We propose a methodology to estimate the density of frozen media (snow, firn and ice) using common offset GPR data. The technique is based on reflection amplitude analysis to calculate series’ of reflection coefficients that we used to estimate the dielectric permittivity of each layer. Applying an empirical equation we determine the vertical density variations for all the acquired GPR traces and, therefore, infer the nature of frozen materials, from fresh snow, to granular snow, firn and ice, and to estimate the snow water equivalent (SWE), which is an essential parameter to determine the actual water volume within a certain frozen body. The proposed technique was tested in the Eastern Glacier of Mt. Canin (Eastern Alps) and validated by direct snow density measurements, obtained using samples collected from manually dug snow pits, and with data from the literature. Despite the necessary approximations, the average value of density for different levels is calculated with acceptable accuracy. Moreover, this analysis can help to find and locate debris or moraines embedded within the ice bodies.
