This shovild be evaluated at both ends of the wave spectrvun being 

 considered, since the values of A and X vary considerably over this range. 

 It should also be evaluated for ARLIS II (20m thick) as well as for the pack 

 ice (2m thick). 



Figiare 2k- shows the "displacement" spectra computed from the "accel- 

 eration" spectra in figure 17 and plotted in terms of twice the mean wave 

 amplitude squared per frequency bandwidth. Representative amplitude maxima 

 for observed long- and short-period waves were taken from this plot and used 

 for evaltiating equation (21). 



The values chosen were for periods of 67 seconds (2237 to 0007Z) and 

 11.1 seconds (03^5 to 0515Z) . These correspond to average amplitudes of 

 1.86cm and 0.032cm, respectively. For these periods the following sets of 

 values were used: 



T = 67 sec T = 11.1 sec 



X = 2,560m (shallow water wavelength) X = 193m 



A = 1.86cm A = 0.032cm 



h = 2m, 20m h = 2m, 20m 



^= 0.3 H-= 0.3 



Although Robin used 5 x 10 dynes/cm as an approximate value for 

 E (Young's modulus for sea ice), more recent data suggest that 1 x 10^ 

 dynes/cm^ is probably closer to the actual value for sea ice, particularly 

 for ARLIS II ice (Langleben and Pounder, I963) • 



From these vsilues, the following maximum stresses were obtained: 



For T = 67 sec , 



1.23 X 10 dynes/cm for ice 2m thick and 

 1.23 X 10^ dynes/cm^ for ice 20m thick. 



For T = 11.1 sec 



3.72 X 10^ dynes/cm2 for ice 2m thick and 

 3.72 X 10^ dynes/cm^ for ice 20m thick. 



xs 



'xs 



These values are considerably less than 2.2 to 3*9 x 10° dynes/cm^^ the 

 range of stresses that Butkovich (1956) found were required to fracture sea 

 ice beams under similar conditions in the laboratory. Tabata (1955) indicat- 

 ed that the major part of sea ice deformation would be elastic under short- 

 period stresses. It is therefore reasonable to conclude, in the same manner 

 as Robin, that the ice can be treated as an elastic plate for a2J. wave ampli- 

 tudes and periods encountered in this experiment. 



It is now desirable to see if the micropressure waves in fact have 

 sufficient vertical force to bend the ice and generate waves beneath it. In 

 order to do this, it is necessary to show that the oscillating vertical 

 pre s store on the ice is enough to bend the ice at least as much as the gravi- 

 meter records for that particular period. 



Only the pack ice system will be considered in this connection, for 



37 



