Impact of Turbulence on Hurricane Intensity Forecast

Abstract

Hurricane intensity prediction and track forecasts are very sensitive to tur- bulence within the Hurricane Boundary Layer (HBL). In the Advanced Research Weather Research and Forecasting (WRF-ARW) model the effect of the sub- grid scale (SGS) turbulence can be represented by varying the magnitude of the model grid-size (∆x) and Smagorinsky constant (Cs). The effect of turbu- lence on the hurricane intensity has been investigated by simulating Hurricane Danielle (2010) using WRF-ARW model. The properties and characteristics of the turbulence within the HBL has been investigated by a Large Eddy Simu- lation (LES) of the idealised Hurricane using WRF. The kinetic energy spectra computed for a high-resolution domain of grid-size ∆x = 62m showed that the size of the maximum energetic turbulent eddies lies between 700m-3000m and matches well with the estimated horizontal turbulence mixing length scale Lh ≈ 750m − 1500m. Defining a filter scale of ≈ 1.5km matching with the resolution of the current hurricane forecast models, the flow in the HBL was spectrally decomposed into the filter scale and sub-filter scale (SFS) motions. The SFS turbulent motions were then used for diagnosing the turbulence prop- erties within HBL. The estimated Turbulent Kinetic Energy (TKE) budget shows that shear production is the dominant mechanism for generating tur- bulence, but it is also largely balanced by the advection within the HBL. The magnitude of the gradient of the explicitly estimated SFS turbulence stress tensor is two order of magnitude larger than that of the turbulence parame- terization schemes. In general the parameterization schemes of WRF-ARW model underestimate the turbulence effects on the resolved scale within HBL.