70 



BJ0RN HELLAND-HANSEN 



[REP. OF THE "MICHAEL SARS" NORTH 



of b metres, we can first, by means of Table A and Fig. 

 24, find the temperature which the water obtains when 

 moved from a metres to the surface; and then, with the 

 latter temperature as argument and by means of Table 

 B and Fig. 25, find the change for a removal from the 

 surface to b metres. 



The temperature which the water attains adiabatically 

 when the pressure is altered is called the potential tem- 

 perature at the new pressure [Helland-Hansen, 1912]. 

 It may be detoned by 0: 







Tm 



+ <5r 



The removal may be indicated by means of indices 

 in this way: when water with a temperature of Ta at the 

 depth of a metres is removed to a depth of b metres 

 its potential temperature referred to the latter depth be- 



comes : 



= 



Ta 



_I- 



The potential temperature referred to the surface (one 

 atmosphere of pressure) is 0. 



We have: 



= 



(ra + dr) + 

 The values of dr 



dr' 



and dr' are found in the 



a — > o o — > b 



tables. It must be emphasized that the argument for 

 (5t is ra, while for dr' it is {r a + ^r). 



a — > o o — y h a — > o 



In the deep part of the oceans the variations in sali- 

 nity are so insignificant that they have no influence upon 

 the compressibility. They are, therefore, of no conse- 

 quence in determining the value of 6r. In the case of 

 adiabatic equilibrium between the levels a and b we have: 



Ta = = Tb + dr 

 b — >-a b — >-a 



Tb 



— 



Ta 



= 



a — > h — > 



None of these equations hold good if the vertical 

 variations of temperature do not correspond to an adia- 

 batic equilibrium. As an example we may take some obser- 

 vations from great depths in the Pacific. G. Schott [1914] 

 has tabulated a number of observations from water near 

 the bottom at various stations in the Philippine Deep and 

 the New Pomerancan and the Bougainville Deeps, the 

 observations from the two latter deeps being treated 

 together. The observations are taken in 1907—1913 by 

 means of reversing thermometers. Schott has calculated 



the mean temperature at different levels. His results for 

 depths of 5000 metres and downwards are quoted in the 

 second column of the following table (rm)- As these 

 temperatures are not the results of serial observations 

 along a vertical they do not claim to give an absolutely 

 correct representation of the strictly vertical distribution 

 of temperature. The chief result seems, however, to be 

 quite certain : the temperature in these deeps increases 

 vertically downwards from 5000 metres to the bottom. 

 By means of Tables A and B and Figs. 24 and 25 we 

 have computed the potential temperatures referred to the 

 surface and to two other levels (a and b); the results 

 are recorded im columns 3—5 of the table. 



In the Philippine Deep the salinity is uniform and 

 equal to 34-68 "/oo from 5000 metres downwards, accord- 

 ing to Schott. In this Deep a minimum of temperature 

 /// situ is found at 5000 metres, while the values of 

 show a minimum at 6000 metres. This means that the 

 water between 5000 and 6000 metres is in a state of 

 positive stability. The values of increase from 6000 

 metres to the bottom, which means a state of instability 

 (an "overadiabatic" distribution of temperature). In co- 

 lumns 6—8 of the table above, the 0-values for the water 

 at 6000 metres in the Philippine Deep are subtracted 

 from the corresponding values for the water at the other 

 depths. We see that the differences given in the three 

 columns are not the same. Comparing the water at 

 6000 metres {a) with that at 9788 metres (b) we find the 

 following: when the water from both depths is raised 

 to the surface the difference of potential temperature be- 

 comes 0-305°, while if referred to 6000 metres the diffe- 

 rence is 0-318°, and if referred to 9788 metres it is 0-323°. 

 We should have adiabatic equilibrium if the temperature 

 at 6000 metres were 0-318° higher than it actually is, 



