Research on Refined Forecast Metheod in Shenyang City
LI Dian;WANG Li-li;CHAI Xiao-ling;ZHANG Zhi;BAN Wei-long;ZHANG Shuai;ZHANG Ge;Shenyang Meteorological Bureau of Liaoning Province;Jinzhou Meteorological Bureau;
Using the high resolution of WRF(weather research and forecasting model)model data and combining with the actual observation data,this paper tested temperature and precipitation forecast products from 2011 to 2014.The results indicated that :1regular temperature prediction accuracy was 52.8% ~76.0% in Hunnan Station.With the increasing of forecasting periods,forecast accuracy rate was on a downward trend. Regional automatic station forecast accurate rate was the highest in Liaoning University,that in Olympic Sports Center was the lowest.The temperature prediction of the mean absolute error from 1.5 ℃ to 2.5 ℃ in Hunnan Station.With the increasing of forecasting periods,mean absolute error was larger. The average absolute error was inversely correlated with the accuracy rate. Regional automatic station in the mean absolute error and the accuracy rate had no obvious correlation. 2TS score was higher in Hunnan Station automatic,the maximum value was 0.571.The weather forecasting accuracy rate of Hunnan Station mode in each time was relatively high. 3 After the correction, it was increased an average of 4. 4 % in less than or equal to 2 ℃temperature accuracy,less than or equal to 1 ℃ temperature accuracy rate increased by 8%.Regional automatic station in Liaoning University and Olympic Sports Center were negative,and others were positive.