Chang et al.: The Antarctic Oscillation index as an environmental parameter for predicting catches of ///ex argentinus 
203 
Figure 1 
Map of the study area (34-55°S, 50— 70°W) and spatial distribution of 
annual mean catch per unit of effort (CPUE), measured in metric tons 
(t) per vessel per day, of Argentine shortfin squid ( Illex argentinus) from 
1986 through 2010 in the southwest Atlantic. The dashed line indicates 
the 200-m isobath, and the 2 black stars indicate the 2 locations on the 
Patagonian shelf where subsurface seawater temperatures were taken: 
36°S, 53°W (north) and 50°S, 60°W (south). The area of the Antarctic Os- 
cillation (AAO) is shown between 40°S and 65°S. 
Chen et al., 2007b). Variability in recruit- 
ment strength of Argentine shortfin squid 
could be explained partly by cold events 
in the southwest Atlantic, which could be 
further connected to the SST anomalies 
in the Pacific (Waluda et al., 1999). Tem- 
peratures of the sea surface and water 
column could also affect spatiotemporal 
distribution of Argentine shortfin squid 
and production of the fishery (Waluda et 
al., 2001b; Bazzino et al., 2005; Sacau et 
al., 2005; Chen et al., 2007b). 
The Antarctic Oscillation (AAO) re- 
flects large-scale changes in atmospheric 
mass between mid- and high-latitude 
surface pressures in the Southern Hemi- 
sphere (Gong and Wang, 1999) and in- 
fluences precipitation, wind, sea ice, 
and SST variability (Silvestri and Vera, 
2003; Turner et al., 2007; Justino and 
Peltier, 2008; Vasconcellos and Cavalcan- 
ti, 2010). The AAO may, therefore, affect 
variability in abundances and distribu- 
tion of Argentine shortfin squid that 
inhabit the South Atlantic. However, in 
no studies has the AAO been considered 
as an environmental factor to examine 
variation in abundance of Argentine 
shortfin squid. Our hypothesis is that 
the AAO could be an important environ- 
mental factor that affects abundance of 
Argentine shortfin squid. We give rela- 
tionships between the catch per unit of 
effort (CPUE) of this squid species and 
both subsurface seawater temperatures 
and regional atmospheric forcing, AAO, 
with and without effects of time-lags. 
Because of environmental conditions that may af- 
fect the abundance and distribution of squids during 
their life cycles, a number of researchers have sug- 
gested incorporation of environmental factors into 
squid stock assessment methods (Dawe et al., 2000, 
2007; Georgakarakos et al., 2002; Sakurai et al., 
2002; Pierce and Boyle, 2003; Waluda et al., 2004; 
Peel and Jackson, 2008; Pierce et al., 2008; Kidokoro 
et al., 2010). Rodhouse (2001) indicated that recruit- 
ment variability in several squid species could be ex- 
plained partly by environmental variability reflected 
in synoptic oceanographic data and suggested that 
the ability to predict recruitment in advance gives 
managers and vessel operators the advantage of be- 
ing able to plan ahead. 
The main environmental factor that could affect the 
physiology of squid is seawater temperature. Previous 
studies have demonstrated the potential effects that 
sea-surface temperature (SST) can have on abundance 
and distribution of squid at different life cycle stag- 
es (Waluda et al., 1999; Waluda et al., 2001a, 2001b; 
Materials and methods 
Fishery data 
The study area, at 34-55°S and 50-70° W, covered all 
the fishing sites of Taiwanese squid jiggers and likely 
included the primary distributional range of Argentine 
shortfin squid in the southwest Atlantic (Chen et al., 
2007a) (Fig. 1). Fishing logs of jiggers were compiled by 
the Fisheries Agency (Council of Agriculture, Executive 
Yuan, Taiwan), and a data set for the period 1986-2010 
was prepared for this study. Each data record included 
the date and site (latitude and longitude) of daily fish- 
ing effort, and the daily catch amount in kilograms. 
The CPUE was calculated as the catch in metric 
tons (t) of squid per vessel per day (t-v^-d -1 ) (Chen et 
al., 2007a, 2007b; Waluda et al., 2004; Wu et al., 2009; 
Fang et al., 2013) and was standardized by a relative 
fishing power method where unit catches of paired 
vessels that operated under similar fishing conditions 
were analyzed (Chen and Chiu, 2009). In each fishing 
