organization, have been unable to maintain dense enough arrays for 
a long enough time to gather statistically significant data. The signals 
are complex, and a sophisticated measuring program is required to 
read them. The problem would be difficult enough if all oceanic fluc- 
tuations were a broad spectrum of linearly superimposed internal 
waves, but, as mentioned above, there is undoubtedly a significantly 
nonlinear domain. Oceanographers need to evolve some fairly elabo- 
rate measuring arrays, with limited regions heavily instrumented. 
They are in the position of radio astronomers who need a radio tele- 
scope of a novel design, a facility quite beyond the capability of a 
single individual to design, build, and operate. The oceanographic 
community has been too concerned with conventional research and 
fund-raising and has devoted insufficient attention to exciting new 
scientific projects such as a viable buoy program. 
A graduated program for measuring and identifying regular oscilla- 
tions in a typical deep-sea area is described in appendix II. This is 
one of several proposals which have emerged in the last few months 
from groups interested in buoy programs. 
Air-Sea Boundary. In order to predict large-scale atmospheric 
behavior for periods longer than a day or two, vertical fluxes of heat, 
momentum, and water vapor must be specified at the surface, both on 
land and in sea. Research and development along several independent 
lines are needed. 
The spectral structure of atmospheric turbulence is being deter- 
mined, and direct measurements of vertical fluxes are being made with 
rapidly responding sensors mounted on fixed platforms, aircraft, or 
submarines. Temperature and wind velocity sensors exist in experi- 
mental form. Interesting work is under way at a few institutions, but 
adequate humidity sensors have yet to be developed. This lack repre- 
sents an important constraint on air-sea interaction research. Mean 
profiles measured from fixed platforms and buoys are also being used 
at a few institutions to estimate vertical fluxes. 
However, in order to relate vertical flux measured at a point by either 
of the above methods to the “synoptic” scale commensurate with the 
weather prediction problem, measurements using integral methods 
over extended areas are needed. These require a carefully planned 
and coordinated program of research utilizing fixed platforms, buoys, 
aircraft, and possibly submarines. To date such programs have not 
been initiated. 
Methods of isotopic and surface chemistry have recently been ap- 
plied to the air-sea boundary, and these offer some interesting oppor- 
tunities which should be exploited. 
A substantia] effort has been directed to the study of surface waves, 
particularly with regard to nonlinear actions and generation by wind. 
Studies have been dominantly theoretical; the need is for adequate 
47 
