Simulation of annual plankton productivity cycle in the Black Sea by a one‐dimensional physical‐biological model

Abstract

The annual cycle of the plankton dynamics in the central Black Sea is studied by a one‐dimensional vertically resolved physical‐biological upper ocean model, coupled with the Mellor‐Yamada level 2.5 turbulence closure scheme. The biological model involves interactions between the inorganic nitrogen (nitrate, ammonium), phytoplankton and herbivorous zooplankton biomasses, and detritus. Given a knowledge of physical forcing, the model simulates main observed seasonal and vertical characteristic features, in particular, formation of the cold intermediate water mass and yearly evolution of the upper layer stratification, the annual cycle of production with the fall and the spring blooms, and the subsurface phytoplankton maximum layer in summer, as well as realistic patterns of particulate organic carbon and nitrogen. The computed seasonal cycles of the chlorophyll and primary production distributions over the euphotic layer compare reasonably well with the data. Initiation of the spring bloom is shown to be critically dependent on the water column stability. It commences as soon as the convective mixing process weakens and before the seasonal stratification of surface waters begins to develop. It is followed by a weaker phytoplankton production at the time of establishment of the seasonal thermocline in April. While summer nutrient concentrations in the mixed layer are low enough to limit production, the layer between the thermocline and the base of the euphotic zone provides sufficient light and nutrient to support subsurface phytoplankton development. The autumn bloom takes place sometime between October and December depending on environmental conditions. In the case of weaker grazing pressure to control the growth rate, the autumn bloom shifts to December–January and emerges as the winter bloom, or, in some cases, is connected with the spring bloom to form one unified continuous bloom structure during the January–March period. These bloom structures are similar to the year‐to‐year variabilities present in the data.