1984 



a oHIGH TIDE 



* *LOW TIDE 



Figure 6. — Median catches per seine haul of chum 

 salmon at high tide (solid line) and low tide stations 

 (dashed line), 1984-86. Solid triangles indicate dates of 

 hatchery releases. 



ity after release than unmarked fish as a result of 

 the added stress of marking and the debilitating 

 bacterial disease th?t afflicted this release 

 group. As a result of higher mortality of marked 

 fish, R was low and M was probably overesti- 

 mated, leading to overestimation of the popula- 

 tion (A^,, Equation (1)). 



Higher loss rates were found for small than for 

 large fish hatchery chum salmon in Netarts Bay 

 from the decline in the natural logarithm of catch 

 per effort of fin-cHpped fish (Table 3). In 1986, 

 the residence half-life (the time for the catch 

 rates to decrease by one-half (Myers and Horton 

 1982)) was 7.4 days for fish released at 1.0 g, 4.9 

 days for fish released at 2.2 g and <2 days for 

 fish released at 6.5 g. None of the largest fish 

 was captured 2 days after release or during sub- 

 sequent sampling. Presumably these large fish 

 emigrated rapidly from the estuary. An anomaly 

 in the trend for loss rates to be positively cor- 

 related with size of juvenile chum salmon re- 

 leased arose for the 1.9 g right ventral (RV) 



FISHERY BULLETIN: VOL. 87. NO. 3. 1989 



clipped fish in 1984. Their residency half-life was 

 16 days, about the same as for 0.75 g fish re- 

 leased on the same day, and three times that of 

 1.9 g fish released in 1985 (Table 3). The signifi- 

 cantly higher (P < 0.01, analysis of covariance) 

 residency half-lives of fish released in 1984 than 

 in other years may have arisen because these 

 fish were released earlier in the spring. In other 

 years, early release groups also had longer half- 

 lives than later groups, but the slopes of the 

 catch vs. time were not significantly different (P 

 > 0.05). 



Juvenile chum salmon actively maintained 

 themselves in the bay. Residency half-lives of 

 marked chum were 10-30 times longer than pre- 

 dicted from random loss with tidal flushing. 

 Assuming that the mean intertidal volume of 

 Netarts Bay is 75% of the total volume at MHW 

 (Glanzman et al. 1971; Kreag 1979) and does not 

 reenter the bay on subsequent tidal cycles, the 

 half-life of water in Netarts Bay is <0.5 day. 



Growth 



Instantaneous gi'owth rates in weight of fin- 

 clipped chum salmon averaged 1.(5-2.3% body 

 weight per day (Table 3). No differences (analy- 

 sis of covariance, P > 0.05) were found in gi'owth 

 in weight among these groups released within or 

 among years. However, linear growth rates esti- 

 mated from changes in fork length over time 

 indicate that fish released at a smaller size grew 

 more rapidly in 1984 (0.48 mm/d for 46 mm fish 

 vs. 0.41 mm/d for 52 mm fish) and in 1986 (0.53 

 mm/d for 48 mm fish vs. 0.33 mm/d for 62 mm 

 fish). Growth rates were similar for both 56 mm 

 vs. 59 mm fish in 1985. Slopes derived from 

 linear regi'ession of mean individual lengths of 

 all fish during a sampling period vs. elapsed 

 time in days did not differ significantly (P > 

 0.05) from the rate of increase of lengths of 

 marked fish in 1984, 1985, or 1986. These slopes 

 were also similar among years. Increases in 

 mean length probably reflected gi'owth. None of 

 the regressions of FL vs. time showed trends for 

 decreasing size late in the spring that would be 

 due to emigi-ation of large individuals from the 

 bay. 



Biomass-Production 



The biomass of juvenile chum salmon in 

 Netarts Bay, estimated from population num- 

 bers (A^,) and average weight of the marked 

 groups of fish during the week after release of 



562 



