324 REFRACTION AND REFRACTIVE INDEX MODELS 



Plain. Aircraft were equipped with a wet-bulb and dry-bulb psychrom- 

 eter, mounted on the })ortside above the wing. Readings were taken 

 three or four times on each horizontal flight leg of 2 or 3 min duration. 

 Special lag and airspeed corrections were applied, resulting in accuracy 

 of ±0.1 °C. It was found that, under the variety of conditions in which 

 observations were made, the aircraft flights were more or less parallel to 

 the surface isobars; hence, the sea-level pressure as recorded at the beach 

 site was considered to hold over the whole track covered by the aircraft. 

 The relationship used for calculating the pressure, P, in millibars at a 

 height h in feet was: 



P(h) = Po - h/30 



where Po is the surface pressure. This approximation (determined by 

 averaging the effect of the temperature and humidity distributions on 

 pressure in a column of air) resulted in a maximum error in the refrac- 

 tivity of 0.5 percent at 900 m. Radiosonde ascents at Hokitika on the 

 west coast of South Island and Paraparaumuo and Auckland on North 

 Island were used to supplement the aircraft measurements, particularly 

 in the altitude levels above 1 km. 



The observations, diagrams, and meteorological records were studied, 

 and a profile of unusually heterogeneous nature was chosen. The 

 synoptic situation for the morning of November 5, 1947, was selected, as 

 it revealed a surface-ducting gradient near the coast with an elevated layer 

 about 100 km off shore. A cross section of the area from Ashburton to a 

 point 200 km offshore was plotted with all available data, and isopleths of 

 modified refractive index, AI, were drawn to intervals of 2.5 units 



M = N -^ {Keh)lO', (8.10) 



where Ke = (15.70) (10-8)/m, and A^ is as defined in chapter 1, (1.20). A 

 simplified version of the lower portion of this cross section with the corres- 

 ponding M curves is accompanied by a sketch of the general location of 

 the experiment in figure 8.10. Some smoothing was necessary, particu- 

 larly near the sea surface and in those areas where aircraft slant ascents 

 and descents caused lag errors in altimeter readings and in temperature 

 and humidity elements. Isopleths over land were plotted above surface 

 rather than above sea level, with an additional adjustment in the scale 

 ratios of height and distance in an attempt to simplify the reading of 

 values from the diagram. 



8.2.3. Cape Kennedy 



The second area studied was the Cape Kennedy to Nassau path for the 

 period of April 24 to May 8, 1957. This material was supplied by the 

 Wave Propagation Branch of Naval Research Laboratories and the 



