406 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



Since no direct measurements of this term were 

 available, it was estimated by subtracting the 

 rate of change of surface temperature from the 

 rate of change of temperature caused by the net 

 heat exchange across the sea surface. The advec- 

 tion can also be obtained graphically by measuring 

 the difference between the seasonal rate of change 

 of the temperature curve and the seasonal heat 

 exchange curve, as in the advection diagrams of 

 figure 22. This method was applied to three- 

 degree square areas in the Hawaii region and 

 revealed four distinct advection periods. The 

 first period during March and April was one of low 

 or warm advection and the others were periods of 

 cold advection during June-July, October-No- 

 vember, and December-January. 



The advection diagrams also suggested different 

 climatic zones. In the northwest portion of the 

 region, area A of chart V, the primary advection 

 peak occurs during June-July and a secondary 

 peak during December-January. In the south- 

 east portion of the survey region, area B of chart 

 V, the primary advection peak occurs during 

 October-November and the June-July and De- 

 cember-January periods are absent. In the 

 intermediate areas the diagrams of chart V exhibit 

 varying magnitudes of these advection peaks, 

 which suggest a transition from one to the other 

 climatic zones. Area B also corresponds approxi- 

 mately with the area south of the depth of mixed 

 layer boundary in which the times of maximum 

 and minimum depths differed from those to the 

 north. Thus, the depth of mixed layer boundary 

 and the advection peaks appear to be closely 

 associated with seasonal changes in water motion. 



In order to illustrate the phj'sical meaning of 

 heat advection, the change of temperature caused 

 by advection can be added to the temperature 

 at the beginning of the month in order to obtain 

 the "intrinsic" temperature at the end of the 

 month. If the two temperatiu-e distributions are 

 then plotted on the same chart, the displacements 

 of isotherms are then equivalent to boundary 

 movements, as shown in chart VI, for the four 

 advection periods in the Hawaii region. Of 

 particular importance are the June-July and 

 October-November charts, since the independently 

 observed salinity changes mentioned above both 

 support and supplement the information obtained 

 from heat budget considerations. Fu-st, within the 

 island area, the June-July advection period coin- 



cides with the northward motion of the salinity 

 boundary and the declining salinity which reaches 

 a minimum in July. Later, the October-November 

 heat advection period coincides with the south- 

 ward retreating salinity boundary and increasing 

 salinity within the islands. In addition, the 

 October-November advection peak in the south- 

 east portion of the survey region coincides with 

 the rapid salinity decline at 13° N. The salinity 

 decline at 21° N. and at 13° N. can only be 

 explained by salt advection, since, at 21° N., 

 evaporation mmus precipitation is positive 

 throughout the year and increasingly positive 

 at 13° N. during November (fig. 19). 



The displacement of the 26° C. "intrmsic" 

 isotherm and the spring and autumn movement 

 of the salinity boundaries as mdicated bj^ the 

 35 °/oo and 34 °/oo isopleths are illustrated in 

 figure 28. The June-July displacements (fig. 

 28A) are best explained by an intensified westward 

 component of flow between 15° and 25° N. The 

 October-November displacements (fig. 28B) are 

 best explained by an intensified southward com- 

 ponent of flow m the northern half of the region, 

 and an intensified westward component of flow 

 in the southeastern portion. 



It is now apparent that the trough of low advec- 

 tion during the cold advection period in figure 23A 

 is associated with the transition zone between two 

 clunatic regions. In addition, one can postulate 

 that the trough coincides with the core of the 

 California Current Extension and that the areas 

 of high advection on both sides of the trough are 

 the areas of the seasonally moving current bound- 

 aries. The latter can also be expressed as the 

 areas through which the boundaries of the season- 

 ally dilating and contracting North Pacific 

 Central and North Pacific Equatorial sj^stems 

 move. 



The surface temperature distribution and its 

 seasonal changes primarily reflect the seasonal 

 changes m the heat exchange across the sea 

 surface. Features in the temperature distribu- 

 tion which may be due to the surface circulation 

 are therefore obscured, except for the tongue- 

 shaped area of lower temperature protruding 

 westward south of the islands durmg the summer 

 months (chart II), reflecting the increased west- 

 ward flow. 



The surface salmity distribution, on the other 

 hand, appears to be more closely related to the 



