SEA-SURFACE TEMPERATURE 35 
Table 40. Amplitudes and phase angles of diurnal and semidiurnal oscillations of sea-surface 
temperature from observations on Carnegie, Gauss, and Challenger 
(Carnegie groups II, XIV, XV, XVII, XVIII, and XXI omitted; 

Carnegie 
Ampli- Phase 
tude angle 


See Ad. Schmidt) 


[eet Me Gauss eee eas Challenger 





: Ampli- Phase Ampli- Phase 
oe ee ace eee mete | Xo. 
ays 
Palate te "Ta tala lel" eel Rea Re en 
35 N-15 N 0.11 nae 200 a 1325022 ane ne a1 SIF) ea Cee eiiwee! Lees . 
15N- 5N 0.18 0.06 302 25 307 0.31 0.13 230 66 16S ee a es 
5S$-25S 0.12 0.05 241 0 1215 0.18 0.04 237 50 Be ted acpi terre ne ceee, cence icnens 
All 
latitudes 0.12 0.04 228 27 248° 0.17 0.06 225 52 2019 0.19 0.03 226 48 651° 
@ Atlantic Ocean. } Pacific Ocean. 
oceans. © All oceans. 
It should be remarked that the investigations car- 
ried out on board the Carnegie took place either in the 
tropics or during the summer months in higher lati- 
tudes. Thus all observational work was done under con- 
ditions where such temperature relations would be ex- 
pected, and for this reason the results given in this sec- 
tion are not presented as being wholly representative of 
_ average conditions throughout the year. 
The difference between mean sea and air tempera- 
tures on the Carnegie was never as great as 270. In only 
two areas, the Gulf Stream and the Gulf of Panama, 
-were mean sea-surface temperatures more than 1:0 
higher than mean air temperatures. A maximum mean 
difference of 1°6 was recorded within the Gulf Stream, 
as might be expected in view of the high water tempera- 
tures of the current and its comparative narrowness. A 
difference of 1°5 between mean sea and air tempera- 
tures in the Gulf of Panama may be explained by the fact 
that during the entire twelve days of this series, the wind 
was consistently from the southwest --from a region in 
which sea-surface temperatures only a few hundred 
miles away were as much as 8° lower than in the Gulf. 
Thus, air considerably colder than Gulf water was con- 
stantly being imported. 
It is also interesting to note that of the means for 
the Groups which include that part of the cruise from 
Japan to San Francisco, those for air temperature appear 
to have been slightly higher than those for sea-surface 
temperature. Differences were small: from 0°1 to 0°7. 
The winds usually had a southerly component during this 
part of the cruise. It may be assumed that air masses 
were Tp (Tropical Pacific), or at least greatly modified 
Npp (Transitional Polar Pacific). 
In one other area, that centered off the coast of Chile 
approximately on the western edge of the Peruvian Cur- 
rent, the mean air temperature was 0°11 higher than the 
rather low mean sea-surface temperature. 
As shown by Visser [24, p. 12] and Braak [36], sea- 
surface temperatures in the tropics are usually higher 
than air temperatures, throughout all months of the year. 
Visser noted a difference of +0°84 as the mean for all 
months during the three cruises of the Snellius (1929- 
1930) in the Netherlands East Indies; and Braak, on his 
voyage between Batavia and Ambon, found a mean differ- 
ence of +1:05. 
¢ Atlantic and Pacific oceans. 
d Atlantic and Indian 
Table 41. Variation with latitude of mean difference 
between temperatures ofsea and air, Carnegie, 1928-29 
Range in Mean Range in Mean 
latitude sea - air latitude sea - air 
° °o X63 ° ° Xe 
>45N +0.97 5N- 58S + 0.43 
45 N-35 N -0.23 5 S-15S +0.14 
35 N-25 N +0.24 15 S-25S +0.34 
25 N-15 N +0.13 25 S-15S +0.21 
15 N- 5N +0.50 35 S-45 S -0.16 
PGR rtninee +0.256 
Variation of Sea-Surface 
and Air-Temperature Differences with Latitude 
As indicated in figure 14, the mean sea-surface 
temperature exceeds the mean air temperature through- 
out all ranges of latitude except between latitudes +35° 
to +45°. Apparently these two ranges mark the discontinu- 
ity between the warm southern and cool northern waters. 
It may be noted that the mean sea-surface tempera- 
ture for all days (21°93) is exceeded by the mean air 
temperature for all days (22°70). This condition may be 
explained by the fact that the recording of sea-surface 
temperature began shortly after the Carnegie left Hamp- 
ton Roads, whereas the recording of air temperature did 
not begin until after the Carnegie left Hamburg. There- 
fore, the mean values for all days are not comparable, 
in that the Carnegie mean sea-surface temperature for 
all days is affected by the greater number of observa- 
tions in the higher latitudes (35° to 45° north). 
The mean difference, however, between sea-surface 
and air temperatures (sea minus air) for all ranges of 
latitude is +0°256. 
Diurnal Variation 
of Sea- and Air-Temperature Differences 
From the corrected hourly means of sea and air 
temperatures for the various Groups, a study of the di- 
urnal variation of the temperature differences has been 
made. From the literature concerning previous investi- 
gations along this line, it was expected that the diurnal 

