湖北省乡镇温度预报方法初探

为了提高精细站点温度预报水平,本文在中尺度模式精细化温度预报的基础上,结合MOS分县客观预报结果,采用带海拔高度的距离权重温度插值方法,经过灰色预测模型修正温度预报误差,对湖北省717个四要素自动观测站的72小时内日高低温度进行了精细化预报,并且利用乡镇自动站温度观测进行了检验。结果显示：对于没有历史观测资料的站点,采用此预报方法,预报效果明显高于模式预报结果,基本接近客观预报方法,可以进行业务应用。对湖北的四要素自动观测站温度检验的结果还表明：鄂西北的预报误差最大,江汉平原的预报效果最好;从分月预报情况来看,高低温度都是夏季和初秋预报效果最好,低温预报冬春季预报效果最差,湖北各个区域趋势基本都一致,而高温的预报却没有一致的趋势。

The Study of Township Temperature Forecast in Hubei Province

To improve fine temperature forcasting,based on the result amending of MM5 by grey dynamic model,and using the temperature interpolation methods with altitude and distance weight,the high and low temperatures in 72 hours at 717 automatic weather stations with 4 meteorological elements in Hubei Province,are forecasted on the basis of the MOS technique and the mesoscale model and tested by observations. The results show that the test score in this method to the stations without historical data is better than ...