FISHERY BULLETIN: VOL. 72, NO. 4 



Table 1. — List of stations used in Figures 1-7. 



'a: STD digital data logger; b: STD analogue ctiarts; c: Nansen-bottle casts. 



Temperature and salinity data used for comput- 

 ing geopotential anomaly were collected with in 

 situ salinity-temperature-depth recorders (STD) 

 except for those from EASTROPAC cruise David 

 Starr Jordan 30, on which a breakdown of the STD 

 necessitated the use of Nansen bottles during the 

 last half of the cruise. Most of the STD data were 

 obtained from digital data loggers, but some were 

 digitized from analogue charts (Table 1). The 

 method of processing STD data is described in the 

 EASTROPAC atlas (Love, 1972a). 



From these temperature and salinity data, 

 geopotential anomaly at the sea surface was com- 

 puted with reference to 500 db. Because of rela- 

 tively homogeneous water at depths greater than 

 500 m in the intertropical ocean, this reference is 

 believed to be adequately deep for estimating the 

 geostrophic current at the sea surface. The unit of 

 geopotential anomaly is chosen to be joule per 

 kilogram (abbreviated J/kg; equivalent to 

 dynamic decimeter). 



Computed geopotential anomaly was plotted on 

 each map, and smooth isopleths were drawn at 

 intervals of 1.0 or 0.5 J/kg. The base map is a 

 Mercator projection and was adapted from U.S. 

 Navy H. O. 526 and 823. 



THE SURFACE GEOSTROPHIC 

 FLOW 



The surface geostrophic flow revealed by the 



1076 



maps of geopotential anomaly is described below. 

 A comparison of these maps with other maps 

 based on long-term averages of set and drift of 

 ships is also made. The atlas of the monthly aver- 

 age surface currents in the eastern North Pacific 

 published by the U.S. Navy Hydrographic Office 

 (H. O. 570, 1947) and the atlas of the quarterly 

 average surface currents in the South Pacific pub- 

 lished by the Meteorological Office (M. O. 435, 

 1939) are pertinent to the comparison. There are 

 also monthly drift charts by Cromwell and Ben- 

 nett ( 1959) for the northern hemisphere and those 

 by Puis (1895) and Wyrtki (1965) for both hemi- 

 spheres. The charts by Cromwell and Bennett are 

 simply a different presentation of the H. O. 570 

 charts for the area east of long. 120°W and south of 

 lat. 30°N. Puis' charts show the current direction 

 and relative intensity for the area between lat. 

 20°N and 10°S. Wyrtki's charts are based on aver- 

 ages over 1-degree squares, but it is not clear how 

 he smoothed or interpolated the original 1-degree 

 averages to obtain the current patterns shown on 

 his charts. 



Various sources of errors and disparities in the 

 geostrophic calculation and set-and-drift observa- 

 tions should be kept in mind. Geopotential anom- 

 aly computed from oceanographic data may in- 

 clude short-period density fluctuations that are 

 not necessarily associated with fluctuations of the 

 actual current. The calculation of geostrophic flow 

 neglects the direct frictional effect of wind stress. 



