Search

This study proposes a model for predicting tropical cyclone (TC) intensity evolution by utilizing the conditional Wasserstein generative adversarial network with gradient penalty (CWGAN-GP) for TC-related risk assessment. The TC intensity evolution is treated as a random variable, which is the output of a nonlinear system conditioned on a series of input random variables that characterize both the TC state (e.g., translation speed and potential intensity) and environment (e.g., vertical wind shear and ocean mixed layer depth). To represent the high-dimensional non-Gaussian probabilistic characteristics, the system is modeled by the CWGAN-GP and calibrated concerning the dataset of 1010 historical TCs from the western North Pacific basin. The over-ocean and overland branches of the model are calibrated separately and then merged. The suitability of the established model is validated by comparing the predicted TC intensity to the observations and the results of existing models such as those based on linear regression. Numerical results indicate that the proposed model successfully replicates the probabilistic properties such as the non-Gaussian marginal distribution of the intensity change and the input-output joint distribution and moments between the TC intensity and predictors, which the linear regression model cannot provide. Further application examples are also carried out by using the model to simulate the evolution of historical TC events and the extreme TC intensities in Southern China, demonstrating its potential role in the TC hazard assessment. • A deep-learning model for predicting tropical cyclone intensity evolution is proposed. • The CWGAN-GP is used to model TC intensity change every 6 h conditioned on environmental variables. • Numerical results demonstrate that the proposed model can replicate the probabilistic properties of TC intensity. [ABSTRACT FROM AUTHOR]

In this article, a new typhoon hazard analysis method is presented by introducing the probability density evolution theory. In the suggested method, the randomness in the typhoon occurrence is represented by the combination of a homogeneous Poisson process and a joint probability density function for both the typhoon genesis time and the genesis position. Then a set of physical models, including a typhoon intensity model, a typhoon track model and a boundary layer wind field model, is introduced to describe the typhoon process. The typhoon intensity model with high efficiency is used to simulate the evolution of typhoon intensity given the environmental variables. The typhoon motion is simulated by the beta-advection model where the steering flow is the linear combination of the large-scale flows at 850 hPa and 250 hPa, and the beta drift is solved from the non-divergent barotropic vorticity equation. At last, the typhoon induced surface wind at the interested engineering site is calculated with the help of a three-dimensional non-linear boundary layer wind field model. On the basis of the above physical models, the generalized density evolution equation is adopted to solve the probability distribution of the annual maximum typhoon surface wind speeds. The typhoon surface wind speeds associated with specified recurrence intervals are estimated consequently. As an example, the typhoon hazard at ten selected cities on the southeast coast of China is assessed. The surface wind records from the observation stations are collected to estimate the cumulative distribution function of the annual maximum wind and are compared to the analyzed results. The good comparison between the historical records and the analyzed results demonstrates the adequacy of the proposed method. • Physical models are used to generate the synthetic history of typhoon. • The probability density evolution theory is used to investigate the propagation of randomness. • The proposed method is used to assess the typhoon hazard of several cities on the southeast coast of China. [ABSTRACT FROM AUTHOR]

Typhoon track model plays an important role in typhoon wind hazard estimation. In this article, a beta-advection model is developed to simulate the typhoon track over the western North Pacific (WNP). The typhoon motion is driven by two effects, the steering flow and beta drift. The steering flow is modeled as the linear combination of the 850 ​hPa height large scale flow and 250 ​hPa height large scale flow. The spatially varying weights are estimated from the 20th Century Reanalysis dataset for each 5 ​° ​× ​5 ​° square in WNP. The large scale flows at both heights are simulated as multivariate random process based on the fact that the energy spectrum of the geostrophic turbulence follows a "-3" power law. As another driving force of typhoon motion, the beta drift is calculated by solving the non-divergent barotropical vorticity equation. The adequacy of the present model is validated by the comparison between the typhoon translation velocity inferred from the best-track dataset and that simulated by the model. Finally, integrated with an intensity model, the beta-advection model is used to analyze the statistics of typhoon key parameters and typhoon wind hazard along the southeast coast of mainland China. • The typhoon motion is modeled by combining the large scale steering flow and the beta drift effect. • The kinetic analysis is used to remove the typhoon structure in the reanalysis data. • The beta advection typhoon track model is used to assess the typhoon hazard along the southeast coast of China. [ABSTRACT FROM AUTHOR]

Tropical cyclone intensity model is one of the most important components in the tropical cyclone hazard assessment. In this study, the tropical cyclone fast intensity model, which uses features of the surrounding large-scale environment to predict the evolution of the tropical cyclone intensity, is used to estimate the typhoon wind hazard in the southeast coast of China. The fast tropical cyclone intensity model that consists of an intensity dynamic equation and a moisture dynamic equation is first validated for the western North Pacific basin. The coefficients in the moisture dynamic equation are calibrated by minimizing the error between the model and the observation from the best track dataset. The model is then integrated with a typhoon track model and a parametric radial wind profile to estimate the 50-year and 100-year recurrence-interval surface wind speeds at selected sites. Three different models for the radius of the maximum wind are considered in this analysis. Accordingly, surface wind records are utilized to estimate 50-year and 100-year recurrence-interval winds. The comparison shows a good agreement between the results of the model and that of the surface observations, which validates the adequacy of the model for engineering applications in the southeast coast of China. • The tropical cyclone fast intensity model is validated with the best track dataset. • The tropical cyclone fast intensity model is used for typhoon hazard analysis for southeast coast of China. • The influence of using different models for the radius of maximum wind on the estimated typhoon hazard is investigated. [ABSTRACT FROM AUTHOR]