VERTICAL MOTION OF ATMOSPHERE 241 



is particularly interesting to consider whether local centers of convergence 

 and divergence, ill-defined on Daily Weather Reports, can be distin- 

 guished by observations with microbarographs and whether the results 

 are correlated with signal characteristics. Some results of such an investi- 

 gation, forming part of a more general study of VHF propagation in the 

 United Kingdom, are shown in figure 6.9. The signal records obtained 

 at a frequency of 186 Mc/s on a 140 km path are shown for 12 September 

 1959 together with the pressure variations recorded at transmitting and 

 receiving sites. The steady high signal at 0900 hours begins to fall as the 

 pressure at the transmitter begins to decrease, and at about 1030 hours 

 the median level and fading rate change abruptly as the pressure at the 

 midpoint falls. Between 1200 and 1700 hours the pressure over the whole 

 path is falling, and the signal characteristics remain essentially uniform. 

 However, a significant increase in median level and a reduction in fading 

 rate are evident at 1700 hours as the pressure begins to increase at the 

 transmitter. By about 2300 hours the pressure at the center of the path 

 has reached a steady value and the signal level is again high with negligi- 

 ble fading. The total barometric variation between 0900 and 2200 hours 

 is only 2 mbar. 



There is cbviously a fruitful field of radio-meteorological study sug- 

 gested by the work outlined above, but we have to conclude that the 

 results obtained, while extremely valuable in clarifying a qualitative 

 relationship between signal strength and vertical motion (or stability), 

 are not immediately applicable in the problem of predicting field strength 

 variation. 



6.1.7. Discussion of the Parameters 



The above review emphasizes the need for a critical inspection of 

 representative data, radio and meteorological, with the objective of com- 

 paring the merits of the several parameters and explaining, if possible, 

 some of their relative merits and limitations. Some relevant points have 

 been mentioned already, and these issues are developed m more detail in 

 the following discussion. 



It is important to bear in mind that in the particular problem of predict- 

 ing field strength changes we are concerned with a statistical relationship 

 between two quantities, the median signal level and a radiometeoro- 

 logical parameter. It may be advantageous, of course, to investigate 

 the merits of special parameters developed for particular climatic 

 areas; for example, the value of the gradient, dn/dh, in a semipermanent 

 elevated layer in the trade winds area. Nevertheless, such a parameter 

 must also be statistically "reliable" if it is to be applied in prediction 

 analysis. 



