CMPAS 5 km融合降水产品在老挝南欧江流域的适用性评估

Applicability Evaluation of CMPAS 5 km Precipitation Merged Products in Nam Ou River Basin of Laos

摘要: 利用2021~2023年老挝南欧江流域50个观测站实况降水数据，通过偏差、平均值误差、均方根误差和相关分析等评估指标对5 km三源融合降水进行适用性评估，得到以下结论：融合降水能反应全流域站点实况平均降水量的年际变化趋势，一级~三级西部流域、五级~七级流域融合降水与站点实况降水的空间分布特征一致性较好。一级至四级西南流域以及七级流域融合降水与站点实况降水误差相对较小且相关系数较大，有一定的参考性，此外，融合降水与实况降水相比，大致呈现东部低估、西部高估的特征。就整个流域而言，针对不同降水量等级，融合降水对雨季降水有所高估，对干季降水有所低估。分流域、分降水量等级来看，从一级至七级流域融合降水均对小雨有所高估，其余量级降水有所低估，此外，随降水量等级的增大，融合降水误差越大，低估越明显。

Abstract: Based on the observed precipitation data of 50 observation stations in the Nam Ou River Basin of Laos from 2021 to 2023, the applicability of the 5km three-source fusion precipitation was evaluated through the evaluation indexes such as deviation, mean value error, root mean square error and correlation analysis, and the following conclusions were drawn: Fusion precipitation can reflect the interannual trend of mean observed precipitation in the whole basin. the spatial distribution characteristics of fusion precipitation and observed precipitation in the western basin of first to third grade, and fifth to seventh grade were in good agreement. The error between the fusion precipitation and the observed precipitation at the station is relatively small and the correlation coefficient is large in the first to fourth grade of southwest basin and the seventh basin, which had a certain reference. In addition, compared with the actual precipitation, the fusion precipitation is generally underestimated in the east and overestimated in the west. For the whole basin, according to different precipitation levels, the fusion precipitation overestimates the precipitation in the rainy season and underestimates the precipitation in the dry season. From the point of view of watershed and precipitation level, light rain is overestimated for the fusion precipitation of the first to the seventh basin, and the other basins of precipitation are underestimated. In addition, with the increase of precipitation level, the greater the fusion precipitation error, the more obvious the underestimation.