The role of planetary boundary layer (PBL) mixing in influencing storm size, track, and intensity, is investigated utilizing semi-idealized simulations with the Hurricane WRF model. In previous work (Bu et al. 2017), we demonstrated that significant variations in storm size can result from different PBL schemes, owing in large part to disparities in the vertical mixing of water vapor with respect to magnitude and depth. The potential impact on both storm size and intensity motivates a closer examination of PBL mixing in tropical cyclones, and comparisons with available observations.
For some time, the operational HWRF model has been using a version of the GFS PBL parameterization, which is known to generate excessive boundary layer mixing. In the past, this mixing was restrained via the introduction of an external parameter, and the current version incorporates a more targeted mixing cap. The HWRF model code has been revised to permit two additional PBL schemes, YSU and MYNN, to operate both with the model and with the GFDL surface layer scheme used to make real-time forecasts. YSU is similar in many respects to the operational GFS scheme, but tends to generate much less mixing and shallower boundary layer depths owing to important differences in assumptions. MYNN prognoses turbulent kinetic energy, and is frequently used in the ARW version of WRF. These schemes have been evaluated in semi-idealized simulations and evidence that MYNN deserves further consideration will be presented.
