400 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



declines after the primaiy peak in June-July, but 

 then rises again in August or September to reach a 

 secondary peak in Octobei^November. The ad- 

 vection chart for the October-November period 

 shows no material difference between the mean 

 isotherm pattern for the beginning of these months 

 and the mean intrinsic isotherm pattern for the end 

 of these months. High advection continued in 

 the western half of the region, with the principal 

 change from the previous chart occurring in the 

 southeast portion. There advection was low dur- 

 ing the June-July period and then increased to 

 1° C. mos."' in October, as shown in the charac- 

 teristic advection diagram of area B (chart V). 



The displacement speeds of the 26° and 27° C. 

 isotherm in the southern portion of the region are 

 as high as 0.9 knot, in the northwest portion they 

 are about 0.5 knot and, in the northeast portion 

 the 25° isotherm was displaced at a speed of about 

 0.2 knot. 



To explain this advection picture one can 

 postulate that the westerly flow of the June-July 

 period slackened and shifted to the southeast 

 portion of the region. This is coincident with the 

 intrusion of lower salinity water into that area, as 

 indicated in figure 9 showing the seasonal salinity 

 variation at 13° N. One can further postulate 

 that with the relaxation and southward shift of the 

 flow from the east, higher salinity water from the 

 northwest reoccupies the region from which it had 

 been displaced during the primaiy advection 

 period. This, again, is in agreement with the 

 salinity data as shown in figure 8. 



The final advection period, as indicated by the 

 characteristic advection diagrams, occui-s during 

 December and January in all the areas except in 

 area A4 where it is delayed until February and 

 area B where it is absent. The mean temperature 

 for December and January at the beginning of 

 these months, and the mean intrinsic temperature 

 for the end of these months are beginning to 

 assume the distribution of the March-April period. 

 The chart shows displacements of the 25° and 26° 

 C. isotherms in the central portion of the region 

 of up to 360 miles or at a rate of about 0.5 knot. 

 To the north, the 24° C. isotherm is displaced by 

 up to 120 miles or at a rate of less than 0.2 knot. 



The primary advection during June and July 

 can be associated with an intensification of the 

 tradcwind system and the secondary period during 

 October and November with its relaxation. 



However, there appeare no wind pattern which 

 could be associated with the December-January 

 advection peak. In the vicinity of the main 

 Hawaiian Islands the salinity reaches a peak 

 during the December to February period and 

 northeast of the Marshall Islands the horizontal 

 temperature distribution at 400 ft. (Robinson, 

 1954) suggests intensified circulation. This points 

 to an inertia! surge in the high salinity circulation 

 sj'stem as the explanation for the advection peak. 

 Although the water movement as suggested by 

 the advection periods has not been observed 

 directly, present knowledge of the general Pacific 

 circulation, and the sketchy salinity data, are 

 compatible with the advection model. In the 

 next section, therefore, it will be shown that the 

 surface temperature can be used to monitor 

 physical processes. 



C. Characteristic Heating Curve 



The simplified heat budget, equation 6, shows 

 the rate of change of temperature to be a function 

 of the two independent variables; the heat ex- 

 change across the sea surface and the heat advec- 

 tion. Since the seasonal variation of the heat 

 exchange across the sea surface and the seasonal 

 variation of the advection are characteristic of 

 the oceanographic climate at any location, the 

 dependent, seasonal variation in the rate of change 

 of temperature should also be characteristic of the 

 location. A graph of the seasonal rate of change of 

 temperature can therefore be called the character- 

 istic rate of change temperature curve, or, simply, 

 the "characteristic heating curve." 



The characteristic advection diagrams of chart 

 V show, with the exception of areas B and Bj, that 

 the heat exchange curves have a regular seasonal 

 shape, with a maximum in June and a minimum in 

 December or January. Changes in advection 

 should therefore be reflected in irregularities of the 

 characteristic heating curves. For example, in 

 April the change of slope in the characteristic 

 heating curve of most areas in the survey region 

 (chart V) indicates a change from negative (warm) 

 advection to positive (cold) advection. In the 

 northeastern portion of the survey region (areas 

 A4 and B3), the pronounced dip in the characteristic 

 heating curve during April and May signifies an 

 initial surge of cold advection, which then slackens 

 during May and June, as indicated by the rapid 

 rise of the rate of change of temperature. 



