3.2.4. Small pelagic regimes and regional interdecadal climate variability

Periods of low (1941 to 1961) and high (1971 to 1977) differences of the SOI-AL series were used as reference periods for computing departures from the seasonal cycle of sea surface temperature (SST), sea level pressure gradient, and thermocline depth within selected systems (Figure 2). Figures 24 and 25 present the results of the analysis for the SST variability. Figures 26 and 27 those of pressure gradient, and Figures 28 and 29 those of thermocline depth.

SST results for most systems where large populations of sardines and anchovies grow suggest that periods of large SOI-AL differences (1971 to 1977) tend to be warmer than those periods when the tropical and the extratropical indices show parallel trends (1941 to 1961, when the SOI-AL differences are relatively small). California, Japan, Humboldt, and the Canary systems seem to behave this way. The only exception is the Benguela system where no difference between periods seems particularly evident. A similar tendency is also noticeable for the Eastern Tropical Pacific, whereas other systems such as Australia, Brazil, and Somali tend to behave in the opposite way.

Figure 24. Seasonal climatology (1: winter, 2: spring, 3: summer, 4: fall) of SST for the nine boxes shown on Figure 2. Base period: 1900 to 1990.

Figure 25. Seasonal departures (1: winter, 2: spring, 3: summer, 4: fall) of SST for the nine boxes shown in Figure 2. Referenced periods: 1941 to 1961 (solid circles) and 1971 to 1976 (open diamonds).

Figure 26. Seasonal climatology (1: winter, 2: spring, 3: summer, 4: fall) of SLP indices for the nine boxes shown in Figure 2. Base period: 1900 to 1990.

Figure 27. Seasonal departures (1: winter, 2: spring, 3: summer, 4: fall) of SLP indices for the nine boxes shown in Figure 2. Referenced periods: 1941 to 1961 (solid circles) and 1971 to 1976 (open diamonds).

Figure 28. Seasonal climatology (1: winter, 2: spring, 3: summer, 4: fall) of thermocline depth for the nine boxes shown on Figure 2. Base period: 1900 to 1990.

Figure 29. Seasonal departures (1: winter, 2: spring, 3: summer, 4: fall) of thermocline depth for the nine boxes shown in Figure 2. Referenced periods: 1941 to 1961 (solid circles) and 1971 to 1976 (open diamonds).

However, the described pattern is not evident from the thermocline depth analysis. Whereas this feature tended to be shallow in California and the Eastern Tropical Pacific during the 1971 to 1977 period, it tended to be deeper in Japan and Canary, and no clear tendencies are evident for the other systems where large fisheries of small pelagics occur (Humboldt and Benguela). At this point, it is not clear whether this lack of a coherent pattern is the result of the particular dynamics of each system or an artifact related to data scarcity. Similar ambiguous results were obtained for the pressure gradient analysis. Despite that differences for most systems seem to occur between the selected periods, no coherent pattern is observed. It is evident that proper evaluation of the interdecadal variability at the regional level will require more detailed approaches than the general analysis intended within this work.