Quantifying successional land cover after clearing of tropical rainforest along forest frontiers in the Congo Basin

摘要

State-of-the-art impact-modeling studies in environmental and climatological sciences require detailed future deforestation scenarios that allow forest to be replaced by a mosaic of multiple successional land-cover types, rather than the simple conversion of forest to a single land-cover type, such as bare soil or cropland. Therefore, not only the amount and location of forest removal has to be known (as is typically provided by scenarios), but also knowledge about the successional land-cover types and their relative areal proportions is needed. The main objective of this study was to identify these successional land-cover types and quantify their areal proportions in regions deforested during the past 37 years around the city of Kisangani, D.R. Congo. The fallow vegetation continuum was categorized in different stages, adapted from existing classifications. Ground-truth points describing the present-day vegetation were obtained during a field campaign and used for supervised and validated land-cover classification of these categories, using the Landsat image of 2012. Areal proportions of successional land-cover types were then derived from the resulting land-cover map. The second objective of this study was to relate these areal proportions to time since deforestation, which is expected to influence fallow landscapes. Landsat images of 1975, 1990, and 2001 were analyzed. Present-day mature tree fallow is less abundant on areas deforested during 1975–1990. The relative areal proportions were used to refine a deforestation scenario and apply it to existing data-sets of LAI and canopy height (CH). Assuming a simple conversion of forest to cropland, the deforestation scenario projected a reduction of grid-cell-averaged CH from 25.5 to 7.5 m (within deforested cells), whereas the refined scenarios that we propose show more subtle changes, with a reduced CH of 13 m. This illustrates the importance of taking successional land cover correctly into account in environmental and climatological modeling studies.