APPENDIX A: PETTERSSON _ 12 — 



land, Bornholm a. o. in autumn has been ascribed to this^ circumstance. But for the 

 open ocean, oiher considerations must be taken into account, and we must distinguish 

 between the real and the apparent debit of heat from the water. If we calculate the 

 difference in the quantity of heat stored up in a column of water of one square metre 

 diameter from the surface down to the under-limit of the Atlantic water at one of the 

 stations of the international investigations at different seasons of the year, we obtain 

 the apparent value of the exchange of heat between the sea and the air or the amount 

 of heat, which the water would have lost or gained supposing, that it had remained in 

 the same place between both observations. In order to know the real heat-debit, we 

 must also take into account the amount of heat, which has been brought to the spot with 

 the waters of the Atlantic current in the meantime. In want of reliable current obser- 

 vations, the following considerations may help to form an idea of the daily interchange 

 of heat between every square meter of the Atlantic area of the Norwegian Sea and the 

 atmosphere. 

 Heat exchange The heat glvcu up by the water is mainly absorbed as latent heat by the water, 



and ^^ which evaporatcs from its surface. According to the most reliable observations^ 

 Atmosphere 3-77 Kilogram water are given off from every square metre of the Norwegian Sea between 

 61° and 66° Latitude. The latent heat corresponding is 2260 units. To such circumstances 

 we must ascribe the fact, that the temperature of the air West of Lofoten is about 27° 

 higher in January than the normal, and likewise the retardation of the winter season, 

 which for that place goes so far, that the coldest day at the island of Rost falls upon 

 the 25* of February. 

 Correspond- But apart from this retardation of the seasons, it may be, that the climatic conditions 



ance between ^f northern Europe would show the same stability as those of the tropical countries if 



the temperature . ,. . r i a i • i i i i ■ i- i • 



of the air and the yearly periodicity of the Atlantic were not troubled by unperiodic perturbations, 

 of the sea ^^j-ga^jy before the commencement of the international investigation , the existence of 

 such perturbations had been proved. The following diagram (Fig. 8) shows the corre- 

 spondency between the oscillations of the surface temperature of the sea in January and 

 February during 30 years at the stations Utsire, Hellisö and Ona at the west coast of 

 Norway and the variations of the mean temperature of the air at Orebro in the central 

 part of Sweden 3. The full-drawn line represents the variations of the temperature of the 

 water, the dotted line those of the air. 



Besides such fluctuations in the amplitude of the period, there frequently occurs a 

 shifting in the phase of the period or a delay in the appearance of the maximum flow 

 of the Atlantic current, which can be delayed from November to January and even 

 longer, 

 influenceofthe The vei'y Opening-year (1902 — 1903) of the international investigation was typical of 



perturbations ^^^^^ abnormal conditions. The full-drawn Une in the diagram (Fig. 9, p. 14) represents the 

 normal annual variation of temperature in the surface-layer of the sea off the Norwegian 

 west coast in Thorshavn (Faroes), Papey (east coast of Iceland) and the named Norwegian 



1 O. Pettersson, Ueber die Beziehungen zwischen hydrographischen und meteorologischen Phäno- 

 menen. Meteorol. Zeitschrift. August 1896. 



2 Of Professor Mohn during the cruise of the Vöringen in 1876—78. 



3 It must be observed, that the mean temperature of every place in the southern part of Sweden 

 shows the same oscillations on a greater or smaller scale as those observed at Orebro. 



