SAUR: SEA LEVEL DIFFERENCES 



Table 2. — Determinations of long-term mean sea level differences from 0/1,000 db geopotential (steric) anomalies 

 at nearby offshore locations. Correlations, r, of monthly sea levels with geopotential anomalies. 



Shore station 



Da to 

 source 



Location of hydrogrophic 



stations 



(see Figure 3) 



Distance from 



shore station 



(km) 



No. of 

 observations 



Average steric 



anomaly, 0/1,000 db 



(dyn cm) 



Correlation, r 

 steric vs. sea level 



1 0/500 db overage plus 49.0 dyn cm for 500/1,000 db overage, based on TWZO data and 1,000 m stations. 



2 0/500 db average plus 45.9 dyn cm for 500/1,000 db average, based on 37 stations in same 1-degree square. 



3 0/400 db average plus 57.9 dyn cm for 400/1,000 db average, based on 25 stations in same 1-degree square. 

 * Correlotion significant at 5% level. 



** Correlation significanf at 1% level. 



the cause of the lower correlation for observa- 

 tions in 1-degree squares 93 and 95. 



The correlations between monthly sea levels 

 and offshore steric levels for the four stations 

 in this study are less than anticipated from the 

 results reported by other investigators, e.g., 

 Shaw and Bonn (1964) and Sturges (1966), as 

 noted in the section. Earlier Studies. It is sus- 

 pected that the lower correlations for the four 

 Pacific stations in the present study may be due 

 in part to comparison of a monthly sea level with 

 a steric level from a single hydrographic obser- 

 vation on a given day. Present available steric 

 data explain only about 40 9r of the variance of 

 monthly sea levels from which the trend has been 

 removed and which have been adjusted to normal 

 pressure. As suggested by Stommel (1958) in 

 relation to sea level at Bermuda, the possibility 

 exists that geostrophically balanced barotropic 

 currents exist which would also affect the month- 

 ly mean sea levels. When time series of frequent 

 hydrographic observations or long-term obser- 

 vations from buoys become available to give a 

 better estimate of continuous changes in steric 

 level, we may then be able to attain a better 

 understanding of the causes of sea level varia- 

 tions. For the present the unexplained varia- 

 bility may limit the use of sea levels and sea level 



differences to interpretation of broad-scale, high- 

 ly anomalous oceanic changes. 



NONSEASONAL SEA LEVEL 

 DIFFERENCES 



The monthly sea level differences used in this 

 study are given in Appendix Tables 1 and 2 for 

 Honolulu-minus-San Francisco and Hilo-minus- 

 Avila, respectively. For each station the linear 

 trends were removed and values adjusted to 

 normal atmospheric pressure. The monthly dif- 

 ferences reflect the adjustment, previously dis- 

 cussed, of the long-term mean sea level difference 

 for all months and years to a mean difference 

 determined from 0/1,000 db geopotential heights. 



VARIABILITY 



The nonseasonal variations are represented by 

 the monthly anomalies, defined as the difference 

 between the value for the month and the normal 

 for the same month. The standard deviations 

 in Table 1 for given months at individual stations 

 exhibit the same characteristics as those reported 

 in an earlier study (Saur, 1962) for six Pacific 



627 



