Using ultrasonic snow depth data to minimaze the error in winter precipitaton measurements

Winter precipitation measurement is more problematic in terms of accuracy than measurement of rainfall precipitation that occur during the rest of the year. The results of rainfall measurements during the winter are significantly influenced by wind, both during the accumulation itself and subsequent redistribution of the snow cover. The aim of this work is to analyze automatically measured data from a heated tipping bucket, ultrasonic snow depth sensor for snow measurement and sensors for measuring snow water equivalent, which aims to determine the dependence of the density of new snow on temperature or humidity and consequently to derive the relation that will serve to refine the quantum estimation of winter precipitation. Based on the correlation and regression analysis of the data, the direct dependence between the density of the new snow and air humidity has not been proven, but the dependence between the density of the new snow and the air temperature has already been proven, though not very significant (Spearman's rank corelation rs = 0.39), this dependence explains approximately 9 % of data variability. The resulting calculated snowfall values in the winter seasons of 2016 and 2017 indicate that the heated tipping bucket has underestimated the amount of precipitation by 50% (2016) and 59%..