A. Maccioni et al., New vegetation indices for remote measurement of chlorophylls based on leaf directional reflectance spectra, J PHOTOCH B, 61(1-2), 2001, pp. 52-61
Citations number
23
Categorie Soggetti
Biochemistry & Biophysics
Journal title
JOURNAL OF PHOTOCHEMISTRY AND PHOTOBIOLOGY B-BIOLOGY
Directional reflectance (R) spectra from 380 to 780 nm for nadir illuminate
d leaves of four different plants (croton, Codiaeum variegatum; spotted ele
agnus, Eleagnus pungens Maculata; Japanese pittosporum. Pittosporum tobira
and Benjamin fig, Ficus benjamina Starlight) were acquired at a viewing ang
le of 30 degrees from the nadir direction. Chlorophyll-a and -b content of
leaves covered a range of 1-60 and 0.5-21 mug/cm(2), respectively. In contr
ast with previous results from hemispherical reflectance measurements, dire
ctional reflectance data does not correlate well with chlorophyll concentra
tion. This is mainly due to the external reflectance (R-R) at the leaf epid
ermis, caused by the mismatch of the refractive index at the air-epidermis
and epidermis-inner layer boundary. The external reflectance can be identif
ied with the blue flat reflectance between 380 and 480 nm. The inner reflec
tance (R-1), obtained by subtracting the external reflectance from the meas
ured spectra, was found to be linearly related to the logarithm of the chlo
rophyll content. Good fitting of the log (Ch1) versus R-1(lambda) curves we
re obtained for R-1 in the green band (around 550 nm) and close to the infl
ection point in the red edge (around 700 nm). The coefficient of determinat
ion, r, of curve fitting improved (up to 0.97) when the normalised inner re
flectance NR1(lambda) = R-1(lambda)/R-1(lambda (0)), with lambda (0) greate
r than or equal to 750 nm, was used instead of the absolute reflectance. Th
e best indices for Ch1, Ch1-a and Ch1-b determination were R-542(1)/R-750(1
), R-706(1)/R-750(1) and R-556(1)/R-750(1), respectively. However, since th
e content of Chi-a relative to Ch1-b was almost constant for the plants inv
estigated, the two last indices must be further validated on leaves with a
high variability in the Ch1-a:Ch1-b ratio. The error in the determination o
f chlorophyll content was found to be of the order of 10%. This value was l
ower than those obtained by applying the vegetation indices previously sugg
ested. Therefore, the normalised inner reflectance in the green and in the
red edge represents a more suitable index for the chlorophyll determination
than those up to now used. (C) 2001 Elsevier Science BY. All rights reserv
ed.