30 



BJ0RN HELLAND-HANSEN 



(REP. OF THE "MICHAEL SARS" NORTH 



surface, a diurnal oscillation of tidal character may just 

 as well occur. 



The diurnal variations at 100 metres and deeper can- 

 not be accounted for by variations in direct heating and 

 cooling. At 300 metres the phase of oscillation (k,) is 

 exactly the same at Stat. 115 as at the Scottish station, 

 but at all other depths there is a marked difference, the maxi- 

 mum in the diurnal variation occurring earlier at Stat. Sc. 

 than at Stat. 1 15. The difference is between 4 and 5 hours 

 at 0, 100 and 400 metres where the amplitudes are quite 

 appreciable at both stations. At 500 metres where the 

 amplitudes are small the phase difference is found to be 

 6 hours. The vertical variations in k, at the two stations 

 are seen from the table p. 27, and are also demonstrated 

 by the curves in Fig. 2. It may especially be noticed that 

 the values of k^ are comparatively high in the upper 300 

 metres and low from 400 metres downwards. The differ- 

 ence in kj between 300 metres and 400 — 500 metres at 

 Stat. Sc. is 12—13 lunar hours, or a difference of phase 

 of approximately 180°. The difference is smaller at Stat. 

 115. It must be kept in mind, however, that the values 

 of kj found refer to the dates of observation only, and can- 

 not claim any general validity, because the analysis in- 

 cludes several diurnal periods of different length and with 

 unknown amplitudes. They may chiefly serve to demon- 

 strate that the phase of the total diurnal variation seems 

 to vary not only from station to station but in vertical 

 direction too. 



The numbers in column 12 of the table p. 27 do not 

 exhibit the same variations vertically as the temperature 

 records in column 4. The former figures show an absolute 

 maximum of diurnal oscillation at 100 metres below the 

 surface at both of the stations 115 and Sc. The amplitude 

 is especially great at Stat. Sc, more than twice the ampli- 

 tude at Stat. 115. At Stat. 115 the amplitude is a little 

 lower at 200 metres than at 100, while at Stat. Sc. it is 

 very much lower at 200 metres than at 100, and less than 

 at 200 metres at the other station. At 300 and 400 metres 

 the oscillations have been still more reduced, having 

 practically the same amplitude at both depths and both 

 stations. At 500 metres we find absolute minima. At 

 600 metres the amplitude at the Scottish station is again 

 comparatively great. 



This seems to indicate that the diurnal variation in 

 the depth of the isothermal surfaces has been most 

 marked at about 100 metres below the surface. It might 

 be suposed to be the effect of a boundary wave along 

 the upper "discontinuity-layer". 



A great many observations were made at the Scot- 

 tish station at 10, 20, 30 and 50 metres, but they show 

 so many sudden variations in temperature and are taken 

 at such irregular intervals of time that it is impossible to 



perform the interpolations necessary for a harmonic ana- 

 lysis, except in the case of 50 metres. The analysis gives 

 the following equation for this depth: 



T^g = 8-72 + 014 cos 15= {t —20-7) + 0-34 cos 30° (t —0-9) 



It follows that the phase of the diurnal variation is 



not very different from that at the surface, but the maximum 



at 50 metres occurs nearly 4 hours later than at 100 



metres. The amplitude is 0-14, so that A =r 0-28, which 



is a smaller temperature variation than was found at the 



surface and 100 metres. It is, unfortunately, impossible 



dz 

 to determine ~j— for 50 metres in a satisfactory manner, 



but there are strong reasons to believe that it has a 

 lower value than at 100 metres, which leads to the con- 

 clusion that a possible vertical motion has possessed a 

 smaller amplitude at 50 metres than at 100. The diur- 

 nal wave should, therefore, have had its maximum 

 development at about 100 metres at the Scottish station 

 and not in the upper "discontinuity-layer" . The vertical 

 temperature-gradient is not particularly great at 100 metres. 

 It is much greater at 400 metres where, however, the ver- 

 tical displacements are small. 



The great amplitude at 600 metres at the Scottish 

 station is remarkable. The amplitude is quite insignificant 

 at 500 metres, as also at the deepest observations at Stat. 1 15 

 (500 metres, with an average temperature which is only 

 0-30° C. higher than at Stat. Sc, 600 metres), k, has also 

 a value at Stat. Sc. 600 metres which differs appreciably 

 from that at higher levels. It may suggest the possibility 

 of a particular oscillation in the deep-water (the "bottom- 

 water") of the Norwegian Sea, if it is not altogether casual. 



25. Semi-Diurnal Variations. 



The semi-diurnal variations at Stat. 115 are different 

 from those at the Scottish station with regard to both 

 amplitude and phase (see Fig. 2 and the table p. 27). 



We have already discussed the variations at the sur- 

 face. At 100 metres at Stat. 115 the amplitude of tem- 

 perature is but slightly higher than at the surface, and the 

 phase is very nearly the same. At the Scottish station 

 the temperature variation is also somewhat greater, but 

 the phase difference is 6 hours (180°). The harmonic 

 constants for 50 metres (the equation section 24) shows 

 the same phase as at 100 metres, but a much greater 

 temperature-amplitude, B being 0-68 against 0-37° C. The 

 conditions may suggest a boundary wave in the upper 

 "discontinuity-layer" at this station as far as the semi- 

 diurnal variations are concerned. 



At Stat. 115 k.^ increases all the way from 100 to 

 400 metres, at first rapidly, then more and more slowly, the 



