These environmental monitoring data is the basis for the systems of decision-making support in ecological management. The snow cover is both a seasonal component of the environment and informative indicator of current air, soil and surface water pollution. The characteristics of the spatial structure of technogenic snow cover pollution is variable according to the snow survey because of the complex conditions of occurrence such as depth, temperature, wind patterns and etc and the diagnostics of snow cover pollution is difficult. The aim of this study is to estimate the snow cover contamination of Academic district of Yekaterinburg to use a hybrid model, combining kriging method and artificial neural networks. The main pollut- ants of snow cover the territory of the district of Yekaterinburg Academic is insoluble forms of aluminum, nickel, cobalt, manganese and dust and soluble form of natrium. The results show a high spatial heterogeneity of snow cover pollution fields, therefore, a high degree of pollution estimate level uncertainty. Uncertainty territories pollution were analyzed for dust by ANN and kriging hybrid models. The hybrid models is found useful for the prediction of the spatial distribution of impurities in snow cover.
|Translated title of the contribution||APPLICATION OF GEOSTATISTICAL AND NEURAL NETWORK APPROACH FOR THE EVALUATION OF SURFACE DISTRIBUTION OF IMPURITIES IN THE SNOW COVER IN THE «ACADEMICHESKY» DISTRICT OF YEKATERINBURG|
|Number of pages||15|
|Journal||Экологические системы и приборы|
|Publication status||Published - 2016|
Level of Research Output
- VAK List