DIRECTIONAL AND POLARIMETRIC EFFECTS IN THE OPTICAL DOMAIN
V.C. Vanderbilt (1) and M. Leroy (2)
(1) NASA, Ames Research Center, Moffett Field (USA)
(2) LERTS, UMR CNES/CNRS, Toulouse (France)
INTRODUCTION.
The directional and polarimetrie properties of the light scattered by objects contributes to the
characterization of those objects. In order to derive canopy parameters from bidirectional reflectance data,
insight and understanding of the non-lambertian scattering behavior is required.
Significant and rapid progress towards understanding directional and polarimetrie effects in the
optical region has occurred in the last five years. The papers presented focused primarily on the properties
of the land surface, including plant canopies, and of the atmosphere, while one paper concerned the ocean.
Research in these areas has rapidly expanded as an accumulating data base of high quality radiometric
measurements documents directional and polarimetrie effects and allows validation of models.
SPECIFIC COMMENT: POLARIZATION.
Since the last congress, significant progress has been made toward addressing key questions in the areas of
aerosols and terrestrial surfaces, cloud properties and ocean color and ocean surfaces. Theoretical
development of models for the polanmetric properties of leaves, plant canopies and the atmosphere has
continued, while analysis of newly available aircraft POLDER polarization data has proven essential for
validating models and assessing the relative importance of competing light polarizing processes in the
atmosphere and on the ground. Rapid progress has been possible in large part because of the POLDER
sensor, a key link in efforts to obtain high quality, multi-spectral, multi-directional, polarization data
measured from aircraft altitudes — and soon from space.
Monitoring of aerosols, an important potential application of POLDER data collected from aircraft
and spacecraft altitudes, has proven feasible for maritime aerosols. Above terrestrial surfaces, monitoring of
aerosols has been demonstrated for turbid atmospheres having moderate to large optical depths but
sometimes has proven difficult for clearer atmospheres. This is because polarization due to the vegetation
and soil of the ground measured through a clear atmosphere adds variability to, and may exceed the
polarization due to the aerosols and thus dominate the polarized signal measured by the sensor. The
polarized reflectance of certain mineralized soils can be as much as 10%, making monitoring of aerosols
above such soils particularly difficult. Estimation of the polarization properties of both the aerosols and the
land surface are areas of active research.
Polarization measurement over terrestrial targets, at near infrared wavelengths and in clear
atmospheric conditions, are expected to provide information about the surface properties, especially the leaf
reflection properties in the important case of plant canopy observations. The angular signatures of the
polarization provided by aerosols and vegetation appear different, providing the possibility that aerosols
may be monitored at smaller phase angles and vegetation at larger phase angles. Simple, single scattering
models based upon the Fresnel equations explain the primary variation of polarization with angle for most
canopies. The magnitude of the polarized reflectance (but not the degree of linear polarization) has been
found to be generally less than 1.5% for vegetation canopies. Our understanding of the polarization
contributed by the land surface has increased rapidly yet in general re mains poor. That the results from the
HAPEX/Sahel experiment, presented at this congress, increased our understanding significantly points to
the need for a survey of the polarized reflectances of diverse land surfaces.
1225
i_J -3
^ ■ Ì
