Brown Bear Predation on Sockeye Salmon at Karluk Lake, Alaska

Abstract

Effects of predation by brown bears ( Ursus arctos ) on sockeye salmon ( Oncorhynchus nerka ) were studied at Grassy Point Creek, a tributary o£ Karluk Lake on Kodiak Island, Alaska, during the summers of 1964 and 1965. In 1964 bears were allowed free access to the stream, but iIl 1965 an attempt was made to exclude them with an electric fence. Bears were efficient predators in the stream, killing up to 79 percent of the salmon in 1964; hoSwever, only 9.6 percent of the dead females sampled were unspawned bear-killed fish. The maximum estimate of eggs lost to bear predatioln in 1964 was about 1 million, compared with a total lo,ss, from all causes, of 8 million potential eggs. As a result of certain behavioral patterns of sockeye salmon, bears usually take spawned-out rather than unspawned females. The ratio of males to females in each year's escapement approached 1:1; the ratio among bear kills was about 3:2. Males acted as a buffer against predatiojn on females. The fence reduced bear predation by two-thirds. It is concluded that bear predation has little adverse effect o!n the production of sockeye salmon. salmon began in 1882; the fishery developed rapidly until about 1907, after which it began a long irregular decline ( Fig. 2 ) . From 1888 to 1907, the annual run averaged about 3.4 million fish; but after 1P7 the runs generally decreased, and from 1946 to 196S, averaged only 0.7 million fish annually. Since l924, when the White Act was prassed ( legislation providing for an escapement of at least SO percent of each year's run to salmon streams in Alaska ), most of the escapements to the Karluk system were of this magnitude, but the numbers of salmon in the runs continued to