We have used correlative analysis between mean December–January–February winter wind velocities, measured at the Xisha Meteorological Observatory (16°50′N, 112°20′E) in the middle of the South China Sea, and mean δ18O data for the corresponding month from Porites lutea coral, collected in Longwan waters (19°20′N, 110°39′E), to obtain a linear equation relating the two datasets. This winter wind velocity for the South China Sea (WMIIscs) can then be correlated to the coral δ18O by the equation WMIIscs = −1.213–1.351 δ18O (‰ PDB), r = −0.60, n = 40, P = 0.01. From this, the calculated WMIIscs-δ18O series from 1944 to 1997 tends to decrease during the 1940s to the 1960s; it increases slightly during the 1970s and then decreases again in the 1980s and 1990s. The calculated decadal mean WMIIscs-δ18O series had a obvious decrease from 5.92 to 4.63 m/s during the period of 1944–1997. The calculated yearly mean WMIIscs-δ18O value is 5.58 m/s from 1944 to 1976 and this decreases to 4.85 m/s from 1977 to 1998. That is the opposite trend to the observed yearly mean SST variation. The yearly mean SST anomaly is −0.27° from 1943 to 1976 and this increases to +0.16° from 1977 to 1998. Spectral analysis used on a 54-year-long calculated WMIIscs-δ18O series produces spectral peaks at 2.4–7 yr, which can be closely correlated with the quasibiennial oscillation band (QBO band, 2–2.4 yr) and the El Ñino southern oscillation band (ENSO band, 3–8 yr). Hence most of the variability of the winter monsoon intensity in the middle of the South China Sea is mainly constrained by changes in the thermal difference between the land and the adjoining sea area, perhaps due to global warming.

Interdecadal versus interannual time-scale variations of catch data of three anchovy fisheries distributed in the Humboldt Current System (HCS) were analyzed during the period 1960–2002, by using first a loess smoother. The loess residual data were considered as the interannual variation. Interdecadal changes were highly coherent for the three fisheries of anchovy distributed in the HCS, while there were differences in terms of interannual variations. The north-central Peru fishery was similar to the south Peru-northern Chile fishery in terms of the interannual component of fluctuation of these fisheries which was significantly and negatively related with El Niño-Southern Oscillations events occurring during the spring of the previous year. Interdecadal synchronous variations in the three anchovy fisheries suggest a common forcing by long-term environmental factors in the HCS, although other causes cannot be excluded. During the last 40 years, coherent changes in anchovy fisheries were observed, and this synchronous pattern is out of phase with large-amplitude sardine regimes. Long-term climate variability seems to be the main cause for interdecadal fluctuations in anchovy fisheries in the HCS, although available time series are too short to prove this link and exclude the hypothesis of nonenvironmental forcing. Large-amplitude regime shifts may be more important to consider than interannual changes since they can affect simultaneously and more dramatically the three anchovy fisheries distributed in the Humboldt Current System.