a comprehensive study of waves on the Arctic Ocean, using a seismograph, tide 

 gage, and gravimeter to examine the spectrum of waves with periods between 

 ]_0-l a»^ io5 seconds. In particular, he observed (l) a phase difference in 

 simultaneous records of two gravimeter s read 400m apart and (2) correlated 

 surface wind speed with the maximum amplitudes of the oscillations recorded 

 by both a seismograph and a gravimeter. Thus, he showed that these oscilla- 

 tions were progressive waves and, by Fourier analyses, computed apparent 

 phase velocities of several of the components. By correlating surface wind 

 speed with maximum oscillation amplitude, he found that when the wind exceeded 

 20 to 2k knots there was a significant positive correlation; below this speed 

 there was no correlation. He therefore suggested 20 to 2k knots as the thresh- 

 old excitation value. 



LeSchack and Haubrich (1964), using power spectrum techniques, examined 

 the characteristics of several long gravity records taken in the Arctic. They 

 showed the similarity between the spectra of two records taken simultaneously 

 at sites more than 1km apart. Representative values for ice displacement as 

 a function of wave period were given. 



Sytinskiy and Tripol'nikov (iS^k) discussed the observations of waves 

 measured with a tripartite array of earthquake seismographs at their stations 

 K-13 and N-l4» They observed progressive waves having the same period range 

 as previously reported and amplitudes averaging 0.02 to 0.025cm, with maximum 

 peaks of 0.1cm in deep water. Maximum amplitude on the continental shelf was 

 0.3cm. These amplitude values were consistent with those computed by LeSchack 

 and Haubrich. 



Sytinskiy and Tripol'nikov attempted to measure phase velocities and wave 

 directions by comparing the records from the tripartite array, presximably by 

 matching peaks of the same phase. The phase velocity values were rather scat- 

 tered but fell mainly between 20 and 6o m/sec. They attributed the velocity 

 scatter to wave interference at the point of observation. They noted, as 

 Hunkins had, the general positive correlation of wave amplitudes with high 

 local surface wind speed. Accordingly, they examined a wave record made diiring 

 a sudden gust of wind and reported the following sequence: 



(1) A short bvirst of energy of about 3- second period due to "bobbing" 

 of the ice computed from equation (l). 



T = 27r(yoh//D'g)'^' (l) 



where T — period (sec) 

 P = ice density 

 h = ice thickness 

 /j' = water density 

 g = gravity 



(2) This "bobbing" was superimposed on a 12- to 15-second wave that 

 seemed to be initiated by the wind gust and continued for several minutes 

 thereafter. They believed that the generation of the 12- to 15-second wave 

 was analogous to oscillations that would theoretically be produced by a load 



