Satellite-based radar measurement are increasingly important for the detection of avalanche activity and the characterization of the snow cover properties on a regional scale. However, they operate with a temporal resolution of several days, which can detect major avalanche cycles, but changes induced by melt-freeze cycles, rain-on-snow events, snow fall, avalanche events and alike are more challenging to deduce. We present a ground-based radar system that measured the snow cover of a complete mountain slope during the winter season 2020-2021 with an hourly resolution and records flowing avalanches with 50Hz sampling. The observed mountain in Austria is a steep slope of southern aspect with avalanche control measures and corresponding artificial avalanche releases after every significant snow fall. We observe significant changes in the radar backscatter signal on different time scales: The rapid avalanche movement is analyzed by means of moving target identification filtering which suppresses the background signal from the resting snow cover, but enhances the rapid signal variations of the flow. Avalanche activity and their deposits can be identified by comparing the radar signal before and after avalanche events. And longer variations like diurnal changes in the backscatter signal are associated with meteorological influences, such as absorption of solar radiation and warming of the snow cover. We analyze the backscatter signal in terms of the flowing avalanches, immediate changes from fresh avalanche deposits and put these into relation to the magnitude of backscatter changes caused by the meteorological factors. We find that the backscatter signal of avalanche deposits can fluctuate significantly across the runout area, which is in accordance with satellite-based avalanche detection procedures that cluster deposits as regions with large radar signal variation. In the future, the developed radar may be used as ground-truth to support snow cover characterization with satellite-based radar missions.