基于BJ-RUC模式预报产品的北京冬半年道面温度预报模型
发布时间：2018-07-31 06:28
【摘要】：选用2012年11月1日至2013年3月30日3 km分辨率BJ-RUC模式输出的气象要素与5个道面站数据(A1027,A1325,A1412,A1414,A1512)温度进行统计分析,按不同起报时次(08、14和05时)分别建立三类逐步回归统计模型预报未来24 h逐时道面温度,选出最优模型预报2013年11月至2014年3月道面温度。结果表明:道面温度与RUC输出的2 m温度、短波辐射显著相关,与长波辐射、湿度次相关;有显著气象因子参与的回归模型预报的道面温度好于仅加入前一天对应时刻道面温度的回归模型,预报准确度可提高25%以上,误差减少1℃以上;滚动筛选不同起报时次预报时段可将模型预报误差控制在±3℃以内,且预报早高峰温度好于晚高峰,白天好于夜间,晴天好于其他天气类型。
[Abstract]:The meteorological elements from the 3 km resolution BJ-RUC mode from November 1, 2012 to March 30, 2013 and the temperature of 5 road surface stations (A1027, A1325, A1412, A1414, A1512) are statistically analyzed, and three stepwise regression models are set up respectively for the prediction of the future 24 h hourly road surface temperature according to the different reporting times (08,14 and 05), and the optimal model is selected. The road surface temperature from November 2013 to March 2014 is reported. The results show that the temperature of the channel surface is closely related to the 2 m temperature of the RUC output, the short wave radiation is significantly related, and the humidity is related to the long wave radiation, and the regression model predicted by the regression model with significant meteorological factors is better than the regression model that only joins the surface temperature of the previous day at the corresponding time, and the prediction accuracy can be raised. It is more than 25%, and the error is reduced by 1 degrees centigrade. The model prediction error can be controlled within + 3 centigrade, and the early peak temperature is better than the late peak, the day is better than the night, and the sunny day is better than other weather types.
【作者单位】： 北京市气象服务中心;
【基金】：北京市科技计划项目(Z151100002115040) 北京市自然科学基金项目(8174083)共同资助
【分类号】：P457.3;U418.41;U492.8
本文编号：2154721
[Abstract]:The meteorological elements from the 3 km resolution BJ-RUC mode from November 1, 2012 to March 30, 2013 and the temperature of 5 road surface stations (A1027, A1325, A1412, A1414, A1512) are statistically analyzed, and three stepwise regression models are set up respectively for the prediction of the future 24 h hourly road surface temperature according to the different reporting times (08,14 and 05), and the optimal model is selected. The road surface temperature from November 2013 to March 2014 is reported. The results show that the temperature of the channel surface is closely related to the 2 m temperature of the RUC output, the short wave radiation is significantly related, and the humidity is related to the long wave radiation, and the regression model predicted by the regression model with significant meteorological factors is better than the regression model that only joins the surface temperature of the previous day at the corresponding time, and the prediction accuracy can be raised. It is more than 25%, and the error is reduced by 1 degrees centigrade. The model prediction error can be controlled within + 3 centigrade, and the early peak temperature is better than the late peak, the day is better than the night, and the sunny day is better than other weather types.
【作者单位】： 北京市气象服务中心;
【基金】：北京市科技计划项目(Z151100002115040) 北京市自然科学基金项目(8174083)共同资助
【分类号】：P457.3;U418.41;U492.8
【相似文献】
相关期刊论文 前5条
1 李迅;尹志聪;丁德平;谢庄;;北京地区高速公路道面结冰特征及气象条件[J];应用气象学报;2012年05期
2 冯涛;李迅;丁德平;;高速公路道面温度预报[J];公路交通科技;2012年05期
3 孟春雷;刘长友;;北京地区高速公路夏季道面高温灾害预报研究[J];防灾科技学院学报;2009年03期
4 张爱英;丁德平;李迅;尹志聪;韩超;;相似离度在北京市道面结冰预报中的初步应用[J];气象科技进展;2012年01期
5 石磊;;冰雪条件下高速公路气象检测与预警系统研究[J];公路交通科技(应用技术版);2012年09期
相关会议论文 前4条
1 李迅;尹志聪;丁德平;谢庄;;北京市高速公路道面结冰的特征分析[A];第28届中国气象学会年会——S7城市气象精细预报与服务[C];2011年
2 张爱英;丁德平;李迅;尹志聪;韩超;;相似离度在北京市道面结冰预报中的初步应用[A];2012北京气象学会中青年优秀论文评选论文集[C];2012年
3 李迅;丁德平;冯涛;韩超;丁谊;尤焕玲;谢庄;;北京道面温度的观测和预报研究[A];第26届中国气象学会年会灾害天气事件的预警、预报及防灾减灾分会场论文集[C];2009年
4 张爱英;丁德平;李迅;尹志聪;韩超;;相似离度在北京市道面结冰预报中的初步应用[A];第28届中国气象学会年会——S7城市气象精细预报与服务[C];2011年
，本文编号：2154721
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2154721.html
