You are here: Home Results Task 5.1.1. Evaluation of new remote sensing and land use predictors

Task 5.1.1. Evaluation of new remote sensing and land use predictors

Task lead: Nick Zimmermann, Eidgenössische Forschungsanstalt, WSL, CH

A new method has been developed to scale models for above-ground vegetation biomass from high spectral resolution (spectro-radiometer) remote sensing data to space-borne Hypmap scenes at a pixel resolution of 30m). Additionally, a statistical method has been developed to calibrate leaf biochemical concentrations of nitrogen, water and carbon using continuum removal and spectral normalization approaches and Hymap hyperspectral data. Also, the method for calibrating land cover continuous fields was improved by demonstrating the importance of design when sampling the spectral space.

Hyper Spectral RS data were recorded for two study area in the French Alps and the Western Swiss Alps. Preliminary analyses showed that both retrieved optical indices (e.g. indices for leaf water content, for leaf chlorophyll content and Leaf Area Index) and raw bands improve the explanatory power of species distribution models and of empirical models of diversity. However, some issues about data extraction have been identified and further analyses are ongoing.

Publications:
  • Huber S, Kneubühler M, Psomas A, Itten KI & Zimmermann NE, 2008. Estimating foliar biochemistry from hyperspectral data in mixed forest canopy. Forest Ecology and Management, 256: 491-501.
  • Mathys L, Guisan A, Kellenberger TW & Zimmermann NE, 2009.Evaluating effects of spectral training data distribution on continuous field mapping performance. ISPRS Journal of Photogrammetry and Remote Sensing 64: 665-673.
  • Psomas A, Kneubühler M, Itten K, Zimmermann NE, in revision. Hyper-spectral remote sensing for estimating aboveground biomass and for exploring species richness patterns of grassland habitats. Int. J. Remote Sensing.
Document Actions