a direct and independent VTEC below its orbit (13 km) over the oceans (the worst case for GPS) UT 19-21UT

13 Example of IGS Final GIM: DOY TEC map 4AnalysisCenters(CODE,ESA, JPL, and UPC) and a Validation Center (UPC) have been providing maps (at 2 hours x 5 deg. x 2.5 deg in UT x Lon. x Lat.), weights and external (altimetry-derived) TEC data. RMS map From such maps and weights the Combination Center (at first ESA, then UPC, and since 28 - UWM) has produced the IGS TEC maps in IONEX format. Units: TECU Example of IGS RAPID GIM: DOY TEC maps RMS maps Units:.1 TECUs

14 Ionosphere sounding by GNSS signals. COSMIC radio occultation Ionosphere sounding by GNSS signals. COSMIC radio occultation Occultation location for COSMIC This map illustrates the typical locations of the COSMIC soundings points during one day.

15 Comparison of IRI profiles with COSMIC and ionosonde data (DIAS) UT 6 1 h, ju km 8 5 ch pr 6 cosmic ionosonde IRI UT 6 1 h, ju km 8 5 ch pr 6 cosmic ionosonde IRI 27 4 ar eb ro ar eb ro Ne 1 5 el/cm Ne 1 5 el/cm UT 6 1 h, ju km 8 5 ch pr 6 cosmic ionosonde IRI UT 6 1 h, ju km 8 ch 5 pr 6 cosmic ionosonde IRI 27 4 ar eb ro ar eb ro Ne 1 5 el/cm Ne 1 5 el/cm UT 6 1 h, ju km 8 ch 5 pr 6 cosmic ionosonde IRI UT 6 1 h, ju km 8 5 ch pr 6 cosmic ionosonde IRI 27 4 ar eb ro ar eb ro Ne el/cm Ne 1 5 el/cm 3 Ex. Geomagnetic Disturbance in October TEC TECU 8 4 Kp Kp index Dst, nt a) b) October UT 9 TECU October 28 TECU TECU Juliusruh (54, 13) DRES (51, 13) Athens (38, 23) ORID (41, 2) Diurnal variations of TEC (red line) over DRES and ORID IGS GPS stations. The crossed line indicates variations of (fof2) 2 /3 over Juliusruh and Athens ionosondes. Blue line corresponds to the average TEC variation

16 Comparison F3/COSMIC electron density with IGS final TEC maps 9 October 28 quiet day 11 October 28 disturbed day 12 UT Comparison F3/COSMIC electron density with IGS final TEC maps 9 October 28 quiet day 11 October 28 disturbed day 14 UT

17 Ionosphere sounding by GNSS signals. COSMIC radio occultation 68% 6 1 NmF2 COSMIC (1 5 el/cm 3 ) y=.994x+.23 R=.986 N= NmF2 Ionosonde (1 5 el/cm 3 ) count NmF2 (%) 68% <x>=.72 % std = 8.42 % <x>= 2.8 km std = km hmf2 COSMIC (km) 3 2 count 6 4 <x>= 1.36 % std = 4.89 % 1 y=.912x+25 R=.949 N= hmf2 Ionosonde (km) hmf2 (km) Monitoring of the TEC fluctuations using GNSS data Monitoring the time-derivative of TEC (ROT, rate of TEC change) is useful for tracing the presence of the ionospheric irregularities. ROT, as a measure of phase fluctuation activity, is calculated using the algorithm, that was proposed by Pi et al. [1997]: ROT i TECk TEC ( t t ) 1 k k i k 1 where i is the visible satellite and k is the time of epoch. In standard RINEX files raw data sampled every 3 seconds. ROT is calculated in units of TECU/min for each visible satellite over GNSS station. The ROT values are calculated and then detrended for all individual satellite tracks for elevation angles over 2 degrees. ROTI is defined as the standard deviation (taken over 5 minutes) of the detrended rate of change of TEC (ROT) [Pi et al., 1997]. Based on the retrieved values of ROT, the ROTI values are calculated over 5-minute periods with running window for each station: ROTI ROT 2 ROT 2 To observe the spatial behavior of the ionospheric fluctuations over the North Pole, we process ROTI data from considered multi-site database and visualize result in the form of ROTI map.

18 The ROTI maps There are more than 7 permanent stations (from IGS, UNAVCO and EUREF databases) have been involved into processing for the ionosphere fluctuation service. Such number of stations provides enough data for representation a detailed structure of the ionospheric irregularities pattern. Thelocationsofthestationsin thenorthhemisphereusedfor ROTImapconstruction Due to strong connections between the Earth s magnetic field and the ionosphere, the behavior of the fluctuation occurrence is represented as a function of the magnetic local time (MLT) and of the corrected magnetic latitude. The grid of ROTI maps in polar coordinates with cell size 2 degree (magnetic local time) and 2 degree (geomagnetic latitude). A quadrant of ROTI map grid. The ROTI maps Each map, as a daily map, demonstrates ROTI variation with geomagnetic local time (-24 MLT). The value in every cell is calculated by averaging of all ROTI values cover by this cell area and it is proportional to the fluctuation event probability in the current sector. If in the cell there are only few ROTI values (less than 3), this grid cell is marked as a blank. This approach will allow to avoid frequent errors of interpolation techniques, related with unrealistic interpolation values over areas with data gaps.

19 The service for creating ROTI maps In the Space Radio-Diagnostics Research Centre (SRRC/UWM) is developing and testing service for automatically download RINEX observational and navigational files, unzip and process data and generate ionospheric products. TheTECfluctuationserviceoperationdiagramandstatus ASG-EUPOS system Presently this segment is composed of the following set of reference stations: - 81 stations with the GPS module, - 18 stations with the GPS/GLONASS module. Products (services)

20 TEC changes over Poland May 2, 21 May 1, 21 May 3, 21 Scintillation measurements July 28, 211 Scintillation receivers: Javad Sigma G3T Septentrio PolaRx3

21 Conclusions 1. Long series of IGS VTEC maps offers a very good source of information about the ionosphere with high spatial and temporal resolution 2. Future improvements are determined by users requirements (the number of users has significantly increased during the last 17 years) years of continuous time series of TEC measurements may be applied to update ionospheric models, e.g., IRI model 4. COSMIC occultation data gives a new opportunity to study/model the ionosphere and to validate IGS TEC maps Conclusions 5. A long time series of accurate global VTEC values are freely available since 1998 for scientific or technical use, with latencies of about 12 days (final product) or 1 day (rapid product). Thanks to the cooperative effort developed within the IGS framework and the international scientific community this open service will hopefully continue its evolution during the next years, sensitive to both new user needs and scientific achievements.

22 Acknowledgments The author is particularly grateful for the GNSS data provided by IGS/EPN and UNAVCO Thank you for your attention

23 Irregularities at Equatorial Regions Pornchai Supnithi Department of Telecommunications Engineering King Mongkut s Institute of Technology Ladkrabang (KMITL), THAILAND pornchai.su@kmitl.ac.th Introduction What is the ionosphere? The electron density also varies with time, location and solar activity km height from the earth surface Ionized (electrons + ions) region which enhances the HF radio propagation (HF Band : 3-3 MHz) Free electrons and ions are due to exterme ultra violet (EUV) and X- ray 2

24 Plasma in the ionosphere Ionospheric plasma consists of the electron and ion fluid immersed in the neutral gas. Neutral gas density > plasma density (until several thousand kilometers in altitude). *** Plasma: Mixture of electrons and ions that behaves as a neutral fluid with electrical properties (mobility, conductivity) Space Daytime atmospheric composition based on mass spectrometer measurements above White Sands, New Mexico (33 N, 16 W) Ionosphere Atmosphere Ionospheric behavior => Plasma Physics + Fluid dynamics 3 Photo Ionization and Recombination Daytime Nighttime Recombination O + + e O or N + + O 2 NO + + O NO + + e N + O O + hv O + 4

25 Layers of Ionosphere Daytime Nighttime Divided into 3 layers: D, E, F The F2 layer is the uppermost layer of the ionosphere which is the most effective layer for long distance HF communications km km F layer:14-4 km E layer : 9-14 km D layer : 5-9 km 5 Ionospheric parameters fof2 is the F2 layer critical frequency (fof2) which are manually scaled from the bottomside ionograms recorded by the ionosonde. hmf2 is the F2 layer peak height, which is computed from 149 hmf2 = -176, M(3)F2 + M (1a) fof2 The propagation factor, M(3)F2: MUF(3) F2 M(3) F2 (2) fof2 MUF(3)F2 is the maximum usable frequency reflected from the F2 layer of the ionosphere which could be received at a distance of 3 km. where ΔM = f1 f2/(fof2/foe - f3) + f4, f1 =.232 Rz , o 2 f2 = 1- Rz12/15 exp(-( /4 ) ), f3 = exp(rz12/41.84), (1b) (1c) (1d) (1e) f4 =.96 (Rz12-25)/15, (1f) is the magnetic dip latitude. foe is the E-layer critical frequency. Rz12 is the 12-month running average of the sunspot number. D. Bilitza, N.M. Sheikh, R. Eyfrig, A global model for the height of the F2-Peak using M3 values from the CCIR numerical maps. Telecommun. J. 46 (9), pp ,

26 Equatorial Ionization Anomaly (EIA) Appleton Anomaly The fountain effect ExB X B E Near the magnetic equator, the daytime eastward electric field affects the plasma The E B drift, which raises the ionosphere to higher altitudes where the ion-neutral collision frequency is reduced. The lifted ionospheric plasma near the density peak subsequently flows down along the arch-shaped magnetic field lines toward the off-equatorial latitudes due to the gravitational acceleration of the Earth and the reduced ion-neutral drag force. On the topside, downward field-aligned plasma flux is induced to reach the diffusive equilibrium. 7 Equatorial Ionospheric Anomaly (EIA) Large scale ionospheric density enhancement around ±15º magnetic latitude 8

27 Noon bitout Noon bite-out in NmF2 or fof2 occur around equatorial regions The daytime upward ExB drifts in the F region moves the ionizations around hmf2 height to higher latitudes Tends to have large peak in the morning (high SA)or afternoon(low SA) Noon Bite-out [Maruyama et al., 213] 9 Zonal electric field V day For electrostatic field, the Maxwell equation, implies, B E t Edl V night Poisson s equation E The voltage difference between the dusk and dawn terminator when the electric field is westward (at night) should equal the voltage drop when electric field is eastward (during the day). c M.Kelly, The Earth s Ionosphere: Plasma Physics and Electrodynamics, 2 nd Ed. 1

28 F region Dynamo Two (N and S) E-region dynamos One F-region dynamo. R 1 P Off-equatorial E-region wind dynamo competes with the F-region dynamo at the equator to determine the voltage differences between magnetic field lines F P Nighttime: is larger F region dominates. Daytime: E-region sources determine the electric field, the F layer acts as a load M.Kelly, The Earth s Ionosphere: Plasma Physics and Electrodynamics, 2 nd Ed. 11 Zonal eastward drift of ion gas Ionosphere Doppler shifts detected by the two beams zonal eastward ion drift Incoherent Scatter Radar Incoherent scatter radar can determine, - plasma temperature - plasma density - plasma composition - ion drift velocity, as a function of altitude and time. Ion velocity perpendicular to B kt B i n M B ( Vi ) E ( ) ( ) g 2 qi n q i B E V i B **** The neutral wind velocity does not appear in this equation. 12

29 Results Zonal Eastward Drift E Vertical Drift ( ) ~2V peak V peak Pre reversal enhancement ~2V peak V peak Day Night Fejer et al. (1991) Sunset Sunrise 13 Thermospheric Wind or Neutral Wind Radiation cooling Solar heating SUN UV/EUV radiations [T. Maruyama, SEALION Symposium, 211] 14

30 Pre reversal enhancement (PRE) g Before sunset: Eastward E After sunset : Westward E Linear growth rate (): [Nishioka et al., 212)] Pre reversal electric field enhancement Enhancement in the eastward electric field Ex g 1 Wy B vin L E x eastward electric field, W y vertical neutral wind, B geomagnetic field, g gravity acceleration, in the ion neutral collision frequency, L scale length of the vertical gradient of the plasma density in the F region The post sunset equatorial F layer undergoes a rapid uplift. high altitude has low v in The positive electron density gradient region of the rapidly rising F layer bottomside becomes unstable Growth of plasma bubble irregularities Rayleigh Taylor instability mechanism. 15 PRE vertical drift Weak SF Strong SF Jicamarca, Peru PRE amplitude increases with Solar Flux (F1.7) Intensity of Spread F is measured From the height extension of the Radar echoes [Chapagain et al., 29] 16

31 Equatorial Spread F Equatorial spread F (ESF) Plasma structure in nighttime equatorial F layer 1. Frequency Spread-F : FSF type 2. Range Spread-F : RSF type 3. Mixed Spread-F : MSF type 17 Equatorial plasma bubbles (EPBs) Flux tube aligned vertically rising plasma depletions Tends to move eastward Elongated along the geomagnetic field line Scale sizes from tens of centimeters to 1s of kilometers The effects can be observed by means of various ranging techniques [T. Maruyama, 212] [E.R.Young et al, 1984] 18

32 EPB development The plasma bubbles develop as with secondary irregularities developing by cascading processes at the steepening density gradients of the bubbles [Tsunoda, EPS, 215] 19 EPB development [Abdu and Brum, EPS, 29] (1) Thermospheric zonal wind (eastward in the evening) PRE The longitudinal/local time gradient in the integrated E layer conductivity near sunset SS. PRE (Rishbeth, 1971; Heelis et al., 1974) PRE The evening F layer height, vertical plasma drift (2) Steep gradient at bottomside F layer R-T instability (3) Thermospheric meridional/transequatorial winds (represented also by the symmetry/asymmetry conditions of the EIA) the integrated Pedersen conductivity of the unstable flux tube 2

33 EPB development The seeds of EPB can be the convection of gravity waves from troposphere or the propagating waves (Kelley at al 1981) such as large scale wave structure (Tsunoda,25&28). 21 increases when Linear Growth Rate Ex g 1 Wy B vin L Enhanced E x (vertical drift) Uplift of F region higher altitudes with lower neutral density, n lower v in 22

34 Current Plasma Bubble Models Ionospheric plasma bubble simulation (J. Retterer, 211) 23 Types of EPB Post sunset EPB Solar max. periods PRE dominates ESF process Post midnight EPB Solar min. periods especially solstice months The role of gravity waves (GW) is evident Occurrence rates increase with decrease of solar flux 24

35 Effects of Plasma bubbles Ionospheric delay gradient Areonautical navigation: Ground-based augmentation system (GBAS) GNSS Satellites S. Saito 211. Plasma bubble Scintillation Ionosphere Amplitude and Phase fluctuation Airplane Differential correction??? Reference Station 25 EPB Observation Satellites (scale size, height) VHF radar (3m) Ionosonde GPS receiver total electron content (TEC) (1 1km), scintillation (3m) Optical sky imager (hundereds of km) HF experiments (LSWS) 26

36 Variability of ESF, EPB Long term variation Pre reversal enhancement (PRE) solar cycle dependent (F1.7, SSN) Medium term variation Short term or day to day variation Due to gravity waves (GW) caused by LSWS in the bottomside F2 layer Not sufficient understanding 27 Long term variation EPB on Global UltravioletImager(GUVI) Optical imager in Brazil [Comberiate&Paxton,21] [Sahaietal.,2]

37 Annual variation Indonesia Seasonal variation was repeated in every year. Occurrence rate was decreasing year by year 29 Medium term variation Monthly and seasonal scale Longitude dependent/magnetic declination/solar declination angle The meridional component of the thermospheric wind (The meridional wind) produce asymmetry in the EIA increase of the field line integrated conductivity ESF can be suppressed by meridional winds but if the P PRE is large, then ESF can still occur. 3

38 Longitudinal dependence of EPB [M. Pezzopane et al., Ann. Geophys., 213] 31 Longitudinal dependence of EPB Chiangmai (18.8 o N, 98.9 o E) Palmas (1.2 o S, o E) [M. Pezzopane et al., Ann. Geophys., 213] 32

39 [Hoang et al., ASR, 21] Magnetic declination: Sao Luis (SL): ~21 o W Ho Chi Minh City (HCM): -2 o N Sao Luis (SL) Ho Chi Minh City (HCM) Monthly mean h F 33 GPS Scintillation DIF Acquisition Tracking Loop Navigation data Scintillation 34

40 GPS Scintillation Scintillation C/No PRN3-1 Ionosphere n 1 n 2 n 3 n 4 The ionospheric irregularities cause the fluctuations in the refractive index (n). The rapid fluctuations in amplitude and phase of GPS signals are resulted. In the worst case, there is loss of lock on satellite signal. n 5 V Amplitude and Phase fluctuation STEC (TECU) STEC (TECU) Time (min) KMITL STFD Time (UTC) 35 S4 measurement Wide Band Power Narrow Band Power Signal Intensity Total S4 S 4 T i i i WBP = I Q NBP I i Q i i1 i1 SI = NBP - WBP 2 2 ESI E SI E 2 SI Detrending: using a 6 th -order low-pass Butterworth filter with a.1 Hz cutoff frequency SI k NBPWBPk NBPWBP, lpf k 36

41 GPS Scintillation 6 PRN PRN3 5 PRN I amplitute 2-2 Signal Intensity raw-si detrended-si C/No Time (min) I-channel amplitude Time (min) The intensity of raw signal and detrended signal Time (min) The C/No estimation 4 PRN3 1 1 PRN3 5 PRN I amplitute Signal Intensity SI Elevation Angle ,5 1,51 1,52 1,53 1,54 1,55 Time (sec) Time (min) The final signal intensity Time (min) The elevation angle 1.8 PRN3 S4-corr S4-total S4 Index Time (min) 37 Rate of TEC change index: ROTI The ROTI is used for ionospheric irregularities detection at one station for one day, defined by Standard deviation of rate of TEC change with 5-minute windows. In this work, we determined.5 TECU/min is the threshold. STEC and ROTI at KMITL station ROT () i STEC( i 1) STEC() i N 1 ROTI ( ROT ( i) ROT ) N i 1 i = Index of time N = Window times (minutes) th September st August 212 KMITL station 15 KMITL station STEC (TECU) 1 5 STEC (TECU) ROTI 1.5 Disturbance ROTI 1.5 Quiet Time (UTC) Time (UTC) 38

42 STEC and ROTI KMITL station September KMITL station 15 KMITL station 15 KMITL station 15 KMITL station 1 KMITL station STEC (TECU) 1 5 STEC (TECU) 1 5 STEC (TECU) 1 5 STEC (TECU) 1 5 STEC (TECU) ROTI.5 ROTI.5 ROTI.5 ROTI.5 ROTI Time (UTC) Time (UTC) Time (UTC) Time (UTC) Time (UTC) 1 KMITL station 1 KMITL station 1 KMITL station 15 KMITL station 15 KMITL station STEC (TECU) 5 STEC (TECU) 5 STEC (TECU) 5 STEC (TECU) 1 5 STEC (TECU) ROTI.5 ROTI.5 ROTI.5 ROTI.5 ROTI Time (UTC) Time (UTC) Time (UTC) Time (UTC) Time (UTC) 39 Statistics of high ROTI levels Number of day Number of day Solar cycle We use the data from September 211,212, 213 and March 212, 213 4

43 GPS observation Bangkok area 22 Sept. 211 (uncalibrated TEC [Tsujii et al., Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 212] 41 VHF radar observation Equatorial Atmosphere Radar (EAR) Location : 1.32 E,.2 S Frequency: 47. MHz, 3.8 MHz Antenna System: Quasi-circular active phased array Beam Direction: 3 deg zenith (Ionospheric irregularity) Observation Range: 1.5km-2km (Atmosphere) 8km - (Ionospheric irregularity) [M. Nishioka, Plasma Buble in the Ionosphere, Kyoto University] 42

44 Equatorial Spread F 5MHz radar (λ = 6m) detects waves of 3-m wavelength 43 3m scale Ionospheric irregularities Echo from irregularity March 24, 24 Plasma bubble occurred around 2: LT. Several echoes appeared and drifted eastward in the speed of 1 15m/s M. Nishioka, Plasma Buble in the Ionosphere, Kyoto University 44

45 Optical Sky Imagers Fukushima et al. (JGR, 215) 45 Optical Sky Imagers 46

46 Spread-F The ESF are categorized into 3 types based on U.R.S.I. Handbook of Ionogram Interpretation and Reduction (W.R. Piggott and K. Rawer, 1972). We consider the 3 types of spread-f as follows : 1. Frequency Spread-F : FSF type 2. Range Spread-F : RSF type The ESF occurrence rate is calculated by 3. Mixed Spread-F : MSF type %Monthly mean ESF occurrence = Sum of ESF occurrence Total number of observed ionogram X 1 47 RSF vs. FSF on ionograms RSF FSF 48

47 Es + Spread F on ionogram Spread F Es 49 RSF Occurrence Rate (CPN Station) 5

48 RSF Occurrence Rate (CMU Station) 51 FSF Occurrence Rate (CPN Station) 52

49 FSF Occurrence Rate (CMU Station) 53 Conjugate points in Southeast Asia [Maruyama et al., 213] 54

50 %RSF Occurrence % % % % This confirms that the plasma bubble is generated around the magnetic equator and then expand to the higher latitude area. Chiangmai (CMU) Chumphon (CPN) Kototabang (KTB) SEP 2% SEP OCT OCT 1% NOV 5% DEC LT NOV DEC data is not available In general, %RSF occurrence at CPN is higher than KTB and CMU The %RSF occurrence at CMU is not over 2% in average. % % % % JAN JAN 15% FEB 2% FEB MAR MAR 15% APR APR 1% LT % % % % The %RSF is not over 2% in average The %RSF is not over 2% in average 2 2 at KTB from OCT to JAN and APR at CPN SEP in NOV, DEC, JAN JAN and FEB SEP JAN The 8higher rate mostly 9occurs The higher rate occurs SEP JAN SEP JAN during the equinoctial months in SEP, FEB and MAR % 2% 4% 1 % OCT 5 % NOV 25% DEC 1% LT OCT NOV DEC % % % % FEB MAR APR LT FEB 2 % MAR 5 % 35% APR % % % % 2 % 2 % LT 1 % OCT NOV DEC % % % % 35 % 35 % APR 2 % LT FEB MAR APR 55 ESF Probability with IRI-212 Model Chumphon Station (CPN) SEP 28 JAN 29 The observed ESF probability = Sum of ESF occurrence Total number of observed ionogram NOV 28 MAR 29 The IRI model overestimates the ESF occurrence probability for all months, most evident in JAN 29. Observation IRI-212 model Note : the data in MAR 29 is only available for 3%. But the IRI model gives closer probability values in SEP 28 than others. Note : the ESF probability of the current IRI model is based on the Brazilian sector and the monthly mean percentage computed by Cubic-B spline interpolation method. Only the RSF type is included in the model since it is related to the cause of the plasma bubble. 56

51 Comparison of ESF Probability with IRI 212 Model : 3 Station ESF Prob. ESF Prob. Chiangmai (CMU) Chumphon (CPN) Kototabang (KTB) SEP 28 SEP 28 SEP 28 ESF Prob. ESF Prob. MAR 29 MAR 29 MAR 29 ESF Prob. ESF Prob. LT LT LT The comparison results of 3 conjugate stations in equinoctial months. For these 2 months, IRI model overestimate the ESF occurrence probability for all months, most evident at Chiangmai (CMU) station. The IRI model gives closer probability values to Chumphon (CPN) station than others. 57 Conjugate Point Equatorial Experiment Conjugate Point Equatorial Experiment (COPEX), Brazil [Abdu et al., 28] 58

52 Vertical Drifts [Abdu et al., 28] 59 Ionospheric effects to GBAS Simplified ionosphere wave front model GNSS Satellites The reference stations provide the differential corrections and integrity information to the receiver that are equipped in the aircraft in the nearby area. Ionosphere Airplane Front Speed V iono_front V airplane h Front Slope Front Width Differential correction c 1 -d d Reference Station Front Slope or Ionospheric delay gradient Causes of Ionospheric delay gradient, 1. Due to the physical ionospheric separation between aircraft and reference station. 2. Due to the ionospheric irregularities (plasma bubbles, SED). Question : How large of the ionospheric delay gradient in the low latitude regions can be? 6

53 Ionospheric delay gradient Single difference method 1 September 211 STEC STEC B B k k K adj _1 1 S R _1 STEC STEC B B k k K adj _ 2 2 S R _2 STEC (TECU) dstec (TECU) dstec STEC STEC STEC 4 3 KMITL STFD 2 1 PRN Quiet time br STFD br KMITL Time (UTC) k k k adj _1 adj _2 k k STEC1 STEC2 br 1 br2 ( ) ( ) dstec Ionospheric disturbance (STEC STFD STEC KMITL ) + (Br KMITL Br STFD ) 61 Differential STEC GPS Satellites In this work, we investigate the ionospheric delay gradient based on two GPS monitoring stations, KMITL ( N, E) STFD ( N, E) Ionosphere STEC STFD STEC KMITL STEC STEC b b adj _ STFD STFD r _ STFD s STEC STEC b b adj _ KMITL KMITL r _ KMITL s 12 km STFD KMITL dstec STEC STEC adj _ STFD adj _ KMITL ( STEC STEC ) ( b b ) STFD KMITL r _ STFD r _ KMITL Offset b r 62

54 Results and discussions STEC (TECU) KMITL Ionospheric disturbance 1 September STFD STEC (TECU) Time (UTC) quiet time until 13 hr ionospheric disturbance from hr 63 Results and discussions STEC (TECU) STEC KMITL STFD PRN2 The differential STEC is computed 1 dstec dstec (TECU) Time (UTC) 64

55 Results and discussions STEC (TECU) 5-5 STEC ~ 2 minutes KMITL STFD PRN29 There is a delay due to the movement of ionospheric disturbance dstec (TECU) dstec Indicate speed 12 km/2 min.= 1 m/s Time (UTC) 65

56 TEC Comparisons with IRI in Asian sector Assistant Professor Dr. Prasert Kenpankho Department of Engineering Education, Faculty of Industrial Education King Mongkut s Institute of Technology Ladkrabang, Bangkok, Thailand prasert.ke@kmitl.ac.th TEC The total electron content (TEC) is an important ionospheric parameter which directly affects the radio wave propagation through the ionosphere. TEC can be derived from the International Reference Ionosphere (IRI) model, called as IRI-TEC. IRI predictions are most accurate in Northern mid-latitudes because of the generally high station density in this part of the world noted by Bilitza and Reinisch (28). IRI215 Workshop 2

57 Asian equatorial latitude For equatorial latitude, there are some experimental observations of the ionospheric measurements used in the IRI model. However, the experimental observation at the eqatorial latitude, Chumphon station, Thailand, is not included in IRI model. IRI215 Workshop 3 Observation Setup Location Latitude Longitude Geomagnetic Latitude Chumphon 1.72 N E 3. Chumphon Magnetic equator IRI215 Workshop 4

58 Observation Setup (Cont.) Ionospheric irregularities 2-3 MHz. FM/CW The ionosondes continuously transmit radio waves from 2 to 3 MHz and receive echoes from the ionosphere to provide the bottom side plasma density profile every 15 minutes Antenna IRI215 Workshop Transmitter/Receiver Computer record 5 Ionogram 8 h F Virtual Height in the F-layer 6 fof2 Critical Frequency in the F2-layer (Ordinary Waves) 4 fxf2 Frequency exceeding critical frequency in the F2-layer (Extra Ordinary Waves) IRI215 Workshop 6

59 GPS Receiver Observation Setup Antenna Amplifier TEC Meter JAVAD Computer Unit IRI215 Workshop 7 GPS receiver h R E Data and Method VTECSTEC =arcsin cos REcos R + h where E where = the zenith angle R E = the mean radian of the Earth = the elevation angle of GPS h = the height of the ionosphere 2 f f STEC - k f f where k = 8.62 (m 3 /s 2 ) ƒ 1 = MHz ƒ 2 = MHz L 1 = the phase of ƒ 1 L 2 = the phase of ƒ 2 1 =.194 m 2 =.2444 m L L IRI215 Workshop 8

60 Data and Method The satellite bias is obtained from NICT based on the GPS Earth Observation Network (GEONET) The receiver bias for a single receiver at Chumpon station is calculated by using the minimum variance method TEC=(STEC - b - b ) cos χ s r where b s = the satellite bias b r = the receiver bias IRI215 Workshop 9 Data and Method IRI TEC from IRI 27 IRI215 Workshop 1

61 Data and Method IRI TEC from IRI-27 IRI215 Workshop 11 Data and Method IGS TEC the two-hour text data of global TEC maps with the different code biases in the IONEX format via the following FTP site: ftp://cddis.gsfc.nasa.gov/pub/gps/products/ionex/. IGS TEC (1N(#32),1E(#57) MAP) IRI215 Workshop 12

62 Results Diurnal variation of the GPS TEC and IRI 27 TEC The largest deviation is evident at 12 LT in 24. For 24 and 25, the large difference is seen at midday. The largest differences between GPS TEC values and IRI-27 TEC values are at the midday time in 24. The GPS TEC values differ from IRI- 27 TEC values at about 15 TECU. The IRI-27 model predicted the TEC data well at pre-sunset hours in 24 and 25. IRI215 Workshop 13 Results Seasonal comparison of GPS TEC, IGS TEC and IRI TEC The 3 seasons including equinox (March, April, September, and October), summer (May, June, July, and August), and winter (January, February, November, and December). The IRI model generally underestimates the observed TEC. The underestimation is more evident at daytime than nighttime, particularly, during the noon bite-out periods. The noon-bite out events are clearly seen on the IRI-27 TEC, but not on the IGS TEC and GPS TEC. The IRI-27 values follow the IGS model in the early morning than any other time of day. The maximum difference between the GPS TEC and the IRI-27 TEC are about 15 TECU during day time and about 5 TECU at the night time. TEC during noontime indicates that the ionosphere at Chumpon or the equatorial area is expanded to cover larger thickness than other times. IRI215 Workshop 14

63 Seasonal Comparison 24 Equinox 25 Equinox TECU Equinox TECU Local time Local time Equinox (March, April, September, and October) At the nighttime, the IRI TEC model generally underestimates the IRI TEC with fof2 option IRI-27 TEC with fof2 measurement IRI-27 TEC Local time The overestimation is more evident at daytime than nighttime IRI215 Workshop 15 Summer TECU Seasonal Comparison Local time Summer Local time Summer TECU Local time IRI215 Workshop Summer (May, June, July, and August) At the nighttime, the IRI TEC model generally underestimates the IRI TEC with fof2 option IRI-27 TEC with fof2 measurement IRI-27 TEC For daytime, the IRI TEC model also generally underestimates the IRI TEC with fof2 option 16

64 Seasonal Comparison Winter TECU Local time Winter Local time Winter TECU Local time IRI215 Workshop Winter (January, February, November, and December). At daytime, IRI TEC model generally underestimates the IRI TEC with fof2 option but overestimates in 26 IRI-27 TEC with fof2 measurement IRI-27 TEC At the nighttime, the IRI TEC model also generally underestimates the IRI TEC with fof2 option during What we found.. At the nighttime, the IRI-27 TEC underestimates the IRI-27 TEC with the fof2 observation (about 5-1 TECU). At the nighttime, the most TEC underestimation is in summer 26 which the IRI-27 TEC underestimates the IRI-27 TEC with the fof2 observation (about 1 TECU). At the noon time, the IRI-27 TEC overestimates the IRI-27 TEC with the fof2 observation (about 5 TECU) in equinox and summer for the year 24 and 26. At the daytime, the most TEC overestimation is in winter 25 which the IRI-27 TEC overestimates the IRI-27 TEC with the fof2 observation (about 7 TECU). In winter during the year 24-25, the IRI-27 TEC underestimates the IRI-27 TEC with the fof2 observation (about 5 TECU). IRI215 Workshop 18

65 Open Problems Investigate TEC by using IRI27 model and IRI212 model. Comparison TEC from IRI models and TEC from observation at Asian equatorial latitudes. IRI TEC or IGS TEC or GPS TEC that is good for Asian equatorial latitudes. AIS-8, Kaliningrad IRI215 Workshop 19

66 AccesstoGNSSdata Prof. Andrzej Krankowski University of Warmia and Mazury in Olsztyn, Poland Space Radio-Diagnostics Research Centre (SRRC/UWM) kand@uwm.edu.pl Outline Agenda GNSS networks as data for the ionosphere monitoring GNSS data generation RINEX data format Compact RINEX. Hatanaka Format. GNSS observation data. Data providers. Total Electron Content (TEC) IONEX access. CDDIS IRI215 Workshop 2

67 GNSS networks as data for the ionosphere monitoring The IGS and other permanent GNSS networks collects, archives, and distributes GPS observation data sets of sufficient accuracy to satisfy the objectives of a wide range of applications and experimentation. The GNSS observations provided by IGS and other permanent station networks, with a 3 s sampling RINEX data. International GNSS Service - IGS IGS polar stations PBO Network Plate Boundary Observatory IGS/EPN POLENET - The Polar Earth (EUREF Permanent Tracking Network) Observing Network GNSS data generation

68 RINEX data format IncludesASCIIfileformatsfor: observation( o ) navigation( n ) meteorological( m ) ionosphericdata( i ) Definedathttp:// Eachfiletypeconsistsofaheadersectionandadatasection Headersectioncontainsglobalinformationfortheentirefileandisplaced atthebeginningofthefile. Containsheaderlabelsincolumns618foreachlinecontainedinthe headersection Theselabelsaremandatoryandmustappearexactlyasperformat description RINEXfilenameconvention: ForsiteSSSS,ondayofyearDDD,sessionTandyearYY: SSSSDDDT.YYo(RINEXobservationfileiethesite sgpsdata) SSSSDDDT.YYn(RINEXnavigationfileiethebroadcastephem) E.g.,hers127.3oisobservationdataforHerstmonceux,day127, session,year23. AllthedatesandtimesinGPST RINEX observation data. Header + Body. 2 OBSERVATION DATA RINEX VERSION / TYPE National GPS Network Ordnance Survey Oct 3 1:25:41 22PGM / RUN BY / DATE Active Station at Ordnance Survey Office Taunton COMMENT TAUN MARKER NAME TAUN MARKER NUMBER National GPS Network Ordnance Survey OBSERVER / AGENCY 8148 LEICA RS REC # / TYPE / VERS 348 LEIAT54 LEIS ANT # / TYPE The following coordinates are NOT APPROXIMATE COMMENT Approx coords replaced by official precise ETRS89 values COMMENT APPROX POSITION XYZ... ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 4 L1 C1 L2 P2 # / TYPES OF OBSERV TIME OF FIRST OBS TIME OF LAST OBS END OF HEADER PRN PRN PRN PRN PRN PRN PRN

69 RINEX observation data. Header + Body. PRN2 PRN3 PRN8 PRN15 PRN17 PRN18 PRN22 2 OBSERVATION DATA RINEX VERSION / TYPE National GPS Network Ordnance Survey Oct 3 1:25:41 22PGM / RUN BY / DATE Active Station at Ordnance Survey Office Taunton COMMENT TAUN MARKER NAME TAUN MARKER NUMBER National GPS Network Ordnance Survey OBSERVER / AGENCY 8148 LEICA RS REC # / TYPE / VERS 348 LEIAT54 LEIS ANT # / TYPE The following coordinates are NOT APPROXIMATE COMMENT Approx coords replaced by official precise ETRS89 values COMMENT APPROX POSITION XYZ... ANTENNA: DELTA H/E/N 1 1 WAVELENGTH FACT L1/2 4 L1 C1 L2 P2 # / TYPES OF OBSERV TIME OF FIRST OBS TIME OF LAST OBS END OF HEADER GNSS observation data. Data providers. The raw GPS data that can be freely downloaded from such data providers: - IGS ftp://cddis.gsfc.nasa.gov; - UNAVCO ftp://data-out.unavco.org/pub/rinex/; - NOAA CORS (ftp://geodesy.noaa.gov/cors); - EUREF ftp://rgpdata.ign.fr; - Natural Resources Canada - RAMSAC CORS of NGI of Argentina - Australian GNSS network ftp://ftp.ga.gov.au; - New Zealand GNSS network ftp://geonet.org.nz Most of the data available via ftp protocol.

70 GNSS observation data. Data structure. International GNSS Service (IGS) Compact RINEX. Hatanaka Format. The "compact" RINEX files, which are generated by a two-step procedure: 1. Actual Hatanaka compression 2. Standard UNIX compression (like zip for Windows systems). This format are nearly 8 times smaller than the original ASCII RINEX files. For the Hatanaka format data (ssssddd.yyd), user can download the Hatanaka Conversion Program "crx2rnx" of Hatanaka to RINEX conversion program from GSI ( All files has a.z extension. RINEX conversion procedure: ssssddd.yyz ssssddd.yyd ssss Station code DDD Day of Year YY - Year ssssddd.yyo

71 GNSS observation data access. CDDIS ftp://cddis.gsfc.nasa.gov/pub/gps/data/daily/215/5/15d/ - Observation files ftp://cddis.gsfc.nasa.gov/pub/gps/data/daily/215/5/15n/ - Navigation files GNSS observation data access. UNAVCO Few types of data assess: - WEB GUI - Java applet - Direct FTP

72 GNSS observation data access. UNAVCO ftp://data-out.unavco.org/pub/rinex/obs/215/5/ - Observation files ftp://data-out.unavco.org/pub/rinex/nav/215/5/ - Navigation files Total Electron Content (TEC) Estimation of absolute Total Electron Content (TEC) using GPS involves two steps: - Levelling the phases to the pseudorange gives the relative TEC - Estimation/removal of instrumental biases (calibration) gives absolute TEC Absolute TEC: TEC = TEC R -( b R + b S ) Relative TEC: TEC R = TEC + <TEC P TEC > ARC b R - Receiver/station bias (TECU) b S - Satellite bias (TECU) TEC P - Differential pseudorange (TECU) TEC - Differential carrier phase (TECU) < > ARC - Average over phase-connected arc TECU TEC P TEC UT (hours)

73 Vertical total electron content estimation RINEX files contains a measurements of dual-frequency signals delays at the frequencies L1=1.6 GHz and L2=1.2 GHz with 3 sec resolution. Dual-frequency radio signals, propagated through the ionosphere, are subject to a differential phase change due to the dispersive nature of the plasma. GPS receiver provides simultaneous measurements of pseudo-range (code) P1 and P2, and carrier phase delays of signals L1(1) and L2(2), which can be written as follows: I c ( t t ) P1 1 s _1 r _ 1 P2 I2 c ( ts _ 2 t r _ 2) L I L 1 1 1N1 2 I2 2N 2 Differential delay of both signals is proportional to the slant TEC: 2 f f f f 2 I TEC M TEC STEC can be transformed into VTEC by use of a mapping function M(): VTEC( rp, t) STEC( PS, PR, t) M ( ) where the zenith angle of the signal path at the piercing point P(r P ) Assuming that the inter-frequency differential delays R and S are known, the geometryfree linear combination can be transformed into VTEC estimation TEC mapping approach For the representation of TEC estimates a spherical harmonics expansion with different degree and order was carried out, as shown in Eq. (1). TEC(, ) 15 n n m P nm (sin)( a nm cos( m) b nm sin( m)) where, are geographic latitude and longitude, Pnm are the normalized associated Legendre functions of degree n and order m; anm and bnm are the unknown SHE coefficients which were derived using GPS TEC observations.

74 TEC mapping approach SHE (spherical harmonics expansion) is one of the most commonly used and approved techniques for TEC mapping. A number of regional TEC maps are based on SHE approach. 1) CODE s Global Ionospheric Maps (GIMs) 2) Australian (IPS) regional TEC maps 3) South American regional ionosphere maps (SAIMs) 4) South African regional TEC maps 5) Regional ionospheric maps over Japan (GEOTEC) TheIONEXformatbody The IONEX (IONosphere interexchange) format allows to store the VTEC and its error estimates in a grid format, in consecutive values at different longitudes- for each latitude grid point. 1 START OF TEC MAP EPOCH OF CURRENT MAP LAT/LON1/LON2/DLON/H LAT/LON1/LON2/DLON/H LAT/LON1/LON2/DLON/H END OF TEC MAP 2 START OF TEC MAP END OF TEC MAP 1 START OF RMS MAP EPOCH OF CURRENT MAP LAT/LON1/LON2/DLON/H END OF RMS MAP END OF FILE

75 VTECmaps:theIONEXformat The IONEX (IONosphere interexchange) format allows to store the VTEC and its error estimates in a grid format, in consecutive values at different longitudes- for each latitude grid point. IONEX header 1. IONOSPHERE MAPS MIX IONEX VERSION / TYPE cmpcmb v1.2 gage/upc 11-may-4 13:1 PGM / RUN BY / DATE ionex file containing IGS COMBINED Ionosphere maps COMMENT global ionosphere maps for day 118, 24 DESCRIPTION IONEX file containing the COMBINED IGS TEC MAPS and DCBs DESCRIPTION EPOCH OF FIRST MAP EPOCH OF LAST MAP 72 INTERVAL 13 # OF MAPS IN FILE COSZ MAPPING FUNCTION. ELEVATION CUTOFF combined TEC calculated as weighted mean of input TEC valuesobservables USED 29 # OF STATIONS 28 # OF SATELLITES BASE RADIUS 2 MAP DIMENSION HGT1 / HGT2 / DHGT LAT1 / LAT2 / DLAT LON1 / LON2 / DLON -1 EXPONENT TEC values in.1 tec units; 9999, if no value available COMMENT DCB values in nanoseconds, reference is Sum_of_SatDCBs = COMMENT DIFFERENTIAL CODE BIASES START OF AUX DATA PRN / BIAS / RMS PRN / BIAS / RMS PRN / BIAS / RMS acor STATION / BIAS / RMS acu STATION / BIAS / RMS... zwen 1233M STATION / BIAS / RMS DIFFERENTIAL CODE BIASES END OF AUX DATA END OF HEADER IONEX access. CDDIS ftp://cddis.gsfc.nasa.gov/gps/products/ionex/ - Ionex files

76 Acknowledgments The author is particularly grateful for the GNSS data provided by IGS/EPN and UNAVCO Thank you for your attention

77