68 DEACON AND WEBB [CHAP. 3 



stratification, Fh and Fm have values around 0.1 (see pp. 70-71). Therefore we 

 may define as a convenient bulk stability parameter which is roughly equal to 

 the Richardson number, 



Eb^ 10 iz ^-^. (33) 



When we take a reference height, 2 = 6 m, and a temperature, T = 290°K, in the 

 vicinity of those commonly encountered in practice, and express d in °C and 

 u in m/sec, then we arrive at a particularly convenient expression for the bulk 

 Richardson number 



i?&6 = 2.o4^^^|f%/ (34) 



Ua^ (m2/sec2) 



D. Vapour Pressure at the Sea Surface 



The air in direct contact with the sea surface has, at least approximately, a 

 water-vapour pressure equal to that of sea-water at the sea-surface temperature. 

 Normally this is approximately 98% of the vapour pressure of pure water at 

 the same temperature. According to Witting {vide Sverdrup, Johnson and 

 Fleming, 1942) the vapour-pressure ratio, sea-water/pure water, is related to 

 chlorinity (CI) as follows : 



v.p. ratio sea-water /pure water = 1 — 0.00097CL 



More recently, Arons and Kientzler (1954) have determined the vapour pressures 

 of sea-salt solutions over a very wide range of temperatures and chlorinities 

 and found reasonably good agreement with Witting's equation for chlorinities 

 up to 20 parts per thousand. 



E. Observational Investigations 



Only recently have refined measuring techniques been applied to the observa- 

 tion of temperature and humidity profiles in the atmospheric boundary layer 

 over the sea ; ^ and such investigations should provide a wealth of data of high 

 quality during the next few years. However, the pioneering investigations of 

 earlier workers using ordinary meteorological instruments have already 

 indicated approximately the magnitude of the humidity-profile coefficient. 



There is an important point of experimental technique which is worth 

 mentioning. By employing a pair of electrical thermometers such as thermistors 

 (the application of which is discussed by Haeggblom, 1959), it is possible to 

 measure the temperature difference between two heights rather than the 

 individual temperatures. This is a highly desirable arrangement for profile 

 studies from the standpoint of instrumental accuracy. Furthermore, in cases 

 where measurements are sampled at intervals during the observation period 



1 An outstanding exception is the much earlier equipment of Johnson (1927). 



