一、引 言 近年来,由于异常台风路径的不断出现,单纯的统计学方法显示了它固有的弱点;但是,由于缺乏足够的大片洋面上的资料,也给动力学方法带来了一定的困难.因此,目前认为最有前途的预报方案是统计学和动力学相结合.国外曾采用“完全预报”(PPM)和模式输出统计学(MOS)两种方法把预报的高度资料引进统计预报方程,但前者导致了预报因子的“过加权”问题,后者又缺乏足够的数值预报资料.因此,Veigas(1966)提出了一个“模拟模式输出统计学”(SMOS)方法,它是用数值预报模式的已知误差特性与那 I. INTRODUCTION In recent years, simple statistical methods have shown its inherent weakness due to the continuous appearance of anomalous typhoon paths. However, due to the lack of sufficient data on the oceanic surface, it also poses some difficulties to the kinetic method Therefore, at present, the most promising forecasting scheme is the combination of statistics and dynamics.Many countries have adopted the methods of “complete forecast” (PPM) and mode output statistics (MOS) to introduce the forecasting height data into the statistical forecasting equation , The former leads to the problem of “over-weighting” of the forecasting factor, and the latter lacks sufficient numerical forecasting data. Therefore, Veigas (1966) proposed a method of “SMOS”, which uses numerical prediction The known error characteristics of the model and that