Contribution à la méthodologie d'estimation de l'humidité du sol à partir de données micro-ondes passives en bande L

Abstract : One of the aims of the future SMOS mission (Soil Moisture and Ocean Salinity) is to monitor the near-surface soil moisture over the continents. The new technology (aperture synthesis) allows us to measure the surface microwave emission at 1.41 GHz for different view angles and for the two polarizations simultaneously. This emission depends on the water present in the 3 to 5 first centimeters of the soil.
However, the presence of vegetation attenuates the soil emission and adds its own contribution ; it is thus necessary to separate the two contributions to the surface emission in the view of estimating soil moisture.
The microwave emission of a vegetation cover is described by the radiative transfer model tau-omega. The inversion of this model, based on a non linear regression, is used to estimate unknown parameters : the soil moisture and the vegetation microwave parameters (optical thickness and single scattering albedo).
This study is based on different data sets acquired over different crop types : corn, wheat, alfalfa, grass, soybean and sorghum. First, we evaluated variations of the vegetation microwave parameters with time, view angle and polarization, for each crop type. This study allows us to establish the hypothesis which could simplify the model in order to improve the soil moisture estimation. Then, we tested the inversion of the tau-omega model with all the data sets simultaneously. Different inversion configurations were compared in the view of choosing the best soil moisture retrieval process, depending on the a priori information. We showed that it is possible to estimate soil moisture with a good accuracy (RMSE= 0.047 m3/m3) using few ancillary information on the soil and vegetation types.
Document type :
Theses
Liste complète des métadonnées

https://tel.archives-ouvertes.fr/tel-00152566
Contributor : Mickaël Pardé <>
Submitted on : Thursday, June 7, 2007 - 10:23:09 AM
Last modification on : Tuesday, May 23, 2017 - 11:21:52 AM
Document(s) archivé(s) le : Thursday, April 8, 2010 - 7:09:35 PM

Identifiers

  • HAL Id : tel-00152566, version 1

Collections

Citation

Mickaël Pardé. Contribution à la méthodologie d'estimation de l'humidité du sol à partir de données micro-ondes passives en bande L. Autre. Université Paris-Diderot - Paris VII, 2003. Français. 〈tel-00152566〉

Share

Metrics

Record views

426

Files downloads

385