Several methods are proposed in order to obtain agrometeorological and geological parameters, such as thermal inertia and evapo-transpiration flux, over large regions of our planet from both visible and thermal infrared data obtained from satellites. These methods are based on minimization techniques which depend on a linear operator that maps the fluxes at the surface to the surface temperature. The methods can be adaptated to the various experimental situations encountered, i.e., for a given set of available input data, they lead to the corresponding set of unknown parameters. For example, they give the thermal inertia and the evapotranspiration flux without a priori knowledge of the heat transfer resistances. To take into account the error of measurements, we propose a stabilization procedure that improves the accuracy of the method.