tends to indicate that the adjustments were 

 caught prior to any low saHnity influence and 

 subsequent elevation of the trough region as 

 observed on the third survey. 



Data of the Banks salinity and volume flow 

 from past years, presented in Table IIA, do not 

 show good correlation with the concept of salinity 

 decrease on the Banks and increase in volume 

 flow. These data are from surveys which were 

 separated m time by a month or more and hence 

 cannot be expected to show a correlation due to 

 the lag time between the arrival of the low salmity 

 water on the Banks and the mixing of the water 

 in unknown proportions in the trough area. The 

 only way to observe this postulated salinity de- 

 crease and volume increase is to start the survey 

 before the water of lowered salinity completes its 

 swing towards the Tail-of-the-Banks and then 

 north to the trough area. At a current velocity 

 of 1 to 2 knots, less than a week is required. 

 Lateral mixing in section U is held to a minimum 

 because of the steep east-west slope of the a, 

 surfaces. This will be discussed in further detail 

 in the section on isentropic analysis. As the 

 current passes the trough area, some alteration 

 of the properties take place in the trough; how- 

 ever, the greatest changes occur while the water 

 completes its trip from north to south and then 

 north to the trough region. Data of previous 

 years do not provide any means for determming 

 short-term volume flow tendencies or the rate of 

 change of the salinity of the Banks water. The 

 significant flow changes observed durmg 1965 

 \\ere the result of closely timed, repeated sections. 



Future Work 



A greater understanding of the salinity-volume 

 flow relationships would enable a temperature- 

 salinity monitoring system on the Banks to uidi- 

 cate salinity reduction and volimae changes 

 occurring to the Labrador Ciu-rent. These could 

 liave implications in iceberg drift prediction. 



Weekly fluctuations oi the Labrador Current 

 might be superimposed on seasonal variations 

 which woidd tend to alias the data collected at 

 monthly intervals thus rendering year to year 

 comparisons misleaduig. Atten^pts will be made 

 during the spring of 1966 to more accurately 

 define the flo\\ changes and to determine wliether 

 or not these changes are of a short- or long-term 

 duration. 



ISENTROPIC ANALYSIS 

 Basic Concepts 



A method of observing water movements and 

 changes in property distribution for comparison 

 with dynamic heights is desired. It would be 

 advantageous to have such a comparison in the 

 hope of verifying and eliminating ambiguities in 

 dynamic height contour charts. Isentropic analy- 

 sis provides this independent method for deter- 

 mining water movement. Rossby et al. (1937) 

 pioneered isentropic analysis of the atmosphere. 

 They showed that atmospheric specific humidity 

 proved valuable for observing the distribution of 

 identifying properties on surfaces of equal density. 

 Parr (1938a) and Montgomery (1938) applied a 

 similar technique to the oceans and found it 

 extremely useful in tracing water mass source and 

 movements. Parr (1938a) made direct compari- 

 sons of isentropic charts with Ice Patrol dynamic 

 height charts and indicated that true water move- 

 ments at the southern end of the Grand Banks, 

 Newfoundland, could better be deduced from the 

 analysis of water circulation on density surfaces. 

 Montgomery (1938) extended his analysis to water 

 movements of the upper layers of the southern 

 North Atlantic Ocean and deduced source loca- 

 tions of the various density layers. 



The implementation of isentropic analysis calls for 

 the determination of the surfaces of constant entropy. 

 According to Montgomery (1938), this surface is 

 best approximated by the surface of constant 

 potential density which at pressures of less than 

 1,000 decibars is equivalent to surfaces of equal 

 sigma-t. This appro .ximation is necessary because 

 no surfaces exist in the ocean along which the 

 mi.xing of water masses can take place without 

 altering the mass distribution and therefore the 

 entropy or energy level of the system. 



A sigma-t or quasi-isentropic surface presenta- 

 tion is useful because flow patterns on these 

 surfaces represent the nearest approach to the 

 true conditions of movement. In addition, the 

 currents tend to flow parallel to the contour lines 

 of the sigma-t surface therefore providing informa- 

 tion on the direction of movement. These are 

 axiomatic truths because of the hydrodynamic fact 

 that flow will take place along levels of equal 

 density unless active mixing alters the energy 

 levels. Some departure will take place because 

 of the effect known as cabbeliiig ' where water 



I Smith (1931) describes this phenomena. Von Xx\ (1963) refers to it as 

 caballing. 



8 



