64 



METEOROLOGY— THEORY 



All values are negative. 



a long period of time. A clieek was made by compar- 

 ing Taylor's data off Newfoundland with computed 

 values. This check was quite good. 



4. The procedure is simple, and consequently the 

 weather officer could readily calculate the M curve. 



However, the entire method may be criticized be- 

 cause : 



1. In the integration of the diffusion equation K is 

 assumed constant while in the application of the in- 



Figure 1. Changes in M curves resulting from modifica- 

 tion of warm, dry air over cool, moist surface. Zero time 

 corresponds to the coast line; M hr, }^ hr, etc., refer 

 to the time the air has been over water. 



tegrated formula K was found to vary with elevation. 

 The values of A' which were used in the final analysis 

 ai'c therefore "ctl'ective values." 



2. K has been considered to be independent of the 

 degree of roughness (probaljly a justifiable assumption 

 over the ocean), the degree of stability, and the wind 

 velocity. These factors were neglected solely because 

 the scant data did not allow a complete analysis of 

 the variation of K. 



These "effective values" should give some indication 

 of the true variation of K. They suggest that K varies 

 linearly with elevation except for a quite rapid increase 

 in about the first 30 ft. Consequently it seems reason- 

 aide to assume that 



-.l['"-"f]' 



dT 



dt dz L ' ' dz . 



-\- q then, from the statement 



If K = pz -\- q then, from the statement that 

 K {8ii/Sz) = constant (eddy stress does not vary with 

 height), the velocity variation with elevation is given 



'',y '^ 



u = a log {z + b) + C. 

 'J'he question now arises: In the laminar layer, is the 

 wind variation with height represented by a logarith- 

 mic law? 



6.1.4 Previous Investigations 



For many years research workers have studied the 

 wind variation near the ground. A few of the conclu- 

 sions will now be presented. 



1. In 1932, Sutton^ assttmed a certain form of the 

 coefficient of correlation between the velocities of the 

 air particles considered at time t and at an interval 

 of time later. This assumption implied that there 

 was a power law for the variation of wind with height. 



u _ /z\" 



Wl Vl/ 2 — 71 



2. In 1933, C'ardington and Giblett- analyzed an 

 extensive series of observations at 4 ft and 143 ft. Of 

 course with only two iioints the observations could be 

 made to fit either a power law or a logarithmic law. 

 If a power law held, then //( is a function of the degree 



