42 



Fishery Bulletin 89(1). 1991 



STUDY 

 AREA 



Washington 



 46" 



22 



Oregon 



21 



2 P) 



19 



y / California 



Cape Mendocino 



48 



44 



42" 



- 4 



Figure 1 



Location of commercial concentrations of pink shrimp Pan- 

 dalus jordani along the U.S. Pacific coast (shaded areas) and 

 statistical areas 18-32. 



18-28 constituted a single stock, based upon an analysis 

 of growth, maturation rates, and coastal oceanographic 

 conditions. We used the same stock unit to allow us to 

 draw upon the summarized sample data from Zirges 

 et al. (1982) for the years 1966-81. 



To evaluate the effect of shrimp density on growth, 

 we compared shrimp carapace length at age for two 

 time-periods representing different levels of population 

 biomass. We use the term "density" in the sense of 

 biomass per unit area rather than the number of in- 

 dividuals per unit area. We used catch-per-unit-effort 

 (CPUE) as our index of shrimp density. 



During the years 1975-78 major improvements in 

 trawl design were implemented by the shrimp industry. 

 Prior to 1976, the predominant shrimp net was a 

 57-foot (headrope) Gulf of Mexico style, Marinovitch 

 trawl, with a 4-foot vertical opening (Zirges and Robin- 

 son 1980). During the years 1975-78 the majority of 



T 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 



66 68 70 72 74 76 78 80 82 84 86 



YEAR 



Figure 2 



Catch-per-unit-effort (kg/single-rig equivalent hour) for the 

 pink shrimp fishery in statistical areas 18-28 for the years 

 1966-88. 



the fleet switched to locally produced 70-90 foot 

 (headrope) box trawls. The new trawls have proven to 

 be much more efficient for pink shrimp. Besides being 

 generally bigger, they open to a height of 12-18 feet, 

 improving fishing for pink shrimp, which come up off 

 the bottom under reduced light conditions. Increased 

 cloud cover and time of day were shown to bring con- 

 centrations of shrimp up off the bottom at least 8 feet 

 (Beardsley 1973). The Oregon box trawls are also con- 

 structed differently from the Marinovitch trawls. They 

 are much longer, employ a slower taper, and are hung 

 with considerable "slack" webbing, all contributing to 

 a much more efficient trawl for pink shrimp. The effort 

 data series is not corrected for these gear improve- 

 ments. Therefore, CPUE data understate the magni- 

 tude of biomass reduction since 1978 (Fig. 2). Accord- 

 ingly, we tested growth for the two time-periods, 

 1966-78 and 1979-88. We considered these periods to 

 be representative of the virgin stock biomass and the 

 reduced biomass, respectively. We used .F-tests to iden- 

 tify significant differences in length at age. 



Four area-months were selected as indices for anal- 

 ysis of age-1 and -2 shrimp growth, based upon the com- 

 pleteness of the time-series data. For age-1 shrimp, the 

 months of April- June were excluded because in some 

 years age-1 shrimp are not fully recruited to the trawl 

 gear in those months. For age-2 shrimp, the months 

 of September and October were excluded because in 

 recent years age-2 and older shrimp comprise a small 

 percentage of late-season catches. Given these criteria, 

 the index area- months selected for age-1 shrimp were 



