Modeling Spatial Structure of Thermokarst Lake Fields in Siberian Arctic by Satellite Imagery Based on Geo-Simulation and Heuristic Approaches

On the basis of geo-simulation and heuristic approaches, a model of the spatial structure of thermokarst lakes fields has been developed, which allows to take into account the lognormal distribution of real lakes in size and tortuosity of their coastal boundaries. As a result, the model represents lakes of all sizes, including small lakes, which are considered as the most intense sources of methane emission into the atmosphere from thermokarst lakes. The log-normal size distribution is confirmed by experimental data based on the integration of data from images of medium and high spatial resolution in the northern territories of Western Siberia. Studies of the statistical properties of the tortuosity of the coastal boundaries of the lakes using high-resolution images of Kanopus-V showed that the distribution of the the degree of tortuosity corresponds to the normal law. The statistical characteristics of the degree of tortuosity of real lakes are determined. On the basis of the heuristic approach, a method of forming model objects of arbitrary shape with a given area and degree of tortuosity in the Cartesian coordinate system has been developed. An algorithm for modeling the spatial structure of the fields of thermokarst lakes is described. Keywords—geo-simulation, heuristic modeling, remote sensing, thermokarst lakes, lake size-distribution, the tortuosity of coastal lakes, modeling algorithm, greenhouse gases, global warming