×

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

691
A MDLTISPECTRAL CANOPY REFLECTANCE MODEL
A. KUUSK
Institute of Astrophysics and Atmospheric Physics
Estonian Academy of Sciences, EE2444 Toravere (Estonia)
ABSTRACT:
The leaf optical model PROSPECT by Jacquemoud and Baret (1990), the soil
reflectance spectrum representation with basis functions by Price (1990), and the
skylight ratio spectral representation by McCartney (1978) have been integrated
into Kuusk's (1993) fast canopy reflectance model. The resulting new multispectral
canopy reflectance model describes the directional reflectance of a homogeneous
vegetation canopy over 400 to 2500 nm with high spectral resolution. The number of
input parameters of the model does not depend on the number of the spectral bands
used. The model has high computational efficiency, and thus it can rather easily
be inverted to determine the vegetation parameters from remote optical measure
ments. The model is tested with the corn reflectance data of Ranson et al. (1985).
KEY WORDS: Canopy Reflectance Spectrum, Directional Reflectance, Model
1. BACKGROUND
Several models have been developed to describe the reflection of optical radiation
from vegetation canopies, for reviews see Myneni and Ross (1991), Myneni et al.
(1989). The need for models with few input parameters and sufficient precision
still stimulates the development of new models (Pinty et al., 1990; Verstraete et
al., 1990; Bégué, 1992; Jacquemoud, 1993; Kuusk, 1993) to be applied in inversion
problems. From measurements of the angular distribution of spectral reflectance,
agronomic variables of a canopy can be determined with the help of canopy
reflectance models (Goel and Strebel, 1983; Kuusk, 1991a). At the same time, it is
rather complicated to apply such models for the inversion of satellite data: it is
difficult to measure the angular distribution of canopy reflectance at satellite
level, and very accurate atmospheric correction is needed. It is much easier to
measure nadir reflectance spectra at satellite level. However, the increasing of
the number of spectral channels is not a solution to the inversion problem, since
the number of input parameters to canopy reflectance models increases correspon
dingly. A possible solution is a model, which can describe simultaneously the
directional reflectance of a canopy over the whole optical region. The first
attempt to develop such a model has been made in the recent paper by Jacquemoud
0993). In its PR0SPECT+SAIL model the wavelength-dependent optical parameters of
the SAIL model are given with the PROSPECT model (leaf parameters) and the
SOILSPECT model (soil parameters), the fraction of diffuse radiation SKYL is
considered constant.
The general outline of the present paper is rather similar: different
submodels are applied to determine the wavelength-dependent input parameters of a
base model. The fast canopy reflectance (FCR) model of Kuusk (1993) has been chosen
for the base model. The FCR model is the successor to Nilson and Kuusk's (1989) and
'»erhoef's (1984) models. It has only 8 input parameters and very high computerefficiency.
computerefficiency. The FCR model describes well the angular distribution of the spectral
directional reflectance, considering the specular reflection on leaf surfaces and
the hot-spot effect. This model has three groups of input parameters: (1) optical