26 



PROCESSING OCEANOGRAPHIC DATA 



In any one study the contour interval should 

 be consistent throughout. The contour spacing 

 should normally be equal but if a particular 

 feature requires intermediate contours they 

 must contrast with the standard contours by 

 different weight or style of line. 



When a section is made up of a large number 

 of lines it is desirable to make every fifth line 

 heavier than the others. In this way individual 

 contours may bo followed more easily by refer- 

 ence to the heavier line. 



Charts showing the dynamic topography of 

 one isobaric surface relative to another, as 

 described in section D. 3 and shown in figure 21, 

 are typical of one method of presenting scalar 

 oceanographic data. A three-dimensional pic- 

 ture of relative currents may be obtained from 

 a series of such charts, each showing the 

 dynamic topography of a different isobaric 

 surface relative to the same reference surface. 



A similar three-dimensional picture would 

 also be obtained from a series of level charts, on 

 each of which are plotted the intersections of 

 isobaric surfaces. The methods are entirely 

 comparable, and either may be used to depict 

 the fields of any scalar quantity. This latter 

 method of showing isopleths of constant tem- 

 perature, salinity, etc., on equilevel sui'faces is 

 most popular because of its ease of construction. 



Vertical Sections 



In vertical sections the horizontal scale is fre- 

 quently distance or time. In either case the 

 scale should be continuous, regardless of the 

 spacing of plotted values. The vertical scale, 

 which represents a depth in the sea, shotild nor- 

 mally be continuous. In rare instances, if the 

 surface features are to be emphasized or in cases 

 where variables become nearly constant at great 

 depths, the vertical scale may change at some 

 depth such as 1,000 meters and the deeper scale 

 compressed. The break in scale should be con- 

 spicuously marked. 



Vertical sections usually depict variables 

 along a straight line of stations. If the stations 

 are not arranged along a straight line the direc- 

 tion of the section should be made clear. This 

 may be accomplished by an inset horizontal 

 chart with reference (station) numbers on both 

 inset and horizontal chart. Another means of 

 representing a change in direction of vertical 

 sections is by changing the direction of the 



plotted section on the page and utilizing pro- 

 jection techniques. This procedure, however, 

 gets involved if the section changes direction 

 several times. In illustration of the vertical 

 section, figure 26 shows isopleths of constant 

 specific volume anomaly, 8, plotted along the 

 section marked A — B on the chart of dynamic 

 topography, figure 21. Figure 27 shows iso- 

 therms in the same section. 



Both sections in figures 26 and 27, as well as 

 figure 21, are based upon the same group of 

 serial observations, and therefore must be 

 drawn in such a way as to be consistent. This 

 fact is obvious where the distribution of a 

 single variable, such as pressure, is illustrated in 

 more than one way. It is less obvious where 

 the distributions of different variables from the 

 same data are illustrated, and it is therefore 

 important to bear in mind the relations between 

 variables. As has been shown above, the slopes 

 of isobaric surfaces are closely related to the 

 distribution of mass. 



Surfaces of constant value of all the mass 

 variables slope in the same direction in the sea, 

 and, except for c, which represents density 

 reduced to a common pressure, form a single 

 family of surfaces. However, specific volume 

 and specific volume anomaly increase upward, 

 density and c, increase downward. Since the 

 mass field is largely dependent upon tempera- 

 ture, isothermal surfaces tend to slope in a 

 similar manner, and currents may usually be 

 deduced from charts of isotherms. 



ACKNOWLEDGMENTS 



The author is indebted to Dr. H. U. Sverdrup 

 for instruction in data processing throughout 

 many years, for preparation of an outline for 

 this report, and for the corrections to the 

 manuscript. Much credit is due Mrs. Anna 

 Strenk for computations of many of the tables, 

 to Miss Margaret Culbertson for assistance m 

 preparation of text and figures, and to Mr. 

 G. L. Prible for final drafting. Many thanks 

 are extended to the following people for con- 

 tributing published and unpublishetl data and 

 for constructive suggestions: D. F. Bumpus, J. 

 N. Carruthers, Townsend Cromwell, G. E. R. 

 Deacon, D. F. Leipper, H. Mosby, M. J. Pollak, 

 D. W. Pritchard, K. 0. Reid, and F. M. Soule. 



