DaUy and Weekly Upwelling Indices, West Coast of 

 North America, 1967-73 



ANDREW BAKUN' 



ABSTRACT 



Daily and weekly indices of intensity of large-scale wind-induced coastal upwelling at selected 

 locations along the west coast of North America are presented for the 7-yr period, 1967-73. The 

 indices are based on 6-hourly computations of the offshore component of Ekman transport using the 

 synoptic surface atmospheric pressure analyses produced by the Fleet Numerical Weather Central to 

 estimate the sea surface stress. The magnitude of offshore transport is considered an indication of result- 

 ant upwelling through the bottom of the Ekman layer. A spatial distortion in absolute magnitude results 

 in noncomparability of numerical values between different locations. 



INTRODUCTION 



Upwelling is a highly variable phenomena which can 

 have important effects on fishery resources. As an aid in 

 accounting for these effects, Bakun (1973) presented 

 monthly indices of intensity of wind-induced coastal upwell- 

 ing at intervals of three degrees latitude along the west 

 coast of North America. This choice of time scale recognized 

 that although important variations in the upwelling process 

 are known to occur on time scales shorter than a month, 

 specification on shorter scales often may not be required or 

 even desired. Some major effects of upwelling occur at the 

 very base of the food chain and are separated from the 

 harvested resource by a considerable time lag and by 

 multiple interacting biological processes. Where such 

 extended linkages are involved, attempts to relate indivi- 

 dual short-term upwelling events to characteristics of a 

 subsequent fishery are likely to be unrewarding. 



However, over longer time scales the mean value may be 

 less important than the frequency content of the distribution 

 from which the mean was formed. For example, a month of 

 continual, fairly steady upweUing activity may have a mean 

 rate of upward transfer similar to another month in which 

 nearly all the upwelling occurs in one short, very intense 

 burst. However, the responses of the biota in the two cases 

 might be quite different. In addition, there are certain 

 applications to which detailed specifications of individual 

 upwelling events are pertinent. For instance, the life history 

 of a particular organism may be such that a crucial process 

 takes place within a definable, short time period. In cases 

 where tactical fishing success is related to upwelling 

 conditions, an indication of short-term variations in upwell- 

 ing may aid in formulating fishing strategy and in 

 interpreting catch data. Multidisciplinary investigations 

 such as -the CUE experiments (O'Brien and Smith 1971) 

 promise new insights into the mechanisms linking upwelling 

 phenomena to fishery resources. Early application of results 

 of such efforts may be facilitated by having descriptive 

 indicators available which are compatible with the short time 



scales of the experiments but which are independent of a 

 particular observational progpram and can be extended to the 

 larger time and space scales involved in most fishery 

 problems. 



This report presents daily and weekly means of upwell- 

 ing indices computed at 6-hr intervals over the 7-yr period, 

 1967 through 1973. The input data are synoptic surface 

 atmospheric pressure fields. The calculation procedures and 

 the locations (Fig. 1) are identical to those used by Bakun 

 (1973). However, in the present case a lower value for the 

 drag coefficent, corresponding to the use of synoptic rather 

 than monthly mean input data, is used to estimate the stress 

 of the wind on the sea surface. 



METHOD 



The procedure for calculating upwelling indices is 

 presented in detail by Bakun (1973). Six-hourly synoptic 

 surface atmospheric pressure fields prepared by Fleet 

 Numerical Weather Central (FNWC) are arranged on the 

 3-degree mesh length grid shown in Figure 1. Finite 

 difference derivatives are formed and the geostrophic 

 wind is calculated. An estimate of the wind field near the 

 sea surface is made by rotating the geostrophic wind 15 

 degrees to the left and reducing its magnitude by 30% to 

 approximate frictional effects. The sea surface stress field 

 is computed according to 



7 = 9. Cd 171 



(1) 



where T is the stress vector, Pu is the density of air, Cd is 

 an empirical drag coefficient, 'Z is the estimated wind vector 

 near the sea surface with magnitude \~Z\ ■ 



The resultant transport in the surface layer of the ocean 

 is calculated from 



M = f T xk 



(2) 



' Pacific Environmental Group, National Marine Fisheries Service, 

 NOAA, Monterey, CA 93940. 



where M is the Ekman mass transport, / is the Coriolis 

 parameter, and Ic is a unit vector directed vertically 

 upward. The component of computed Ekman transport 

 directed off shore is termed the upwelling index and is 

 considered an indication of the rate of upwelling of deeper 



