武威市相对湿度气候特征及预报

Climate Characteristics and Forecast of Relative Humidity in Wuwei City

利用武威市1960-2009年5个气象站平均相对湿度资料,系统分析该区域相对湿度的时空分布特征及变化趋势。结果表明:相对湿度北部小,南部大。各地年、年代相对湿度总体呈减少趋势,相对湿度的时间序列存在着6～8 a的周期变化;相对湿度夏季最大,其次为秋季,春季最小,冬季相对湿度总体上呈增加趋势,其他季节为减小趋势;相对湿度8月、9月达到最大,依次向两端减小,4月最小;一天之内,早晚的相对湿度较大,中午前后的相对湿度较小,相对湿度的日变化与气温的日变化呈反相关。同时,运用2003-2007年逐日20时ECMWF数值预报格点场资料,采用Press准则进行预报因子初选,逐步回归预报方法进行预报因子精选,使用最优子集回归建立相对湿度预报方程,采用CSC双评分准则确定各站逐月平均相对湿度全局最优的显著性预报方程,预报方程通过了α=0.01信度检验。预报方程回代拟合率为70%～77%,预报准确率为65%～72%,达到了一定的预报水平,可为相对湿度的业务预报提供客观有效的指导产品。

In this paper, the spatiotemporal distribution and change trend of relative humidity were systematically analyzed based on the mean relative humidity data observed by 5 meteorological stations in Wuwei City during the period from 1960 to 2009. The result shows that relative humidity was low in the north but high in the south. Annu- al and decadal relative humidity was holistically in a decrease trend, and there was a 6 -8-year periodic change. Relative humidity was the highest in summer, than in autumn, it was the lowest in spring, holistically in an in- crease trend in winter, and in a decrease trend in other seasons. Relative humidity was the highest in August and September and the lowest in April, and it was high in the morning and evening but low around noon. There was a counter correlation between the diumalvariations of relative humidity and temperature. The daily ECMWF numerical forecast grid field data at 20：00 during the period from 2003 to 2007 were used, the forecast factor was initially e- lected with Press criterion, the forecast factor was chosen with stepwise regression forecast method, and the relative humidity forecast equations were developed with optimal subset regression. The overall situation and the most supe- rior significance forecast equations of mean monthly relative humidity were determined with the CSC double grading criterion, and the prognostic equations were at a confidence level ofα=0. 01. Back substitution fitting rate of the prognostic equations varied in a range of 70% -77% , and the forecast accuracy rate in a range of 65% -72%. Objective and effective instruction products could be provided for forecasting relative humidity.