Page 16 - Lloreta_alii_2001
P. 16

relationship between two variables should indeed not depend on the time periods on

                  which it is calculated, otherwise this would mean that the relationship is not reliable


                  and/or the time series are not consistent. However, we recomputed the analyses on a

                  common window greater than 80 years, i.e., 1868-1960, considering seven BFT time


                  series (Bonagia and Medo das Casas being removed since they were never synchronous

                  over such a period with the other BFT time series). The results computed on the


                  common period are highly similar to those computed on different periods, so that the

                  overall conclusion, i.e., the relationship with temperature, holds (see Figure 6).

                        Recent studies put forward that the NAO  could affect Atlantic bluefin tuna,


                  through its impact on recruitment (Marsac, 1999; Santiago, 1998). However, Fromentin

                  (2002a) and the recent ICCAT workshop on environment and tuna recruitment (ICCAT,


                  2002) investigated the NAO/recruitment relationship, using further statistical analyses,

                  and found no clear connection between BFT recruitment and NAO. Our results, based


                  on long-term comparisons, support these last findings. The time delay between NAO

                  and the recruitment period can be a critical question, especially when analyses focus on


                  year-to-year fluctuations and are computed on short time series (which is, however, not

                  the case in the present study). We, thus, investigated delayed effects between Hurrell’s


                  NAO index and BFT time series, using a lag from 1 to 5 years (4 to 5 years

                  corresponding to the age-at-maturity for the East Atlantic BFT). These analyses (results

                  not shown) did not yield to significant relationships with the three NAO indices, so that


                  we may reject, from a statistical viewpoint, a possible connection between long-term

                  fluctuations of the BFT and NAO.


                        Klyashtorin (1998) put forward that the atmospheric circulation index (ACI), that

                  is tightly related to the excess Length Of the Day (LOD), was closely correlated with


                  catches of the main Atlantic and Pacific commercial pelagic fish. This study focused on





                                                            17
   11   12   13   14   15   16   17   18   19   20   21