×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Mesures physiques et signatures en télédétection

an airborne version of POLDER has been built
and several field experiments already
performed (Deuzé et al., 1993, Bréon and
Deschamps, 1993; Goloub et al., 1994). The
airborne POLDER also acquired
measurements during the HAPEX-Sahel
experiment which was an international
program focused on the soil-plant-atmosphere
energy, water and carbon balances in the West
African Sahel (Goutorbe et al., 1993; Prince et
al., 1993).
We herein present results of bidirectional
polarised reflectance measurements acquired
simultaneously with a field instrument, and
from the airborne POLDER. Two surface
reflectance models are proposed that
reproduce accurately the measurements. We
discuss our results in the context of remote
sensing using polarised light.
2- MEASUREMENTS AND PROCESSING
2.a- Surface measurements
An instrument, called REFPOL, has been
developed in the "Laboratoire d'Optique
Atmosphérique", Lille, France, for measuring
the BRDF (Bidirectional Reflectance
Distribution Function) and the BPDF
(Bidirectional Polarisation Distribution
Function) of surfaces in the field. It is
equipped with 4 optical barrels, each of them
carrying a different spectral filter. During the
HAPEX-Sahel campaign, the 4 channels were
centred at 450, 650, 850 and 1650 nm with
passbands of 100 nm for the middle infrared
band and 40 nm for the others. Polarizers
rotate in front of the 4 detectors (3 photodiodes
for the visible bands and one germanium for
the near infrared band) and allow three
successive measurements, with the polarizer
turned by steps of 45°.
The radiometer was mounted on a step-bystep
step-bystep electric engine which allows an
inclination up to 75° on each side of nadir
viewing. An inclinometer provided
automatically the zenith orientation of the
radiometer and this information was stored
with the measurements. The total angular
sampling time from -75° to +75° was of 120s.
The apparatus (engine and radiometer) was
mounted at the extremity of a 1.5m length
boom, at the top of a seven meters high mast.
The field of view (FOV) was approximately 16°
(total) which yields a footprint of 2m diameter
at nadir, so the boom allowed minimal
contamination by the supporting mechanism.
By rotating from +75° to -75°, the radiometer
allowed to sample the angular signature of the
radiance in a vertical plane defined by its
azimuth angle relative to the sun. The angular
signature of the reflectance is derived from
measurements acquired with a varying surface
cover as the radiometer rotates. Thus, a
uniform surface is necessary to derive
meaningful signatures.
The digital counts are converted to radiance
using calibration coefficients. The radiances
are then normalised to the top of the
atmosphere (TOA) solar irradiance. From the
three reflectance measurements, the polarised
reflectance is derived from:
P p = ^[(Pi _ P2) 2 + (P2 ~ P3) 2 ]
2.b- Airborne measurements
The POLDER instrument optical design is
based on the concept of a CCD matrix, a
rotating filter wheel that caries spectral filters
and polarizers, and a wide field of view lens
(Deschamps et al., 1994). On the filter wheel, 9
slots allow successive spectral measurements
with various polarisation directions. The table
below indicates the spectral bands that were
selected for this campaign.
Cent. Wavl. (nm)
450
570
670
865
910
Band Width (nm)
20
10
10
10
20
Polarisation
Yes
No
No
Yes
No
The bidimensional CCD matrix yield a
multidirectional view of the area in one
instantaneous shot. The POLDER field of view
is 43° along track and 51° crosstrack. The field
of view on the ground is, therefore, about
twice the instrument altitude, or 8.2x10.9 km 2 .
The CCD matrix is composed of 384x288
pixels, which yields a spatial resolution of
about 30 meters, and an angular resolution of
approximately 0.1°.
Three measurements are performed for the
polarised channels with the polarizers turned
by steps of 60°. From those three
measurements, one deduce the polarised
reflectance, the total reflectance, and the
polarisation direction.
3- SURFACE POLARISED REFLECTANCE
MODEL
The polarised reflectance of vegetation has
been shown to be generated by specular
scattering on the leaf surfaces. Following
Rondeaux and Herman, if the leaf inclination
distribution is isotropic, the polarised
reflectance pp can be simply written as: