Validation of a New GO Kernel and Inversion of Land Surface Albedo by Kernel.Driven Model (1)

doi:10.11834/jrs.2000S101

Views : 0 下载量: 523 CSCD: 0

PDF
Export
Share
Collection
Album

Validation of a New GO Kernel and Inversion of Land Surface Albedo by Kernel.Driven Model (1)
Vol. 4, Issue S1, (2000)
Published：2000
DOI： 10.11834/jrs.2000S101
稿件说明：

移动端阅览

DOI：

Validation of a New GO Kernel and Inversion of Land Surface Albedo by Kernel.Driven Model (1)[J]. Journal of Remote Sensing, 2000, 4(S1): 7. DOI： 10.11834/jrs.2000S101.

摘要

基于核驱动模型的AMBRALS程序将用于提供MODIS的反照率和二向反射产品;该文提出一个新的几何光学核-LiTransit核

用以取代目前AMBRALS运行程序中的LiSparseR核。当植被及阴蔚总面积在视场中比例接近于;时

新核由LiSparseR核向LiDens。核自然诫。因而新核兼有公Sparse·核面LiDense核的优点

比LiSparseR核更符合几何光学模型的基本原理。对于29组地面观测数据的反演试验表明;LiTrarisit核有与LiSparse核相当的数据拟合能力

外推到大的天顶角时有.ense核的稳定性;与LiSparseR核比较

RossThick-LiTransit的核组合更能反映直入扇出反照率随太阳天顶角变化的趋势。用NOAA AVHRR多时相数据反演的结果

也表明RossThick-LiTransit组合在反演甲更为稳定。

Abstract

Kernel-driven BRDF model was applied in AMBRALS (algorithm for MODIS land surface BRDF and albedo products). In this paper

we derived a new Geometrical-Optical (GO) kernel-LiTransit-to substitute the LiSparseR kernel

which is currently used in AMBRALS. The new kernel is a natural transition from LiSparse to LiDense kernel "when vegetation become filling the field of view. LiTransit has the advantage of both LiSparse and LiDense

and accords more to the basic concept of GO model. Experimental results of 29 ground measured BRDF data sets show; LiTransit has similar data fitting ability with LiSparseR

but more stability when extrapolated to large zenith angles; comparing with Rossthick-LiSparseR

the combination of RossThick-LiTransit shows more accordance with albedo's downward tendency with respect to increasing sun zenith. Our inversion result with NOAA-AVItR data also suggests that the results derived by RossThick-LiTransit are more robust.