Dorn: Environmental covariates of Merlucaus productus growth 



269 



large-scale fishery for Pacific whiting Merluccius -pro- 

 ductus in 1966. 



Environmental influence on growth has been ob- 

 served for many marine fish species (Kreuz et al. 1982, 

 Anthony and Fogarty 1985). Because it can be easily 

 measured and is associated with widespread changes 

 in the aquatic environment, water temperature is the 

 covariate most often studied. Water temperature may 

 have a direct physiological effect on the growth of fish, 

 or it may be indirectly linked to growth. For example, 

 decreases in water temperature occur with the onset 

 of upwelling in many coastal marine environments. On 

 the west coast of North America, the availability of 

 coastal upwelling indices on monthly, weekly, and daily 

 time scales (Mason and Bakun 1986) has made it pos- 

 sible to investigate the direct effect of upwelling on 

 growth, although convincing evidence of a link has not 

 yet been found (Kreuz et al. 1982, Francis 1983, 

 Boehlert et al. 1989). 



Results from the Pacific whiting growth-increment 

 regression show that an environmental covariate, sea- 

 surface temperature, and population density could 

 explain the deviations from a simple baseline model for 

 asymptotic growth. The effect of temperature was 

 greatest on the youngest ages present in the fishery 

 samples and declined as age increased. In contrast, the 

 effect of population biomass on annual growth in- 

 creased with age. Temperature was the most impor- 

 tant covariate, both in terms of its statistical signif- 

 icance and its effect on growth. This association of 

 enhanced growth and reduced sea temperature is con- 

 sistent with what is known about the California Cur- 

 rent, a major eastern boundary current system. Kreuz 

 et al. (1982) found an identical inverse effect on the 

 growth of English sole Parophrys vetulus and Dover 

 sole Microstomus pacificus at two locations off the 

 Oregon coast. 



Two mechanisms are believed to contribute to the 

 high productivity of the California Current system. 

 Coastal upwelling, generated by wind-driven offshore 

 transport in the surface Ekman layer, brings low 

 temperature, low salinity, and nutrient-rich water to 

 the surface (Bakun and Nelson 1977). A second 

 mechanism is the southward advection of water from 

 the Alaskan Subarctic Gyre. This water is characterized 

 by low temperatures, high nutrient content, and a large 

 standing stock of zooplankton (Roesler and Chelton 

 1987). Regardless of which mechanism is dominant dur- 

 ing a particular year, low mean sea-surface tempera- 

 ture during the summer can be expected to be asso- 

 ciated with high productivity. 



The diet of Pacific whiting provides the link between 

 primary productivity and growth. The major prey of 

 Pacific whiting are euphausiids, primarily Thysanoessa 

 spinifera and Euphausia pacifica (Livingston 1983, 



Rexstad and Pikitch 1986). In summer, the abundance 

 and pattern of distribution of these short-hved species 

 (1-2 yr) are closely tied to upwelling and primary pro- 

 ductivity (Simard and Mackas 1989). Rexstad and 

 Pikitch (1986) found that euphausiids comprised 90% 

 by weight of the diet of Pacific whiting 30-44 cm in 

 length collected during a trawl survey in 1983 off the 

 coasts of Oregon and Washington. Above 45 cm, a 

 length which corresponds approximately to ages 5-7, 

 this percentage drops to 20%. In the diet of the older 

 Pacific whiting, euphausiids are largely replaced by 

 small schooling fish and shrimp. These include northern 

 anchovy Engraulis mordax, Pacific herring Clupea 

 harengus pallasi, eulachon Thaleichthys pacificus, pink 

 shrimp Pandalus jordani, and rockfish Sebastes spp. 

 (Livingston 1983). 



This shift in dietary preferences by Pacific whiting 

 may help explain the effect of population biomass on 

 growth. When the biomass of Pacific whiting is high, 

 predation on fish species with multiyear life cycles may 

 become intense enough to reduce their availability to 

 the whiting population. In contrast, euphausiid abun- 

 dance is closely coupled to annual variations in produc- 

 tivity, so whiting predation would likely have little 

 effect on their abundance. 



Some supporting evidence for density-dependent 

 growth of Pacific whiting is found in Dark (1975), who 

 also documented a decline in length-at-age using fishery 

 samples from an earlier period in the fishery, 1964-69. 

 At the time of Dark's research, estimates of popula- 

 tion abundance were not available for Pacific whiting. 

 It is now believed that the 1961 year-class was excep- 

 tionally strong, nearly the same size as the record 1980 

 year-class (Dorn and Methot 1990). Consequently, this 

 earlier decline in length-at-age may also be partly at- 

 tributable to increases in population density as the 1961 

 year-class moved into the population. 



Although weight-at-age is the measure of size typical- 

 ly used in stock assessment models, the analysis in this 

 paper focuses on length rather than weight. A prac- 

 tical reason for this strategy is that most at-sea sam- 

 pling platforms are not sufficiently stable to obtain 

 accurate individual weights offish. Indeed, the weights- 

 at-age used in stock assessment models for Pacific 

 whiting are obtained by first estimating length-at-age, 

 then converting to weight using a length-weight rela- 

 tionship (Dorn and Methot 1990). In addition, growth 

 in length has several characteristics that make it 

 amenable for analytical modeling. Except in very rare 

 instances, changes in fish length are always positive 

 or zero. The annual growth increment in length sum- 

 marizes the growth response of the organism to envi- 

 ronmental conditions that are prevalent throughout the 

 year, or are short-term. In contrast, weight-at-age has 

 a seasonal pattern of increase and decline, associated 



