ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Data-driven feature-oriented modeling of Southwest Monsoon Current

Gadi, R and Vinayachandran, PN and Subramani, DN (2021) Data-driven feature-oriented modeling of Southwest Monsoon Current. In: Ocean Modelling, 168 .

[img] PDF
oce_mod_168_2021.pdf - Published Version
Restricted to Registered users only

Download (10MB) | Request a copy
Official URL: https://doi.org/10.1016/j.ocemod.2021.101912


The multiscale synoptic circulation in the southern Bay of Bengal region is simulated using a high-resolution Feature-Oriented Regional Modeling and Simulation (FORMS) approach. First, the region's prevalent ocean circulation features are identified and mapped onto a FORMS template using a literature survey, and analysis of in-situ and remote sensing data. These features include the Southwest Monsoon Current (SMC), Sri-Lanka Dome, Anti-cyclonic Eddy, and Subsurface Baroclinic Eddy to the east of the SMC. New data-driven feature models are developed for the SMC and Subsurface Baroclinic Eddy using observational transects collected during the Bay of Bengal Boundary Layer Experiment (BoBBLE) of July 2016. Specifically, a new elliptical feature model for the SMC's high salinity core is developed and utilized. The developed temperature and salinity feature models are placed on the FORMS template, and pseudo observation profiles are sampled for use in an objective analysis initialization scheme. The background for the objective analysis is taken to be the NEMO global analysis model data on the day of the synoptic initialization. The FORMS derived initialization fields on 04 July 2016 are used with the MIT-MSEAS Primitive Equation modeling system to simulate the ocean circulation in the region from 04 July to 15 July 2016. The forecast skill of the model is quantified by comparison to the time-series observations collected during BoBBLE (but not used in the FORMS development) and Argo data available in the region during the simulation. Four experimental short-term synoptic simulations at a 3 km horizontal resolution and 70 vertical levels, with and without the different components of the feature models are showcased. Results are utilized to calibrate and improve the feature model, including the addition of a Subsurface Baroclinic Eddy and the development of a new surface bias-correction scheme to obtain a better match with validation datasets. The present data-driven feature modeling approach improves the representation and forecast skill for several critical dynamical features: (i) the initial high salinity core at the time-series location (ii) evolution and expansion of the sub-surface high salinity core; and (iii) salinization and freshening events at the time-series location between 3 m to 20 m. The MSEAS model simulation from the final experiment match the BoBBLE temperature and salinity data (total RMSE of 0.14 psu in salinity and 0.57 C in temperature in between 60-120 m depths and total RMSE of 0.17 psu in salinity and 0.33 C in temperature in between 3-60 m depths) and the Argo measured temperature and salinity (total RMSE of 0.16 psu in salinity and 0.66 C in temperature), showcasing the model's success in providing excellent spatio-temporal representation of the ocean in the region of interest. © 2021 Elsevier Ltd

Item Type: Journal Article
Publication: Ocean Modelling
Publisher: Elsevier Ltd
Additional Information: The copyright for this article belongs to Elsevier Ltd
Keywords: Atmospheric thermodynamics; Boundary layer flow; Boundary layers; Oceanography; Remote sensing; Time series; Time series analysis, Bay of Bengal; Bay of bengal boundary layer experiment; Data assimilation; Feature models; Feature-oriented; Model and simulation; Primitive equations; Regional modelling; Southwest monsoon; Synoptic forecasting, Forecasting, baroclinic instability; baroclinic motion; calibration; data assimilation; monsoon; numerical model; oceanic current; spatial resolution, Bay of Bengal; Indian Ocean
Department/Centre: Division of Interdisciplinary Sciences > Computational and Data Sciences
Division of Mechanical Sciences > Centre for Atmospheric & Oceanic Sciences
Date Deposited: 22 Nov 2021 11:25
Last Modified: 22 Nov 2021 11:25
URI: http://eprints.iisc.ac.in/id/eprint/70533

Actions (login required)

View Item View Item