376 
Fishery Bulletin 106(4) 
determine otolith interpretation. Otolith material corre- 
sponding to a time near birth was isolated by extracting 
the core up to the first 3 years and measured for its 14 C 
content. We analyzed the A U C from the first category 
of otolith cores using statistical methods first reported 
by Kastelle et al. (2008). 
Bomb radiocarbon age validation has been used on an 
increasing number of species and is considered one of 
the best methods to confirm the accuracy of fish ages 
(Campana, 2001). Recent uses in the North Pacific in- 
clude that on the white shark ( Carcharodon carcharias ) 
(Kerr et al., 2006), quillback rockfish ( Sebastes malin- 
ger) (Kerr et al., 2005), canary rockfish (S. pinniger) 
(Piner et al., 2005; Andrews et al., 2007), bocaccio rock- 
fish (S. paucispinis) (Andrews et al., 2005; Piner et al., 
2006), Pacific halibut ( Hippoglossus stenolepis) (Piner 
and Wischniowski, 2004), and Pacific ocean perch (S. 
alutus) (Kastelle et al., 2008). 
Radiocarbon fish age validation relies on a time refer- 
ence provided by production of 14 C from atomic bomb 
testing. The above-ground testing of atomic bombs 
that introduced 14 C into the atmosphere and marine 
environment began in the 1950s and continued into the 
1960s (Kalish, 1993; Nydal, 1993). This caused a rapid 
increase in marine 14 C lasting through about 1970 — an 
increase that is recorded in calcified marine organisms 
and otoliths and provides a necessary time reference. 
To validate ages from a “validation species” (in this 
case Dover sole), a 14 C “reference chronology” is used, 
where the exact time frame of the 14 C increase is con- 
sidered known. Two reference chronologies have been 
developed for the North Pacific Ocean: one from Pacific 
halibut (Piner and Wischniowski, 2004) and one from 
yelloweye rockfish (S. ruberrimus) (Kerr et al., 2004). 
The posited birth years for the validation species are 
calculated from ages estimated by otolith growth zone 
counts and date of collection. Specimens representing 
the validation species are chosen such that the range 
of posited birth years spans the period of rapid marine 
14 C increase. Otolith core material deposited in the 
first one or two years of life from the validation spe- 
cies is analyzed and each core provides one 14 C data 
point. To evaluate the ages, the 14 C from the cores of 
the validation species is plotted with respect to the 
posited birth years and compared to the known 14 C 
values in the reference chronology. If there is a timing 
difference between the 14 C increase in the validation 
species and the reference chronology, then the esti- 
mated ages of the validation species are often assumed 
to be in error. Alternatively, if a timing difference is 
not present, the ages from the validation species are 
considered accurate. In a recent bomb radiocarbon age 
validation study of Pacific ocean perch, a series of new 
procedures was used to compare the 14 C measurements 
in the validation samples to the reference chronology 
(Kastelle et al., 2008). We used the same methods 
here — purposely biasing the ages to be validated by ±0, 
1, 2, and 4 years; standardizing the validation sample 
14 C values to the reference chronology; and evaluating 
the residuals between the validation samples and the 
reference chronology to see if inaccuracies in the age 
estimates were present. 
There are two important assumptions when validat- 
ing fish ages with the bomb radiocarbon method (Cam- 
pana and Jones, 1998; Piner and Wischniowski, 2004; 
Piner et al., 2005; Kastelle et al., 2008). Assumption 1 
is that the validation species must be biologically and 
environmentally similar to the species in the reference 
chronology during the first years of life. If both species 
are receiving their 14 C from the same sources, then the 
magnitude and timing of the 14 C increase should be 
similar (Andrews et al., 2007). A reference chronology 
based on the same species as that being investigated is 
best, and occasionally available (Campana, 1997; Cam- 
pana et al., 2002; Piner and Wischniowski, 2004). As- 
sumption 2 requires that the otolith core used for each 
14 C analysis be uncontaminated and that it constitute a 
closed system. Therefore, an accurate extraction of the 
core without contamination from other carbon sources 
or different years is necessary. Dover sole otoliths pre- 
sented a unique challenge in this regard because of 
their small size. For further information regarding 
radiocarbon age validation studies, one can consult the 
earlier mentioned studies from the North Pacific along 
with Kalish (1993, 1995) and Campana (1997). 
Materials and methods 
Otolith selection and coring procedures 
The Gulf of Alaska (GOA) Dover sole otoliths used in this 
study were collected either during AFSC survey cruises 
or by AFSC fishery observers aboard commercial vessels. 
The survey cruises took place in 1984 and 2005; the 
otoliths were removed from the fish at sea, stored in a 
glycerin and thymol mixture, and archived for future age 
determination. The specimens collected from commercial 
harvests were caught in 1998 and treated similarly, 
except they were first stored dry for up to 3 months before 
storage in a glycerin and thymol mixture. The glycerin 
and thymol mixture is not expected to be a contaminant 
in otolith 14 C measurements (Campana et al., 2003). 
After the archival period, the otoliths were aged at 
the AFSC for stock assessments. The initial ages were 
determined by the break-and-burn method (Chilton and 
Beamish, 1982) with the blind-side otolith. Assumed 
annual growth zones were counted by enumerating the 
translucent zones. The otolith growth over the course 
of one year is assumed to consist of an opaque zone and 
a translucent zone. After growth zones were read, the 
otoliths were archived again for varying durations up to 
14 years. Samples where the initial age estimate placed 
the birth year near the era of marine 14 C increase were 
re-examined by age readers experienced in the interpre- 
tation Dover sole otolith growth zones, and considered 
for possible 14 C measurement. In the re-examination 
process otoliths were re-aged to assign a “final age” and 
were placed into two subjective categories based on the 
ease of interpretation of the growth zones: 
