Numerical study on the hydrodynamic background in coastal aquaculture dominated regions and corresponding interactions in the Yellow Sea

Abstract

The Yellow Sea is an important region for aquaculture in China as the main production area for shellfish and seaweed. The aquaculture organisms sometimes can be the major group in a local ecosystem. e.g., in the Sanggou bay where about 84,500 tonnes of kelp (dry weight) and 100,000 tonnes of shellfish (wet weight) are produced annually from a surface area of ~144 km2 (Zhang et al., 2009; Mao et al., 2018). To maintain the development of the aquaculture industry at such scales and to minimize the negative impact on the natural ecosystem, the knowledge of the biological processes at different scales is necessary for decision-makers in the formulation of policy and management strategies. However, a comprehensive description of the biogeochemical process in the aquaculture-affected regions can be highly complicated. Observations are often limited in time and space to fully describe the environmental variations in the aquaculture areas. Numerical models are capable of resolving the ecosystem processes at an often sufficient spatial and temporal scale, but with an increasing complexity from current models describing the physical environment to ecosystem models trying to describe complicated and often less known processes. In this thesis, we have implemented a hydrodynamic model based on the Regional Ocean Modelling system (ROMS) to provide the background physical information for aquaculture related applications, the Yellow Sea Model. We have collected various observations to validate the model, and the results do reproduce reasonably well the ambient environment in aquaculture areas. The tide is the dominating current component in the Yellow Sea, moving the water back and forth continuously. The tide also provides energy on the shallow shelves creating usually well mixed water masses. In the summer, a tidal mixing front is established around the 20-50 m isobaths bordering on the Yellow Sea bottom cold water mass below the seasonal thermocline in the central Yellow Sea. An associated frontal jet flows along the tidal mixed front, transporting water masses along the shelf breaks. The tidal current also make the tidal mixing front oscillate laterally creating temporal temperature variations in the farm regions of bottom cultured scallops. The assessment index derived from these temperature oscillations is correlated to a massive scallop mortality found in the past years. Our model results are also applied to study the baroclinic tides in the northern Yellow Sea, with a semi-diurnal internal tide being present in the stratified waters in the tidal mixing front region. The baroclinic flow associated with this internal tide contributes to enhance the total current in the bottom layer, thus potentially being important for material transportation to farmed scallops. The baroclinic signals are mostly coherent with the barotropic tides, indicating a local generation and a rapid dissipation. Finally, we have established an ecosystem model for the integrated culture of Pacific oyster Crassostrea gigas and kelp Saccharina japonica in Sanggou bay based on a box model concept. The growth of oysters and kelp is simulated at the individual level based on the dynamic energy budget theory. The hydrodynamic information is included as forcing data to compute volume transportation and nutrient exchange. The model is validated with individual growth data recorded in the aquaculture field and water quality data for nutrients from cruises and mooring devices. The model results show that the intensive aquaculture of these low-trophic species is dominant in the local ecosystem and dramatically impacts the phytoplankton population and nutrient flux. The bay acts as a nitrogen sink during the rapid growth stage of kelp from early spring until the harvest in May. The model enables a stocking density adjustment of the culture organisms, thus providing a tool to predict the dynamic process under different scenarios. The model results support that the actual aquaculture stock density, with 50 oyster ind./m2 and 4 kelp ind./m2, is a balanced choice of production and cost based on decades of practical experience.