90 



OBSERVATIONS AND RESULTS IN PHYSICAL OCEANOGRAPHY 



these earlier expeditions with the Carnegie results. The 

 same applies, to a still greater extent, to the salinities. 

 Numerous observations of the salinity have been made 

 by the Challenger but, as stated before, these have not 

 the accuracy which permits combination with modern 

 data. The following discussion of the physical oceanog- 

 raphy of the Pacific will be based, therefore, principally 

 on the Carnegie data alone. 



Temperature and Salinity 

 Horizontal Distribution 



The values of temperature and salinity, scaled from 

 graphs for each station, have been entered in figures 

 210-233; I-B, and isotherms and isohalines have been 

 drawn in order to bring out the characteristic features. 

 It must be emphasized, however, that at the higher levels 

 these lines have no well-defined physical significance. 

 They do not represent the values at a given moment nor 

 the mean annual values because the observations have 

 been made in different seasons in the different regions. 

 In the upper layers great deviations from the aver- 

 age annual conditions must be expected because of sea- 

 sonal variations in heating and cooling. In the courses 

 of the currents and because of irregular changes, but at 

 greater depths, such variations are probably small and 

 here the lines can be expected to represent the average 

 conditions, although they are based on a small number of 

 observations. 



By means of earlier observations, Schott and Schu 

 (1910) have prepared charts showing the horizontal dis- 

 tribution of temperature at different levels down to a 

 depth of 4000 meters, and Schott (1928) has published a 

 chart showing the distribution of the salinity at the sur- 

 face. In the following we shall undertake some compari- 

 sons between these charts and those derived from the 

 Carnegie data. 



Surface . --At the surface (figure 210: I-B) the tem- 

 perature in the Northern Hemisphere decreases regular- 

 ly from the equator toward the north in the western part 

 of the Pacific. In the eastern part of both hemispheres 

 the isotherms are bent toward the equator but more so 

 in the Southern Hemisphere, where a region of low tem- 

 perature can be followed along the equator from the Pe- 

 ruvian coast toward longitude 150° west. In the western 

 part of the ocean a corresponding bend of the isotherms 

 toward the equator is found to the northeast of Japan, 

 going from Bering Sea down to latitude 40° north. 



The chart by Schott and Schu shows the mean annual 

 isotherms and, therefore, it cannot be expected that the 

 Carnegie data will agree with the chart values because 

 of the annual variation of the surface temperature. The 

 Carnegie observations were made in summer in both 

 hemispheres, for which reason they must be higher as a 

 rule than the means for the year. Comparing the Car - 

 negie data with the corresponding values which can be 

 read off the Schott-Schu chart, we find that the Carnegie 

 temperatures generally are higher. The most striking 

 exception is in the vicinity of the Galapagos Islands, 

 where, at stations 41, 42, 43, 44, and 45, the average de- 

 viation from the chart values is -2.°2. 



It is not possible to reduce the Carnegie observa- 

 tions to the mean of the year because sufficient data as 

 to the annual variation are lacking. The Marine Impe- 

 rial Institute at Kobe, Japan has published mean monthly 



isotherms for the greater part of the North Pacific, but 

 the charts do not cover the entire area in question. For 

 the equatorial regions Puis (1395) has published monthly 

 isotherms, but these do not quite agree with the above- 

 mentioned in the areas where the two representations 

 overlap. In order to eliminate the effect of changes in 

 the currents on the temperature distribution (Helland- 

 Hansen, 1930) it would be necessary, furthermore, to 

 take the salinity variations into account, but isohalines 

 for each month are not available. The foundation for a 

 reduction of the observed temperatures to the mean of 

 the year is, thus, insufficient, but we can draw attention 

 to the character of the differences between the Carnegie 

 values (CJ and the mean annual temperatures as repre- 

 sented by Schott and Schu (SS) and by Japanese charts 

 (Jap) . We also shall examine the differences between the 

 Carnegie data and the corresponding mean monthly val- 

 ues as shown in Japanese charts and in those by Puis. 



Forming mean values we find for the North Pacific, 

 from stations 98 to 140: C-SS = +1?9, 



C - Jap (year) = +2.°0. 



C - Ia£ (month) = +0.°7; 

 for the equatorial regions, from stations 35 to 47: 



C - SS = -l.°l. 



C - Puis = -0.°1; 

 from stations 70 to 74, 93 to 109, 138, 139, and 149 to 

 162: C-SS = +lf5, 



C - Puis = +0.°5. 

 The differences between the Carnegie observations 

 and the values from the Schott-Schu chart are shown in 

 table 7, where they have been arranged in groups accord- 

 ing to the latitude, and where the months in which the 

 observations were taken, are shown. The table also con- 

 tains the differences in salinity according to the Carnegie 

 observations and Schott's chart to which we shall return 

 presently. In this place it will be pointed out that the 

 Carnegie observations give the relatively highest tem- 

 peratures in the middle part of the North Pacific in the 

 months of September and October and the relatively low- 

 est values, aside from the conditions in and near the 

 Gulf of Panama, in the equatorial region in April and 

 May. 



Table 7. Differences in temperature and salinity 



between the Carnegie observations (C) and the values 



from charts by Schott and Schu (SS) and Schott (S) 



Stations 



Latitudes 



Months 

 1928-1929 



Tem- 

 per- 

 ature 

 C-SS 



Salin- 

 ity 

 C-S 



These features, and the fact that the differences are 

 reduced when comparing with monthly charts, indicate 

 that the discrepancies between the Carnegie observations 

 and the annual values according to Schott and Schu prin- 

 cipally are because of the annual variation of the surface 

 temperature, but part of the variation is probably con- 

 nected with accidental changes in the currents. 



