20 



BJ0RN HELLAND-HANSEN 



[REP. OF THE "MICHAEL SARS" NORTH 



IV. LOCAL VARIATIONS IN GENERAL. 



19. Variations limiting the Validity of 

 Oceanographical Observations. 



A question of great importance is, whether or not it 

 is allowable to attribute a general and extended represen- 

 tative value to serial observations from a single station. 

 Variations occur in the sea which restrict the validity of 

 serial observations to the exact time and locality of the 

 measurements. Although it is a problem which is en- 

 countered in most cases when observations of temperature 

 and salinity are to be discussed it is as yet very far from 

 being solved. In the oceanographical literature far-reaching 

 conclusions are sometimes drawn from single series of 

 measurements although a short displacement in locality, 

 or observations at another time, might have given quite 

 different results. 



It is a well known fact that the deep-water (the "bottom 

 water") forming the greater part of the water-masses in the 

 Ocean is remarkably constant in its physical properties. 

 The local and temporal variations are essentially restricted 

 to the upper water-layers above some 2000 metres below 

 the sea surface, and are in many regions displayed only 

 in the strata much nearer to the surface. Temperature, 

 salinity and density of the upper water-layers as observed 

 from a point fixed relative to the sea-bottom are not 

 only subject to seasonal and annual changes in a vari- 

 able degree, but in many places evidently also to rapid 

 changes, within some hours or days. The question of 

 these temporal variations forms an interesting problem in 

 itself, but may also be of importance to an ordinary 

 geographical study of the distribution of temperature, 

 salinity and currents. In the latter case the importance 

 depends upon the local variations in the physical properties 

 of the water-masses. 



It has previously been a common idea that the 

 horizontal variations in temperature, salinity and density 

 at a certain moment have been fairly evenly graded over 

 relatively wide stretches in the great oceans except at some 

 few places, as for instance along the border of the "Gulf 

 Stream" near the coast of the U. S. A. and the Newfound- 

 land banks. This view of the conditions was quite reason- 

 able as long as the observations were scattered over large 

 areas, with long distances between the stations of obser- 

 vation and, consequently, with interpolations for wide 

 gaps. As, however, the number of observations has 

 increased and a dense net of stations has been worked 

 from some deep-sea areas, it has proved that the variations 



in the physical properties and in the movements of the 

 water-layers on the top of the deep-water are often much 

 greater than was previously assumed. In many regions 

 the conditions are often much more "irregular" than they 

 at first seemed to be. Relatively short displacements of 

 the waters may often occasion quite different results with 

 regard to the distribution of temperature, salinity and other 

 elements of observation. A small variation in the posi- 

 tion of the stations may then also lead to mistakes as to 

 the oceanographical changes. It is, therefore, necessary 

 to examine the local variations in the sea before we pro- 

 ceed to a discussion of the temporal variations. 



20. Local Variations. 



At some distance from the continents the vertical 

 variations in temperature, salinity and density from the 

 sea-surface down to considerable depths are generally very 

 small in winter. In summer the heating at the surface 

 causes great variations in temperature and density down to 

 some 50 or 100 metres. Further downwards the variations 

 are mostly smaller. They may in some regions show a 

 marked increase again in the lower water-strata on ap- 

 proaching the uniform deep-water. The isotherms and iso- 

 pycnals exhibit in summer a "step", so to speak, a little below 

 the surface, and sometimes another one, but less pronoun- 

 ced, some distance further down (see for instance Stats. 

 65 and 68, pp. 70* and 71*). The lower "step" is also 

 found in the isohalines. In coastal waters the salinity is 

 often relatively low in the uppermost layers so that great 

 vertical variations in salinity and density may occur all 

 the year round. This is also the case in those regions 

 where Polar currents are met with, as for instance at the 

 Newfoundland Bank (cf. Stats. 70—74, p. 71*). 



We have seen (section 18) that there is upon the whole 

 a marked parallelism between temperatures and salinities. 

 This means that the isotherms and isohalines have much 

 the same course, which is also more or less parallel to 

 the course of the isopycnals. In the eastern part of the 

 North Atlantic some irregularities are caused by the water 

 coming from the Mediterranean. Here the temperatures 

 and salinities at intermediate depths exhibit considerable 

 vertical variations which differ from those in other regions, 

 while the density-variations show no similar peculiarities 

 (see for instance the curves for Stats. 23 and 25 B, p. 66*). 

 Disregarding such irregularities it may be stated as a 



