融合气温实况产品在内蒙古地区的适用性评估分析

Analysis of Applicability Assessment of Merged Temperature Real-Time Products in InnerMongolia

采用邻近插值法,以地面观测站点资料为真值,并采用均方根误差、平均误差、平均绝对误差以及相关系数评估指标,对2021年5KM和1KM两种气温实况融合产品进行了评估。结果表明:5KM和1KM气温实况融合产品的数据质量较高,与站点观测值具有较强的相关性,相关系数均达到了0.99,1KM产品在内蒙古的适用性优于5KM,站点均方根误差大部分在1℃以下,平均误差小于0.05℃。两种气温实况融合产品误差随季节变化明显,夏季温热短暂,均方根误差和平均误差偏小,冬季随着气温的降低,实况产品以高估为主,均方根误差在0.8~1.2℃之间,平均误差在1~2月份达到峰值;1KM和5 KM气温实况融合产品都能较好地监测高温天气,落区与观测相当,但对于40℃以上的高温,表现稍差,其站点数较观测偏少,数据偏小。

In this paper,the ground observation station data were taken as the true value.The proximity interpolation method was used to evaluate the quality of the two temperature merged productsin 2021 by root mean square error,average error,average absolute error and correlation coefficient.The results show that the merged products of 5 KM and 1 KM have high-quality data.There is a strong correlation with the observed values,and the correlation coefficients have reached 0.99.However,the applicability of the 1 KM merged product in Inner Mongolia is better than 5 KM.The root mean square error of most sites is below 1℃,and the average error is less than 0.05℃.The error of the two temperature merged products changes obviously with seasons.In summer,the root mean square error and average error are relatively small due to the warm and short.In winter,the merged products are over estimated as temperatures drop.The root mean square error is between 0.8~1.2℃;the average error peaks in January and February.The two temperature merged products can better monitor high temperature weather,and the data is comparable to the observed value,but the performance is slightly worse for the high temperature above 40℃.Compared with the observed values,the number of sites and data of the merged products is small.