12 RADIO REFRACTIVE INDEX OF AIR 



values being about one-tenth that of other workers, one must recognize 

 that systematic errors undoubtedly contribute to the disagreement in 

 values obtained by different workers. It appears that there exists a real 

 need for a new determination carried out with the greatest possible care 

 and over as large a pressure and temperature range as possible. 



The need for further measurements is particularly evident by the 

 several Essen and Froome determinations. Their values of the dielectric 

 constant have been determined with impressive precision and absolute 

 accuracy over a wide range of frequencies, but their concomitant deter- 

 minations of the constants in the equation for A'^ differ from one another 

 as much as the Smith- Weintraub and the 1951 Essen and Froome values. 

 It is clear that the use of either set of Essen and Froome values permits 

 the calculation of either relative or absolute values of A'' with equal pre- 

 cision. However, the question of comparative absolute accuracy with 

 the results of other workers remains. 



As a further example, drawn from radio geodesy, consider the electrical 

 path length 



R = / ndS, (1.23) 



Jo 



where »S indicates the radio path. If we assume that the measurements 

 are made in a stratum of constant n, then (1.23) becomes 



R = n ds = nS = S{1 -\- N X 10"'), 



Jo 



and 



S = R/{1 -{-NX 10-') c^R{l - N X 10-'). (1.24) 



An error in ^V thus directly produces an error in radio distance determina- 

 tion. In typical radio geodesic field work (see, for example, Wadley, 

 [16]) A^ would be determined from pressure, dry-bulb temperature, and 

 wet-bulb temperature readings. The partial pressure of water vapor may 

 be obtained from Sprung's [19] (psychrometric) formula 



e ^ e's - 0.00067 (T - T') P (1.25) 



where e', denotes the saturation vapor pressure at the wet-bulb tempera- 

 ture, r'. For sea-level conditions of P = 1013 mbar, T = 15 °C and 

 T — T' = 4.1 °C (60 percent RH) one finds, neglecting errors in the 

 constants, that 



AA^' ^ 0.28AP - 4.32A7^ - 6.96Ar. (1.26) 



