Prediction of leaf chemistry by the use of visible and near infrared reflectance spectroscopy

Abstract

The chemical content of dry, ground leaf material sampled from deciduous and conifer tree species from sites in Alaska, Wisconsin, and California was estimated using visible and shortwave infrared spectroscopy. Seven chemical components—sugar, starch, protein, cellulose, total chlorophyll, lignin, and total nitrogen—were analyzed by wet chemical methods and their concentrations regressed against log 1/ρ and first and second differences of log 1/ρ (where π is measured reflectance) at wavelengths selected by stepwise regression. Predictions of chemical concentrations based on cross validation suggest that this technique may be useful for extracting vegetation canopy biochemical information by remote sensing.