Application of packed bed reactor theory and Bayesian inference to upweller culture of juvenile oysters

摘要

The use of upweller culture units in bivalve nurseries is widely practiced as a technique that enhances the ability to rear large quantities in a semi-controlled environment. However, guidance has varied for optimal flow rates, and thus there is a need to develop a more mechanistic assessment. The application of packed bed reactor theory, including axial diffusion models, would improve optimization of these culture methods. The following paper presents a series of controlled experiments to determine the hydrodynamic properties of a packed bed of oysters. The data gained from these experiments was used to develop mechanistic models calibrated through Bayesian inference. Specifically, the Ergun equation and the axial diffusion model were used to predict the experimental data. The Ergun equation was able to predict the hydrodynamic equivalent diameter distribution of oyster shells (mu = 3.18 mm, ( )sigma = 0.74 mm). This oyster shell diameter and void ratio distribution gained through the Ergun equation were used in the relationship of axial diffusion and superficial velocity. The mean axial diffusion coefficient in the oyster bed was estimated 1.65 x 10(4) m(2)/s at 0.01 m/s and 7.26 x 10(4) m(2)/s at 0.08 m/s. The use of Bayesian inference allows for greater understanding of the credibility of individual parameter distributions (i.e., rates and physical attributes) within these mechanistic formulations. This work establishes a baseline methodology to systematically evaluate and optimize bivalve upweller culture systems.