南海夏季风建立日期的预报研究

利用1958-2001年ECMWF和1948-2006年NCEP的全球再分析气象资料,研究南海夏季风爆发迟早与前期气象要素场之间的关系。结果表明,南海夏季风建立日期与当年2月份全球风场(u、v、ω分量)、位势高度场、气温场等气象要素场之间都存在显著相关区。把各要素场中的显著高相关区视为影响季风建立日期的关键区,并定义关键区内的要素平均值为建立日期的预报因子,从中选取10个与建立日期关系最为紧密的预报因子建立预报方程,对季风建立日期进行预测。分析发现,有10个预报因子与季风建立日期有密切联系,对季风爆发迟早有一定的指示能力和预测作用。最后,经过SAS系统的分析和试验选择最好的回归预测方案,利用10个预测因子建立南海夏季风建立日期回归预报方程(简称回归预报方程)。对回归预测方程进行回报实验发现,拟合得到的南海夏季风建立日期与许多学者确定的建立日期之间的相关系数都达到95%以上的置信度,拟合误差基本控制在2候以内,有一定的预报能力。可以说,该方法是作为预报南海夏季风爆发迟早的一种新的尝试。

Forecast of the Onset Date of the South China Sea Summer Monsoon

Using two kinds of the monthly mean reanalyzed data of the ECMWF (1958-2001) and NCEP/NCAR, the relationship between onset dates of the South China Sea (SCS) summer monsoon and the preceding meteorological fields is analyzed. It is found that there is a marked correlation between the onset dates and the wind fields (u,v,ω), geopotential height fields, air temperature fields in February. The forecast zones are defined as high correlation coefficient zones. Ten forecast factors are selected from the forecast zones. The results indicate that the onset dates are closely related to the factors that can forecast and diagnose SCS summer monsoon onset. Using these forecast factors and SAS statistic analysis system, the regression forecast equation is established to forecast the monsoon onset date. The correlation coefficients between the fitting onset dates of SCS summer monsoon based on regression forecast equation and onset dates from different references are over the 95% significance level. The regression forecast equation can forecast monsoon onset dates with low error. This method is a new way to forecast SCS summer monsoon onset dates.