492 PATTULLO [CHAP. 22 



changes in the pressure field on the floor of the deep sea. Changes in air pressure 

 are followed rather quickly by compensating changes in sea-surface elevation, 

 so that the total mass of air plus water resting on the sea floor remains a con- 

 stant (Groves, 1957). The effects of phenomena (6) and (c) may be computed 

 either separately or as a combined effect by using measurements of temperature 

 and salinity at various depths. The results are usually added together and called 

 steric levels since they represent seasonal changes in level related to changes in 

 the volume of the water only. Therefore, (a), (b) and (c), taken together, comprise 

 all those changes and only those changes in sea-level that can occur with no 

 net change of pressure anywhere on the sea floor, as stated above. Lisitzin and 

 Pattullo (1961) have examined the average magnitude of the isostatic term 

 in the Pacific Ocean, and have compared it with the observed variation in sea- 

 level in the same ocean during the I.G.Y. Their results show that, over most of 

 the open ocean, the recorded variations in sea-level and the isostatic variations 

 agree well. Exceptions to this generality occur at certain places near coastlines 

 and in the tropical western Pacific. The largest differences between recorded 

 and isostatic levels occur along the mainland coast of Mexico, northeastern 

 Australia, and, occasionally, the coast of North America near the border 

 between Canada and the United States. The pressure and steric terms attain 

 the same amplitudes, but not at the same localities. Pressure effects are a 

 maximum between 40° and 60°N ; steric levels show largest deviations between 

 40°N and 40°S. 



The implications of this result are that the variation in total mass over any 

 part of the deep-sea floor must be small (a finding of some interest to geo- 

 physicists), and that seasonal changes in the mass of water at a given location 

 can be estimated with some reliability from atmospheric pressure variations 

 alone. 



This pressure effect is a rather clear-cut case where one can readily discuss 

 cause and effect, since the changing forces accompanying changes in atmos- 

 pheric pressure can be computed if the field of pressure is known. "Causes" of 

 the steric deviations are not so easy to identify. For example, Pattullo (1957) 

 has shown that about two -thirds of the seasonal change in heat content in the 

 oceans of the Northern Hemisphere can be accounted for by local heating 

 alone. However, many problems still remain unsolved as to the relationships 

 between wind and thermal structure ; one must consider at least the following : 

 set-up (which may add more warm surface waters to the area), currents (per- 

 manent and transient), and mixing. That is, one can determine values for (b) 

 quite readily, but its causes are not yet well understood. 



The isostatic term is by no means equal to the observed deviations every- 

 where. The island areas of the tropical Pacific show marked, sometimes erratic, 

 variations in both recorded and steric levels and the agreement between the 

 two cannot be considered satisfactory (the pressure term is relatively un- 

 important at these latitudes). These are regions of complicated hydrography 

 and topography ; it appears that more years of observation and of detailed 

 analyses will be necessary before we will understand precisely what is happening 



