52 



DEACON AND WEBB 



[chap. 3 



observations. However, for these conditions, (17) approximates to the more 

 convenient expression 





(18) 



which is equivalent to the relationship proposed by Monin and Obukhov 

 (1954).! The full line in Fig. 2 is for a = 4.5 but this appears to apply over the 

 range of Bi from —0.05 to +0.15 at the most. 



Fig. 2. Observations of Km = u*I{z Sujdz) related to the Richardson number Ri; over 

 grassland. 



O Results from Swinbank's eddy correlation apparatus. 



A Rider's (1954) results by the isolated test surface method. 



The vertical lines indicate the standard errors of the means. 



Equation (18) with a = 4.5 



Rossby-Montgomery formula (17) with ct = 9 



Ellison's equation (20) with y= 18. 



A convenient way of applying (18) in the evaluation of u^ from wind- 

 profiles obtained under stability conditions falling within the prescribed limits 

 is as follows. Assuming that Kh = Km under such conditions, the Richardson 



1 Monin and Obukhov used, however, the stabihty parameter zjL. For the near-neutral 

 conditions to which (18) refers, zjL closely approximates to Ri. From wind- and tempera- 

 ture-profiles over the steppes, Monin and Obukhov found a value of a = 0.6, which is very 

 much smaller than the value of ~4.5 corresponding to ct = 9. This has been shown by 

 Taylor (1960) to be largely a result of Monin and Obukhov having fitted their relationship 

 over a much wider stability range than appears to be justifiable. 



