Leber et al.: Marine stock-enhancement potential in nursery habitats of Mugil cephalus 



467 



tial behavior of stocked fish can also be directly affected 

 by handling stress and temperature change, either of 

 which could cause an increase in susceptibility to preda- 

 tors shortly after release (Fuiman and Ottey, 1993). 



The difficulty in quantifying initial mortality is in 

 obtaining an unbiased early sample of abundance to 

 establish a baseline recapture rate for comparison 

 with recapture rates in later samples. Dispersal of 

 striped mullet from the point of release is not an 

 immediate event (Leber et al., personal observ. ). The 

 first few hours after a release should be the very time 

 when released fish are most vulnerable to predators. 

 However, immediate sampling in an open environ- 

 ment to determine initial abundance in samples vio- 

 lates assumptions of basic mark-recapture models 

 (Ricker, 19751, primarily the assumption that marked 

 fish mix randomly within the study area. 



Without a reliable, unbiased method to estimate 

 postrelease survival in open coastal ecosystems, how 

 do we gauge benefits from hatchery releases? In ef- 

 fect, the best indicator of hatchery impact is total 

 annual catch of cultured fish in a fishery. Thus, as- 

 sessment of marine stock enhancement requires un- 

 biased estimates of CPUE, total fishing effort, and 

 proportion of cultured fish in the fishery. Accurate 

 catch statistics for marine organisms are expensive 

 and often lacking (e.g. Shomura 7 ), as in Hawaii for 

 striped mullet. The difficulty of assessing total hatch- 

 ery impact is exacerbated when cultured fish are 

 released unmarked, which has often been the case 

 with marine fishes (discussed in Richards and 

 Edwards, 1986). 



Because a fishery on juveniles is lacking, some 

 measures besides fishery landing statistics are 

 needed to evaluate survival of cultured fish during 

 the nursery phase. Consequently, researchers in 

 coastal systems report recovery rates and percent 

 cultured fish in samples of cultured and wild fish 

 (hatchery contribution) as early indicators of en- 

 hancement effect. Percentage of cultured fish can be 

 a good relative indicator of release impact but only 

 when reported with CPUE for wild and cultured fish. 

 Hatchery contribution is a function of abundance of 

 wild fish, release magnitude, survival, dispersal, size 

 of hatchery and wild fish, and environmental carry- 

 ing capacity. 



Hatchery contribution is sensitive to variation in 

 any of the above six factors. Thus, alone, it has low 

 information content as an indicator of success. In fact, 

 year to year comparisons of hatchery contribution 



7 Shomura, R. S. 1987. Hawaii's marine fishery resources: 

 yesterday (1900) and today (1986). Honolulu Laboratory, 

 Southwest Fisheries Center, Natl. Mar. Fish. Serv., Admin. Rep. 

 H-87-21. 



can be misleading. Three examples below illustrate 

 the relationship among hatchery contribution (%C), 

 CPUE of cultured (C) and wild fish (W) in samples, 

 and actual hatchery effect on recruitment: 



Example 1 %C in year 1 and year 2 are equal, but 

 W is greater in year 2: hatchery effect 

 on magnitude of recruitment actually 

 increased; increase is not apparent from 

 comparing %C but is when C and W are 

 compared between years; 



Example 2 %C is greater in year 2 but W is less: 

 actual hatchery effect on abundance 

 could be identical in years 1 and 2; 



Example 3 %C and W are both greater in year 2: 

 substantial increase in hatchery effect 

 (i.e. greater hatchery impact than real- 

 ized by merely comparing %C between 

 years). 



Lacking an unbiased indicator of actual survival 

 following hatchery releases, we submit that recruit- 

 ment, hatchery contribution, and recovery rate (no. 

 captured/no. released) provide a good preliminary in- 

 dication of enhancement effect if all three of these 

 are presented. Each is needed to evaluate and con- 

 trol enhancement effectively. 



Variation in fishery yields is driven largely by re- 

 cruitment dynamics (see Frank and Leggett, 1994, 

 for recent review). The CPUE of cultured and wild 

 fish reflects the magnitude of recruitment, which 

 should have a direct relevance to fishery yields, if food 

 and habitat are available. Also, without CPUE data, 

 there is no indication of how the whole population (cul- 

 tured and wild individuals) is changing from year to 

 year in response to natural recruitment processes, en- 

 hancement, and other management strategies. 



Hatchery contribution from pilot studies can be 

 used as an index to help in planning release magni- 

 tude at levels that would not swamp wild stocks with 

 cultured fish. Hatchery contribution is easily computed 

 from CPUE data (i.e. %C = [C x 100%] /[C + W])- 



Because cost effectiveness is directly related to 

 survival of cultured fish in the wild, optimizing re- 

 covery rate should also be a key consideration in 

 enhancement programs (Kitada et al., 1992). Recov- 

 ery rate provides a relative measure of survival and 

 is a good indicator (if standardized by sampling ef- 

 fort) for comparing performance of cultured fish. In 

 comparisons of recovery rates from different years, 

 differences in release magnitude are factored out. 



The above three indicators can be used to evalu- 

 ate release effects during the juvenile stage, before 

 fish enter the fishery, to obtain a preliminary esti- 

 mate of stock enhancement success. Thus, a way of 



