Presented on March 29, 2023, by

Prof. Yuqing Wang
Department of Atmospheric Sciences
International Pacific Research Center
School of Ocean and Earth Science and Technology
University of Hawaii at Manoa

Abstract:

In this presentation, some applications of the newly developed time-dependent theory of tropical cyclone (TC) intensification will be briefly introduced and discussed. One of the applications is to help determine the key environmental factors and their relative importance/contributions to the observed TC intensity change. This has been done by introducing an environmental ventilation parameter into the time-dependent theory. The environmental ventilation parameter is a multiplication of individual ventilation parameters induced by all individual factors and is quantified by machine learning algorithm. Six environmental factors are evaluated, including the large-scale environmental vertical wind shear (VWS), the climatological ocean heat content (COHC), the divergence in the upper troposphere (D200), the mid-level relative humidity (RHMD), the gradient of maximum potential intensity (dMPI) along the TC track, and the TC translational speed (SPD). The results show that the environmental VWS contributes the most to the slowing down of TC intensification or the weakening of TCs. Other environmental factors contribute about equally but secondary compared with the environmental VWS. The second application of the time-dependent theory is to real TC intensity forecasting. This is done by considering the environmental ventilation parameter as an unknown latent variable and is determined by a combination of the Bayesian hierarchical model (BHM) and neural network (NN) and ensemble mean based on selected environmental factors. With the determined environmental ventilation parameter, the time-integration of the time-dependent equation of TC intensification rate can provide TC intensity forecast at required lead times. Our preliminary results for TCs over the North Atlantic show that the model has a good skill in predicting TC intensity change up to 10 days, indicating that the scheme has a potential to be used for real-time TC intensity forecasting after implementation and further testing in an operational real-time setting in the near future.