Monsoonal alternation of a mixed and a layered structure in the phytoplankton of the euphotic zone of the banda sea (Indonesia):A mathematical analysis of algal pigment fingerprints

The chlorophyll a concentration of the upper 25 m of the euphotic zone of the Banda Sea was 5 times higher in August 1984, in the upwelling season during the southeast monsoon, than in February 1985, during the northwest monsoon when a deep chlorophyll maximum layer was presented at 40–80 m. Similarity between stations was calculated by means of a cluster analysis of the concentrations of 4 different chlorophylls and 6 carotenoids. High similarity over large areas, one to the west and one to the east of the Banda Arc, is evidence of the importance of large-scale phenomena influencing the phytoplankton during the monsoons. All samples taken in the deep chlorophyll maximum layer during the northwest monsoon were clustered as a separate group.Multiple regression analysis of chlorophyll a and the four most important accessory pigments suggests that in August 1984 (upwelling season) the fucoxanthin-containing group (mainly diatoms) contributed 50% to the chlorophyll a in the euphotic zone; green algae and Prymnesiophyceae contributed each 20%, Cyanobacteria only 9%. In February 1985, during the downwelling season, pico- and nanoplankton containing 19'-hexanoyloxyfucoxanthin (Prymnesiophyceae) and zeaxanthin (Cyanobacteria) each contributed 40% of total chlorophyll a, both near the surface and in the deep chlorophyll maximum layer; 19'-butanoyloxyfucoxanthin (a marker of certain Prymnesiophyceae and Chrysophyceae) was only present at depth. On the assumption that Chl. b is a pigment confined to green algae, this group contributed 14% of the Chl. a in the deep maximum. Other green-algal pigments were, however, scarce at depth (lutein, violaxanthin), which suggests presence of Prochlorophyta.Shifts in pigment ratios (zeaxanthin : Chl. a; Chl. b : Chl. a), probably related to light adaptation, hinder the use of algal pigment fingerprints for estimating quantitative composition of natural phytoplankton; indeed, differences between samples in pigment signature can have both a floristic and an ecophysiological background. Statistical analysis of pigment fingerprints may further be complicated due to presence of intermediates of pigment synthesis and breakdown, and poor taxon specificity of fucoxanthins and possibly even of Chl. b and zeaxanthin