44 
Fishery Bulletin 116(1) 
120 
Figure 3 
(A) Estimates of the number of recruits (individuals between 5 and 6 months old) 
of European anchovy (Engraulis encrasicolus) in the Gulf of Cadiz, Spain, during 
1988-2004, determined from the study model. (B) Population size (lines) versus 
catches reported by the International Council for the Exploration of the Sea (bars) 
for European anchovy in the Gulf of Cadiz. Dotted lines indicate the 5 and 95 
percentiles, and the solid line indicates the median of the posterior distribution. 
Quarterly reported catches were transformed to preserve the monthly scale. 
al. (2009). Therefore, the numerical burden necessary 
to approach convergence would render intractable if 
we had formulated this model with weekly time steps. 
Although our computational time demanded by our ap¬ 
proach is still large, it is tractable. In particular, with 
the advances in parallelization and computing power 
available for research, this tractability can be used 
with this new approach in the coming years as a tool 
to run models that resolve fine scales at short opera¬ 
tional times. 
Admittedly, it would have been more precise to use 
monthly varying F values, but constant F was chosen 
for our study to achieve convergence of parameters in 
reasonable time. In spite of constant F, the model re¬ 
sulted in a better coherence between modeled and ob¬ 
served abundances: Figure 4 of this article, in contrast 
to Figure 7 of Ruiz et al. (2009), illustrates this fact for 
the seasonal variations and interannual tendencies of 
juveniles in the Gulf of Cadiz. This improvement is evi¬ 
dent if we assess the performance of Ruiz et al. (2009) 
versus the present model as shown in Tables 3 and 4. 
These tables synthesize deviations between model out¬ 
puts and all available data. These data include those 
that are incorporated in the modeling exercise (catch. 
