Lightning Characteristics Analysis And Prediction Of Hubei Province

Lightning is a transient discharge phenomenon occurs in the natural atmosphere, and accompanied by sound, light. In modern life, lightning not only resulted in building damage, the forest fire, but also the more serious harm to human life. With the rapid development of socio-economic and technology, acceleration of urbanization, the increase in the diversity of building materials, development of intelligent degree in new building, computer and communication systems have extensive application of electronic control system,lightning has more serious damage to electricity, radio, television, aerospace, posts and telecommunications, national defense construction, transportation, petrochemical, electronics industry and other system electronics, electrical equipment especially the more damage events of microelectronic control electromagnetic system and losses caused by lightning disasters has expanded each year.Hubei Province is located in China’s hinterland of the Yangtze River at29°O5’-33°20’N,108°21’-116°07’E. It is in east of Anhui Province, west of Sichuan, Chongqing, north of Henan Province, south of Hunan Province, Northwest and the junction of Shaanxi Province. Developed land and water transport and warm and humid, suitable for human life and interaction in Hubei Province form the large province with the center status of political, economic, cultural and speed development of industrial and agricultural. Hubei Province has the characteristics of the transition from subtropical to warm temperate, cold and warm air often formed in the intersection of the Yangtze River in summer.Basin in plum front many thunderstorms happened. In order to avoid mitigate economic losses and casualties caused by lightning disaster and ensure the normalization of Hubei, we should study on the spatial and temporal distribution lightning and its related factors deeply and improve forecasting capabilities.This paper uses the thunderstorm observation data,TRMM satellite lightning imaging sensor(LIS) observations of the lightning data and lightning location system data(LLS) lightning data to analyze the spatial and temporal characteristics of thunderstorms and related parameters in Hubei Province. The results show that the thunderstorm days in Hubei province have the downward trend and exist16a,8-9a long cycle, and3a,4a alternating short cycle. Lightning activities occur mainly from March to September, the most of which is summer, the least of which is December and January. The characteristics of lightning diurnal variation is that summer is daytime-oriented type,spring belongs to night-oriented type. Calculating flash density spatial distribution, it is show that the flash density calculated by thunderstorm days> the flash density calculated by LLS> the flash density calculated by LIS.The LIS flash density and LLS flash density have the same trend, but are inconsistent with thunderstorm days’, the high value of flash density area is located in central west and east, but thunderstorm days mainly located in the northwest and southeast of Hubei; lightning density distribution is significantly affected by terrain elevation, lighting occurs mainly in altitude in the transition interval of300-400m and400-500m,500m later decrease with the altitude increase.In thunderstorm anomalous years, the relationship between the average atmospheric circulation and the geo potential height at500hPa from June to August suggests that an obvious negative anomaly exists in eastern Asia. The subtropical high is relatively weak and to the east in active thunderstorm years, and vice verse in less active thunderstorm years. Durative El Nino events are likely to induce the abnormal increment of spring thunderstorm frequency while in both ENSO warm and cold event each year or adjacent years is prone to thunderstorms anomalies.This paper also presents an ensemble approach to thunderstorm days time series fitting which integrating ARIMA with NN. The results show that MAPE is reduced from16.83%to0.45%, the fitting accuracy of the integrated model was significantly higher than a single ARIMA model, and there maybe22thunderstorm days in Wuhan in2011.