1 



1 10 The Three-dimensional Temperature Distribution and its Variation in Time 



The diurnal temperature variations always decrease towards higher latitudes; the 

 maximum occurs at 14.00 h and the minimum at 04.00-05.00 h. The diurnal course 

 corresponds almost exactly to a pure sine curve. KuHLBRorx (1938) obtained the same 

 results from a study of the daily temperature records of the "Meteor" Expedition by 

 the elimination of the effect of changes in position of the vessel. The average daily 

 amplitude for all areas of the South Atlantic amounts to only 0-26 °C. This value, 

 which was obtained by averaging all days without any selection, is somewhat smaller 

 than the value obtained by Wegemann (1920) from the "Challenger" observations 

 and by Meinardus (1929) from the "Gauss" observations. 



The heating of the sea surface begins soon after sunrise due to the absorption of 

 solar radiation in the uppermost layer of the water, but the largest part of the added 

 heat is used for the evaporation of water (about two-thirds) and only a small part 

 remains for a temperature rise. The temperature thus rises only slowly to the maximum 

 at 14.00 h. After sunset the temperature fall continues due to outgoing radiation. 



There are very few measurements of the depth to which the diurnal temperature 

 variation penetrates. The only information for 50 m depth is given by the hourly 

 observations at the anchor stations. However, for these depths near the thermocline 

 the influence of tides through the associated vertical currents (internal tide waves) 

 cannot be entirely excluded. Table 43 contains some values for the diurnal temperature 

 variation at 50 m depth showing that for these depths the amplitude is less than 

 0-05 °C. Aime has made measurements of the diurnal temperature variation at different 

 depths off the Algerian coast. It is, however, not entirely certain that all the observed 

 changes can be attributed to the diurnal cycle; however, if this assumption is made it is 

 found that the nocturnal cooling at 14 m is one-fifth of the surface amplitude and that 

 the heating during the day, which is three to four times stronger than the nocturnal 

 cooling, falls to a tenth at 28 m. Schmidt (1925) calculated from this decrease of 

 temperature the vertical exchange coefficient as 35-40 g cm"^ sec~^. The observations 

 of Knott on the "Pola" Expedition in the eastern Mediterranean show a decrease in 

 the amplitude to a tenth at 29 m, which corresponds to an exchange coefficient of 

 42 g cm~^ sec~^. Since the ocean covers more than two-thirds of the surface of the 

 Earth it can be said that over much the largest part of the surface the diurnal tem- 

 perature variations remains less than half a degree. Therefore, the considerably 

 greater diurnal temperature variations of the continents play only a minor part in the 

 total heat budget for the Earth. 



{b) The Annual Temperature Variation 



Changes in temperature over longer periods can be investigated in two different 

 ways. They can be recorded as "individual" temperature changes in a water mass 

 which is followed in its course in the ocean; they are then described by reference to 

 "oceanographic" co-ordinates. On the other hand, they can be followed at fixed 



