In view of these open questions, we want to study these soil moisture effects in two L-band airborne campaigns . The low frequency, short revisit times, small spatial baselines, and (in one campaign) absence of vegetation cover are expected to reduce the impact of these additional influences such as the topography and vegetation-related processes. The soil moisture effects, by contrast, are expected to be more dominant and thus detectable. In particular, this allows us to address the question of the sign, magnitude and statistical significance of these effects. We do so by using regression techniques whereby we describe the interferometric observables as a function of the change in soil moisture Δmv. Furthermore, we want to assess the plausibility of the different conjectured mechanisms that could describe these effects. This assessment is made by comparing their predictions with the empirically found impact of soil moisture on the interferometric data. As the applicability and relevance of these explanations are not well understood, we focus on the differences between these explanations rather than particular models and parameterizations. This analysis is conducted for different polarizations, as the sensitivity to soil moisture is not necessarily identical. In most previous studies (both observational and models), the polarimetric aspect was not addressed explicitly, often due to lack of suitable data or because the proposed physical explanations did not involve any polarimetric differences ( De Zan et al., 2014 and Nolan, 2003a).