ATLANT. DEEP-SEA EXPED. 1910. VOL i) PHYSICAL OCEANOGRAPHY AND METEOROLOGY 



83 



well known, but such a sudden transition as our cliarls and 

 figures suggest has evidently not been perceived before, li 

 must be admitted, however, that we posses no serial observ- 

 ations along sections straight across the Great Atlantic 

 Current, so it cannot be directly proved that the transitions 

 are as sudden as stated here, but the "Micliael Sars" observ- 

 ations strongly support our view, which is further strength- 

 ened by dynamical calculations. It is probable tiint the 

 horizontal gradient of temperature is siTialler at the surface 

 than at some intermediate depths, and it is possible that 

 the abruptness of the transition has been exaggerated as 

 far as the 100 metres level is concerned, but scarcely with 

 regard to the deeper levels. 



The conditions in the eastern part of the North 

 Atlantic are more complicated than in the western part 

 as has repeatedly been mentioned. The Great Atlantic 

 Current is split up in several branches. One of these goes 

 northwards, so that the sea east of about 20° W. is warm 

 even N. of 50° N., as compared with the sea further to 

 the west. Fig. 31 illustrates the horizontal distribution of 

 temperature along the meridian of 18° W., from 60° to 

 30° N. The curves are smoothed. They do not show all 

 the local variations represented in Figs. 28 and 29, but 

 only the main features. The curves demonstrate clearly 

 the following conditions between 60° and 30° N. In the 

 northern region the temperatures are much higher in the 

 eastern part of the North Atlantic than they are in the 

 western part at all depths between the surface and 1000 

 metres (the same holds good of greater depths too). In 

 the southern regions (S. of about 48° N.) the conditions 

 are reversed as far as the upper 500 metres are concerned, 

 but from 600 metres downwards it is warmer in the eastern 

 than in the western part of the sea. The latter phenomenon 

 is accounted for by the influence of the Mediterranean 

 upon the Atlantic. The same influence explains the fact 

 that the difference in temperature between 400 and 1000 

 metres is much smaller in the eastern areas than in the 

 western (about 3° C. against 7°), and that the vertical 

 gradient of temperature deeper down, when approaching 

 the deep water, is much greater in the former areas than 

 it is in the latter (cf. section 37). 



37. The Vertical Distribution of Temperature. 



The vertical variations of temperature are, on the 

 whole, much greater than the horizontal. We have dis- 

 cussed the variations in the deep water in section 35 and 

 shall here deal only with the troposphere of the North 

 Atlantic. In nearly all regions the maximum temperature 

 is at the surface and the minimum in the deep water, so 

 the total vertical variation of temperature within the tropo- 



sphere is generally equal to the surface temperature minus 

 3°. This total variation reaches its maximum in the tro- 

 pics, corresponding on an average to about 1°C. per 100 

 metres. It decreases with increasing latitude and may be 

 subject to considerable seasonal variations. 



We shall only deal here with a few general consid- 

 erations. In the following discussion we shall use the 

 expression mean variation to denote the average vertical 

 variation of temperature in 1/100° C. per metre (or whole 



degrees per 100 metres) between two levels 



('°^-;)- 



When 



we reckon the positive depths downwards, the mean varia- 

 tion is generally negative, as the temperature usually decrea- 

 ses with increasing depths. 



In some regions the temperature may become nearly 

 uniform from the surface down to several hundred metres 

 at the end of the winter cooling. With perfect mixing the 

 mean variation will become slightly positive on account 

 of the adiabatic effect, which may be calculated by means 

 of the tables in section 34, p. 67. At medium latitudes in the 

 North Atlantic the adiabatic rise of temperature downwards 

 will only amount to about 0-01 °C. per 100 metres (the 

 mean variation about 001) and may therefore be neglected. 



On the "Michael Sars" Expedition .some stations were 

 worked in April in the sea S. of Ireland. At Stats. 4, 7, 

 8 and 9 the difference of temperature between the surface 

 and 200 metres was from 0-30 to 0-35^ C, and the mean 

 variation, on an average, — 0-17. Between 200 and 400 

 metres the mean variation only amounted to between — 0-03 

 and — 008, the average being — 0-06. At 200 metres the 

 temperature ranged from 10-25° to 10-65° C. at the stations 

 in question, and at 400 metres from 10-10° to 10-52" C. 

 It is very probable that the vertical convection in winter 

 (March) in this region extended to such a depth that the 

 temperature was uniform from the surface to 400 metres 

 or more. Even if the seasonal variations are small below 

 200 metres (cf. section 32), some propagation of heat from 

 above may take place in early spring when the virtual 

 coefficient of temperature conductivity is relatively great 

 because the water-masses are almosi in a state of neutral 

 equilibrium. Stat. 93 was worked in the same region on 

 the 25th of July. At this station the difference of tempera- 

 ture was 4-72° C. between the surface and 200 metres. At 

 200 metres the temperature was 10-38° and at 400 metres 

 10-24° C, the mean variation between these depths thus 

 amounting to — 0-07. This variation and also the tempera- 

 tures themselves were practically the same as at the stations 

 in April. 



In our area "B", off the Bay of Biscay (Fig. II), so 

 many stations have been worked that we may attempt a 

 computation of the average differences of temperature from 

 level to level at various seasons, and find the correspond- 



