Wikindx Resources

Journal Article: BibTeX citation key:  Baret2009
Baret, F., Weiss, M., Allard, D., Guarrigue, S., Leroy, M., Jeanjean, H., Fernandes, R., Myneni, R., Privette, J. L., Morisette, J., Bohbot, H., Bosseno, R., Dedieu, G., Di Bella, C., Duchemin, B., Espana, M., Gond, V., Gu, X. F., Guyon, D., Lelong, C., Maisongrande, P., Mougin, E., Nilson, T., Veroustraete, F. & Vintilla, R. (2009) VALERI: a network of sites and a methodology for the validation of medium spatial resolution land satellite products. IN Remote Sensing of Environment, .
Added by: Fanny Lefebvre 2010-11-19 19:16:54
 B  
Categories: General
Creators: Allard, Baret, Bohbot, Bosseno, Dedieu, Di Bella, Duchemin, Espana, Fernandes, Gond, Gu, Guarrigue, Guyon, Jeanjean, Lelong, Leroy, Maisongrande, Morisette, Mougin, Myneni, Nilson, Privette, Veroustraete, Vintilla, Weiss
Collection: Remote Sensing of Environment

Number of views:  747
Popularity index:  37.96%
Maturity index:  submitted

 
Abstract
Validation is mandatory to quantify the reliability of satellite biophysical products that are now routinely generated by a range of sensors. This paper presents the VALERI project dedicated to the validation of the products derived from medium resolution satellite sensors (www. avignon.inra.fr/valeri/). It describes the sites used, and the methodology developed to get the high spatial resolution map of the biophysical variables considered, i.e. LAI, fAPAR and fCover that can be estimated from ground level gap fraction measurements.
Sites were selected to represent , with the other validation projects, the large variation of biomes and conditions observed over the Earth’s surface. Each site is about 3×3 km² in size and should be flat and relatively homogeneous at the medium resolution scale. For each site, the methodology used to generate the high spatial resolution biophysical variable maps is described. It is mainly based on concurent use of local gound measurements and a high spatial resolution satellite image, generally SPOT-HRV. Local ground measurements should be representative of an elementary sampling unit (ESU) that has approximately the same size as a SPOT-HRV multispectral pixel. The ground measurements mainly consist of gap fraction measurements achieved with LAI-2000 measurements or hemispherical photographs. The ESUs are selected over the whole 3×3 km² site in order to sample the vegetation types observed and to allow the derivation of variograms. A transfer function is subsequently established over the ESUs to relate the ground measurements of the biophysical variables with those of the high spatial resolution satellite image. Finally, co-kriging is applied to generate the high spatial resolution map of the biophysical variables over the 3×3 km² area.
The methodology presented in this paper can serve as a basis for validating medium resolution satellite products. Finally these methodological aspects are discussed and conclusions drawn on the limitations and prospects of beforementioned validation activity.
Added by: Fanny Lefebvre