given by the empirical relationship (17). 



The estimates which led to (22) and (23) are based partly on 

 theoretical considerations and partly on empirical evidence. It 



follows from (23) that the more the wave velocity differs from the wind 



2 2 



velocity, the greater is the value of y • The relation y = f (p) 



seems to be a better approximation than the assumption of a constant 



2 

 value Y at all wind velocities and all stages of wave development. 



In their paper [1], Sverdrup-Munk call special attention to the fact 



2 

 that the value of y would probably be greater than the assumed con- 



2 — "^ 



stant value y = 2.6 x 10 -^j if the wave velocity differs too much 



from the wind velocity. By the relations (23) and (22), an attempt 



has been made to account for these circumstances, and in the following 



2 

 table some numerical values for y are given as calculated by means 



of (23): 



P 0.4 0.^ 0.6 0.7 0.8 0.9 1.0 1.1 1.2 

 Y^'IO'^ 5.62 4.55 4.05 3.64 3.27 3.01 2.78 2.68 2.57 



Per unit area we have 



w 



-^" "^eff = P'C(?)v2 , (24) 



if we introduce the abbreviated form 



C(P) = Y^(P^) + f T6^(l - P„)^ - f irfi/d - p^*)2 (25) 



for the effective coefficient of resistance. 



It may be mentioned in this connection that the determination 

 of the sheltering coefficient from experiments by Sir Thomas Stanton, 

 according to Sverdrup-Munk [1], leads to a numerical value which ap- 

 parently is in good agreement with our assumption s = 0.095. The 

 sheltering coefficient s can be evaluated from these measurements 



64 



