| 摘要: |
| 再分析气象数据为数据缺乏地区的生态环境演变、水文与水资源研究提供新的数据来源。本文以开都河上游为例，定量评估CMADS数据集多时间尺度精度及其在SWAT模型中的适用性，从两个角度进行精度评估：一是评估CMADS数据集降水量、最高和最低气温在开都河上游的表现；二是使用CMADS数据集和地面观测气象数据分别驱动SWAT模型进行模拟精度比较，探究再分析气象数据在数据缺乏地形复杂地区的可替代性。研究结果表明：（1）CMADS数据集在气温探测上表现优秀，相关性均在0.9以上；逐日降水量的探测表现较差，相关性为0.59，月尺度降水相关性显著提高（CC=0.84）；（2）CMADS数据集和地面观测数据在开都河上游径流模拟中，均取得令人满意的结果，表明SWAT模型在开都河上游具有很好的适用性。（3）CMADS数据在率定期存在基流高估，对较高的峰值出现低估，较低峰值出现高估的现象；验证期低估峰值、略高估基流。CMADS数据径流模拟存在部分不够精确的情况，但总体符合实际径流量变化。表明该数据集可以为气象数据缺乏的高寒山区径流模拟提供替代数据。 |
| 关键词: CMADS；SWAT；径流模拟；精度评价；开都河 |
| DOI： |
| 分类号:K90 |
| 基金项目:国家自然科学基金NSFC联合(U1703341),新疆维吾尔自治区重点实验室开放课题(2020D04039) |
|
| AccuracyandApplicabilityAssessmentofCMADS Dataset Based on the SWAT Model intheUpperReaches of KaiduRiver |
|
CHEN ShiXue, YUSUFUJIANG Rusuli
|
|
College of geography and tourism,Xinjiang Normal University
|
| Abstract: |
| Reanalysis data provides new data sources for researches of eco-environmental evolution, hydrology and water resources in data-poor areas. In this paper, we quantitatively assess the multi-timescale accuracy of the CMADS dataset and its applicability in the SWAT model using the upper Kaidu River as an example research area, and evaluate the accuracy from two perspectives: firstly, we assess the performance of the CMADS dataset in terms of precipitation, maximum and minimum temperatures in the upper Kaidu River; secondly, we compare the simulation accuracy of the SWAT model using the CMADS dataset and ground-observed meteorological data driven separately. To explore the substitutability of reanalyzed meteorological data in areas with complex topography where data are lacking. The results show that ：(1) the CMADS dataset performs well in temperature detection, with correlations above 0.9; the daily precipitation detection performs poorly, with a correlation of 0.59, and the monthly-scale precipitation correlation is significantly higher (CC=0.84); (2) both the CMADS dataset and the ground observation data achieve satisfactory results in the simulation of runoff in the upper Kaidu River, indicating that (2) Both the CMADS dataset and the ground observation data achieved satisfactory results in the runoff simulation of the upper Kaidu River, indicating that the SWAT model is well applicable in the upper Kaidu River. (3) The CMADS data overestimated the base flow, underestimated the higher peak and overestimated the lower peak during the rate period, and underestimated the peak and slightly overestimated the base flow during the validation period, while the CMADS data runoff simulation was not accurate in some cases, but generally conformed to the actual runoff variation. It shows that this dataset can provide alternative data for runoff simulation in alpine mountains where meteorological data are lacking. |
| Key words: CMADS; SWAT; Simulation?of?runoff; Accuracy evaluation；Kaidu?River |
