摘要
SMAP卫星的二级(L2)土壤水分数据是直接反演结果,能够从模型、算法、参数等多方面体现其对土壤水分反演的综合能力。在这一级别下,SMAP设计了包括L2_SM_P(36km)、L2_SM_P_E(9 km)和L2_SM_SP(3 km和1 km)在内的多种尺度的土壤水分数据,能满足不同的实验和应用需求。以ISMN地面实测土壤水分数据作为对比参照,以偏差(Bias)、均方根误差(RMSE)、无偏均方根误差(ubRMSE)和相关系数(R)作为分析指标,分析了SMAP L2土壤水分数据和ISMN实测数据间的差异表现。结果显示:在不同静态条件下(气候类型、土壤性质和植被类型),植被对差异的影响最大,土壤性质的影响最小;在不同动态条件下(土壤水分、植被光学厚度和地表温度),植被光学厚度和土壤水分对差异影响较大,地表温度的影响较小;在4种SMAP L2土壤水分数据中,9 km数据与ISMN实测数据的差异最小,其次是36、3、1 km尺度的数据;结合静态条件和动态条件来看,36 km和9 km尺度的数据与ISMN实测数据的差异情况类似,3 km和1 km数据差异情况类似。
The level 2(L2)soil moisture data of SMAP satellite is a direct retrieval result,which can reflect its comprehensive ability of soil moisture retrieval from models,algorithms,parameters and other aspects.At this level,SMAP designed soil moisture data at multiple scales including L2_SM_P(36 km)、L2_SM_P_E(9 km)and L2_SM_SP(3 km and 1 km),the soil moisture data can meet different experimental and application requirements.In this paper,the difference characteristics between SMAP L2 soil moisture data and ISMN measured data are studied and analyzed by using the ISMN ground measured soil moisture data as reference,Bias,root mean square error(RMSE),unbiased root mean square error(ubRMSE)and correlation coefficient(R)as analysis indicators.The results show that under different static conditions(climate type,soil property and vegetation type),vegetation has the largest impact on the difference,while soil property has the smallest impact;Under different dynamic conditions(surface soil moisture,vegetation optical depth and surface temperature),vegetation optical depth and surface soil moisture have a greater impact on the difference,while surface temperature has a smaller impact;Among the four SMAP L2 soil moisture data with different spatial scales,the difference between the 9km data and the ISMN ground measured data is the smallest,followed by the 36km data,3km data and 1km data scales;According to the static and dynamic conditions,the differences between the 36km and 9km scale data and the ISMN ground measured data are similar,and the differences between the 3km and 1km data are similar.
作者
黄健庭
杨娜
马超
Huang Jianting;Yang Na;Ma Chao(School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454000,China)
出处
《遥感技术与应用》
CSCD
北大核心
2022年第6期1392-1403,共12页
Remote Sensing Technology and Application
基金
NSFC-区域创新发展联合基金重点项目(U21A20108)。
