English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Tropical cyclone (TC) forecast models have numerous components that must be properly implemented to achieve accurate forecast of TC track and intensity. The simulation of upper ocean response underneath a TC is of particular importance to TC intensity forecasts. This is because a TC relies on the upper ocean as a source of latent heat energy for its formation and maintenance of its intensity. When a large wind stress applies at the ocean surface underneath a TC, it enhances upper ocean vertical turbulent mixing, promotes deepening of the mixed layer and entrainment of colder water from below the thermocline, and reduces the sea surface temperature (SST). Even a slight SST decrease can significantly reduce the latent heat flux to the atmosphere and weaken the storm. For this reason, coupling of an ocean model to the atmospheric model used for TC forecasting is essential. Accuracy of upper ocean modeling strongly depends on the quality of the ocean model initialization, the wind forcing (air-sea momentum flux) specification, and the upper ocean mixing scheme. Recent studies suggest that the latter two can be significantly modified by different sea states (surface wave fields) and that accurate upper ocean modeling requires coupling of a wave prediction model to the TC-ocean model, as well as implementation of the sea-state dependent upper ocean mixing scheme (Langmuir Turbulence, or LT parameterization). In this thesis we investigate both ocean initialization methods and ocean mixing schemes in the Message Passing Interface Princeton Ocean Model (MPIPOM) underneath ideal and realistic TC wind fields and evaluate their performance against available observations. The second chapter evaluates the Feature-Based (FB), HYbrid Coordinate Ocean Model (HYCOM) and the Navy Coupled Ocean Data Assimilation (NCODA) ocean model initializations in a standalone version of MPIPOM. The model ocean is compared to satellite sea surface temperature (SST) and Airborne eXbendable BathyThermograph (AXBT) data before, during, and after Hurricane Edouard has passed the region. Both statistical and visual comparisons are used to evaluate the methods. All three initialization methods are shown to produce reasonable initial ocean temperature fields as well as reasonable SST cooling responses, with some notable biases. The third chapter investigates the effects of implicit (sea-state independent) and explicit (sea-state dependent) Langmuir turbulence parameterizations, based on the modified K Profile Parameterization scheme, on the upper ocean underneath TCs. The MPIPOM is coupled with the WaveWatch III (WW3) surface wave model to simulate the complex surface wave fields generated by TCs. Both idealized and Hurricane Edouard wind fields are utilized. The air-sea momentum flux budget is also implemented, accounting for the change in momentum flux to ocean currents due to growing or decaying wave fields. These experiments are performed on an idealized horizontally uniform ocean to clarify the impacts of different mixing schemes. In agreement with previous studies, the sea-state dependent mixing schemes are shown to significantly modify the spatial distribution of SST cooling underneath TCs. The final chapter utilizes the same model setup as the third chapter with two realistic ocean initializations (FB and Real-Time Ocean Forecasting System, RTOFS), a realistic bathymetry, and the Hurricane Edouard wind forcing. The Feature-Based initialization method is found to be more accurate than the RTOFS method. The SST cooling patterns agree well with those in the idealized cases of the previous chapter, as expected. Visual and statistical comparisons with satellite SST and AXBT observations suggest that the sea-state dependent Langmuir turbulence mixing scheme is more skillful than the implicit scheme in predicting the cold wake generated by the storm. The results therefore suggest that the coupling of both wave and ocean models to a TC model, with an explicit sea-state dependent mixing scheme, is needed for accurate predictions of SST and upper ocean structure underneath TCs.

Evaluation of Updated MPIPOM-TC Ocean Model Initializing and Sea-State Dependent Mixing Scheme