Development of Rations for the Enhanced Survival of Salmon, 1984-1985 Progress (Annual) Report.

It is believed that hatchery feed nutritional quality plays an important role in determining the health and fitness of smolts. Commercial fish meal, the major source of protein in salmon rations, is subject to heat damage during drying and chemical interaction of fat oxidation products with meal proteins. Protein bioavailability is reduced and dietary stress may be introduced into hatchery feeds. The basic hypothesis of this investigation is that ration protein quality can influence the survival of smolts and the ultimate return of adults. Improved hatchery production would be better able to reestablish natural runs of salmon in the Columbia River and its tributaries and maintain and improve the genetic integrity of specific stocks. The general approach being used to prove this hypothesis involves a comparison of the hatchery growth response, survival and return of coho and chinook salmon reared on nutrient dense rations containing a very high quality fish protein complement and commercial ration relying on commercial fish meals as a source of protein. Coded wire tagging experiments are being conducted on replicate brood years of test and control fish to determine the influence of ration protein on survival. Project rearing and release of tagged fish to date include 1982 and 1983-brood replicates of coho salmon and 1983 and 1984-brood replicates of fall chinook (tule stock) salmon. The 1984-brood year replicate of coho salmon is presently being reared and has been tagged for release in April 1986. Planning was completed for rearing a 1985-brood replicate of fall chinook (upriver bright stock) salmon. This report covers the rearing and release of the 1983-brood coho and the 1984-brood fall chinook (tule stock) replicates. Duplicate lots of coho salmon were reared on two test rations containing vacuum dried salmon and hake meals and a control ration composed of the Sandy hatchery supply of Oregon pellet feed system rations from 1 June 1984 to release on 30 April 1985. A computed 57,981 fish/pond replicate (2.14 g average fish weight) were reared to a 28.75--32.67 g average fish weight of 56,272 to 57,334 fish/pond released, 25,827--26,673 possessed a recognizable coded wire tag. Fall chinook salmon (tule stock) were reared on a test ration containing vacuum dried salmon meal and a control ration composed of the Bonneville Hatchery supply of Oregon feed system rations from 28 December 1984 to 13 May 1985. Fish were initially ponded at 0.61 and 0.64 g/fish in two lots of 504,766 and 564,113 fish/pond: one pond was supplied the test ration and the other the control. The two lots of fish (1.43--1.58 g/fish) were split into duplicate ponds of 215,480 to 260,957 fish each on 31 January 1985 and reared to 5.98 to 7.18 g/fish at release. Of the 150,774 to 213,251 fish/pond released, 78,962 to 60,242 possessed recognizable coded wire tags. The growth response of both coho and fall chinook salmon between mid February of 1985 to release in May was altered from that expected by poor test ration palatability. Palatability problems were traced to one of two lots of herring oil used to prepare rations. One lot of oil was not properly stabilized with antioxidants by the supplier. Although it possessed no chemical characteristics of autoxidation, it did possess a very high potential for oxidation when incorporated into a region which could not be adequately stabilized with antioxidants. Ration preparation and oil evaluation protocol have been initiated to guard against a future occurrence