Forest ecosystem chlorophyll content: implications for remotely sensed estimates of net primary productivity

Current methods for estimating photosynthesis and hence net primary productivity (NPP) of forest ecosystems from remote sensing are based on the relationship between (i) the fraction of incident photosynthetically-active radiation absorbed by the canopy (fPAR) and (ii) spectral indices (e.g. NDVI). However, ground-based estimates of fPAR used to quantify this relationship for a specific vegetation type are derived from measurements of canopy structure only (e.g. using light interception methods such as hemispherical photography). Using a coupled leaf-canopy model of radiative transfer, we demonstrated that NDVI is highly sensitive to both canopy foliar and understorey chlorophyll content, which could account for significant errors in remotely sensed estimates of fPAR and hence NPP.