Microwave Imaging of Soil Water Diffusion Using the Linear Sampling Method

This letter deals with the monitoring of the volumetric water content of a soil column in a fully controlled environment by means of a noninvasive microwave imaging system. Indeed, soil moisture is an important piece of information to improve fluid flow modeling or to better understand the water uptake by plant roots. In this letter, we address the problem of recovering the footprint of soil moisture evolution with respect to time using a built-in laboratory microwave setup coupled to a robust qualitative microwave imaging method: the linear sampling method (LSM). The evolution of the water content in the soil is ruled by the Richards equations and stored at different time steps. The associated maps of soil water content are converted into permittivity maps using the Dobson model. Electromagnetic scattered fields are then computed with finite-element software. We have tested the LSM in a situation that can be encountered in agricultural soils where the water content is not homogeneous. We show that the evolution of the soil water content can be qualitatively monitored with the LSM. We also point out that the source is more precisely located by considering the evolution in time of the singular system of the multistatic matrix (multiple signal classification method).