Scintillation and TEC Study of the High Latitude Ionosphere Over Casey Station, Antarctica

During the summer season 1997-1998 a GPS Ionospheric Scintillation Monitor (GISM) was installed at Casey station, Antarctica, 66.28° S, 110.24° E, -80.38 invariant latitude. This system is capable of tracking up to 11 GPS satellites at the L1 frequency of 1575.42 MHz. The purpose of the GISM receiver is to automatically record scintillation parameters of amplitude and phase at a 50 Hz rate averaged over 60 seconds. These measurements will be used to map scintillation activity associated with the cusp, auroral oval and other ionospheric features such as patches. In conjunction with scintillation measurements, 1998 Total Electron Content (TEC) data were recorded by the Australian Surveying and Land Information Group (AUSLIG), using the L-band signals from GPS satellites for the Antarctic stations Casey, Davis, Mawson, and the Sub-Antarctic station Macquarie Island. From these data, 3-5 days are being selected for each month for Casey centered on the Regular World Days (RWD). These results are being used to investigate the diurnal and seasonal variations in TEC, and other features such as patch occurrences

1 INTRODUCTION

Time variations in phase, amplitude and angle of arrival of radio waves propagating through the ionospheric medium are known as ionospheric scintillations. Scintillations are due to the diffraction caused by irregularities in motion relative to the ray path. With solar activity due to reach its maximum in the year 2000, scintillation activity is expected to increase accordingly. This paper describes scintillation and TEC measurements being made at Casey station, 66.28° S, 110.24° E, in Antarctica.

1.1 Amplitude Scintillation

Amplitude scintillation is obtained by monitoring the index S₄. The S₄ index is derived from detrended signal intensity of signals received from satellites [1]. The S₄ index, which includes the effects due to ambient noise, is defined as the normalized root mean square of the power P divided by the average power P as follows:

where áñ represents the average values over a 60-second interval. Note: equation 1 is referred to as the Total S₄.

1.2 Removing the Effects of Ambient Noise

The Total S₄ defined in equation 1 has a significant amount of ambient noise associated with it, which needs to be removed before further analysis. This is achieved by estimating the average signal-to-noise density (S/N_o) over a 60-second interval. This estimate is then used to determine the expected S₄ due to ambient noise (also known as S₄ correction) as follows:

Replacing the S/N_o with the 60-second estimate, , gives the S₄ due to noise. Hence, subtraction of equation (1) and (2) yields equation (3), which is the S₄ with the effect of ambient noise removed.

Phase measurements are obtained by monitoring the standard deviation, sDj , and the power spectral density of the detrended carrier phase from signal received satellites. The spectral slope is measured above 1 Hz from detrended carrier and the spectral strength is measured at 1 Hz from detrended carrier.

The detrending of the carrier phase is achieved by passing the raw 50 Hz phase measurements through a sixth-order high-pass Butterworth filter. This removes all the low frequency effects below its cut-off of 0.1 Hz [2].

The sDj ’s are computed over 1, 3, 10, 30 and 60 second intervals every 60 seconds using the 50 Hz detrended phase measurements. The 1, 3, 10, 30 and 60 seconds sDj ’s are further averaged over the 60-second interval.

2 TOTAL ELECTRON CONTENT

The electron density is obtained by counting the number of electrons in a vertical column with a cross-sectional area of 1 m², extending from the GPS satellite to the observer. The electron density thus obtained is termed as the Total Electron Content (TEC) [3], and is mathematically expressed as:

                                                                                                          (4)

where N is the electron density m^-3 and p is the propagation path between the satellite and the detector. A TEC unit is defined as 1 ´ 10¹⁶ electrons m^-2. TEC measurements are generally derived from satellite radio signals observed at various angles, and are normally expressed as an equivalent vertical TEC by dividing the slant TEC by the secant of the elevation angle at a mean ionospheric height (i.e, 350 – 400 km). Here, vertical TEC is used, as it is more easily comparable than slant values at various angles.

3 EXPERIMENTAL SET-UP

Figure 1. This is an illustration of the operational configuration for the GISM. GPSCard^TM PC Series Installation and Operational Manual page 21. Printed in Canada 1995 by NovAtel Ltd.

The GISM was installed at Casey station and it consists of three major components (See figure 1.).

1. GPS antenna (NovAtel’s model 503 GPS antenna) which is mounted onto a platform located on the roof of the Casey science building.
2. GPS receiver (NovAtel’s 3951R GPSCard) installed into computer, which is located in the Casey Atmospheric Space Physics (ASP) lab.
3. GSV-3003 5 or 10 MHz Conversion Board provides an externally generated reference clock signal for the NovAtel GPSCard, generating the 20.473 MHz reference signal required by the GPSCard. The signal is phase locked to either the on board 10 MHz OCXO, or an externally applied 5 or 10 MHz local station frequency.

The receiver (GPSCard) is capable of tracking up to 11 GPS C/A signals at the L1 frequency and measures amplitude and phase at a 50-Hz rate, and code/carrier divergence at a 1-Hz rate for each satellite tracked. Menu-based programs GSV4000.EXE and ISMVIEW.EXE run the system. The first consists of the following sub-menus: Log Selection, allowing the selection of binary data logs to be recorded onto the PC hard disk. The bandwidth of the 6^th order Butterworth filters can be modified for both the phase high pass filter (HPF), and the amplitude low pass filter (LPF) used to detrend the raw phase and amplitude 50-Hz data respectively. This bandwidth can be varied between 0.001 and 5.000 Hz, and is set to 0.1 Hz. For the purposes of this campaign, the channel assignment is so that all 11 channels are operational. The GISM is capable of monitoring the S4, sDjand the power spectral parameters of scintillation activity.

Screen Parameter Selection displays only four parameters on the PC screen. It cannot accommodate any more than that. However, these choices can be changed at anytime. The last menu is the Display mode, which enables the viewing of the data in real time. This option however, was not available for use during this campaign.

The raw data is stored in logging files known as SIN and DIV, and the processed in ISMR. The SIN data log contains the raw 50 Hz rate phase and amplitude measurements and the DIV data log records raw code/carrier divergence (C/No), due to the ionosphere. The average and standard deviation of the C/No are computed every minute. Signal-to-noise ratio (S/No) is also recorded every second.

The ISMR data log collects the reduced raw measurements from SIN and DIV every minute on the minute along with some other useful parameters. After processing the raw data can be saved, however, for this campaign only the processed ISMR data is saved and used.

The ISMVIEW.EXE program permits the extraction of the ISMR data logs for specified satellites into individual files for further processing

4 PRELIMINARY RESULTS

The GISM has been running continuously now for one year. The processed data is collected monthly via the use of the ISMVIEW program. As previously mentioned the receiver records the parameters S_4T S_4No and S₄. A sample of such data is illustrated Figure 2.0 for September 98, day 266, satellite PRN 18. The S₄ index here shows that there still appears to be some noise on the edges of the two passes, possibly due to multipath. There is, however, a prominent peak at the center of the second pass around 18 UT hours, with S₄ measuring at 0.25 which could be indicative of some ionospheric disturbance (i.e, auroral activity or patches). The true source of this increase is yet to be determined. The 60-second sigma data (phase scintillation) for the same satellite also show some scintillation activity around the same time.

The Lock Time of the satellite signal is a useful parameter to have access to, for the simple reason that it allows the Lock Time of the receiver to be observed for each satellite pass. This helps in the initial observation of the noise content of the pass. If the receiver loses lock, or does not maintain lock long enough (i.e, < 240 seconds), then the data can be disregarded. This is the time it takes for the detrending high-pass-filter to re-initialise lock of the carrier phase signal. Looking at the lock time for PRN 18 in Figure 2.0, it can be seen that the receiver remains continuously locked on during this tracking period. Therefore, the increase that is observed around 18 UT must be associated with some sort of disturbance activity.

Figure 3. Plot of the Vertical TEC day 266 and satellite pass for PRN 18. The dotted line in the polar plot represents the location of Casey Station.

The corresponding TEC for this scintillation data is represented in Figure 3.0. It too shows disturbed conditions around 18 UT hours when the Interplanetary Magnetic Field (IMF) is southward, Bz<0. The TEC observed in Figure 3.0 may be related to gravity waves or patch activity [4], [5].

Patches are known as regions of enhanced ionisation, and are observed to drift antisunward across the polar cap when the IMF is directed southward (Bz<0). Not only do they occur when the IMF Bz component is negative, but also during moderately disturbed conditions (Kp greater than 4) [6], and during quiet periods [7]. The seasonal occurrence of patches has been observed to have a peak occurrence in April with a pronounced minimum in July, a Maximum in August and minimum occurrence near mid-summer (December-January) [8]. Patches of smaller size and steep density gradients at their edges are associated with intense scintillation and interfere with transionospheric radio signals. Based on previous studies increases in ionisation are defined to be at least 3 TECU or more.

During late August 1998 magnetic storm conditions were observed for days 238 and 239. The Kp and Disturbance Storm Time (DST) are illustrated in Figure 4.0. A steady decrease in the DST was observed towards the end of day 238, reaching a maximum decrease of –188 around 09 UT on the following day, with a corresponding Kp value of 7.7. The maximum Kp observed for day 239 was 8.0. Figure 5.0 (day 238) shows a plot of vertical TEC for PRN 14 along with the satellite passes in both UT and MLT to show there positions relative to both the geomagnetic and magnetic poles. This particular sample of TEC data was taken at the beginning of the storm activity and clearly shows an enhancement in the ionisation at 17 UT of 10 TECU with a Kp value of 5.7, possibly associated with a patch occurrence. The Kp values continue to rise during day 239 with numerous TEC enhancements observed as shown in Figure 6.0 for PRN 6. A noticeable increase can be seen during 07-10 UT with a peak in TEC of 6 TECU at 09 UT. IMF data is now needed to determine if these observed increases are associated with patches. Unfortunately, IMF was not available for these days.

With the approaching solar maximum, scintillation effects on GPS are becoming more frequent, with GPS receivers losing lock on satellites. At this stage however, scintillation levels observed have been no higher than 0.25 for the high latitude region. So far the receiver has performed well in keeping locked on to a tracked satellite, although there have been a few cases where it loses lock at the beginning and end of a satellite pass at low elevation angles. The cause of the scintillation observed in Figure 2.0 is still to be determined. The magnetic storm occurrence for August shows some increase in the ionisation over the two-day period. These enhancements have been observed to range anywhere form 6 – 10 TECU with Kp values ranging from 5.7 to 8.0. Further analysis is still needed to verify the cause of their high latitude enhancements in ionisation.

ACKNOWLEDGEMENTS

Thanks to AUSLIG for supplying the GPS TEC data, and a special thanks to the Australian Antarctic Divisions wintering physicist for the continual operation of the GISM. This research is being funded by an ASAC (Antarctic Science Advisory Committee) grant with support from the Australian Antarctic Division and the Cooperative Research Center for Satellite Systems (CRCSS). An APA (Australian Postgraduate Award) scholarship and the CRCSS support N.M. Shilo.

REFERENCES

[1] A. J. Van Dierendonck, John Klobuchar and Quyen Hua, Ionospheric Scintillation Monitoring Using Commercial Single Frequency C/A Code Receivers, Proceedings of ION GPS-93, Salt Lake City, UT, September 1993

[2] A. J. Van Dierendonck, Pat Fenton and John Klobuchar, Commercial Ionospheric Scintillation Monitoring Receiver Development and Test Results, Proceedings of ION 52^nd Annual Meeting. Cambridge, MA., June 19-21, 1996.

[3] J. A. Klobuchar, Ionospheric effects on GPS, GPS World, April 1991.

[4] H.M. Beggs, E.A. Essex and D. Rasch, Antarctic polar cap total electron content observations from Casey station, Journal of Atmospheric and Terrestrial Physics, Vol. 56, No. 5. pp 659-666, 1994.

[5] Huang, C.S., D.A. Andre and G.J. Sofko, Observations of Solar Wind Directly Driven Auroral Electrojets and Gravity Waves, Journal of Geophys. Res, Vol. 103.(A10), pp 23,347-23,356, 1998

[6] E.J. Weber, J. Buchau, J.G Moore, J.R. Sharber, R.C. Livingston, J.D., Winningham, and B.W. Reinisch, F Layer Ionisation Patches in the Polar Cap, Journal of Geophys. Res., 89, 1683-1694, March 1, 1984.

[7] S. Basu, E.J. Weber, and G.J. Bishop, Plasma structuring in the polar cap, J. Geomagn. Geoelectr., 42, 763-776, 1990

[8] Tate. B.S, Shilo. N.M.S, Essex. E.A., An Investigation of the Ionosphere in the Southern High Latitude during Low Sunspot Numbers., Acta Geod. Geoph. Hung., Vol. 33(1), pp. (1998) 83-90.