Assessment of forest canopy chemistry with ISM

• Main Authors: Gastellu-Etchegorry, J.P., Zagolski, F., Pinel, V., Giordano, G., Romier, J., Marty, G., Mougin, E.
• Format: Conference Proceeding
• Language: English
• Subjects: Absorption; Chemical analysis; Chemistry; Infrared spectra; Needles; Nitrogen; Optical imaging; Powders; Spectral analysis; Spectroscopy
• DOI: 10.1109/IGARSS.1994.399324

An ISM (Imaging SpectroMeter) airborne campaign was organised in 1993 over a pine plantation (Les Landes, SW France). ISM is a spectrometer that operates in the near and mid infrared portions of the spectrum [800-3200 nm]. The study was aimed to assess the capability of this airborne spectrometer for obtaining information about the chemical composition (nitrogen, lignin, cellulose) of a tree canopy. Samples (needles) were collected in the field at the time of the ISM survey: 21 parcels were sampled with 5 samples per parcel. Once these samples were dried and reduced in powder, they were analysed for determining their chemical concentrations. Some relationships appeared between the age of the trees and the needle chemical concentrations. Spectral analyses of samples was conducted with an InfraAlyser 450 that operates with 19 spectral bands centred on the absorption bands of chemicals. Stepwise analyses of reflectances were conducted for determining statistical relations that relate chemical concentrations to reflectances (lignin r=71%; nitrogen r=78%; cellulose r=75%). ISM data were geometrically corrected for obtaining spectra of parcels where samples were collected. The objective was to determine if most efficient wavelengths for predicting chemical concentrations are the same at the laboratory and remote sensing level. This comparative analysis was conducted through regression analyses (lignin r=78%; nitrogen r=78%; cellulose r=84%).< >