Geospatial Modeling of Canopy Chlorophyll Content Using High Spectral Resolution Satellite Data in Himalayan Forests

Abstract

Chlorophyll is an important pigment in plants, and considered as an essential element to be estimated for understanding florescence, assessing carbon sequestration potential, primary productivity, etc. at different levels in biological hierarchy. Several biophysical and climate models require chlorophyll as an important input parameter to simulate global carbon and primary productivity. Present study is an attempt to model canopy chlorophyll content (CCC) in Himalayan forests using (1) ground survey (2) Hyperion EO-1 Hyper spectral data and (3) PROSAIL Radiative Transfer model across different forest systems. Sampled chlorophyll content index (CCI) values were converted to leaf level chlorophyll content using existing model. Spectral regions from 515 nm to 850 nm were used for spectral modeling and upscaling. Waveform analysis approach which includes Band Depth (BD), Continuum Slope (CS), Band Maximum (Bmax), Band Centre (BC) and band area (BA) was followed for estimating total canopy chlorophyll content (CCC). The highest canopy chlorophyll (2.63 g/m2) was found in temperate broadleaved forests of Oak (Quercus leucotrichophora) and with mean 0.82 g/m2. Bauhinia species dominates sub-tropical forest and showed higher canopy chlorophyll content (mean=0.70 g/m2) as compared with the Sal dominated moist tropical forest (mean=0.42 g/m2). BD and BA were found to be highly correlated and may be used as surrogate for each other. BD was highly correlated with CCC (R2=0.92). Modelled spectra have shown encouraging results only for low ranges of CCC, however, at higher ranges of CCC the correlation became poor.