A new insight into oceanography with multivariate and time-series analyses on the 1990-1999 planktonic foraminiferal fluxes in the Bering Sea and the central subarctic Pacific
A nine year-long study on planktonic foraminifer fluxes was conducted in the Bering Sea (Station AB) and the central subarctic Pacific (Station SA). Mathematical analyses were carried out in order to decipher the consequences of temporal variation of foraminifer fluxes and their faunal assemblages with the oceanographic variability. As a result of R-mode (true) factor analysis, four factors were recognized: Warm Water Factor; Seasonal Factor; Temperature Factor; and Oligotrophic Factor. The temporal variations of these four factors appeared to correspond to the oceanographic changes occurred at each of the stations. Among the temperature-related factors, Temperature Factor showed reasonable correlation coefficients with SST anomaly at both stations (Station AB: $ r=0.57 $; Station SA: $ r=0.83 $). Notable high scores of Warm Water Factor occurred during 1997 at Station SA. During this period, a significant suppression of the other measured biogenic particles such as diatoms occurred. This extraordinary event observed at Station SA may be a reflection of the large scale climatic shift. The temporal variation of Oligotrophic Factor at each of the stations demonstrated different oceanographic features which controlled the local primary production. At hemipelagic Station AB, the temporal variation of the SST anomaly, which is a measure of controlling the subsurface nutrient supply to surface layer, was the significant feature influential to the local primary production. On the other hand, the power balance between the Alaskan Stream and the Subarctic Current influenced most to the local primary production at pelagic Station SA. The Maximum Entropy Method revealed that temporal variation of foraminifer fluxes, their relative abundances, and the factor scores fluctuated with significant periodicity. Most fluxes of foraminifer taxa at Stations AB and SA showed significant seasonal (six month, and twelve month-long) and inter-annual (twenty to forty month-long) cycles. Seasonal cycles were attributed to the occurrence of the primary flux maxima, which occurred in fall (Station AB) and spring (Station SA), and the secondary maxima appeared in spring (Station AB) and fall (Station SA). The periodicities of the relative abundances were the most significant for the twelve month-long cycle. The nine year-long monthly means of the relative abundances and their twelve month-long cycles indicated the preference of each of the foraminiferal taxa in a specific condition. Contrary, the periodicity of four factors tended to appear inter-annually, except for Seasonal Factor. It was the only factor with twelve month-long cycle. The other factors appeared to have twenty to forty month-long cycles. The inter-annual cycle of these factors may be a reflection of the large scale climatic shift occurred in the Bering Sea and the subarctic Pacific during 1990 to 1999.
九州大学理学部紀要 32(1), 73-96, 2007-02-01