Abstract.— We describe a simple 

 biostatistical model of reproductive 

 success (logarithm of recruits/ 

 spawner) applied to three coastal pe- 

 lagic fish stocks off southern Cali- 

 fornia: northern anchovy Engraulis 

 mordax, Pacific sardine Sardinops 

 sagax, and chub mackerel Scomber 

 japonicus. We used the model to de- 

 tect possible influences of gross cli- 

 matic conditions and contaminant 

 loadings (particularly of metals and 

 organochlorines) on reproduction in 

 these three stocks. Data included 

 several decades of annual estimates 

 of recruitment and stock size, 

 monthly measures of climate, and 

 annual estimates of contaminant 

 loadings; the model included a com- 

 pensatory stock-size component be- 

 fore adding environmental effects. 

 The study was meant to generate, 

 rather than test, hypotheses. For the 

 chub mackerel stock, we detected cli- 

 mate influences, but no contaminant 

 influences, on reproductive success, 

 which was usually high during con- 

 ditions typical of El Nino-Southern 

 Oscillation events. For the northern 

 anchovy stock, we detected no cli- 

 mate or contaminant influences on 

 spawning success; however, the 

 negative results may reflect low sta- 

 tistical power, rather than absence 

 of contaminant influences. For 

 spawning success in the Pacific sar- 

 dine stock, we detected no consis- 

 tent climate influences, but we found 

 a strong negative correlation with 

 contaminant loadings. This result is 

 consistent with the hypothesis that 

 contaminant loadings accelerated 

 the collapse of the Pacific sardine 

 stock while it was under stress from 

 severe overfishing. Although many 

 scientific questions about validation 

 of models, mechanisms of action, and 

 identity of specific deleterious con- 

 taminants remain to be answered, 

 the observed data are well described 

 by the hypothesis of contaminant- 

 mediated decline. 



Detection of contaminant and 

 climate effects on spawning success 

 of three pelagic fish stocks off 

 southern California: Northern 

 anchovy Engraulis mordax, Pacific 

 sardine Sardinops sagax, and chub 

 mackerel Scomber japonicus* 



Michael H. Prager 



Southeast Fisheries Science Center, National Marine Fisheries Service, NOAA 

 75 Virginia Beach Drive, Miami, Florida 33 149 



Alec D. MacCall 



Southwest Fisheries Science Center, National Marine Fisheries Service, NOAA 

 3 I 50 Paradise Drive, Tiburon, California 94920 



Manuscript accepted 29 January 1993. 

 Fishery Bulletin, U.S. 91:310-327 1 19931. 



In recent years, there has been in- 

 creasing concern about possible ef- 

 fects of contaminants on the nation's 

 living resources, particularly ex- 

 ploited fish and shellfish populations. 

 Although contaminants such as 

 heavy metals and synthetic organo- 

 chlorines are commonly detectable in 

 coastal sediments, waters, and organ- 

 isms (Mearns et al. 1991), it has been 

 difficult, if not impossible, to deter- 

 mine whether sublethal levels of con- 

 taminants have influenced the pro- 

 ductivity of coastal fish stocks. Mass 

 mortalities and deformities of fish 

 and shellfish have occurred after ex- 

 posure to high levels of contaminants, 

 but mortality and morbidity in natu- 

 ral populations attributed to chronic 

 low-level exposures is difficult to 

 prove conclusively, even though it 

 may be strongly supported by circum- 

 stantial evidence (Sindermann 1978). 

 In addition, the question of popu- 

 lation-level effects has been largely 

 unexplored. For individuals of most 

 species, the physiological effects (if 

 any) of chronic sublethal exposure to 

 contaminants are poorly understood. 

 For populations, the effects of such 

 exposure are not known. Nonetheless, 



evidence from many studies of indi- 

 viduals suggests that populations 

 should be affected also. 



This paper describes exploratory 

 models of the effects of contaminants, 

 climate, and spawning-stock size on 

 three exploited fish populations off 

 southern California. With a rich body 

 of data, this area is unusually well 

 suited to such a study. The waters 

 off southern California contain el- 

 evated levels of numerous contami- 

 nants, including PCBs, DDT, and sev- 

 eral toxic metals (McCain et al. 1988). 

 Estimates of historical contaminant 

 loadings have been made by Sum- 

 mers et al. (1988), and data on cli- 

 matic conditions are available from 

 public records. Major coastal pelagic 

 fisheries have been monitored by the 

 California Department of Fish and 

 Game since about 1930. 



Two main conceptual approaches 

 might be used to model the effects 

 of contaminant exposure on pop- 

 ulations. The first approach is a 

 "bottom-up" one: constructing a 



"Contribution MIA-91/92-26 of the Miami 

 Laboratory, Southeast Fisheries Science Cen- 

 ter, National Marine Fisheries Service, NOAA. 



310 



