MODIS LAI产品真实性检验与偏差分析

发布时间：2019-07-02 20:26

【摘要】：当前对MODIS LAI产品的真实性检验工作中,更多的是关注遥感产品在数值与趋势上与地表真值的一致性程度,很少工作能够全面分析遥感LAI产品偏差来源以及不同来源的偏差对全局偏差的贡献率。本文在对MODIS LAI产品进行真实性检验基础之上,进一步分析了MODIS LAI产品偏差来源。将遥感产品真实性检验偏差来源分解为反演模型,反射率数据和冠层聚集效应3个方面,并定量分析各个偏差源对真实性检验结果的影响。以河北省怀来玉米为研究对象,结合实测LAI数据和Landsat 8 OLI(Operational Land Imager)数据建立NDVI LAI半经验模型,得到LAI参考数据,据此对MODIS LAI产品进行真实性检验及偏差分析。研究表明,该区域MODIS LAI产品存在明显的低估现象,参考数据和MODIS LAI数据均值分别为3.53 m2/m2和2.33 m2/m2,MODIS产品低估为34.14%。在各个偏差因素中,反射率数据的差异对结果影响最大,即MODIS地表反射率数据与Landsat 8 OLI地表反射率数据的差异造成的偏差占总偏差的57.50%;聚集效应的影响次之,占总偏差的28.33%;模型差异对结果的影响最小,占总偏差的14.17%。本研究对遥感产品真实性检验及其不确定性分析具有一定的借鉴意义。
[Abstract]:At present, in the authenticity inspection of MODIS LAI products, more attention is paid to the consistency between remote sensing products and surface true values in terms of value and trend. Few work can comprehensively analyze the sources of deviation of remote sensing LAI products and the contribution rate of deviation from different sources to the global deviation. On the basis of testing the authenticity of MODIS LAI products, this paper further analyzes the sources of deviation of MODIS LAI products. The deviation sources of remote sensing product authenticity test are decomposed into three aspects: inversion model, reflectivity data and crown aggregation effect, and the influence of each deviation source on authenticity test results is quantitatively analyzed. Taking Huailai corn in Hebei Province as the research object, combining the measured LAI data and Landsat 8 OLI (Operational Land Imager) data, the NDVI LAI semi-empirical model is established, and the LAI reference data are obtained. based on this, the authenticity and deviation analysis of MODIS LAI products are carried out. The results show that there is obvious underestimation of MODIS LAI products in this region, the average values of reference data and MODIS LAI data are 3.53 m2/m2 and 2.33 m2 鈮，

本文编号：2509207