716 DABBYSHIRE [CHAP. 20 



Hatherton (1960) has analysed microseisms recorded at Scott Base during 

 the International Geophysical Year. He found that during January and 

 February when the Ross Sea is clear of ice but the outer ocean is still covered, 

 the microseisms were of short period, 1 to 3.5 sec, and the maximum amplitude 

 varied as T^-^. Later on during March and April, when all the sea is free of ice, 

 longer-period microseisms of 4-10 sec are recorded and the maximum amplitude 

 varies as T^-". The shorter ones can be related to the shorter-period sea waves 

 generated within the Ross Sea area. The longer ones are due to swell from the 

 neighbouring ocean, and the different relation between amplitude and period 

 is attributed to the selective period attenuation of microseisms in crossing the 

 continental shelf. The two-to-one period ratio between microseisms and sea 

 waves appears to hold in all cases. 



The work of Bath in Scandinavia has already been mentioned. Besides his 

 conclusions on the effect of cold fronts crossing the coast, he also showed that 

 the microseisms in Scandinavia were local in origin and not coming from the 

 Atlantic. An examination of Fig. 7 suggests that this is due to the effect of 

 refraction. 



A completely different approach to the subject has been made by Nanney 

 (1959) who has published work to show that the incidence of microseisms is 

 slightly correlated with that of earthquakes. 



A good deal of microseismic work has been done in the U.S.S.R. An interest- 

 ing paper was published by Savarensky, Lysenko and Komplanetz (1958) on 

 microseisms recorded at Raibach on the shores of Lake Issik-Kul. This lake is 

 about 200 km long and 50 km wide at the widest part. The amplitude of short 

 period microseisms (1.5-3 sec) increased very rapidly with increase in local 

 wind speed, particularly when this was westerly. A west wind could produce 

 high waves on the lake which would be reflected off the steep rocks lying round 

 the lake. There is a lag of about 9 h between the time of maximum microseism 

 intensity and maximum wind intensity and this can be explained by the time 

 required for the wind to build up the highest waves and form a stationary wave 

 system. The two-to-one period ratio appears to hold. It was also possible to 

 estimate the reflection coefflcient of wave energy from the rocks. It is 

 about ^0". 



Rykunov and Prosvirnin (1958) have studied microseisms from distant 

 storms and have tracked storms in the Atlantic and off the coast of Scandinavia 

 from stations ranging from Murmansk to the Black Sea. They find that the 

 direction finding technique is improved by taking into account refraction, as 

 was done by Darbyshire. 



Prosvirnin and others (1959) also found that for stations near Scandinavia 

 there are two azimuths of maximum amplitude for the line joining the record- 

 ing station to the microseism source, one being normal and the other tangential 

 to the Scandinavian coast. This effect is attributed to the continental shelf and 

 the mountain ranges, which tend to be ahgned normal to the coast. These 

 conclusions have been checked and confirmed by model experiments. 



I 



