Radiative transfer modeling within a heterogeneous canopy for estimation of forest fire fuel properties

• Published in: Remote sensing of environment 2004-08, Vol.92 (3), p.332-344
• Main Authors: Kötz, Benjamin, Schaepman, Michael, Morsdorf, Felix, Bowyer, Paul, Itten, Klaus, Allgöwer, Britta
• Format: Article
• Language: English
• Subjects: airborne imaging spectrometry; Animal, plant and microbial ecology; Applied geophysics; bidirectional reflectance; Biological and medical sciences; canopy structure; chlorophyll content estimation; coniferous canopy; danger assessment; Earth sciences; Earth, ocean, space; equivalent water thickness; Exact sciences and technology; foliage water content; forest fire; Fundamental and applied biological sciences. Psychology; General aspects. Techniques; hyperspectral data; Imaging spectroscopy; Internal geophysics; leaf-area index; radiative transfer; reflectance model; remote-sensing data; Teledetection and vegetation maps; vegetation water
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2004.05.015

Imaging spectrometer data were acquired over conifer stands to retrieve spatially distributed information on canopy structure and foliage water content, which may be used to assess fire risk and to manage the impact of forest fires. The study relied on a comprehensive field campaign using stratified systematic unaligned sampling ranging from full spectroradiometric characterization of the canopy to conventional measurements of biochemical and biophysical variables. Airborne imaging spectrometer data (DAIS7915 and ROSIS) were acquired parallel to the ground measurements, describing the canopy reflectance of the observed forest. Coniferous canopies are highly heterogeneous and thus the transfer of incident radiation within the canopy is dominated by its structure. We demonstrated the viability of radiative transfer representation and compared the performance of two hybrid canopy reflectance models, GeoSAIL and FLIGHT, within this heterogeneous medium. Despite the different nature and canopy representation of these models, they yielded similar results. Subsequently, the inversion of a hyperspectral GeoSAIL version demonstrated the feasibility of estimating structure and foliage water content of a coniferous canopy based on radiative transfer modeling. Estimates of the canopy variables showed reasonably accurate results and were validated through ground measurements.