The Influences of Boundary Layer Mixing and Cloud-Radiative Forcing on Tropical Cyclone Size

Abstract

Tropical cyclone (TC) size is an important factor directly and indirectly influencing track, intensity, and related hazards, such as storm surge. Using a semi-idealized version of the operational Hurricane WRF (HWRF) model, we show that enabling cloud-radiative forcing (CRF) and enhancing planetary boundary layer (PBL) vertical mixing can both encourage wider storms by enhancing TC outer core convective activity. While CRF acts primarily above the PBL, eddy mixing moistens the boundary layer from below, both making peripheral convection more likely. Thus, these two processes can cooperate and compete, making their influences difficult to deconvolve and complicating the evaluation of model physics improvements, especially since the sensitivity to both decreases as the environment becomes less favorable.Further study shows not only the magnitude of the eddy mixing coefficient, but also the shape of it, can determine the TC size and structure.

Sensitivity of the Simulated Tropical Cyclone Intensification to the Boundary‐Layer Height Based on a K‐Profile Boundary‐Layer Parameterization Scheme

“…This behavior could produce two consequences. One is that by enhancing the boundary‐layer height, greater water vapor transport near the top of the boundary layer can bring the air there closer to saturation, which can encourage more active convection activities in the outer region (Figure ; Bu et al, ). This may help increase upward transfer of surface enthalpy flux in the outer region, but there are complex competitions between the eyewall and outer‐core convections, since the outer‐core convection could undermine inward energy transport in the boundary layer (Hence & Houze, ; Li & Wang, ).…”

“…Besides, the spiral rainbands tend to become more active as the boundary‐layer height is increased. This could be interpreted from the result that the enhanced eddy diffusivity leads to greater moisture in the upper part of the boundary layer, which, as articulated by Bu et al (), favors the activities of outer‐core convection.…”

“…Figure 10 shows azimuthally averaged tendencies of water vapor and potential temperature for varying boundary-layer height at an interval of 600 m. From Figures 10a-10d, vertical diffusivity has an effect of drying the air in the lower part of the boundary layer while moistening the air aloft, since the water vapor decreases monotonically with height ( Figure 11). Consistent with Bu et al (2017), Figures 10a-10d show that the drying in the lower part of the boundary layer and moistening in the upper atmosphere become more pronounced and vertically expanded with increasing the boundary-layer height. This is as expected since vertical mixing increases steadily with the boundary-layer height (Figure 5b).…”

“…It belongs to a K‐profile parameterization scheme, similar as the Medium Range Forecast (MRF) model scheme in the fifth‐generation Pennsylvania State University‐National Center for Atmospheric Research Mesoscale Model (MM5), and the GFS scheme in the HWRF model. Note that there are some important differences among these schemes, including the handling of the turbulent Prandtl number, diffusion above the boundary layer, and specification of the critical bulk‐Richardson number (Bu et al, ). Although the K‐profile parameterization scheme has its limitations as cautioned by Kepert (), it has been proved to work well in modeling the intensity and structure of tropical cyclones (e.g., Ma et al, ; Ma, Fei, Huang, et al, ; Ma, Fei, Cheng, et al, ; Nolan, Zhang, et al, ; Nolan, Stern, et al, ; Zhang et al, ), and performs satisfactorily in the HWRF model (Gopalakrishnan et al, ; Zhang et al, , ).…”

“…Nonetheless, using an axisymmetric numerical model, Bryan and Rotunno (2009b) and Bryan (2012) found that the maximum intensity of a tropical cyclone is weakly affected by the vertical diffusion. These inconsistent conclusions by previous studies could be attributed to two-dimensional or three-dimensional models being used, different time lengths of model integration, or different model physics, especially the treatment of atmospheric radiation (Bu et al, 2017).…”