[57] It should be noted that the calculated phase velocity (about 1000 m s-1 ) for the height hs=hd is close to the velocity determined in the approximation of the plain front and to the speed of sound at heights of the electron concentration maximum in the ionosphere. (Figure 6) [Li et al., 1994].
[58] It is worth comparing the results of the spatial-time processing in the approximation of the spherical front (Tables 3 and 4) to the estimates of the mean velocity of the SAW propagation Va determined from the delay Dt=t ext-t0 and known length of the path between the earthquake epicenter and subionospheric point. The path length is determined along the great circle arc. The velocity value obtained in such a way varies within 352-483 m s-1 and 477-907 m s-1 for ID on 25 September 2003 and 4 June 2000, respectively. On the average it is by a factor of 1.5 less than the value of Vr in Tables 3 and 4.
[59] The difference is easily explained by the fact that the path length L is determined along the Earth surface whereas the detecting is conducted in the vicinity of the main electron concentration maximum at a height of 400 km, making the path longer in our case by a factor of 1.5.
[60] Thus the method of determination of the phase velocity Va used earlier [Calais and Minster, 1995] was giving wrong (underestimated) evaluations of the disturbance propagation velocity. This method did not take into account rather complicated propagation mechanism discussed in this paper. In this mechanism at the first stage the acoustic disturbance propagates within a narrow cone of the zenith angles up to ionospheric heights and then in the form of a spherical wave diverges with the radial velocity close to the speed of sound at these heights.
[61] In this section the values of the phase velocity obtained of about 1000 m s-1 agree to the data of other experiments on determination of the ID velocity during earthquakes [Afraimovich et al., 2001a, 2001b, 2002a, 2002b]. These data are compiled into Table 5.
[62] Thus it is found that the ionospheric disturbances generated in the moment of the main shock during the 4 June 2000 and 25 September 2003 earthquakes have a form of a spherical wave diverging with a velocity of about 1000 m s-1 from the "secondary source" located over the epicenter at the level of the ionospheric F2 layer (300-400 km), the "switching on" of the source being delayed relative the moment of the main seismic shock by ~10 min. The results agree with theoretical models according to which the atmospheric disturbance propagates within a narrow cone of zenith angles up to ionospheric heights and then diverges in the form of a spherical wave with the radial velocity close to the speed of sound at these heights.
Powered by TeXWeb (Win32, v.2.0).