ALBERS and ANDERSON: DIET OF PACIFIC COD 



September information suggests that cod consum- 

 ed an estimated 142 t of pink shrimp over the same 

 period. 



DISCUSSION 



Cod Diet 



Pink shrimp was the dominant food item identified 

 by frequency of occurrence (63%) and percent 

 volume (31%). In the Gulf of Alaska near Kodiak 

 Island, Jewett (1978) reported pink shrimp occur- 

 ring in 4% of the Pacific cod examined, and Hunter 

 (1979) found pink shrimp occurring in 24% of the 

 cod, representing 16% of the diet by weight. Hunter 

 also identified pink shrimp as the dominant food 

 item in his study. However, our study exhibited a 

 higher percent frequency of occurrence for that 

 species. Unlike our study, Jewett and Hunter's 

 studies included examinations of cod from offshore 

 areas, which were not regions of high pink shrimp 

 density (Gaffney 19773). 



The prey size ranges of walleye pollock and snow 

 crab in our study were similar to those found by 

 Hunter (1979). In our study, the maximum length 

 of walleye pollock consumed by cod was 25 cm FL, 

 although the majority were between 6 and 20 cm. 

 Hunter (1979) reported that cod around Kodiak 

 Island were feeding on groundfish (including wall- 

 eye pollock) between 2 and 24 cm. The snow crab 

 consumed by cod in Pavlof Bay ranged from 5 to 45 

 mm CW, which is similar to the size range of snow 

 crab (1-40 mm) found in cod stomachs by Hunter 

 (1979). Jewett (1978) reported a greater size range 

 of snow crab occurring in cod stomachs (from 1.8 

 to 70 mm), yet 78% were between 7 and 23 mm. The 

 maximum size of cod examined by Jewett was 92 cm 

 TL (total length) compared with 69 cm FL in our 

 study, and this difference probably accounts for his 

 observation of larger snow crab. 



We believe that cod were not feeding on selected 

 sizes of shrimp. The size ranges and mean sizes of 

 pink shrimp and humpy shrimp consumed by cod 

 were similar to those found in the trawl (Tkble 4). 

 However, cod did consume small (<10.0 mm CL) pink 

 shrimp that were not captured by the trawl. We 

 believe this is due to trawl bias toward larger shrimp. 

 No significant difference was found between pink 

 shrimp length distributions from cod stomachs and 



trawl samples in two of the three surveys, indicating 

 that cod were not feeding on selected sizes of shrimp. 

 There was a significant difference (P = 0.009) for 

 the September 1981 survey. In this sample either cod 

 selected slightly for smaller shrimp or the trawl 

 caught slightly larger shrimp. 



Estimate of Pink Shrimp 

 Biomass Consumed 



The estimated pink shrimp biomass in Pavlof Bay 

 decreased by 1,501 1 between the May and Septem- 

 ber surveys in 1981. During this period we estimated 

 that Pacific cod consumed between 142 and 875 t 

 of pink shrimp. Since Pavlof Bay is believed to con- 

 tain a geographically isolated stock of pink shrimp 

 (Anderson 1981) and because the bay was closed to 

 shrimp fishing in 1981, cod predation is responsible 

 for at least part of the biomass decline 



The estimate of pink shrimp biomass consumed 

 over the 112-d period using the May survey data was 

 733 t more than was estimated using September 

 data. Two of the parameters used to calculate con- 

 sumption estimates were responsible for this dif- 

 ference In May the mean weight of pink shrimp con- 

 sumed daily by one cod was about double the amount 

 in September. Pink shrimp were more abundant and 

 made up a larger percentage of the diet in May than 

 in September. Additionally, cod biomass was 

 estimated to be almost three times higher in May 

 than it was in September causing the consumption 

 estimate to be higher in May (Tkble 5). 



We belive that consumption of pink shrimp by cod 

 probably lies toward the high end of the calculated 

 range (142-875 t). Biomass estimates were probably 

 conservative for pink shrimp and Pacific cod. Bio- 

 mass was calculated on the assumption that all cod 

 and shrimp were on bottom and all those in the path 

 of the trawl were caught. This is not true for cod 

 or shrimp. For example, Edwards (1968) reported 

 that up to 49% of the gadoids in the path of a trawl 

 avoid capture. Also, an estimate of the catchability 

 of shrimp with the high-opening shrimp trawl was 

 about 56% (Alaska Department of Fish and Game 

 1982'*). If the cod biomass estimate was conservative, 

 the consumption of pink shrimp by cod would be 

 higher than calculated. Further, if the pink shrimp 

 biomass estimate was conservative, the calculated 



^Gaffney, F. G. 1977. Kodiak pandalid shrimp research. Com- 

 mercial Fisheries Research and Development Act, Project No. 

 5-36-R. Unpubi. manuscr., 76 p. National Marine Fisheries Ser- 

 vice, NOAA, Wash., DC 20235. 



■•Westward Region Shellfish Staff, Alaska Department of Fish 

 and Game 1982. Westward Region Shrimp Fishery Management 

 Plan. Unpubi. manuscr., 70 p. Alaska Department of Fish and 

 Game, Kasheruaroff, Mission Road, P.O. Box 686, Kodiak, AK 

 99615. 



607 



