Sensitivity analysis of vertical mixing schemes in a regional domain using modular Ocean model.

In this study, a regional modelling approach is adopted using the Modular Ocean Model version 5 (MOM5) with implementation of open boundary condition (OBC) in the Bay of Bengal (BoB) region. The domain is given an initial stratification profile of climatological temperature and salinity and is forced using seasonally varying wind and air-sea fluxes at the surface. A sensitivity analysis is performed by changing the vertical mixing schemes. Changes in vertical mixing schemes have significant impact on the ocean system, including the surface properties. Two different vertical mixing models are used: K-profile parameterization (KPP), and k- ∈ turbulence model. Inter-comparison between these two experiments suggests that k- ∈ model performs seemingly better than the KPP model. With k- ∈ mixing scheme, the simulated sea surface temperature (SST) and mixed layer depth (MLD) are closer to observations. Model generated turbulent kinetic energy (k) and dissipation ( ∈) are different in the two experiments because of different representations of k and ∈ in the two models. The results show the importance of regional modelling and the use of different mixing schemes. Employing a physics-based vertical mixing scheme, such as the k- ∈ turbulence model, shows improvement in the model physics and the model simulated ocean parameters. [ABSTRACT FROM AUTHOR]