Snow stratigraphy from upward-looking radar

Snow slab avalanches involve the release of a cohesive slab over an extended plane of weakness. In-situ observations of snow stratigraphy are necessary for reliable prediction of snow-slope instability, conventionally obtained from destructive methods in snow pits. This hampers acquiring information about temporal snowpack evolution and often exposes the investigator to avalanche risk. Therefore, in-situ information from slopes during high-instability periods are seldom obtained. This project investigates, employs and advances the application of remotely operated upward-looking radar systems to non-destructively image and characterise the local physical properties of the snowpack, and assimilation thereof into an existing model of snowpack evolution. The radar is mounted underneath the snowpack and buried with increasing snow height. As weak layers in the snowpack have different structure or physical properties than their surroundings, transmitted waves traveling upwards through the snowpack are partially reflected. Analyses of reflected waves received by the radar yield a physical characterisation of the snowpack (internal layer number and properties, snow thickness, density). Repetition of such measurements on a daily basis allows for quasi real-time and destruction free monitoring of the development of snowpack stratigraphy and concurrent information to estimate avalanche danger. This project provides an initial step for autonomous monitoring of snow stratigraphy also in potentially unstable slopes without risks for the investigators.

Forecasting snow avalanche danger in alpine regions is of major importance for the protection of infrastructure in avalanche run-out zones. Inexpensive measurement devices capable of measuring the temporal evolution of snow height and layer properties in avalanche starting zones may help to improve the quality of risk assessment. The goal of the project MUSI was to observe the snowpack properties with upward-looking radar systems during the course of a winter season. The project was a cooperation within the D-A-CH network between the Institute of Environmental Physics (UNI Heidelberg), the Alfred-Wegener-Institute for Polar and Marine Research (Bremerhaven), the WSL Institute of Snow and Avalanche Research (SLF, Davos) and the Institute for Electronic Engineering (FH JOANNEUM).The Austrian contribution to the project was the development of an upward-looking L-band (1-2 GHz) FMCW radar, its installation in the bottom of the SLF measurement area on Weißfluhjoch (Davos) and the data analysis. While the radar subsequently was covered by snowfalls it recorded reflections from the overlying snowpack. Thereby it automatically recorded the evolution of the snow surface and variations within the snowpack during three winter seasons. From the reflection magnitudes and phase (first-time with FMCW snow measurements) we generated radar plots, which clearly indicated the temporal evolution of the snow surface and that of internal snow layers. The results showed that the temporal evolution could be observed near completely during the full three winter seasons. This applied just as for the critical wet snow periods. Important surface events for risk assessment, like snowfall, melt, melt-freeze cycles, wet snowpack, were clearly detectable and coincided with conventional measured snow and environmental parameters in the measurement area.For the automatic detection of the snow height from the FMCW data we developed an automated and a semi-automated (which used the phase information) snow surface tracking algorithm. The comparison between the results of this algorithms and the snow height measured with a laser snow-depth sensor showed, that the phase information improved the accuracy of the snow height detection.Results suggest that the upward-looking FMCW system may be a valuable alternative to conventional snow-depth sensors for locations, where fixed installations above ground are not feasible.