Predictability of ROMS-Ocean State Ocean Model using Information Theory

Aakash Sane1, Baylor Fox-Kemper2, David S Ullman3, Lewis M Rothstein4 and Christopher R Kincaid3, (1)Brown University, Providence, United States, (2)Brown University, Department of Earth, Environmental and Planetary Sciences, Providence, United States, (3)University of Rhode Island, Narragansett, RI, United States, (4)University of Rhode Island- Narragansett Bay, Narragansett, RI, United States
Abstract:
The Ocean State Ocean Model (OSOM) is an implementation of the Regional Ocean Modeling System (ROMS) covering Rhode Island estuaries which includes the Narragansett Bay, Mt. Hope Bay, and nearby regions including the shelf circulation from Long Island to Nantucket. Our focus is: on modeling the physical aspects of the Bay in order to build a forecast and prediction system; use the model to find the estuarine time scales and compare them to previous studies.

Perturbed ensemble simulations with altered initial condition parameters (temperature, salinity) are combined with concepts from Information Theory to quantify the predictability of the OSOM forecast system. Predictability provides a theoretical estimate of the potential forecasting capabilities of the model in the form of prediction time scales and enhances readily estimable timescales such as the freshwater flushing timescale. The predictability of the OSOM model is around 10-40 days, varying by perturbation parameters and season.

Freshwater flushing time scale and total exchange flow was calculated for the OSOM model. The freshwater flushing time scale was found to be ~20 days and varies with the choice of the estuary boundary. The predictability time scales and flushing time scales reveal important dynamics of the tracers involved and elucidate their role in driving the estuary.

Back to: The Dynamics of Buoyancy-Driven Flows in Estuaries, Continental Shelves, and Polar Seas IV Posters