【摘要】 基于2008-2013年全球天气预报系统(GFS)预报场资料,构造了30个云预报因子,分析了其与低云量的相关关系,采用多元逐步回归预报方法建立了黄河河套周边地区低云量0～168 h逐小时精细化区域预报,评估了预报模型的预报效果.结果表明,低云量与水汽类因子整层湿度和整层相对湿度具有较强的正相关性;与大气可降水量、低层温度露点差和总温度具有较强的负相关性;与GFS云类预报因子有较强的正相关性.在预报方程中水汽类预报因子和GFS云量因子具有较高的引用频数.多元逐步回归方法低云量预报值与实况的差值明显低于直接使用GFS低云量的预报结果,在夏季逐步回归方法对低云量有较大改进,上半年对低云量的订正效果强于下半年.逐步回归方法对西北部低云量的订正能力高于东南部地区,部分地区的订正能力在20%以上.更多还原

【Abstract】 Based on data from the global weather forecast system(GFS) forecast field from 2008 to 2013,cloud forecasting factors were constructed and the correlations with low cloudiness analyzed. Multivariate stepwise regression forecasting method was used to establish the low cloudiness around the Hetao Area. 0-168 h refined regional forecast by hour and the forecast results of the forecast model were evaluated. The results showed that low cloud cover had a strong positive correlation with the entire and relative humidity of the water vapor factor; it was related to the atmospheric precipitation and had a strong negative correlation with the low-level temperature dew point difference and the total temperature, and a strong positive correlation with the GFS cloud forecast factor too. The water vapor forecast factors and GFS cloud cover factors had higher reference frequencies in the forecast equation. The difference between the forecast and observed values of low cloud cover was significantly lower than the direct use of GFS low cloud cover. The multiple stepwise regression method in summer had greatly improved the low cloud cover, and the correction of the low cloud cover in the first half of the year was stronger than in the second half. The stepwise regression method had a higher correction capacity for low cloud cover in the northwest than in the southeast, and the correction capacity in some areas was more than 20%.更多还原