Abstract:

Based on the daily surface observation data, sounding data and NCEP/NCAR FNL reanalysis data during 2002-2014, the distribution characteristics of meteorological factors during fog appearance were analyzed in Fuyang of Anhui. Then the six forecast factors were selected, and their thresholds and negative indexes were determined by the cumulative frequency method. On this basis that the potential forecast model of fog in Fuyang of Anhui was built by Perfect Prognosis and index overlap methods based on the fine mesh numerical products of European Centre for Medium-Range Weather Forecasts (EC). The assessment of fog potential forecast during 2013-2014 showed that the forecast model had obtained a good forecast effect with the threat score rate of 0.49, the accuracy rate of 0.91, the false rate of 0.38 and the fail rate of 0.30. Therefore, this model was worthy of application in fog potential forecast.

Key words: fog, Perfect Prognosis method, index overlap method, potential forecast

摘要：

利用2002～2011年安徽阜阳市逐日地面、高空观测资料和NCEP/NCAR FNL再分析资料，分析大雾发生时各预报因子的分布特征，确定预报因子的阈值及消空指标，建立基于PP法思想和指标叠套方法的阜阳市大雾天气潜势预报模型。通过2013～2014年逐日业务化运行检验，基于EC细网格数值预报产品的大雾天气潜势预报模型取得了较好的预报效果，其TS评分为0.49、准确率为0.91，说明该模型具有较好业务应用价值。

关键词: 大雾, PP法, 指标叠套法, 潜势预报