Using machine learning to classify stratigraphic layers of snow according to the snow micro pen device

Abstract

The observation of snow cover on the site of the meteorological observatory by the staff of the Geographical Faculty of Moscow State University has been conducted for a long time. The article describes the features of snow accumulation and stratigraphy of the snow cover. At the time of the third cyclone that came to Moscow on the night of December 14, since the beginning of the snow accumulation, there was a large height of snowdrifts and mark of 49 cm was recorded at the MSU weather station. The difficulties of classifying layers in the snow column have been investigated and are being investigated by many glaciologists, which is also considered in this paper. Machine learning methods were used to classify stratigraphic layers of the snow column according to measurements from the snow micro pen device. The shapes of ice crystals in the snow column resulting from metamorphism (rounded, faceted, thawed) differ both in density and in the parameters obtained as a result of processing data from the Snowmicropen device (MPF(N) is the average resistance force SD(N) is its standard deviation, and cv is its covariance). This makes it possible to cluster the processed device data and type new measurement data of the device without involving the results of direct manual drilling. The data obtained from the device were processed, and by comparing with the data of direct snow stratigrafy survey, a comparison of the classified stratigraphic layers of the snow column was made. In the future, according to the available classified data of the device of stratigraphic layers of the snow column, by clustering K-nearest neighbors, it turned out to be possible to classify stratigraphic layers according to the new obtained data of the device without involving additional manual survey.