L. Serrano et al., Estimation of canopy photosynthetic and nonphotosynthetic components from spectral transmittance, ECOLOGY, 81(11), 2000, pp. 3149-3162
Spectral transmittance signatures (expressed as absorbances) were studied a
s a potential indicator of photosynthetic and nonphotosynthetic contributio
ns to the canopy-absorbed photosynthetically active radiation (PAR). An ana
lytical approach was made under laboratory conditions using synthetic canop
ies in an integrating sphere. This approach provided the basis for identify
ing spectral (absorbance-based) features and indices to estimate green (pho
tosynthetic) and nongreen (structural and dead materials) contributions to
canopy absorbance. A strong relationship was found between the amplitude of
the first derivative of the absorbance (A(RE)) and green area, while the i
ntegrated absorbance in the PAR region (A(PAR)) mainly responded to variati
ons in total area. The ratio A(RE)/A(PAR) was closely correlated to the fra
ction of photosynthetic area to total area (i.e., the canopy green fraction
). Similarly, the ratio and normalized difference of the absorbances at 680
and 900 nm (A(SR) and A(NDVI)) closely tracked variations in the canopy gr
een fraction. Subsequently, these indices were tested in field plots with c
ontrasting structural characteristics. Under field conditions, A(RE) was a
good indicator of green biomass. The indices A(SR) and A(NDVI) were also re
liable indicators of green biomass but were affected by changes in sampling
conditions. As in the lab study, A(RE)/A(PAR) was a good indicator of cano
py green fraction. Thus, ground-based measurements of canopy spectral trans
mittance provided a tool for determining the photosynthetic contribution to
canopy-absorbed PAR by correcting for nonphotosynthetic canopy components.
Moreover, A(RE) showed a strong correlation with conventional vegetation i
ndices derived from spectral reflectance measurements. This technique could
be a useful tool for plant ecophysiology studies and a field-validation me
thod for remote-sensing studies.