Zeidberg et al The fishery for Loligo opa/escens from 1981 through 2003 



t/l o 



O) o 



c o 



by the frequencies determined from the 

 spectral analysis. These splits resulted in 

 three separate means for CPUE in APR 

 (7.5-year frequency) and five means for 

 the OCT region (4.5-year frequencies). 

 Anomalies of CPUE from these means 

 were compared to the climatic indices, 

 and had significant linear regressions to 

 NIN03, SOI, and UI anomalies, but ex- 

 plained less than 5% of the variance (data 

 not shown). 



1981-2003 squid fishery data 



In comparisons of landings (Fig. 3A) by 

 month, the six areas fell into two catego- 

 ries: APR and OCT. Effort in vessel-days 

 and CPUE showed similar trends. The 

 Loligo opalesce/TS squid fishery generally 

 occurs from April through November in 

 APR. Although landings peak in May, by 

 then there are so many vessels in oper- 

 ation that CPUE has dwindled to half 

 that of April (Fig. 3, B and C). There is a 

 second landings pulse in August. 



In the five areas of the OCT grouping, 

 landings typically begin in October, build 

 to a peak in January, and diminish to 

 lows in August (Fig. 3A). A large uni- 

 modal pulse of squid landings occurred in 

 November for all areas except SCB. The 

 SCB had a bimodal recruitment pulse, the 

 two largest recruitment events in all of 

 California: one in November and a larger 

 one in January. In SD, like SCB, land- 

 ings peaked in January, but there was no 

 strong November signal in this region. 



CPUE by month for APR was typically half that of 

 OCT. The APR CPUE varied between 8 and 20 tons/ 

 vessel-day, for months with more than seven vessel- 

 days, whereas CPUE for the southern California (OCT) 

 areas ranged from 17-36 tons/vessel-day (Fig. 3C). 



Time-series analysis 



The periods of the ten highest peaks of the variance 

 spectrum were determined for all six areas (Table 2). 

 The largest peaks from the spectral analysis occurred 

 at periods of 372 and 356 days, or roughly 1 year for 

 all areas. There was a 7.5-year peak in the MB and CC 

 areas. There was a 4.5-year peak (for all areas except 

 MB) that was similar to the period of the four El Nifio 

 events that occurred in this area during the 22-year 

 period. There was a 3.7-year peak for all areas except CC 

 and SM. The seven-day cycle is most likely a stochastic 

 factor of fishermen working within weekend closures 

 because data before 1998 did not have this periodicity. 

 There was no 28- or 14-day cycle in any of the areas; 

 this finding likely indicates that spawning squid do not 

 respond to tidal currents or lunar light. 



10 



iC 



- a o.. o- ° 



o O • 



•o •» o ° 



— *l- — I 1 1 1 1 1 1 1 1 , ^ 



1 2 3 4 5 6 7 8 9 10 11 12 



Month 



Figure 3 



Fishery data for Loligo opalescent:, summed by month for 1981-2003. 

 (A) Landings (metric tons). (B) Effort in vessel-days (VD). (C) Catch 

 per unit of effort (tons/VD). Monterey Bay (April [APR] — black circles) 

 and southern California (October [OCT] — unfilled circles). Scale of 

 y-axis changes between A, B, and C. Largest landings occurred one 

 month later in May in the APR region and in November in the OCT 

 region when SST was 11.7"C and 16.1°C, respectively. 



The most significant cross correlations of time lag 

 analysis for CPUE to SST are listed in Table 3. In all 

 cases of biological significance, CPUE lagged SST by 

 4, 5, or 10 months. MB CPUEs were highest (in May) 

 when SST was low four months earlier (Jan), and hence 

 gave a negative correlation. In all other regions, the 

 four or five month correlation was positive, with CPUE 

 high (Nov) when SSTs were high four months earlier 

 (Jul). For the southern California areas there was a 

 negative correlation with SSTs 10 months earlier (Janl. 

 Therefore, cold winters and warm summers correlate 

 with larger landings. Recruitment of spawning adults 

 to the fishery occurs during the productive seasons in 

 both APR, MB, and OCT. Productivity in APR and MB 

 co-occurs with the spring-summer upwelling season, 

 and in OCT productivity correlates with winter storms 

 that lead to a deeper mixed layer. There were signifi- 

 cant cross correlations with SOI, NIN03, and UI, but 

 not with the anomalies of SOI and UI. Interestingly 

 correlations for NIN03 were greater than those for SOI 

 (Fig. 4), indicating that the CPUE of Loligo opalescens 

 is more closely related to the "oceanic teleconnection" 

 than to "atmospheric teleconnection," 



