Simulating forest cover change in the northeastern U.S.: decreasing forest area and increasing fragmentation

Context: Understanding how the Northern Forest landscape has changed and is likely to change, both in terms of forest extent and forest configuration, has important implications for management. Objectives: We examined historical changes in forest pattern and extent to: (1) characterize recent forest cover change and potential drivers of that change, (2) identify areas vulnerable to future forest loss, (3) assess the impact of such loss on forest fragmentation, and (4) examine correlations between projected forest loss and socioeconomic variables to help inform future planning. Methods: We developed a cellular automata model to simulate changes in forest land cover in the Northern Forest region from 2015 to 2075. The model was parameterized from observed historical trends (1985 to 2015) and correlating spatial variables using Bayesian Weights of Evidence. Using our model outputs, we identified areas most vulnerable to change, and impacts of these changes on forest fragmentation. Results: Though we find an overall trend of decreasing forest area across the region, rates of change vary spatially and temporally, with an overall increase in forest cover between 2000 and 2015. Areas most attractive for development (e.g. high population density, low slope and elevation) were most likely to experience deforestation. Forest fragmentation increased during observed and simulated time steps, even during an observed period of net forest regeneration. Conclusions: Forest loss and fragmentation due to development represent a formidable threat to the Northern Forest. Historical trends indicate that simply increasing forest extent is not sufficient to restore forest connectivity in the region.