LAUSR

Time series of snow depth bear relevant information for environmental studies in Alpine areas. In addition, due to its sensitivity to air temperature, snow depth dynamic is a robust indicator of climate change effects in mountainous areas. Unfortunately, collecting accurate snow depth data is a difficult task and often time series are obtained by merging data from different sources. Especially in the past, observations were not standardized, such that changes in the operator and in the equipment were sources of inhomogeneities in the time series. To overcome this problem and make the most efficient use of the available information, homogenization techniques may be used. However, a standardized approach for homogenizing snow depth data is currently lacking, despite its importance in climatic and hydrological studies. We evaluate the performance of the Standard Normal Homogeneity Test (SNHT) to homogenize mean seasonal snow depth data collected in the Province of Trento, Northeastern Italy. The proposed algorithm showed good performance in both detecting breakpoints and identifying homogeneous time series. Breakpoints have been detected in about 20% of the analysed time series. A homogeneity analysis on mean seasonal snow depth datasets is hence recommended before performing climatological studies.