Modeling spectral and directional soil reflectance as a function of moisture content in the solar domain ; Modélisation de la réflectance spectrale et directionnelle des sols en fonction de la teneur en eau

Soil water content (SMC) is a key variable at the Earth's surface. Its estimation is essential in many fields such as agriculture, hydrology, climate, defense or planetology. Since the 1970s, airborne or satellite remote sensing has established itself as an effective tool for mapping soil water content over large areas. Water can be detected at different depths depending on the wavelength range: solar range (0.3-3 μm), thermal infrared range (3-12 μm), or microwave range (0.5-100cm).The solar domain, which is the subject of this work, gives access to surface humidity because the depth of penetration of light into the ground does not exceed one to two millimeters. The main objective of this thesis is to develop a physical model, taking into account the spectral and directional effects, validated on a wide variety of soils, which makes it possible to relate the reflectance of the soil in the solar domain to its humidity. A new version of the MARMIT model called MARMIT-2 has been developed. This represents a wet ground like a dry ground surmounted by a thin layer of a mixture of water and mineral particles, this mixture constituting a major specificity of MARMIT-2. Spectral reflectance of moist soil is expressed as a function of soil reflectance dry and three input parameters: the thickness of the water layer, the cover fraction of the water layer and the volume fraction of soil particles in the water layer. MARMIT-2 was tested on a database of 225 soil samples whose reflectance spectra were measured in the laboratory at several humidity levels. The prediction of wet soil reflectance spectra has been improved, with an RMSE going from 2.2% for MARMIT to 0.9% for MARMIT-2.Considering that these "smooth" samples are not representative of the much more complex natural soils, and in order to adapt MARMIT-2 to airborne or satellite hyperspectral images, we studied the directional effects of soil roughness on the reflectance by performing BRF (bidirectional reflectance function) simulations of rough and humid soils using ...