Results of Statistical Analysis of the Morphometric Characteristics of Mountain Glaciers in Russia

Abstract

Based on the statistical analysis of data from the electronic Catalogue of Glaciers of Russia, a study of the morphometric characteristics of 5863 mountain glaciers had been performed. The values of pair correlations of three characteristics (length, perimeter, area) of glaciers are at the level of 0.8-0.9, both for the entire data set and for different types of glaciers and different glaciated regions. Correlations remain high even when a random sample of glaciers of different scales is taken. The correlation analysis showed that the relationship of morphometric characteristics of mountain glaciers can be optimally described by power functions. In the «area-length» relationship, finding a trend line (using the least-squares method) for the entire dataset yields an equation with power of 1.582. As the smallest glaciers are excluded from the general sample, the power in the equation becomes 1.618, while the constant multiplier of the equation remains unchanged. For other samples from the dataset, the least-squares method constructs different trend lines, but the resulting equation remains the most optimal not in terms of maximum R2, but in terms of total area estimates and approximation in the area of large glaciers. In this case, the resulting equation indicates that the ratios of the morphometric characteristics of mountain glaciers are scalable and, as the study has shown, are applicable to many types of glaciers. For regional studies, it is important that the total values of the actual and calculated glaciation areas coincide with an error of less than 1%. Despite the high accuracy of total glaciation area estimates, the area of individual glaciers may differ significantly from the actual area and can only be considered as probabilistic.