Forest is an important component of the ecosystem, which has been under increasing pressure in recent decades to provide raw material to emerging industries. Under these circumstances, it is necessary to develop methodologies for assessing the economic pressure on forest resources. In order to be able to assess forestry changes at regional level, a complex methodological approach is required, which implies GIS-based methods for obtaining quantitative information and fractal analysis, in order to obtain qualitative information. Applying GIS methods was designed to extract information on the spatial dynamics of the forest fund from post-processed satellite imagery and to obtain the basis of the fractal analysis. However, in order to obtain qualitative information about the degree of homogeneity / heterogeneity, fragmentation / compacting, anisotropy / isotropy and complexity of the deforested forest areas, a non-Euclidean complex analysis was applied based on the fractal analysis methods. It has been identified that in the years with very intense deforestation (2001, 2014 and 2016) the largest increases in the degree of complexity, heterogeneity, anisotropy, but insignificant increases of the fragmentation of the forest areas occurred. In antithesis, the years with low deforestation (2002, 2003 and 2005) were characterized by the smallest increases in complexity and heterogeneity (close to 0), decreases in anisotropy, but more pronounced increases in forest fragmentation. Thus, we have shown that the fractal methodology along with the GIS is very useful and provides additional information on forest dynamics.