Estimation of Liquid Water Content and Density in the Surface Layer of the Snowpack from the Phase and Amplitude of SFCW Radar Signals

The precise determination of liquid water content (LWC) and density in the surface layer of the snowpack is crucial for understanding hydrological, energetic, and mechanical processes in snow-covered environments. The surface quality of the snowpack controls energy exchanges with the atmosphere and influences the electromagnetic response of the medium. However, the simultaneous and non-intrusive estimation of density and LWC remains challenging due to the strong interdependence between these parameters and the limited ability of many methods to separate them.

This work presents a method for the simultaneous estimation of density and liquid water content in the surface layer of a snowpack using a Stepped Frequency Continuous Wave (SFCW) radar operating in the 0.6–6 GHz range. The methodology is based on identifying, within the Fourier transform of the received signal, the peak corresponding to the air–snow interface. From this peak, two fundamental quantities are extracted (amplitude and phase) which are used to estimate the electromagnetic and physical properties of the snowpack surface.

Phase differences of the reflection peak are used to estimate LWC, as liquid water is the only constituent of the snowpack that introduces a significant imaginary component to the refractive index within the considered frequency range. In this interval, the complex permittivity of water exhibits high values, with a dominant effective imaginary part, while air introduces no losses and ice has an imaginary component at least three orders of magnitude smaller than that of water. Consequently, the accumulated phase shift of the reflected signal is directly controlled by the presence of liquid water, allowing small variations in LWC to be detected in the phase of the reflection peak.

The amplitude of the reflection peak depends on the total material content at the surface, as all constituents contribute to the real part of the effective refractive index. The amplitude is influenced by both snow density and LWC. Since the liquid water fraction is obtained beforehand from the phase, the relative proportions of air and ice can be estimated. From this information, the dry snow density is calculated, and through a volumetric balance, the total density of the surface layer and the LWC are determined.

The method is supported by preliminary calculations of the reflection coefficient Γ, which are used to derive calibration relationships for both phase and amplitude. Validation is carried out using synthetic snow structures representative of different surface conditions, including variations in dry snow density, liquid water content and layer thickness. In addition, initial field experiments have been conducted, showing responses consistent with the synthetic analysis and demonstrating the applicability of the approach under realistic conditions.

The results indicate that the combination of phase and amplitude constitutes a robust, non-intrusive tool for in situ monitoring of the snowpack, with the potential to detect early-stage compaction, melting, refreezing and rainfall events on snow.