NOAA - NATIONAL MURINE FISHERIES SERVICE 



PACIFIC ENVIRONMENTAL GRDUP 



MONTEREY. CALIFORNIA 



NET HERT EXCHANGE 



1 UATTS/M««2 ) 



LONGSHORE TIMESERIES GRIO SECTIDN I 1 ) 



LONG TERM MEAN 



JRN FEB liRR RPR MAY JUN JUL RUG SEP OCT NOV DEC 



--.'■-":■ - ■ -- ..'..._-.:■ ;e= ::~ ■■:. :-: : 



1 14°W; lat. 27°N, long. 1 15°W). A 23-yrtime series (1950-72) of 

 monthly values was computed for each 1 ° square. For these partic- 

 ular grid points, the computed values were based on an average of 

 10 to 30 observations/mo. Spectrum analysis was used to demon- 

 strate the predominance of the annual cycle and to possibly identify 

 other frequency bands accounting for substantial variance. 



Power spectra of Q N at each location (Fig. 21A, B) show strong 

 peaks at frequencies representing the annual cycle and suggest 

 lesser peaks corresponding to a semiannual component. Referring 

 to the 95% confidence limits for 10 degrees of freedom, it is clear 

 that the variance associated with the annual cycle is at least an 

 order of magnitude larger than the average level of variance at 

 lower or higher frequencies. Within the Southern California Bight 

 (Fig. 21A), no significant long period fluctuations are apparent. 

 Near Punta Eugenia (Fig. 2 IB), there is a general increase in spec- 

 tral energy density at low frequencies, from a relative minimum at a 

 frequency corresponding to a 2-yr cycle to a relative maximum for 

 periods exceeding 8 yr. This feature is also evident in power spectra 

 of monthly anomalies ofQ x (Fig. 21C), for which the seasonal var- 

 iation in the time series has been removed by subtracting out the 

 long-term (1950-72) monthly mean values. The increase in vari- 

 ance at low frequencies reflects the general 2- to 6-y r persistence in 

 the sign of the annual Q s anomalies at these locations. 



Time series of net heat exchange may be characterized by large 

 interannual differences, persistence, and long-term trends. However, 

 these features may not be spatially coherent. Coherence functions were 

 computed to test the significance of correlations between time series of 

 Q N in adjacent 1 ° squares. In the Southern California Bight (Fig. 2 1 A), 

 coherency squared consistently exceeds the 95 % significance level for 

 10 degrees of freedom only in the frequency band from 0.06 to 0. 1 1/ 

 mo (0.7 to 1.3/yr). The spread over several frequency bands reflects 

 the nonsinusoidal shape of seasonal fluctuations which tend to have a 

 faster rate of change in spring and fall than in summer or winter. Near 

 Punta Eugenia (Fig. 21B), coherency squared is significant (95% 

 level) at frequencies corresponding to semiannual and annual periodic- 

 ities and at frequencies <0.03/mo (0.4/yr). Low frequency coherence 

 remains significant at the 95 % level after seasonal variations have been 

 removed (Fig. 21C). The lack of significant coherence between adja- 

 cent squares in the Southern California Bight probably reflects the 

 large uncertainty in net heat exchange calculations based on surface 

 observations archived in the TDF-1 1 file, but also may be related to 

 significant spatial variations over distances <200 km in this region. 



These data demonstrate the relatively large amplitude of the 

 annual cycle in midlatitude atmospheric properties, which suggests 

 that the long-term monthly mean heat exchange fields (Appendix I) 

 represent the expected seasonal variability. More extensive time 

 series calculations have been published by Clark et al. (1974). 

 However, the wide variations and lack of persistence in patterns of 

 net heat exchange anomalies computed from their data, in addition 

 to our own results, support the conclusion that merchant vessel data 

 may not always be sufficiently reliable to compute statistically sig- 

 nificant indices of the month-to-month or interannual variations in 

 air-sea interaction processes (Husby 1980). 



Figure 20.— Seasonal cycle of nel heal exchange near the coast. Means 

 of nel heat exchange were computed by month for the I " squares imme- 

 diately adjacent to the coast. Units are W/m 2 . The contour interval is 

 50.0 \\ m-. Negative values are shaded and indicate nel heat loss from 

 ocean to atmosphere. 



fornia Bight (lat. 32°N. long. 117°W; lat. 32°N. long. 118°W;lat. 

 32°N, long. U9°W) and near Punta Eugenia (lat. 27°N. long. 



Heat Budget of a Coastal Upwelling Region 



The principal processes affecting the seasonal variation of air-sea 

 heat transfer over the California Current are related to seasonal 

 changes in the atmospheric circulation patterns which modify the 

 cloud cover and the oceanic circulation. Coastal upwelling has a 

 marked effect on the turbulent fluxes of heat, particularly off north- 



28 



