Upper Ocean Response to Hurricane Wind Asymmetries

Hurricane Gilbert (1988) was one of the strongest storms in the Atlantic in recent history. During its passage in the western Gulf of Mexico, a warm core eddy was positioned at a distance of about 4 R_max (radius of maximum winds, 60 km) from the storm-track. High resolution observations were acquired by deploying Airborne eXpendable Current Profilers (AXCPs) and Airborne eXpendable BathyThermographs (AXBTs) before, during and one and three days following the storm passage. Hurricane Gilbert was going through eye wall replacement in the western Gulf of Mexico, that resulted in a very broad wind field with dual wind maxima. Using flight level winds, buoy measurements and background winds from ECMWF, realistic surface winds were estimated using the Hurricane Research Division Wind Analysis package. The oceanic measurements are used to initialize the Miami Isopycnic Coordinate Ocean Model (MICOM) in the Gulf of Mexico. This model uses explicit mixed layer physics with different entrainment mixing mechanisms to investigate the upper ocean response including the heat, mass and momentum balances in the vicinity of the mesoscale features. By using realistic wind forcing, the numerical simulations address the effect of dual wind maxima and the associated wind asymmetries on the ocean response and the modulation of air-sea fluxes.

With the ocean model initial conditions being the same, symmetric (by symmetric here we mean the vector sum of storm centered symmetric component and the translation speed) and total winds are used to force the model and the results are compared with observations. While the thermal response for the realistic oceanic initial condition compares well with the AXCP data for both total and symmetric wind fields, the area averaged surface fluxes in the directly forced region increase by about 20 to 30 % for the total wind forcing. A similar increase in fluxes are also seen by using realistic initialization of the ocean model with the eddy in the domain with a marked decrease of 0.8 to 1.1° C in the rms differences between simulated and observed temperatures in the upper ocean. Thus, mesoscale wind asymmetries and realistic oceanic conditions enhance surface fluxes significantly.