396 



Fishery Bulletin 104(3) 



Bin numbers 



10 



20 30 



Distance (km) 



40 



50 



40 S 



60 



Figure 4 



Histogram (gray bars, left-hand scale) and cumulative frequency 

 (bold line, right-hand scale) showing percentage of 2002 tra- 

 jectories intersecting bins from the 1999 coverage boundary. 

 Bottom scale shows distance around the coverage boundary in 

 kilometers. Arrow above histogram shows boundary location 

 nearest Platform Irene. 



offshore side (bins 22-39), and the remainder crossed 

 the northern side (6%) or the southern side (1%). 



To compare results more directly between years, a 

 histogram of crossings was generated from the 2002 

 trajectories, using the 1999 coverage boundary. The 

 1999 coverage boundary was completely contained 

 within the 2002 coverage boundary. With the 2002 

 trajectories, a peak in the histogram again occurred 

 along the offshore boundary (Fig.4, left-hand scale), 

 this time at bin 12 compared with bin 11 when the 

 1999 trajectories were used. A second, but much 

 smaller peak occurred along the 50-m isobath at bin 

 3, consistent with the small peak along the 50-m iso- 

 bath of Fig. 2B. Table 1 and the cumulative histo- 

 gram (Fig. 4, right-hand scale) showed that 18% of 

 trajectories crossed the 50-m isobath, 66% crossed 

 the offshore side of the coverage boundary, and the 

 remainder crossed either the northern (7%) or south- 

 ern (8%) sides. 



The time required for trajectories to cross the cover- 

 age boundary, defined in our study as the residence 

 time, varied between years and mainly depended on 

 the size of the coverage area. In 1999 the mean and 

 standard deviation for the residence time was 19 ±12 

 hours, and the maximum was 86 hours (Table 1). In 

 2002 they were 47 ±34 hours and the maximum was 

 163 hours. When the 2002 trajectories were computed 

 over the 1999 coverage boundary, residence times were 

 comparable to the 1999 values: 22 ±18 hours and a 

 maximum of 116 hours. 



Discussion 



Because of limitations in spatial coverage, the HF radar- 

 derived trajectories could not be used to examine the 

 full range of length and time scales over which actual 

 trajectories may extend. We used a trajectory data set 

 resulting from the release of Argos drifters in the region 

 to examine these scales. Drifters were deployed in the 

 Santa Barbara Channel and Santa Maria Basin at 

 irregular intervals from October 1992 through Decem- 

 ber 1999 as part of a circulation study conducted by the 

 Scripps Institution of Oceanography (SIO; see Dever 

 [1998] and Winant et al. [2003] for a description of the 

 drifter data set). Drifter positions were obtained up to 

 six times per day, typically for 40 days, and had a spa- 

 tial accuracy of about 1 km. Several trajectories ended 

 earlier when the drifters beached. 



No drifters were released at Platform Irene although 

 many approached the platform after release elsewhere 

 (130 drifters were deployed to the north of Platform 

 Irene, and 440 were deployed to the south). To approx- 

 imate trajectories originating at Platform Irene, all 

 drifters released during all seasons for all years, and 

 approaching within 10 km of the platform, were identi- 

 fied. This distance is a compromise between proximity 

 to the platform and ensemble size; 93 trajectories ap- 

 proached within 10 km of Platform Irene (white circle 

 in Fig. 5 is 10 km in radius and is centered on Platform 

 Irene). Of these, 34 were released north and 59 south of 

 Platform Irene. The ensemble of trajectories beginning 



