集成方法在热带气旋路径和强度预报中的应用

Application of Consensus Method to Forecast of Tropical Cyclones

利用多元线性回归、递归正权决策和算术平均方法,分别建立热带气旋路径和强度多模式集成预报方程,并用2004年西北太平洋和南海生成的热带气旋资料进行预报试验,同时对各家预报和集成预报进行了检验分析。结果表明,用集成预报方法作热带气旋路径和强度预报是可行的,预报年总平均误差有一定程度的减小,最大预报误差也有所减小,出现较大预报误差的概率明显减小。集成方法的预报能力一般好于集成方法中采用的任意单一预报,体现了集成方法的优越性。

Several kinds of numerical forecast products are available operationally at the National Meteorological Center （NMC） of China. It is necessary to apply the consensus method to improve the efficiency of using these objective and subjective data. The consensus methods for using certain forecast products to forecast the tracks and intensities of tropical cyclones are experimented on the cyclones occurred over the western North Pacific and South China Sea in 2004. Three methods are used to build the consensus forecast equations respectively. The objective and quantitative verification results show that the forecast precision could be improved observably with the consensus method. This method can help forecasters make more accurate movement and intensity predictions of tropical cyclones.