Forested land-cover patterns and trends over changing forest management eras in the Siberian Baikal region

Remote sensing observations over areas of the former Soviet Union suggest that there may be important ongoing influences on forested landscapes resulting from divergent land use and forest management associated with the Soviet versus post-Soviet eras. As the Russian Federation implements its new Forest Code and associated regulations, knowledge of existing forest patterns and trends, plus the development of methods with which to understand the landscape-level influence of different forest management strategies is increasingly important. We developed spatial–temporal models and projections of forest patterns and trends over Soviet and early post-Soviet forest management eras for a study site in the Lake Baikal region in southern Siberia. We used Landsat-derived land-cover data, logistic regressions, and Markov and cellular automata methods (CA–Markov) to characterize patterns and trends 1975–1989 and 1990–2001, and to develop predictive scenarios through 2013. Relationships of forest types (Conifer, Mixed, Deciduous) and Agriculture to other explanatory environmental variables indicated mostly consistent forest–environment relationships, but some different spatial relationships between eras were found for Cut and Regeneration disturbance types. Landscape proportional trends showed greater differences between eras. Cut proportions observed via Landsat in 2001 were approximately 74% lower, and the area of Conifer observed was approximately 14% higher, than modeled proportions predicted for 2001 using 1975–1989 Soviet era transition rates. The proportion of Cut projected for 2013 was about 80% lower when based on early post-Soviet era probabilities. Overall, modeled results indicate that should early post-Soviet trends continue, low rates of logging, some agricultural abandonment, re-growing forests especially near access routes, increases in deciduous cover, along with continued or increased fire events in mixed and conifer forests will define the landscape. Should forest management change, for example to Soviet era rates and patterns of harvest, different outcomes are projected. More broadly, results highlight the real and prospective effects that divergent management strategies can have on forested landscapes, and demonstrate that land-cover data combined with emerging spatial–temporal modeling methods provide an approach to understand and project the complex and ongoing influences associated with changing forest management at landscape scales.