Известия Томского политехнического университета: Инжиниринг георесурсов (Apr 2018)
Lake size-distribution in cryolithozone based on satellite images of medium and high resolution
Abstract
Relevance of the research is caused by the need to determine the kind of law of lake size-distribution for modeling and predicting the dynamics of thermokarst lakes fields and methane reserves in them in vast areas of the Arctic and subarctic zones in Siberia and other regions of the world. The main aim of the research is to determine the kind of law of thermokarst lake size-distribution in the cryolithozone of Western Siberia in a wide range of their sizes on the basis of a synthesized histogram of lake distributions obtained from medium and high-resolution space images. Methods: remote methods of studying the fields of thermokarst lakes using satellite imagery; method of test (key) sites; method for determining zones of active thermokarst from satellite imagery; methods of geoinformation systems for spatial analysis of data on thermokarst lakes derived from satellite imagery; statistical methods in determining the kind of distribution law of random variables from empirical data using the Pearson criterion. Results. The paper considers briefly the methodological issues of synthesizing two histograms of the lake size-distribution, one of which is obtained from mid-resolution space images, another one - from high-resolution imagery on a limited set of test sites in the study area. The authors have used the images of the average resolution of LandSat-8 and high-resolution images of Canopus-V, BKA and Alos to construct histograms. Remote studies on high resolution images were carried out on 66 test sites, located fairly evenly in the territory of the cryolithozone of Western Siberia. As a result, the authors obtained the synthesized histogram of lake size-distribution in a very wide range of their sizes from 50 m2to 20000 hectares. Statistical analysis showed that the empirical distribution of lakes by area, according to the Pearson criterion, corresponds to the lognormal distribution law. The authors determined the parameters of the lognormal law of lake areas distribution by empirical data.
