VAN CLEVE and BEVAN: DECLINE OF KARLUK SALMON RUNS 



found that had stream-type scale nuclei indicating they 

 had entered the sea as fry, ii has been rather generally 

 considered that the great majority of the offspring of 

 these helow-the-lake spawners perish. 



Rounsefell in 1958 was apparently unaware 

 of the work of the International Pacific Salmon 

 Fisheries Commission on the Fraser River sock- 

 eye salmon that had revealed the peculiar life 

 history of several of the lai'gest runs in that sys- 

 tem which spawn below the lakes where the 

 young are reared so that the fry have to migrate 

 upstream into the nursery lakes either 

 immediately after emergence or after a brief 

 period of growth to small fingerlings. However, 

 he did express doubt that the progeny of these 

 fish were wasted as Gilbert and Rich (1927) 

 thought. 



The Karluk River for the most part is difficult 

 to observe and would require much more effort 

 to study than the lake and its tributaries. How- 

 ever, evidence of the importance of spawning 

 in the Karluk River is overwhelming. The dif- 

 ferent estimates of numbers of sockeye salmon 

 spawning in the Karluk River below Karluk Lake 

 listed by Rounsefell (1958) and shown in Table 

 1, for between 1917 and 1932, are as great as 

 400,000 in 1926, i.e., more fish than were record- 

 ed for any other part of the watershed. Rutter, 

 quoted by Chamberlain (1907) observed large 

 numbers of fry and small fingerlings in the upper 

 Karluk River and 47 yr later in 1950, 1951, and 

 1952, Walker saw fry and fingerlings moving 

 up the Karluk River through the weir. He noted 

 two upstream migrations. The first was of fry 

 which averaged 40 mm or less in length and 

 occurred in late June and early July. The second 

 was of fingerlings which varied from 48 to 72 

 mm in length and began in late July and conti- 

 nued through August. Moreover, Walker and 

 Bevan (see footnote 8) stated that the sockeye 

 salmon population spawning in the Karluk was 

 the largest population in the system. They also 

 said that the decline in this race could explain 

 the decline in size of smolts and increase in 

 numbers of 4-yr smolts mentioned by Barnaby 

 (1944) as evidence of a decrease in productivity 

 of Karluk Lake. Bevan considered the negative 

 correlation between the number of sockeye re- 

 turning from the even year runs, which com- 

 pete with pink salmon for spawning room 

 mostly in the river below Karluk Lake, as evi- 



dence of the importance of the Karluk River 

 sjiawning. While Bevan's tagging experiments 

 of 1948 and 1949 were not designed to identify 

 the spawning location with the time the fish 

 were in the fishery, the recoveries did show the 

 l^resence of spawners in the Karluk River in 

 late August and early September. Finally, Burg- 

 ner et al. (1969) estimated there are 126.000 

 redd sites in the Karluk River as opposed to 

 48,000 in the rest of the watershed. While the 

 total spawning capacity of 348.000 they esti- 

 mated is far below the numbers known to have 

 spawned in the Karluk system in the past, the 

 capacity of the river is still about three times 

 greater than that of the rest of the system. Fur- 

 thermore, the slow-moving section of the Karluk 

 oi:)posite the portage must provide an ideal loca- 

 tion for the growth of the fry to fingerlings as 

 similar localities do in the Chilko and South 

 Thompson Rivers in the Fraser River system. 



The loss of the midseason races of sockeye 

 salmon in the Karluk is well authenticated. 

 Chamberlain (1907) mentioned that the Karluk 

 run had a single peak in 1903 and earlier 

 reports of McDonald (1894) and Moser (1899) 

 described the intensity of the fishery that finally 

 depleted the run. Evidence of its decline came 

 early and was met by increasingly severe regu- 

 lation of the commercial fishery which was pre- 

 vented from using barricades in about 1890, was 

 excluded from the Karluk River in 1889 or 1890, 

 and was excluded from the Karluk Lagoon in 

 1918. The gradual loss of the midseason peak 

 was demonstrated by Thompson (1950) but 

 Thompson and Bevan (1954) showed that the 

 fishery was still concentrated in late July and 

 early August, as late as 1950. Rounsefell dis- 

 agreed with Thompson's conclusions, that dif- 

 ferent races of sockeye salmon exist in the Kar- 

 luk basin, but in effect agreed with Thompson 

 by recommending added protection of the mid- 

 season fish. Rounsefell's conclusions have been 

 refuted by Owen et al. (see footnote 4) and later 

 by Hartman and Raleigh (1964) and by Card 

 and Drucker usee footnote 5). Owen et al. 

 showed that no change had occurred in either 

 Karluk Lake or in the spawning habitat which 

 might have caused the decline in the Karluk 

 sockeye salmon runs, but demonstrated that 

 changes had occurred in different sections of the 

 spawning population, especially in the midsea- 



645 



