OCEANOGRAPHIC CLIMATE OF HAWAIIAN ISLANDS REGION 



401 



Again, in most areas of the survey region, the 

 rapid decline in the characteristic, curve, after 

 reaching a peak in May-June, is associated with 

 the June-July advcction peak. The slackening of 

 the primary advcction is reflected in a change of 

 slope in the characteristic heating curve or a 

 secondary peak as in areas Aj, A3, A4. 



In areas A, A,, Aj, A4, and B3, both the October- 

 November and the December-January (February 

 for A4) dips in the characteristic heating curves 

 can be interpreted as advectiou peaks at those 

 times. In areas Bj and B2, these advcction periods 

 overlap in such a way as to produce only a single 

 minimum in the characteristic heating curve. 



Generall}^, the characteristic heating curve is 

 expected to be sensitive to cold advcction during a 

 period of rising heat exchange and most sensitive 

 to warm advcction during declining heat exchange. 

 During the winter months in areas where heat ex- 

 change is small, the characteristic heating (cool- 

 ing) curves would be very sensitive to changes in 

 advcction. This is illustrated in the characteristic 

 advcction diagram of area Ai, where the secondarj^ 

 advcction peaks produce large changes in the char- 

 acteristic heating cm-ves during October-Novem- 

 ber and December-January. 



The characteristic advection diagrams of chart 

 V are useful in discovering gross climatic features 

 and in'delineating climatic regions. More useful, 

 however, are the characteristic heating curves 

 drawn from data regularly collected at fixed moni- 

 toring stations, particularly if featm-es of the curve 

 can be related to other contmuously monitored 

 events or processes. Gross climatic processes, as 

 discovered in the characteristic advection curves, 

 should again be reflected in the shape of the heat- 

 ing curves. 



Figure 25 represents the characteristic heating 

 curve as drawn from mean data collected in the 

 vicinity of Koko Head, Oahu, from 1951 to 1958. 

 It differs from that of figure 22B primarily during 

 the winter months, when it forms only one mini- 

 mum in December, whereas tlie latter forms a 

 minimum m November-December and another 

 dip in the characteristic temperature curve during 

 January-February. The March-April and the 

 June peaks are in phase on both diagrams. These 

 differences are to be expected, since the curve of 

 figure 22B represents data collected in a three- 

 degree square covering both sides of the island 

 chaui. The Koko Head cm-ve, on the other hand, 



o 



JAN MAR mX JUL SEP NOV JAN MAR 



Figure 25. — Mean (1951-58) characteristic heating curve 

 for the vicinity of Koko Head, Oahu. 



represents data collected only at a point on the 

 "island barrier" within a large circulation system. 



The mean characteristic heating curve at Oahu, 

 then, shows that the rate of change of temperature 

 becomes positive at the end of February, indicating 

 warm advection. The commencement tune is 

 similar to that in area A2, just to the west of Oahu. 

 The rapid rise m the rate of change of temperature 

 dm-ing May can be interpreted as a period of low 

 advection, as in figure 22B where the rate of change 

 of temperatm-e is the same as the net heat exchange 

 across the sea siu-face. Since, during this time of 

 the year, the isotherms are approximately parallel 

 to the islands (see May temperatiu-e chart II), 

 this means that the flow is also parallel to the 

 island chain. Then, again as in figure 22B, the 

 rate of change of temperature reaches its maximum 

 in June, slightly later than m area A2 and earlier 

 than in area B3. This is also apparent in area A4, 

 where a combination of maximum rates of change 

 of temperatures occurring either duruig May- 

 June or June-July produce the wide peak in the 

 cliaracteristic heating cm"ve. 



Since the flow during the June-Jul}' advection 

 period is from east to west, at a low angle of in- 

 cidence within the island chain, one can postulate 

 that its deflecting effect is at a maximum. How- 

 ever, as the incident flow acquires a southerh^ 

 component and the angle between it and the island 



