MONTHLY SEA LEVEL DIFFERENCES BETWEEN THE HAWAIIAN 

 ISLANDS AND THE CALIFORNIA COAST 



J. F. T. Saur' 



ABSTRACT 



Time series of monthly sea level differences, adjusted for linear trends and the isostatic 

 effect of atmospheric pressure, are computed for Honolulu-minus-San Francisco (1905- 

 69) and Hilo-minus-Avila (1947-59 and 1961-67) as an index- to broad scale changes in 

 current around the eastern limb of the North Pacific anticyclonic current gyre. 



The normal seasonal cycle of sea level differences imply a net southeastward surface 

 current that is strongest from April through October and weakest from December through 

 February and the range is about 20% of the average. Nonseasonal differences for Hono- 

 lulu-minus-San Francisco and Hilo-minus-Avila show a highly significant correlation 

 over the 240 months of coincident records. Inferences regarding nonseasonal large-scale 

 long-term changes in geostrophic current are drawn from the 65-year Honolulu-minus- 

 San Francisco smoothed monthly anomalies in which variability of less than 6 months was 

 suppressed, as follows: 1) nonseasonal changes in current speed exceed ± 10% of the 

 normal current over one-third of the time and range up to 54% of the normal monthly 

 current, 2) the periods 1922-38 and 1950-54 were eras of low variability as compared 

 with greater variability in the rest of the record, and 3) periods of weakest circulation 

 were in 1911, 1918, 1941, 1957-58, and 1967, and of strongest circulation in 1915, 1920-21, 

 1943-44, 1948-49, and 1959. 



For more than a decade there has been a rapidly 

 growing interest of fishery scientists and biolog- 

 ical oceanographers in the role of the environ- 

 ment in fishery problems. One objective is to in- 

 clude environmental effects in models of popula- 

 tion dynamics and in fishery forecasting pro- 

 cedures. However, to do this, environmental 

 characteristics which are significant to fisheries 

 must be determined and set forth in quantitative 

 form. One empirical approach to these complex 

 environmental problems is to compile historical 

 oceanic and atmospheric data into time series 

 to gain a knowledge of the mean seasonal cycles 

 and variability, identify periods of highly anom- 

 alous environmental conditions for further 

 study, and seek an understanding of cause-ef- 

 fect relations which bring about the observed 

 changes. Such an understanding would be use- 

 ful for efficient monitoring of the oceanic envi- 



' National Marine Fisheries Service, Southwest Fish- 

 eries Center, La Jolla, CA 92037. 



ronment and in predicting environmental con- 

 ditions. 



Oceanographic data with good time and space 

 distribution for time-series studies of environ- 

 mental changes of a few months to several years 

 are woefully scarce. Only sea-surface temper- 

 ature observations taken by ships as a part of 

 the marine weather observations have long-term 

 continuity in time as well as oceanwide distri- 

 bution. As a result, these have been used ex- 

 tensively in air-sea interaction studies and to 

 indicate oceanic changes. 



To augment the sea temperature data, mean 

 sea levels compiled from continuous tidal records 

 at coastal and island stations provide another 

 source of long time-series data on ocean varia- 

 bility. The statistical characteristics and inter- 

 relations of the sea levels, atmospheric pressure, 

 and temperature of many coastal and island sta- 

 tions in the Pacific have been described in several 

 papers by Roden (1960, 1963, 1966) using auto- 

 correlation, spectral, and coherence techniques. 



Manuscript accepted April 1972. 



FISHERY BULLETIN: VOL. 70, NO. 3, 1972. 



619 



