Global Wave Forecasts Using Spacecraft Data 



27. Ewing, G.C., "Oceanography from Space," Woods Hole Oceanographic Ref- 

 erence 65-10, 1965 



28. Moore, R.K., and Pierson, W.J., "Measuring Sea State and Estimating Sur- 

 face Winds from a Polar Orbiting Satellite," presented in Miami at the 

 International Symposium on Electromagnetic Sensing from Space, Univer- 

 sity of Miami (in press, 1965) 



DISCUSSION 



John W. Miles 



Institute for Geophysics and Planetary Physics, LaJolla Laboratories 



University of California, San Diego 



LaJolla, California 



I shall confine my comments to the discrepancies between the theoretical 

 (15)* and observationally inferred (17,18,24) values of the energy-transfer coef- 

 ficient, say /3, that governs the exponential growth of wind-generated waves. 

 The most likely causes of these discrepancies appear to be (a) the neglect of 

 wave -induced components of the Reynolds stresses in the turbulent shear flow 

 and (b) the departure of the mean wind velocity from the logarithmic profile as- 

 sumed in the original theoretical model. Both of these discrepancies are likely 

 to be especially significant for those parts of the spectrum in which the critical 

 layer, say z = z^,, at which wave speed equals wind speed, is at a relatively 

 large elevation, say kz^, ^ l; this last condition appears to be typical in the 

 neighborhood of the spectral peak for a fully developed sea. On the other hand, 

 recent laboratory measurements, by Hsu and his colleagues at Stanford, appear 

 to yield values of /s that are consistent with the original theoretical calculations 

 for kz^, « 1. 



It seems quite unlikely that the profile discrepancy, (b) above, could account 

 for an order-of-magnitude increase of 8 in /i, relative to the value calculated on 

 the hypothesis that the mean flow can be regarded as laminar. It is, however, 

 entirely plausible that the wave-induced turbulent Reynolds stresses could ac- 

 count for this discrepancy, at least for kz^ « i, although it seems unlikely that 

 the energy-transfer processes associated with the quasi-laminar and turbulent 

 mechanisms would be effective in the same proportion for all kz^. 



Further progress in our basic understanding of the physical processes in- 

 volved in the generation of waves by wind almost certainly depends on controlled 

 laboratory experiments. Further theoretical development also is possible, but 

 our very limited understanding of turbulence suggests that this development is 



'Numbers refer to references given by Pierson et al. 



529 



