Retrieval of forest biophysical variables by inverting a 3-D radiative transfer model and using high and very high resolution imagery

• Published in: International journal of remote sensing 2004-12, Vol.25 (24), p.5601-5616
• Main Authors: Gascon, F., Gastellu-Etchegorry, J.-P., Lefevre-Fonollosa, M.-J., Dufrene, E.
• Format: Article
• Language: English
• Subjects: Animal, plant and microbial ecology; Applied geophysics; Biological and medical sciences; Earth sciences; Earth, ocean, space; Exact sciences and technology; Fundamental and applied biological sciences. Psychology; General aspects. Techniques; Internal geophysics; Sciences of the Universe; Teledetection and vegetation maps
• ISSN: 0143-1161; 1366-5901
• DOI: 10.1080/01431160412331291305

Obtaining detailed observations of the amount and condition of vegetation is an important issue for describing, understanding and modelling the role of the biosphere in the global carbon cycle. Here, multispectral optical imagery was used for retrieving biophysical variables through the inversion of a 3-D radiative transfer model. Two inversion procedures are presented: a classical procedure for high resolution imagery and an innovative procedure specifically designed for very high resolution imagery (resolution around 1 m). They were tested with SPOT ('Satellite Pour l'Observation de la Terre') and Ikonos images, respectively. One of the objectives was to assess to which extent the inversion of high and very high resolution satellite imagery can help in assessing how Fontainebleau forest (France) was damaged by a very strong storm on December 1999. Retrieved biophysical variables are: Leaf Area Index (LAI), Crown Coverage (CC) and leaf chlorophyll concentration (C ab ). Compared with ground measurements, SPOT-derived LAI has a root mean square error (RMSE) of around 1.4 at stand scale. This is not accurate enough to quantify the effects of the storm. However, LAI variation was assessed at a forest scale. On the other hand, the innovative procedure applied to Ikonos data led to more accurate results. For example, the relative error between estimated and ground measured LAI was improved, on average, from 23% (using 20 m resolution imagery) to 6% (using very high resolution imagery).