The possibilities of creating an innovative educational and scientific center for monitoring forest resources in Siberia on the basis of the Department of Space Facilities and Technologies of the Siberian State University of Science and Technology named after Mikhail Fedorovich Reshetnev are discussed, with the aim of training highly qualified engineering personnel and conducting promising scientific research in the field of monitoring, modeling, forecasting and management of forest resources. Methodological solutions and algorithms for three-dimensional modeling of forest structure and dynamics based on laser scanning data, digital aerial and space photography are proposed. These methods contribute to operational monitoring and can significantly reduce the cost of monitoring the condition and use of forest resources over the vast territory of Siberia. Remote sensing data is presented in the form of a geotransformed database and digital photo map, compatible in formats with computer-aided design systems and with the main geographic information systems – ArcView, ArcINFO, MapINFO. The innovative monitoring center will be used for operational state control and monitoring of forest management, the state of forest lands, forest management and forest inventory, solving problems of ecology and environmental management, geoecology, formation of a forest resource inventory, aerospace methods for studying natural resources and territories, information technology. Solving these problems will allow for the training of highly qualified specialists. The center's specialists plan to create information technologies for remote sensing of natural objects with the aim of import substitution of foreign software products. The main scientific directions of the created center: development and research of methods for system analysis of large-scale multidimensional remote sensing data based on nonparametric decision-making algorithms and parallel computing technologies; testing hypotheses about the distributions of large-volume remote sensing data based on nonparametric nuclear-type pattern recognition algorithms; detection of compact groups of large-volume remote sensing data corresponding to unimodal fragments of the joint probability density of multivariate random variables.