YE Yanting1,2,3 , FANG Weihua1,2,3
(1. Key Laboratory of Environmental Change and Natural Disaster, Ministry of Education of China, Beijing Normal University, Beijing 100875, China; 2. State Key Laboratory of Earth Surface Processes and Resources Ecology, Beijing Normal University, Beijing 100875, China; 3. Academy of Disaster Reduction and Emergency Management, Ministry of Civil Affairs & Ministry of Education, Beijing Normal University, Beijing 100875, China)
Tropical cyclone intensity scales are officially classified according to the maximum sustained winds. But the potential destructiveness of a tropical cyclone is not only determined by its intensity, but also other parameters, especially like radius of maximum wind, rainfall and duration, after landing. Therefore how to integrate all the parameters together to reflect the compound hazard severity is of great importance, especially when some of these parameters are inter-dependent and some are not.
In this paper, a copula based approach is employed to describe the dependence between tropical cyclone parameters to derive a multivariable severity of tropical cyclones. Firstly the treatment of all parameters is discussed to obtain proper indicators, and the cumulative probability distribution functions (CDF) of the selected indicators are fitted separately with extreme value theory model, using the modeling data on wind field and best track dataset of tropical cyclone over northwest Pacific Ocean from 1949 to 2012. The return periods of each cyclone by different indicators are then estimated, and it can be found that the return period of one cyclone may vary dramatically by using different parameters. Secondly, in order to estimate the compound severity, the joint probability distribution is fitted with copula function by using the previously fitted CDFs of each parameter as marginal probability functions. The return periods of each landing tropical cyclone are then estimated with the joint probability distribution, to represent their compound severity. Lastly, statistical tests are employed to assess the reliability and performance of the presented copula models. It demonstrated that t-copula model may have significant advantages over the other models, particularly, with respect to extremes. And the time series of the compound hazard severity by using t-copula is analyzed.
It is founded that the compound hazard severity, which integrates the wind, rainfall, and duration together, can better represent the overall destructiveness of landing tropical cyclones over coastal areas. The linkage between compound hazard severity and its impact is also discussed by using historical statistic disaster loss data from 1949 to 2010 of China.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner