Abstract：In order to understand the intelligent grid temperature prediction effect and error characteristics in gansu region,this article conducted a test and evaluation of the daily highest and lowest temperature forecasting products of the ECMWF model from 78 national meteorological stations in Gansu Province from March 2022 to February 2023,the national grid guidance report SCMOC issued by the China Meteorological Administration,and the provincial grid forecast SPCC in Gansu Province.The results showed that the accuracy of SCMOC and SPCC grid forecasts for daily maximum and minimum temperatures was significantly higher than that of ECMWF models,and the quality of daily highest and lowest temperature forecasts for the first two models was significantly improved compared to ECMWF models.The spatial distribution of the accuracy of the highest and lowest daily temperatures in SCMOC and SPCC is relatively consistent.The poor forecasting performance in the Ganmin mountainous area is mainly related to the large amplitude of temperature changes in the mountainous area,which increases the uncertainty of the forecast;ECMWF,SCMOC,and SPCC have weaker predictive strength for cooling compared to actual conditions,with SCMOC's predictive range for cooling being closer to actual conditions than ECMWF and SPCC.The research aim to provide reference for the localized application of grid temperature forecasting products in Gansu.
PAN L J,XUE C F,ZHANG H F,et al.Comparison of three verification methods for high-resolution grid precipitation forecast[J].Climatic and environmental research,2017,22(1):45-58.
2024, Vol. 41 
