244 TRANSHORIZON PARAMETERS 



diverge most at the present time. Unfortunately, adequate radio data 

 are not available for several areas of interest (e.g., equatorial Africa) and 

 in these cases we can only propose tentative explanations based on a 

 critical examination of existing results, for similar but not identical con- 

 ditions. 



An examination of available data, in published and unpublished reports, 

 has shown two particular examples which deserve further study; these 

 are (1) climatic conditions in which stable elevated layers are persistent 

 during certain seasons of the year, and (2) equable climates in which the 

 annual range of N s, AN, and field strength is relatively low. Attention 

 has already been drawn to an example in the first category; namely, the 

 path from San Diego to Santa Ana, Calif. [2]. The well-defined coastal 

 inversion in this area occurs at a height of about 0.7 km, and the asso- 

 ciated stable layer has a strong influence on radio field strength. Here 

 the correlation between field strength and A^ gradient for heights up to 

 0.7 km above the surface is small and negative, (i.e., opposite to the general 

 trend). The correlations with N s, however, and with N differences to 

 heights above the base of the inversion, are about 0.8. This result sug- 

 gests an explanation of some of the results, already mentioned, for Dakar 

 and Leopoldville. Consider the profile of figure 6.10 typical of ascents 

 made through daytime inversions in equatorial and Mediterranean areas. 

 (For comparison purposes, an exponential reference atmosphere is also 

 shown.) 



At Dakar, for example, an elevated layer is observed in 40 to 50 percent 

 of the daytime soundings during August, frequently with the base of the 

 inversion above 1 km height. In these conditions, at least, the high N s 

 values measured in August are not accompanied by high values of AN 

 (0 to 1 km), for the lapse rate, dN/dh, below the inversion is generally less 

 steep than would be expected for the given value of N s- It would be 

 valuable, therefore, to examine in more detail the distribution of the 

 height of the layer, particularly in the summer months. If the data given 

 by Misme contain an appreciable fraction of such profiles, then the poor 

 Ns — AN correlation he discusses is not surprising. The correlation be- 

 tween N s and signal level, however, could still be significant, as in the 

 San Diego-Santa Ana link [2]. Further radio data are obviously required 

 for areas in which elevated inversions are persistent. 



In many temperate regions, a somewhat different situation exists. In 

 western Europe, for example, the annual range of monthly median values 

 of A" s is 10 to 20 N units; similarly, the variation in the monthly median 

 values of field strength is also small and frequently lies within the esti- 

 mated measurement error. Figure 6.11 shows some results which illus- 

 trate these features. Monthly median values of relative field strength 

 are shown for the months January-November, 1959, for a 300-km path at 

 a frequency of 174 Mc/s; the terminals being located at Lille (northern 



