720 



Fishery Bulletin 102(4) 



enough that we could assume that 226 Ra activities were 

 similar among all samples and that use of an average 

 was valid (0.0643 [±0.0035] dpm/g). This approach is 

 acceptable because 226 Ra activities measured in pre- 

 vious radiometric studies on Pacific rockfishes were 

 relatively constant. For example, the activity of cored 

 yelloweye rockfish (S. ruberrimus) otoliths had a mean 

 226 Ra activity of 0.0312 (±0.0026) dpm/g (n=18; Andrews 

 et al., 2002), and the rougheye rockfish (S. aleutianus), 

 another deepwater species (to 730 m; Love et al., 2002), 

 had a similar otolith core 226 Ra activity averaging 0.065 

 (±0.003) dpm/g (Kastelle et al., 2000). 



Accuracy and uncertainty of ages estimates 



Radiometric activities measured in blackgill rockfish 

 otoliths generally agreed with expected activity ratios 

 for 210 Pb and 226 Ra (Fig. 4), confirming the validity of 

 growth-zone-derived age estimates. In addition, a direct 

 comparison between radiometric age and predicted age 

 resulted in a strong agreement (r 2 =0.89; Fig. 5), which 

 was further supported by slope and elevation tests that 

 revealed no significant difference from a 1:1 agreement 

 line. 



The most critical sources of error involved in age 

 estimation, prediction, and radiometric age determina- 

 tion were the following: 1) age estimate uncertainty, 2) 

 regression error associated with predicted ages, and 3) 

 analytical uncertainty associated with the radiometric 

 aging technique (TIMS and a-spectrometry). Conven- 

 tional aging techniques are inherently subjective (Boe- 

 hlert, 1985; Campana, 2001) and thus create uncertain- 

 ty associated with an estimated age. This uncertainty 

 is transferred to the prediction model, where the natu- 

 ral variability associated with individual otolith weight 

 must also be considered. For most samples, however, the 

 error bars either overlapped or were in contact with the 

 agreement line (Figs. 4 and 5), further confirming the 

 concordance of radiometric age with predicted age. 



Implications for management 



When considering the longevity of rockfishes for which 

 a maximum age has been reported (Munk, 2001; Cail- 

 liet et al., 2001), a longevity exceeding 90 years places 

 the blackgill rockfish within the top 2Q ( 7c of long-lived 

 rockfishes. There is a trend for rockfishes that may 

 indicate that longevity increases as maximum depth of 

 occurrence increases, and physiological adaptations to 

 the environmental conditions of deep-sea living could 

 provide an explanation (Cailliet et al., 2001). The con- 

 firmed longevity and the maximum depth of occurrence 

 (~800 m) for the blackgill rockfish provide further sup- 

 port for this concept. 



Longevity in the rockfishes has been central to its 

 evolutionary success in relation to other marine teleosts. 

 The suite of life history characters implicit with a long 

 lifespan (slow adult growth, late age-at-maturity, low 

 adult natural mortality) represent a "slow and steady" 

 adaptive strategy, whereby the energy allocated towards 



individual growth is prolonged, eventually contributing 

 to greater fecundity (due to larger size at maturity) 

 over the lifespan of the individual. This reproductive 

 strategy serves to propagate genetic material across 

 several generations, as well as to diffuse the effect of 

 mortality associated with each reproductive event (Lea- 

 man, 1991). In this sense, longevity may act to buffer 

 the species against short-term (El Nino) and long-term 

 environmental change (Pacific Decadal Oscillations), 

 and the stochasticity inherent in the Pacific Ocean 

 system (Moser et al., 2000). 



In the absence of fishing pressure, the genetic con- 

 tribution of a slow-growing, longer-lived species may 

 be more conserved in the collective species' gene pool 

 i Munk 2 1. In the presence of fishing pressure, however, 

 this "slow and steady" adaptation may be detrimental 

 (Musick, 1999). Although modeling fish populations for 

 the purpose of management typically involves some or 

 all of these parameters, the focus is often on deter- 

 mining sustainable biomass and this approach largely 

 ignores the unknown effects of changes in age struc- 

 ture due to removal of the oldest individuals from the 

 population (Craig, 1985), as well as a loss of genetic 

 diversity that could prevent full recovery of severely 

 depleted populations (Hauser et al, 2002). Given the 

 current depressed condition of many heavily fished rock- 

 fish stocks, species-specific life history characteristics, 

 such as longevity, growth rate, and age-at-maturity 

 estimates, should be given thorough consideration in 

 the development of an effective management strategy. 

 Management regulations that account for these charac- 

 teristics, such as a limited fishing season, or designa- 

 tion of harvest refugia (Yoklavich, 1998), would provide 

 a stronger basis for conservation and sustainability of 

 the resource. 



Acknowledgments 



We wish to thank John Butler, Don Pearson, and Cindy 

 Taylor of the Southwest Fisheries Science Center, Mark 

 Wilkins, Jerry Hoff, Waldo Wakefield, and Bob Lauth 

 of the Alaska Fisheries Science Center, and Bob Lea 

 of CDFG for donating specimens for this study. Mary 

 Yoklavich (NMFS). Di Tracey and Larry Paul (NIWA, 

 New Zealand), Kristen Munk (ADFG), Don Pearson 

 (NMFS), Tom Laidig (NMFS), and Steve Campana 

 (DFO, Canada) provided valuable insight into black- 

 gill rockfish growth patterns. Patrick McDonald of the 

 Oregon Department of Fish and Wildlife aged otolith 

 sections. Pete Holden at the University of California, 

 Santa Cruz, measured radium in the refined samples 

 using TIMS. The comments and suggestions of three 

 anonymous reviewers were greatly appreciated. This 

 work was supported by the National Sea Grant College 

 Program of the U.S. Department of Commerce's National 

 Oceanic and Atmospheric Administration under NOAA 

 Grant number NA06RG0142, project number R/F-182. 

 through the California Sea Grant College Program; and 

 in part by the California State Resources Agency. 



