This study examines how the structure and amount of cloud ice water content are related to rates of tropical cyclone (TC) intensification using CloudSat profiling radar measurements and simulations from the Hurricane Weather Research and Forecasting (HWRF) model. Observational studies have demonstrated the signal of TC intensification in the passive satellite measurements of frozen water concentration. However, the vertical and horizontal resolution of passive satellite observations are limited. CloudSat measurements and HWRF simulations provide high-resolution data sets of ice water content to better understand its relationship with the rate of TC intensification.  It is found that rapidly intensifying TCs have larger ice water content compared to TCs with slower intensification rates. Similar results are obtained even after accounting for the effect of initial TC intensity. Such precursors of rapidly intensifying TCs may be used to better understand and improve the prediction of TC intensification.

Boundary layer processes are known to impact tropical cyclone (TC) structure and intensity change. However, uncertainties in the turbulence parameterization of the planetary boundary layer (PBL) under high-wind conditions remain challenging, partly due to limited observations. This study presents a novel framework for evaluating and improving PBL schemes using large-eddy simulations (LES). This framework allows for an evaluation of eddy viscosity (Km), turbulent stress, and wind profiles from simulations with different types of PBL schemes. Our analyses show that the high-order MYNN scheme performs well under different high-wind and thermodynamic conditions. Among the first-order schemes, the GFS-EDMF PBL scheme tends to produce an excessively deep inflow layer due to Km. Using LES data as a benchmark, we improved the performance of the GFS-EDMF scheme by modifying the definition of boundary layer height and  the “shape parameter” of Km profiles. Our idealized simulations suggest that the modified GFS-EDMF scheme could potentially improve TC structure forecasts.