A study of ocean spray effect on tropical cyclones dynamics

Very little progress has been made in hurricane intensity prediction in the last 15-20 years. One of the reasons for this is the lack of understanding and representation of physical factors contributing to the hurricane wind intensification. It was proposed that ocean spray-wind interaction plays a significant role in the physical mechanism responsible for such intensification. Specifically it has been reported recently (Barenblatt, et al., 2005; BCP hereafter) that the presence of ocean spray in the vortex of tropical cyclone (TC) leads to a significant reduction in turbulent intensity and consequently to a sharp flow acceleration. We reexamined Barenblatt's mathematical model and found that some of their conclusions are unjustified and contradictory. Specifically it is calculated that the flow velocity magnitude increases by 20% at most for large water droplets, which is significantly below of almost an order of magnitude increase claimed by the authors. We found much stronger flow acceleration for small water droplets, the case which is not studied in BCP. The flow lubrication effect for this case however is not nearly as strong as suggested in BCP.

We also extended the theoretical model by considering different mechanisms of ocean spray production, positive feedback of wind acceleration, more consistent turbulence closure models and some other contributing factors. The sensitivity of ocean spray lubrication effect to the theoretical model has been investigated. Specifically we found that the lubrication effect is very sensitive to the influence of spray stratification on the turbulence mixing length. We account for this effect by employing full prognostic equations for turbulent kinetic energy and dissipation rate with no mixing length assumption. Currently we carry out a numerical study of an ideal hurricane model. A series of numerical experiments with and without spray has been performed and the results are compared for different theoretical models. An effort to develop a proper spray parameterization based on the theoretical consideration and the results of numerical experiments is undertaken. Consequently, we will incorporate the spray parameterization in the realistic Weather Research and Forecast (WRF) numerical model in order to improve the accuracy of TC intensity prediction.

This work is supported by a grant from the National Oceanic and Atmospheric Administration, Educational Partnership Program under the cooperative agreement NA06OAR4810187.