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

Proceedings of the Symposium on Global and Environmental Monitoring

W. Lüdeker, K.P. Günther
DLR - Oberpfaffenhofen, Institute of Optoelectronics, D- 8031 Wessling , West-Germany
In order to investigate the blue shift of the red edge in the reflectance spectra theoretically, the stocha
stic leaf model of Tucker and Garratt (1977) was used. This approach treats the radiative transfer of dif
fuse light as transitions of light with weighted probabilities according to the theory of homogeneous
Markov processes with a finite number of states. The transition probabilities are determined by the opti
cal parameters (e.g. specific absorption coefficients of different pigments, scattering coefficients and
refractive index of cell membranes), the pigment concentrations and the geometrical dimensions of the
leaf (e.g. total leaf thickness, thickness of palisade and spongy parenchyma and the geometrical cross
section of chloroplasts and cells).
Assuming realistic leaf parameters our results showed that the modeled reflectance spectra were not in
agreement with measured data and the spectra presented by Tucker and Garratt (1977).
Based on these findings the model was reinvestigated. By introducing a new radiation state and by revi
sing the transition probabilities of Tucker and Garratt a successful modeling of reflectance spectra is
possible assuming realistic leaf parameters.
In order to investigate the blue shift of the red edge the model parameters were varied systematically. It
was shown that
- a reduced light absorption and / or
- a shift in the absorption spectra of chlorophyll a. and b
are responsible for the observed blue shift of the red edge.
A reduced light absorption is not only caused by a reduced pigment concentration but also by a reduced
light path. In turn the light path may be reduced by a decrease of the thickness of the palisade and / or
spongy parenchyma or may be reduced by a decrease of the scattering coefficients.
The shift of the absorption bands of chlorophyll a or b may occur to a minor extent when the chloro
phyll-protein aggregates make a phase transition from polymeres to dimers or monomers due to environ
mental stress. By traeting leaves with the herbicide DCMU in the laboratory the observed blue shift may
be attributed to a shift of the absorption bands by partially extracting the chlorophyll pigments from the
Key Words: Leaf reflectance, stochastic model, Markov chain, blue shift, red edge
Observing the reflectance spectra of vegetation in
the visible domain a variation of the steep rise of
the near-infrared reflection edge at about 690 nm
can be seen. In the last few years the application of
high resolution imaging spectroscopy stimulated the
remote detection of this phenomenon, called the
blue shift of the red edge, in order to monitor vege
tative chlorophyll status and leaf area index inde
pendently (Horler et al. 1983) and plant stress indu
ced e.g. by geochemical metal deposition (Collins et
al. 1983). The trace metal-induced stress retards the
pigment synthesis and in turn decreases the pigment
absorption leading to higher reflectance. Since the
spectral shift is related not only to soil geochemi
stry but also to tree species ( Singhroy et al. 1985) a
uniform forest stand is best suited for studying the
blue shift. A comparison of in situ and airborne
spectral measurements of the blue shift in Vermont
and West-Germany over homogeneous forest stands
(Rock et al. 1988) showed that the blue shift is
associated with damage in spruce needles. The au
thors suggested that the shift of the red edge to
shorter wavelengths may be due to an absorption li
ne narrowing when the chlorophyll a concentration
decreases in contrast to the explanation of a shift in
the position of the maximum of the chlorophyll
absorption. Also Buschmann and Lichtenthaler (1988)
correlated the observed blue shift of the red edge