Abstract. The melting layer designates the transition region from solid to liquid
precipitation, and is a typical feature of the vertical structure of
stratiform precipitation. As it is characterised by a well-known signature in
polarimetric radar variables, it can be identified by automatic detection
algorithms. Though often assumed to be uniform in space and time for
applications such as vertical profile correction, the spatial variability of
the melting layer remains poorly documented. This work aims to
characterise and quantify the spatial and temporal variability of the melting
layer using a method based on the Fourier transform, which is applied to
high-resolution X-band polarimetric radar data from two measurement campaigns in
Switzerland. It is first demonstrated that the proposed method can accurately
and concisely describe the spatial variability of the melting layer and may
therefore be used as a tool for comparison. The method is then used to
characterise the melting layer variability in summer precipitation on the
relatively flat Swiss Plateau and in winter precipitation in a large inner
Alpine valley (the Rhone valley in the Swiss Alps). Results indicate a higher
contribution of smaller spatial scales to the total melting layer variability
in the case of the Alpine environment. The same method is also applied to data from vertical scans in order to study the temporal variability of the
melting layer. The variability in space and time is then compared to
investigate the spatio-temporal coherence of the melting layer variability in
the two study areas, which was found to be more consistent with the
assumption of pure advection for the case of the plateau.