A Modeling Study Of The Phytoplankton Dynamic In The Bohai Sea

Great progress has been made in the research of ecosystem dynamic models during the last two decades. Physical-biological models of various levels of sophistication have been developed for different regions of the ocean. Firstly the features of different marine ecosystem models are summarized and discussed in this paper. Different functions of the representative biological process and the effect of circulation, water temperature, transparence and solar radiation on the ecosystem are described and analyzed. Based on the historical data, this paper demonstrates the variation and distribution of chlorophyll-a, primary production and nutrient in the Bohai Sea and analyses the features of the physical process. According to the analysis of the historical data, a three-dimensional ecosystem model based on the cycles of phosphate and nitrate is developed for the Bohai Sea, which is coupled with a three-dimensional physical transport model. Sensitivity analysis indicates that the variation of the phytoplankton biomass is sensitive to phytoplankton maximum growth rate, phytoplankton mortality rate, zooplankton grazing rate and remineralization rate of benthic detritus. Light, temperature and nutrient can all affect the production of phytoplankton. If their variations do not change the relation of the limitation, then the variations have little effect on the ecosystem; otherwise they can change the system greatly. The biological process and transparence can both affect the occurrence of the phytoplankton biomass peak and its amplitude. While the variation of the river discharge and the physical transport process can only change the relative amplitude of the peak, but have no effect on its occurrence. It can be concluded that transparence, solar radiation, biological process and the vertical mixing affect the local variation of the ecosystem, while the horizontal advection can spread the local variation of the system to adjacent field and thus affect the distribution of the phytoplankton and nutrient. The model is then used to simulate the variation and distribution of chlorophyll-a, primary production and nutrient in 1982 and the simulation is validated by the data in 1982/1983. The concentration of both DIN and phosphate decreases from spring to summer and increases from autumn to winter in all the areas. The variation is a response to the consumption of the phytoplankton. In spring, the phytoplankton biomass increases as the temperature increasing and reaches the highest peak in summer. The concentration of nutrient drops to the lowest level during the same period. After the phytoplankton bloom of summer the dissolved inorganic nutrients increase gradually as the input of river increase and the decay of the thermocline. The variations of the primary production are same in different regions of the Bohai Sea. The high value of primary production appears in July and August and the lowest value appears in December and January. The primary production in the Laizhou Bay has two peaks and has one in the other part of the Bohai Sea in 1982. Water temperature has great influence on whether the peak of the primary production will appear in spring. Among the four parts of the Bohai Sea, the primary production is the highest in the Laizhou Bay and the lowest in Bohai Bay. The low transparence in the Bohai Bay is the main cause of the lowest primary production in it. So physical factors, such as temperature and transparence have a great effect on the variation and distribution of the primary production. The variation of the nutrient in the Central Bohai Sea is relatively stable while the variation is strong in coastal area of the Bohai Sea. The high concentration locates in the Bohai Bay in winter. The phosphate concentration in the northwest part of Liaodong Bay maintains a high level in the whole year. The value of phytoplankton biomass of the whole Bohai Sea is low in winter and the high value of the biomass first appears in the coastal area in spring. The high value expands from Laizhou Bay and Bohai Bay to the Central Bohai Sea in summer and the concentration of chlorophyll-a is about 1 mg/m3 in the Central Bohai Sea. The isoline of 1 mg/m3 moves to the coastal area again in autumn. The Huanghe river has great influence on the distribution of nitrate and phytoplankton concentration in the Laizhou Bay. The vertical distribution of nutrients is homogeneous in the whole year, except in the summer the concentration at the bottom layer is greater than that in the surface. This is due to the occurrence of thermocline in the Bohai Sea. The stratification does not decrease the concentration of chlorophyll-a in summer but has an effect on the vertical distribution of chlorophyll-a. The stability of water with enough supply of nutrient is propitious to the production of phytoplankton. Production and respiration are the most important sink and source of nutrients. The remineralization of the detritus pool is an important source of nutrient regeneration. It can compensate 23 percent of the consumption of nutrient from the production process. The net nutrient budget is -3.05 kilotons of P and 31.6 kilotons of N on the whole. The net carbon budget is 110 kilotons and the Bohai Sea is a weak sink of CO2 of the atmosphere. 13.7% of the gross primary production transfers to high trophic level. Remineralization at the bottom is the mechanism of transferring the nutrient from organic form to inorganic form. Finally a biological model, coupled with a three-dimensional physical transport model is described. Simulated distributions of plankton and nutrient are obtained from this model with nitrate input from a river. The simulations are then resampled and the data are used in numerical experiments to assess the ability of using an adjoint data assimilation approach for estimation the poorly known parameters of this ecosystem model. The spatial resolution and temporal resolution of the data are complementary in the assimilative model, thus the improvement of either of them can result in better recoveries of the model parameters. The assimilation of phytoplankton data is essential to recover the model parameters. Observational data collected during a Sino-German cooperation are taken as an example. Twin experiments, using simulated data of the same type and spatial and temporal distribution as that of the investigation, demonstrate the feasibility to estimating the model parameters with thein situ observations. Some advice is given about the designing of sampling strategies for making measurements in a project like ours.