|
Vegetation and soil temperatures have long been recognized as an indicator of water availability. When plotting the twodimensional
distribution of temperature and vegetation index (T-VI) corresponding to an area with well distributed
vegetation cover and moisture content, one gets a triangular or trapezoidal shape. Iso-moisture lines range between two
edges of the distribution: the dry edge and the wet edge. This description is based on a simplified representation of the
thermal and radiative properties of the soil/vegetation structure and of the heat and mass transfer. A large number of
parameters, in addition to soil moisture, are actually influencing the apparent soil/canopy temperature. Stretching the
temperature distribution along only one dimension like the VI is indeed not enough for allowing an unambiguous
determination of soil moisture. We propose to improve the identifiability of soil moisture by introducing an additional
observable parameter: albedo or CAI (Cellulose Absorption Index). Albedo was chosen to separate areas according to
the absorbed solar radiation whereas CAI was chosen to separate areas according to the fraction of senescent vegetation.
The aim of this study is to analyze the benefit of adding a third index to the classical T-VI empirical method for soil
moisture mapping. The proposed procedures were applied on remote sensing data obtained during two airborne
campaigns: HyEurope 2007 and AgriSAR 2006. In the first case, when applying the 2D trapezoïd method, the coefficient
of determination between the inferred moisture index and the gravimetric moisture content reaches 0.61. It slightly
diminishes to 0.59 when adding albedo whereas it increases to 0.69 when adding CAI instead. Therefore adding albedo
doesn't seem to provide an improvement, at least for the considered cropland. The introduction of the Cellulose
Absorption Index seems to be more promising. The crops during AgriSAR 2006 campaign presented in July a too high
LAI whereas the soil surface was generally very dry. These conditions prevented to draw any significant conclusion.
Better results are expected to be met for areas presenting lower vegetation cover as for example semi-arid regions.
|
