development of tec fluctuations in antarctic ionosphere...

A C T A G E O P H Y S I C A P O L O N I C A

Vol. 53, no. 2, pp. 205-218 2005

STUDY OF TEC FLUCTUATIONS IN ANTARCTIC IONOSPHERE DURING STORM USING GPS OBSERVATIONS

Andrzej KRANKOWSKI1, Irk I. SHAGIMURATOV2, Lubomir W. BARAN1 and Ivan I. EPHISHOV2

1Institute of Geodesy, University of Warmia and Mazury ul. Oczapowskiego 1, 10-957 Olsztyn, Poland

e-mail: [email protected] WD IZMIRAN,

Prospekt Pobedy 41, 236017 Kaliningrad, Russia e-mail: [email protected]

A b s t r a c t

GPS observations carried out at Antarctic stations belonging to the IGS net-work were used to study TEC fluctuations in the high-latitude ionosphere during storms. Dual-frequency GPS phase measurements along individual satellite passes served as raw data. Ionospheric irregularities of a different scale develop in the auroral and polar ionosphere. This is a common phenomenon which causes phase fluctuations of GPS signals. We distinguished TEC variations related to ionospheric structures of a spatial scale bigger than 200-300 km. In the diagram of temporal variations of TEC along satellite passes, the structure of TEC corresponds to a time scale longer than 15-30 min. We attribute the variations in a time scale smaller than 15-30 min to TEC fluctuations related to small-scale ionospheric irregularities. We used the rate of TEC index (ROTI) expressed in TECU/min as a measure of TEC fluctuations. Large-scale ionospheric structures cause an increase in horizontal gra-dients and difficulties with the carrier phase ambiguity in relative GPS positioning. In turn, the phase fluctuations can cause cycle slips. At polar stations MCM4, CAS1, DAV1 we detected ionospheric structures of TEC enhanced 3-5 times rela-tive to the background, whereas TEC increased to 10-30 TECU in about 10-15 min. The structures were observed during a storm, as well as during moderate geomag-netic activity. The structures can be probably attributed to polar cap patches.

mailto:[email protected]

A. KRANKOWSKI et al. 206

During storms the intensity of phase fluctuations increased. The occurrence of phase fluctuations was even detected during the active storm period of 31 March 2001 at a middle-latitude station OHIG (located at 49° corrected geomagnetic lati-tude).

Key words: ionosphere, total electron content, ionospheric storms.

1. INTRODUCTION

The structure of the high-latitude ionosphere is very complicated and varied. Strong changes in the ionosphere occur during geomagnetic disturbances. Dramatic changes take place very frequently in the auroral and polar ionosphere. In this region, irregu-larities at a different scale are common, which causes fluctuations in the total electron content. In the paper we distinguished two types of total electron content (TEC) fluc-tuations. In the first type, large-scale fluctuations (LSF) of TEC are caused by iono-spheric irregularities whose scale is bigger than 100-300 km. These ionospheric struc-tures occur as deep spatial variations of TEC. Fluctuations of the other type are irregu-larities whose size is about ten kilometers, causing phase fluctuations of GPS signals. Small irregularities can co-exist with large-scale structures.

Several studies have used GPS observations from a single site or local network to monitor TEC fluctuations and related irregularities in the high latitude ionosphere (Coker et al., 1995; Aarons, 1997; Aarons et al., 2000) and equatorial regions (Kelly et al., 1996; Beach and Kintner, 1999; Mendillo et al., 2000). Correlation between amplitude scintillations and TEC fluctuations was analyzed (Basu et al., 1999, and Bhattacharyya et al., 2000). The possibility of using the International GPS Service (IGS) to monitor global effects in the ionosphere was presented by Pi et al. (1997). These studies concerned irregularities whose scale was below 100 km.

In this paper we present an analysis of the development of TEC fluctuations in March 2001. Two great geomagnetic storms took place on 20 and 31 March. The storm of March 31 was one of the severest in the last decade. The Kp index reached a

Fig. 1. Kp variations during March 2001.

DEVELOPMENT OF TEC FLUCTUATIONS IN ANTARCTIC IONOSPHERE

207

maximal value of 9 and ΣKp made up 60. The Dst index reached maximum magni-tudes with an extremely high value of about –360 nT. The geomagnetic conditions during March 2001 are presented in Fig. 1 as variations in the Kp index.

2. TEC ESTIMATION TECHNIQUE

The absolute TEC and instrumental biases were estimated using a single site algorithm (Baran et al., 1997). To determine the ionospheric TEC, a geometry-free linear com-bination of GPS observables was used. In most cases, the total vertical electron con-tent (VTEC) is modeled. The vertical TEC is approximated by a spherical layer with infinitesimal thickness. The height of this layer, approximately 350 to 400 km, corre-sponds to the average altitudinal position of the electron density profile peak, hm. The value of TEC is a function of terrestrial longitude and latitude.

The slant TEC and vertical TEC with first-order approximation can be described by the following geometrical factor:

cosVTEC TEC z′= ⋅ (1)

where is the zenith angle of a satellite at a height hm (Fig. 2). For the zenith angles less than 80°, the error of approximation is small and vanishes at the zenith.

z′

From Fig. 2 the relation

sin sinE

E m

RzR h

′ =+

z (2)

can be obtained, where RE is the mean radius of the Earth, hm is the height of the ionospheric layer, z΄ and z are the zenith angles at the iono-spheric point IP and at the observation site, re-spectively.

The relationship between the ionospheric de-lay and TEC, and the difference between the dual-frequency code P and phase Φ measurements may be written as

cos P PTECP M A

z∆ ε= + +

′ , (3)

cosTECM A

z Φ Φ∆Φ ε= + +′

. (4)

Here TEC is the vertical electron content, M is a scale factor, Pε , Φε are the noise terms, AP , AΦ

Fig. 2. Geometry of the TEC esti-mation.


are the equipment biases (AΦ contains phase ambiguity), z is the zenith angle of the ray at the sub-ionospheric point.

The algorithms of the bias estimation are discussed by Lanyi and Roth (1988), Ciraolo (1993), Wanninger et al. (1994), and Georgiadiou (1994). The biases were de-termined for every individual station using GPS measurements for all satellite passes over a given station during a 24-hour period. The absolute TEC for all satellite passes observed over a single station during 24 hours is calculated using this procedure.

3. TEC DATA BASE

GPS observations carried out at Antarctic IGS stations were used to study the devel-opment of TEC fluctuations in the high-latitude ionosphere. Standard GPS measure-ments with 30 s sampling provide the detection of irregularities whose size is bigger than 6-10 km. Table 1 presents the geographic and corrected geomagnetic coordinates of the stations. The broad longitudinal area of Antarctic stations enables us to study the time development of TEC fluctuations.

The dynamics of the high-latitude ionosphere is controlled by the geomagnetic field. Table 1 shows that geomagnetic latitudes are essentially different from geo-graphic ones. Thus, in the Antarctic region we can choose among middle-latitude (OHIG), auroral (SYOG, MAW1) and polar (MCM4, CAS1, DAV1) stations.

Table 1 Antarctic International GPS Service stations

Geographic Corrected geomagnetic IGS stations

latitude longitude latitude longitude MLT

midnight

O’Higgins (OHIG) –63.32 -57.90 -48.93 12.23 03h47m Casey (CAS1) –66.28 110.52 -80.66 159.10 18h19m McMurdo (MCM4) –77.84 166.67 –79.95 325.00 07h07m Sanae (VESL) –71.67 –2.84 –61.67 43.56 01h46m Syowa (SYOG) –69.01 39.58 –66.65 72.51 23h55m Mason (MAW1) –67.60 62.87 –70.68 91.49 22h39m Davis (DAV1) –68.58 77.97 –74.91 101.92 21h56m

4. LARGE SCALE FLUCTUATION OF TEC

For the analysis of spatial and temporal changes in TEC during a storm we used high-precision dual-frequency GPS phase measurements that provide more precise meas-urements of TEC than the group-delay ones.


209

Spatial and temporal variations in TEC are clearly seen in the time variations of TEC along individual satellite passes. Figures 3a and b give an example of TEC varia-tions for individual satellite passes, as observed at different stations during a storm (19 and 31 March, respectively) and a quiet period (17 and 26 March, respectively). The vertical TEC (in units of 1016 el/m2) is plotted as a function of universal time (UT). This represents a part of the diurnal TEC pattern sampled by satellites at these times. These figures show also satellite ionospheric tracks at an altitude of 450 km, in geo-graphic coordinates (cross line). Because the GPS satellites are on 12 sidereal hour or-bits, the tracks repeat day by day (the only difference is that the satellites arrive 4 min earlier each day). The plot approximately indicates the satellites ionospheric trace for two days under consideration.

The series of large-scale fluctuations (LSF) as enhancement of TEC are clearly shown as temporal patterns (Fig. 3a and Fig. 3b). At the polar stations (CAS1, MCM4, DAW1) TEC patterns demonstrate great variability both during disturbed and quiet days. During a storm the intensity of the fluctuations increases dramatically. The TEC increase can exceed the factor of 2-8, and the enhancement of TEC can exceed 10-20 TEC units to relative phone. At lower latitudes the intensity of LSF decreases. We at-tribute the TEC enhancement to the occurrence of polar cap ionospheric patches. Polar cap patches are large regions of enhanced F region plasma density. They were ob-served to travel through ionospheric polar caps, under the influence of high-latitude convection (Pederson et al., 2000). Discrete F region electron density is enhanced by the factor of 2 or more. Patches are typically considered to be of the order of 100-1000 km in horizontal extent. The traveling speed of the patch is between 300-900 ms-1 (Rodger and Rosenberg, 1999). Thus, in the temporary pattern showing the variations in TEC along satellite passes, the duration of patch occurrence can be 10 min or more.

The patterns of TEC fluctuation demonstrate similar structures at spaced stations. It is well seen in Fig. 3a for DAV1 and MAW1. Very similar patch structures show the TEC variation for PRN 15 and PRN 17 at CAS1 stations (top panel of Fig. 3a) for March 2001.

Deep variations in TEC are observed very frequently at polar stations. Analyses of data for MCM4 stations show that patch-like structures (about 90% cases) were registered during March-April 2001. Over an auroral station (VESL), while on a quiet day TEC demonstrated a smooth run, during a storm LSF were often observed. The amplitudes of TEC fluctuations in this region were smaller than at polar stations. At a middle-latitude station (OHIG) large structures could be detected during a storm, which we attribute to the occurrence of the main ionospheric trough. At that time, the horizontal gradients in the ionosphere over the OHIG station increased. The storm time gradients can even be opposite to quiet geomagnetic conditions (Fig. 3b).


Fig. 3a. TEC variations for satellites observed at different stations (CAS1, DAV1, MAW1,VESL) during storm (19 March – dashed line) and quiet period (17 March – solid line).


211

Fig. 3b. TEC variations for satellites observed at different stations (CAS1, DAV1, MCM4,OHIG) during storm (31 March – dashed line) and quiet period (26 March – solid line).


5. PHASE FLUCTUATIONS OF GPS SIGNALS

TEC fluctuations, also called phase fluctuations, are caused by the presence of me-dium- and small-scale irregularities in the ionosphere. Dual-frequency phase meas-urements with a 30 s interval are usually used to estimate phase fluctuations. In our study, TEC data at a rate of change of 1 min, ROT, were used to estimate phase fluc-tuations. The use of these relatively infrequent samples enables to study irregularity structures of the order of kilometers. When using ROT we avoid the problem of phase ambiguities.

As a measure of ionospheric activity we also used the rate of TEC index (ROTI) based on standard deviations of ROT (Pi et al., 1997):

22 ROTROTROTI −= . (5)

ROTI was estimated at a 10-min interval. Figure 4 demonstrates the occurrence of phase fluctuations (using ROT) on 29-31

March 2001, with the most disturbed day – 31 March. The plot shows variations in raw phase fluctuations for all passes of satellites observed at MAW1, CAS1 and MCM4 over a 24-hour interval during quiet and disturbed days. The top panel demon-strates the behavior of the geomagnetic field at the Mawson station. In the pictures, the mark points out to the location of the geomagnetic midnight. During the storm the in-tensity of fluctuations strongly increased, relative to quiet conditions. The diurnal variations in phase fluctuations reached a minimum about 10-14 UT at the MAW1 sta-tion. At the MAW1 station, located on the equator edge polar cap (Φ = 70°), the oc-currence of phase fluctuations very well correlates with geomagnetic activity.

At the polar stations CAS1 and MCM4 the picture of development of phase fluc-tuation is more complex. During the disturbed day of 31 March, the occurrence of phase fluctuations at the stations under consideration is similar, so the development is controlled by UT. However, sometimes the occurrence of fluctuations at CAS1 and MCM4 is different. It is well visible for the quiet day of 30 March. The diurnal occur-rences of phase fluctuations at the CAS1 and MCM4 stations on 29 and 30 March were not alike. It can be explained by a higher intensity of phase fluctuations, ob-served simultaneously at all satellite passes.

At polar stations, phase fluctuations were observed all day. At a middle-latitude station (OHIG) fluctuations occurred only during the storm of 31 March. It is evident that on 31 March the auroral oval of irregularities expanded until middle latitudes.

Figures 5a, 5b and 5c show the latitudinal occurrence of phase fluctuations over the Antarctic region on quiet and disturbed days. The intensity of fluctuations is de-noted by different symbols. Figure 5a illustrates the occurrence of phase fluctuations over a polar (CAS1) and auroral (MAV1) station. As shown by Aarons et al. (2000), the dominant factor in the development of phase fluctuations during quiet periods is the location of a station relative to the auroral oval. During a quiet day of 26 March


213

0 6 12 18 24

6800

7200

7600

8000

H, n

T

0 6 12 18 24 0 6 12 18 24

0

5

10

15

20

25

30

N s

atel

lite

0 6 12 18 24 0 6 12 18 24 0 6 12 18 24

0

5

10

15

20

25

30

N s

atel

lite

0

5

10

15

20

25

30

N s

atel

lite

6800

7200

7600

8000

0

5

10

15

20

25

30

0

5

10

15

20

25

30

0

5

10

15

20

25

30

March 29, 2001, Kp=28Mawson (-70.7°, 91.5°)

March 30, 2001, Kp=21Mawson (-70.7°, 91.5°)

March 31, 2001, Kp=61Mawson (-70.7°, 91.5°)

MAW1(-70.7°, 91.5°), March 29, 2001 Mawson (-70.7°, 91.5°), March 30, 2001 Mawson (-70.7°, 91.5°), March 31, 2001

CAS1 (-80.7°, 159.1°), March 29, 2001 CAS1 (-80.7°, 159.1°), March 30, 2001 CAS1 (-80.7°, 159.1°), March 31, 2001

MCM4 (-80.0°, 325.0°), March 29, 2001 MCM4 (-80.0°, 325.0°), March 30, 2001 MCM4 (-80.0°, 325.0°), March 31, 2001

Fig. 4. The magnetic activity at Mawson and the phase fluctuations occurrence at different sta-tions on March 29-31. The plots show the phase fluctuations for individual satellites, the markpoints out location of geomagnetic midnight.


6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

CAS126.03.2001

MAW126.03.2001

MAW130.03.2001

CAS130.03.2001

MAW131.03.2001

CAS131.03.2001

Fig. 5a. Location of TEC fluctuations derived from GPS measurements in geomagnetic localtime and corrected geomagnetic latitude for 26.03.2001, 30.03.2001 and 31.03.2001 at differ-ent stations in the south hemisphere. The intensity of fluctuations is indicated with the follow-ing symbols: crosses represent fluctuations between 0.3 and 0.5 TEC/min, rhombs – biggerthan 1.5 TEC/min.


215

6

12

18

0

80 60

6

12

18

0

80 60 40

6

12

18

0

80 60

6

12

18

0

80 60 40

6

12

18

0

80 60

6

12

18

0

80 60 40

OHIG26.03.2001

MCM426.03.2001

OHIG30.03.2001

MCM430.03.2001

OHIG31.03.2001

MCM431.03.2001

Fig. 5b. Location of TEC fluctuations derived from GPS measurements in geomagnetic localtime and corrected geomagnetic latitude for 26.03.2001, 30.03.2001 and 31.03.2001 at differ-ent stations in the south hemisphere. The intensity of fluctuations is indicated with followingsymbols: crosses represent fluctuations between 0.3 and 0.5 TEC/min, rhombs – bigger than1.5 TEC/min.


6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

6

12

18

0

80 60

SYOG26.03.2001

DAV126.03.2001

DAV130.03.2001

SYOG30.03.2001

SYOG31.03.2001

DAV131.03.2001

Fig. 5c. Location of TEC fluctuations derived from GPS measurements in geomagnetic localtime and corrected geomagnetic latitude for 26.03.2001, 30.03.2001 and 31.03.2001 at differ-ent stations in the south hemisphere. The intensity of fluctuations is indicated with followingsymbols: crosses represent fluctuations between 0.3 and 0.5 TEC/min, rhombs – bigger than1.5 TEC/min.


217

the maximal intensity was observed around the magnetic local midnight. Geomagnetic storms modify the diurnal patterns of phase fluctuations, extending the time of devel-opment and increasing their intensity. At the polar station CAS1, the intensity during a storm increased more than at the auroral station MAV1. The development of fluctua-tions over the DAV1 station was more clearly controlled by geomagnetic activity. Figure 5b illustrates the occurrence of phase fluctuation over a middle-latitude (OHIG) and polar (MCM4) station. During quiet and moderate magnetic activity, phase fluctuations were very weak at OHIG; more intensive fluctuations occurred only on the most disturbed day, i.e., 31 March. They appeared during the greatest storm when the oval of irregularities developed until middle latitudes.

Figure 5c demonstrates the occurrence of phase fluctuations at the lower (SYOG) and higher latitude edge of the auroral oval. The intensity of fluctuations was essen-tially lower over SYOG than over DAV1. During a stormy day the fluctuations were registered all the time, and their intensity also markedly increased in the storm time.

6. CONCLUSIONS

The occurrence of TEC fluctuations depends on the geomagnetic latitude of the site. In the Antarctic region the difference between geomagnetic and geographic coordinates of the site may be bigger than 10 degrees. Hence, midlatitude, subauroral, auroral and polar stations can appear there.

Maximal TEC fluctuations were recorded at polar stations. During a storm the variations in TEC reached 10-40 TECU. The enhancement of TEC exceeded 2-8 times the relative phone. Deep variations of TEC observed along individual satellite passes are related to polar patches.

At lower latitudes, fluctuations of GPS signals are attributed to small- and mid-dle-scale irregularities. The intensity of phase fluctuations depends on geomagnetic activity. During the maximal phase of the storm of 31 March 2001, fluctuations of moderate intensity were observed at a middle-latitude station (OHIG). The develop-ment of TEC strongly correlated with the geomagnetic field variations at the Mawson station. The ionospheric gradients increased essentially during the storm. The irregular gradients sometimes exceeded the regular ones. During the storm, this can be the cause of major errors while determining phase ambiguities of GPS observations in the Antarctic region.

R e f e r e n c e s

Aarons, J., 1997, GPS system phase fluctuations at auroral latitudes, J. Geophys. Res. 102, A8, 17219-17231.

Aarons, J., B. Lin, M. Mendillo, K. Liou and M. Codrescu, 2000, Global Positioning System phase fluctuations and ultraviolet images from the Polar satellite, J. Geophys. Res. 105, A3, 5201-5213.


Baran, L.W., I.I. Shagimuratov and N.J. Tepenitsina, 1997, The use of GPS for ionospheric studies, Artificial Satellites 32, 1, 49-60.

Basu, S., K.M. Groves, J.M. Quinn and P. Doherty, 1999, A comparison of TEC fluctuations and scintillations at Ascension Island, J. Atmos. Solar-Terres. Phys. 61, 1219-1226.

Beach, T.L., P.M. Kintner, P.M., 1999, Simultaneous Global Positioning System observations of equatorial scintillations and total electron content fluctuations, J. Geophys. Res. 104, A10, 22,553-22,565.

Bhattacharyya, A., T.L. Beach, S. Basu and P.M. Kintner, 2000, Nighttime equatorial iono-sphere: GPS scintillations and differential carrier phase fluctuations, Radio Science 35, 1, 209-224.

Ciraolo, L., 1993, Evaluation of GPS L2-L1 biases and related daily TEC profiles, Paper pre-sented at the workshop “Modelling of the Ionosphere for GPS Application”, Neuster-litz, September 20-30, 1993.

Coker, C., R. Hunsucker and G. Lott, 1995, Detection of auroral activity using GPS satellites, Geophys. Res. Lett. 22, 23, 3259-3262.

Georgiadiou, J., 1994, Modeling the ionosphere for an active control networks of GPS stations, LGP Series, Publs. Delft Geod. Comp. Centre 7.

Kelley, M.C., D. Kotsikopoulos, T.L. Beach and D. Hysell, 1996, Simultaneous GPS and ra-dar observations for equatorial spread F at Kwajalein, J. Geophys. Res. 101, A2, 2333-2341.

Lanyi, G.E., and T. Roth, 1988, A comparison of mapped and measured total ionospheric elec-tron content using global positioning system and beacon satellite observations, Radio Science 23, 4, 483-492.

Mendillo, M., B. Lin and J. Aarons, 2000, The application of GPS observation to equatorial aeronomy, Radio Science 35, 3, 885-904.

Pedersen, T.R., B.G. Fejer, R.A. Doe and E.J. Weber, 2000, An incoherent scatter radar tech-nique for determining two-dimensional horizontal ionization structure in polar cap F region patches, J. Geophys. Res. 105, A5, 10637-10655.

Pi, X., A.J. Mannucci, U.J. Lindqwister and C.M. Ho, 1997, Monitoring of global ionospheric irregularities using the worldwide GPS network, Geophys. Res. Lett. 24, 18, 2283-2286.

Rodger, A.S., and T.J. Rosenberg, 1999, Riometer and HF radar signatures of polar patches, Radio Science 34, 2, 501-508.

Wanninger, L., E. Sardon and R. Warnant, 1994, Determination of the total ionospheric elec-tron content with GPS-difficulties and their solution, Proceedings of the Beacon Satel-lite Symposium, 11-15 July 1994, Aberystwyth, Wales, UK.

Received 14 September 2004 Accepted 31 January 2005

development of tec fluctuations in antarctic ionosphere...

Documents