Evaluation of Competition and Predation as Limits to Juvenile Kokanee and Sockeye Salmon Production in Lake Ozette, Washington

We examined temporal and spatial patterns of feeding, distribution, and relative abundance of potential predators and competitors of juvenile kokanee and sockeye salmon Oncorhynchus nerka to determine which interactions might limit smolt production in Lake Ozette, Washington. Kokanee, the landlocked form of sockeye salmon, dominated the limnetic planktivore fish community, potentially creating greater intraspecific than interspecific competition. However, bioenergetic simulations and cladoceran egg-ratio analysis indicated that consumption demand by adult and juvenile kokanees and juvenile sockeye salmon could be satisfied by less than 1% of the instantaneous production of the preferred large Daphnia pulicaria throughout the growing season. Juvenile kokanees and sockeye salmon represented 72% of the annual diet by volume of large (≥300-mm fork length), limnetic northern squawfish Ptychocheilus oregonensis during winter-summer. Forty percent of the diet of large (≥300-mm fork length), limnetic cutthroat trout Oncorhynchus clarki consisted of age-0 and age- 1. O. nerka (40-140 mm fork length) during spring and summer. The per-capita predation by cutthroat trout on juvenile salmonids was 25 times greater than predation by northern squawfish. Simulations indicated that 138,900 age-0. O. nerka (17.4% of the estimated fry production) were eaten for every 1,000 large cutthroat trout in the lake, whereas only 5,600 age-0 fish (0.7% of the fry production) were eaten per 1,000 large northern squawfish. Lower predation by northern squawfish was due to the small percentages (2-8%) of these predators that inhabited the limnetic zone and interacted with age-0 and age-I O. nerka during spring and summer. We conclude that competition is unlikely to limit O. nerka production, even with an intensive enhancement program. However, predation could undermine enhancement efforts if piscivore populations are sufficiently large; therefore, piscivore abundance must be assessed before total predation losses can be estimated.