2019-2021年红河州智能网格温度预报检验评估

Test and Evaluation of Temperature Forecast in Honghe Prefecture Intelligent Grid from 2019 to 2021

为综合评估智能网格预报业务的产品质量,基于站点观测和格点实况分析资料,开展对红河州2019—2021年省级指导智能网格预报温度要素的检验评估,并与欧洲中心(ECMWF)和GRAPES模式进行对比。分析结果表明:智能网格省级指导温度预报效果良好,最高气温、最低气温预报与格点实况的空间分布基本一致,但存在最高温偏低2~3、最低温偏高1~2℃的现象。智能网格省级指导最高温/最低温预报准确率达72.1%/69.9%,较ECMWF和GRAPES模式提高5~9个百分点;省级指导最高温/最低温平均绝对误差达1.2℃/1.3℃,较ECMWF和GRAPES模式提高0.4~0.5℃。省级指导产品的温度预报在数值模式的基础上有显著提升,有较好的业务应用价值。

In order to comprehensively evaluate the product quality of the intelligent grid forecasting business,based on the site observation and grid analysis data,the inspection and evaluation of the temperature elements of the provincial guidance intelligent grid forecasting in Honghe Prefecture from 2019 to 2021 was carried out,and compared with the European Center(ECMWF)and the GRAPES model.The analysis results show that the intelligent grid provincial guidance temperature prediction effect is good,and the maximum temperature and minimum temperature prediction are basically consistent with the spatial distribution of the grid actual situation,but there is a phenomenon that the maximum temperature is 2~3℃lower and the minimum temperature is 1~2℃higher.The accuracy rate of the provincial maximum temperature/minimum temperature forecast guided by the intelligent grid is 72.1%/69.9%,which is 5~9 percent points higher than that of the ECMWF and GRAPES models.The average absolute error of the provincial guidance maximum temperature/minimum temperature is 1.2℃/1.3℃,which is 0.4~0.5℃higher than that of ECMWF and GRAPES.Based on the numerical model,the temperature prediction of provincial guidance products has been improved significantly,and has a good value of application.