
Retrieval of land surface parameters from airborne POLDER bidirectional reflectance distribution function during HAPEX‐Sahel
Author(s) -
Roujean JeanLouis,
Tanré Didier,
Bréon FrançoisMarie,
Deuzé JeanLuc
Publication year - 1997
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/97jd00341
Subject(s) - remote sensing , environmental science , inversion (geology) , albedo (alchemy) , bidirectional reflectance distribution function , photosynthetically active radiation , atmospheric correction , ground truth , leaf area index , reflectivity , geology , optics , structural basin , computer science , physics , machine learning , art , paleontology , ecology , photosynthesis , botany , performance art , biology , art history
Spaceborne instruments such as POLDER (polarization and directionality of Earth's reflectances) provide multidirectional estimates of the surface reflectance (after atmospheric correction). A better description of the surface cover is needed for an improvement in climate modeling. This paper describes how the inversion of a three‐parameter bidirectional model from the measurements yields some variables of interest such as the surface albedo, the leaf area index, the fraction of vegetation, and the daily integrated photosynthetically active radiation absorption. Airborne measurements of POLDER were made during the HAPEX‐Sahel experiment over various surface types and for different atmospheric conditions. These measurements together with some “ground truth” allow a validation of the proposed method. The reflectance bidirectional signature of the Sahelian landscape shows some classical features such as a general increase toward backscattering and the hot spot phenomenon. The bidirectional model allows an accurate representation of the observed signatures. However, a large range of angular measurements is necessary to constrain the model inversion and to decrease the uncertainty in the three‐parameter retrieval. The surface variables derived from these three parameters are found to be in agreement with the in situ measurements. Obviously, a validation on other landscapes with a higher vegetation coverage is now needed. The technique described in this paper will be applied to the spaceborne POLDER data sets.