Improved short-term predictions of ƒ0F2 using GPS time delay measurements


  • Z. Houminer,

  • H. Soicher


Reliable HF communications along short-, medium- and long-range paths require propagation assessment. Such assessment could be facilitated with the monitoring of ionospheric characteristics by continuously available passive means, i.e., measurement of the total electron content (TEC) using satellite-emitted signals without a need for burdening the electromagnetic spectrum. With ubiquitous Global Positioning System (GPS) providing instantaneous time delay, or equivalently, TEC, values when needed, an assessment of HF propagation conditions may be available on a near-real-time basis. Both TEC and the peak electron density of the ionosphere, which determines the ordinary upper frequency limit (ƒ0F2( for HF sky wave vertical propagation, vary strongly with solar and geomagnetic parameters. Their ratio, the equivalent slab thickness, may vary to a lesser degree and hence be modeled with greater accuracy. A slab thickness model combined with real-time TEC measurement anywhere on the globe may possibly yield an improved HF parameter prediction algorithm. To test the efficacy of the hypothesis, one has to ascertain the correlation, as exhibited by the correlation coefficient, between the TEC daily variability about the monthly mean and the ƒ0F2 variability. To determine such correlation, a study compared Faraday TEC data as well as GPS-generated TEC data collected in Israel and with corresponding ƒ0F2 values obtained from vertical sounder measurements near the appropriate subionospheric location in Cyprus. The analysis shows that for large percentages of the time, very good correlation exists between TEC and ƒ0F2 short-term variations. The correlation coefficient varies between 0.7 or better during winter and summer months to about 0.5–0.6 during equinox months. A study of the diurnal dependence of the correlation indicates that a better correlation exists during daytime than nighttime. There was no indication that the coefficient is dependent on geomagnetic activity or on protonospheric electron content during the period of this study.