Phytoplankton in the Beaufort and Chukchi Seas: Distribution, dynamics, and environmental forcing

Time-series of remotely sensed distributions of phytoplankton, sea ice, surface temperature, albedo, and clouds were examined to evaluate the variability of environmental conditions and physical forcing affecting phytoplankton in the Beaufort and Chukchi Seas. Large-scale distributions of these parameters were studied for the first time using weekly and monthly composites from April 1998 to September 2002. The basic data set used in this study are phytoplankton pigment concentrations derived from the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), ice concentrations obtained from the Special Sensor Microwave Imager (SSM/I) and surface temperature, cloud cover, and albedo derived from the Advanced Very High Resolution Radiometer (AVHRR). Seasonal variation of ice cover was observed to be the dominant environmental factor as the ice-edge blooms followed the retreating marginal ice zones northward. Blooms were most prominent in the southwestern Chukchi Sea, and were especially persistent immediately north of the Bering Strait in nutrient-rich Anadyr Water and in some fronts. Chlorophyll concentrations are shown to increase from a nominal value during the onset of melt in April to a maximum value in mid-spring or summer depending on location. Large interannual variability of ice cover and phytoplankton distributions was observed with the year 1998 being uniquely associated with an early season occurrence of a massive bloom. This is postulated to be caused in part by a rapid response of phytoplankton to an early retreat of the sea-ice cover in the Beaufort Sea region. Correlation analyses showed relatively high negative correlation between chlorophyll and ice concentration with the correlation being highest in May, the correlation coefficient being −0.45. 1998 was also the warmest in the 5 years globally and the sea-ice cover was least extensive in the Beaufort/Chukchi Sea region, partly because of the 1997–1998 El Niño. Strong correlations were noted between ice extent and surface temperature, the correlation coefficient being highest at −0.79 in April, during the onset of the bloom period.