2 The Water Masses of the World Ocean Some Results of a Fine Scale Census L V Worthington 2 1 Introduction In his original brief monograph Helland Hansen 1916 introduced the concept of a water mass as being defined by a temperature salinity T S curve He found that over a large area of the eastern North Atlantic a normal T S curve could be drawn He showed that variations from this curve could be attributed to the intrusion of alien water masses that had originated elsewhere The use of the T S diagram has been almost universal in physical oceanography since Helland Hansen introduced it It is not only a powerful descriptive tool but observers at sea routinely plot T S diagrams and use them as a check on the tightness of their sampling bottles and the correct function of their thermometers The term water mass has been very loosely used by numerous authors According to Sverdrup Johnson and Fleming 1942 a water mass is defined by a seg ment of a T S curve and a water type by a single value of temperature and salinity that usually falls on a T S curve Thus a T S curve is made up of an infinite number of water types These definitions will be adhered to in this chapter as far as is possible Oceanographers have used other methods to describe the ocean both before and after Helland Hansen s introduction of the T S diagram and these methods will be briefly discussed below but I will deal primarily with the world water masses as defined by T S diagrams The most important advance in water mass analysis since Helland Hansen came with the introduction by Montgomery 1958 Cochrane 1958 and Pollak 1958 of the volumetric T S diagram in which the volumes of all the world water masses were estimated The volume of the world ocean including adjacent seas is 1369 x 106 km 3 Montgomery Cochrane and Pollak were able to divide the individual and world oceans into bivariate classes of temperature and salinity each of which contained an assigned volume For example the most abundant class found by Montgomery 1958 in the world ocean was T 1 0 1 5 C S 34 734 8 o he calculated that this relatively small class contained 121 x 106 km 3 or 9 of the water in the ocean Wright and Worthington 1970 produced a volumetric census of the North Atlantic that was a direct descendant of Montgomery s 1958 work This later census was motivated by the introduction pioneered by Schleicher and Bradshaw 1956 of the very accurate salinometers based on the measurement of electrical conductivity The precision of data obtained with these salinometers enabled Wright and Worthington 1970 to divide the North Atlantic into much smaller classes than Montgomery and his colleagues had used Wright and Worthington s smallest class below 2C was 0 1 C x 0 01Oo Fifty of these classes make up one of 42 L V Worthington Montgomery s classes 0 5 C x 0 1 o This fine scale census had clear advantages over the coarser scale census that inspired it and in consequence I undertook a census of the world ocean water masses using the finescale classes that Wright and Worthington 1970 introduced The greater part of this paper will be devoted to the presentation of the results of this census with some discussion of the formation of these water masses 2 2 Methods of Describing the Oceans The simplest and the most universally used method of describing the oceans has been the preparation of vertical profiles of temperature salinity dissolved oxygen or some other variables constructed from oceanographic sections made across an ocean or part of an ocean from a ship or a number of ships Ocean wide temperature profiles have been drawn by oceanographers since Thomson s 1877 treatment of the Challenger sections but the standard of excellence for this kind of presentation was set by Wiist and Defant 1936 in their atlas of the temperature salinity and density profiles from the Atlantic Meteor expedition of 19251927 and by Wattenberg 1939 who prepared the oxygen profiles These vertical profiles were drawn in color with detailed bottom topography provided The atlas by Wist and Defant 1936 provided the model for Fuglister s 1960 atlas of vertical profiles of temperature and salinity from the transatlantic sections made by various ships and observers during the International Geophysical Year Later Worthington and Wright 1970 drew similar profiles for sections made by the Erika Dan in the northern North Atlantic in 1962 They also included dissolved oxygen profiles modeled on those of Wattenberg which Fuglister had been unable to do because of the poor quality of oxygen analyses made from Woods Hole ships during the International Geophysical Year Probably the finest example in this form is that of the vertical profiles by Stommel Stroup Reid and Warren 1973 for the transpacific sections at 28 S and 43 S from Eltanin in 1967 These profiles are shown in six color plates the variables are temperature salinity oxygen phosphate nitrate and silicate The station and sample bottle spacing for these sections were carefully planned so as not to miss any important baroclinic gradient or variation in nutrient concentration Composite vertical profiles are often drawn from data provided by a number of ships from different years or even different decades Such sections are of course less useful for dynamical studies but sometimes provide an excellent description of the water A fine example is that of Wiist s much cited north south temperature salinity and oxygen profiles in the Atlantic Wiist 1935 plate XXIII Reid 1965 used the same method to construct zonal and meridional profiles of temperature salinity oxygen and phosphate across the Pacific Helland Hansen and Nansen 1926 drew vertical profiles of temperature and salinity superimposed on each other in their work on the eastern North Atlantic This method has been followed by others notably Tait 1957 I find such profiles difficult to read but that may be idiosyncratic Helland Hansen and Nansen 1926 also introduced vertical profiles of the anomaly of salinity in this they were followed by Iselin 1936 My own preference Worthington 1976 figures 18 19 and 20 is to draw vertical salinity anomaly charts preferably in color with selected isotherms included I feel that this method illustrates most clearly the vertical juxtaposition of different water masses As Wuist 1935 remarked horizontal charts of vari ables are of limited use in water mass analysis because the various layers rise and sink above and below a fixed level Nevertheless the best early
View Full Document
Unlocking...