Multi-scale remote sensing based estimation of leaf area index and nitrogen concentration for photosynthesis modelling

Abstract

Leaf area index (LAI) and leaf nitrogen concentrations (N) are two important quantities controlling the photosynthetic rates of vegetation canopies. While the green LAI is closely related to the absorption of light used in photosynthesis, fertilization rates determine the leaf nitrogen concentrations which, in turn, govern the maximum photosynthetic capacity of the leaves. Using airborne multi-spectral images from mid-June in Denmark, it was found that nitrogen concentrations are strongly correlated with spectral reflectance in the green and far-red spectral bands. In contrast, the LAI correlates strongly with the near-infrared reflectance, the Enhanced Vegetation Index and the Normalized Difference Vegetation index. Because of the feasibility of the new generation of satellites, such as Terra-MODIS and Envisat-MERIS, to measure reflectance in a narrow green band, these results suggest that independent quantities of nitrogen concentrations and LAI may also be derived using data from such sensors. Due to the predominance of small fields in Denmark, the application of multi-scale resolution remote sensing data is in the present study used to transfer the regression equations established at field level to lower-resolution satellite data such as MODIS (500 m). Because of temporal variations in leaf specific weights, it is found that the satellite observations are related to the areal (rather than mass) nitrogen concentrations. The remote sensing based estimates of LAI and N are finally applied for photosynthesis modelling and compared with atmospheric CO/sub 2/ fluxes recorded by the eddy covariance technique.