606 KOSsiTEK [chap. 16 



Daly's estimate of 5 m eustatic lowering of sea-level during the last few thousand 

 years leads to a decrease of 30 msec in the length of the day for a rapid growth 

 of ice-caps ; this is in accord with the 27 msec derived from ancient observations 

 of eclipses. 



Munk and Revelle have also put numerical values to the wandering of the 

 north pole of rotation, and hence to the variation of latitude, caused by glacial 

 eustasy. They find it can be a sensitive indicator of the source of melted ice, 

 1 cm rise in level displacing the pole "by | ft towards Chicago, or 2.8 ft towards 

 Greenland, according to whether the ice melts on Antarctica or Greenland". 

 Astronomical evidence of a rather doubtful nature suggests a movement of 

 10 ft towards Greenland over 25 years at the beginning of this century, in 

 accordance with a eustatic rise of 10 cm/century stemming from Greenland ice ; 

 recent confirmation of this shift has been found by Markowitz {in litt.) after 

 re-analyzing 50 years of latitude observations. 



B. The Pole Tide 



Secular and apparently irregular changes in latitude have been referred to 

 on page 605, but a regular movement of the pole of rotation was predicted by 

 Euler on theoretical grounds in 1765. The period of the Eulerian nutation 

 should be 10 months. "In 1891 a variation in latitude was discovered by 

 Chandler, but its period was 428 days instead of the expected 305 days. The 

 explanation was soon given by Newcomb (1892): Euler's theory applies to a 

 rigid earth ; for the actual case, the elastic yield of the solid earth and the fluid 

 yield of the oceans have to be taken into account, and these increase the period 

 of the free nutation from 10 to 14 months." 



The above quotation is taken from the introduction to a paper by Haubrich 

 and Munk (1959) who have used the results of a spectral analysis of sea-level 

 observations to search for the pole tide, that oscillation of the sea which is 

 theoretically a direct consequence of the Chandlerian nutation and the eccentri- 

 city of the earth. Their theory indicates that the equilibrium form of the pole 

 tide, when corrected for a yielding earth, should have an amplitude of the 

 order of 5 mm. Thus, as with the search for the nodal tide, the greatest difficulty 

 in determining the pole tide comes from the unwanted, usually random varia- 

 tions, or "noise", and this is a point to which perhaps insufficient attention has 

 been given in previous attempts to confirm the pole tide. Haubrich and Munk 

 find that there is a weak response, barely above noise level, corresponding to 

 the 14-month tide in the average monthly sea-level data for 11 stations. The 

 data for some of the stations, however, when examined separately, show no 

 peak at 0.84 cycles per year. Its frequency is well identified, but, from a com- 

 parison with a spectral analysis of the corresponding astronomically observed 

 variations of latitude, the mean amplitude is twice that expected from the 

 equilibrium theory. It is unfortunate, therefore, that this comparison cannot 

 be used to throw further light upon the values of the Love numbers. 



