We evaluated the use of the Green-Red Vegetation Index (GRVI) as a phenological indicator based on multiyear stand-level observations of spectral reflectance and phenology at several representative ecosystems in Japan. The results showed the relationships between GRVI values and the seasonal change of vegetation and ground surface with high temporal resolution. We found that GRVI has the following advantages as a phenological indicator: (1) “GRVI = 0” can be a site-independent single threshold fordetection of the early phase of leaf green-up and the middle phase of autumn coloring, and (2) GRVI can show a distinct response to subtle disturbance and the difference of ecosystem types.

Chlorophylls (Chl) are important pigments in plants that are used to absorb photons and release electrons. There are several types of Chls but terrestrial plants only possess two of these: Chls a and b. The two pigments form light-harvesting Chl a/b-binding protein complexes (LHC), which absorb most of the light. The peak wavelengths of the absorption spectra of Chls a and b differ by c. 20 nm, and the ratio between them (the a/b ratio) is an important determinant of the light absorption efficiency of photosynthesis (i.e., the antenna size). Here, we investigated why Chl b is used in LHCs rather than other light-absorbing pigments that can be used for photosynthesis by considering the solar radiation spectrum under field conditions. We found that direct and diffuse solar radiation (PAR_dir and PAR_diff, respectively) have different spectral distributions, showing maximum spectral photon flux densities (SPFD) at c. 680 and 460 nm, respectively, during the daytime. The spectral absorbance spectra of Chls a and b functioned complementary to each other, and the absorbance peaks of Chl b were nested within those of Chl a. The absorption peak in the short wavelength region of Chl b in the proteinaceous environment occurred at c. 460 nm, making it suitable for absorbing the PAR_diff, but not suitable for avoiding the high spectral irradiance (SIR) waveband of PAR_dir. In contrast, Chl a effectively avoided the high SPFD and/or high SIR waveband. The absorption spectra of photosynthetic complexes were negatively correlated with SPFD spectra, but LHCs with low a/b ratios were more positively correlated with SIR spectra. These findings indicate that the spectra of the photosynthetic pigments and constructed photosystems and antenna proteins significantly align with the terrestrial solar spectra to allow the safe and efficient use of solar radiation.

Understanding of the ecophysiological dynamics of forest canopy photosynthesis and its spatial and temporal scaling is crucial for revealing ecological response to climate change. Combined observations and analyses of plant ecophysiology and optical remote sensing would enable us to achieve these studies. In order to examine the utility of spectral vegetation indices (VIs) for assessing ecosystem-level photosynthesis, we investigated the relationships between canopy-scale photosynthetic productivity and canopy spectral reflectance over seasons for 5 years in a cool, temperate deciduous broadleaf forest at ‘Takayama’ super site in central Japan. Daily photosynthetic capacity was assessed by in situ canopy leaf area index (LAI), (LAI × V_cmax [single-leaf photosynthetic capacity]), and the daily maximum rate of gross primary production (GPP_max) was estimated by an ecosystem carbon cycle model. We examined five VIs: normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), green–red vegetation index (GRVI), chlorophyll index (CI) and canopy chlorophyll index (CCI), which were obtained by the in situ measurements of canopy spectral reflectance. Our in situ observation of leaf and canopy characteristics, which were analyzed by an ecosystem carbon cycling model, revealed that their phenological changes are responsible for seasonal and interannual variations in canopy photosynthesis. Significant correlations were found between the five VIs and canopy photosynthetic capacity over the seasons and years; four of the VIs showed hysteresis-type relationships and only CCI showed rather linear relationship. Among the VIs examined, we applied EVI–GPP_max relationship to EVI data obtained by Moderate Resolution Imaging Spectroradiometer to estimate the temporal and spatial variation in GPP_max over central Japan. Our findings would improve the accuracy of satellite-based estimate of forest photosynthetic productivity in fine spatial and temporal resolutions, which are necessary for detecting any response of terrestrial ecosystem to meteorological fluctuations.