Contemporary land use/land cover types determine soil organic carbon stocks in south-west Rwanda

Soil organic carbon (SOC) constitutes a large pool within the global carbon cycle. Land use change significantly drives SOC stock variation. In tropical central and eastern Africa, how changes in land use and land cover impact on soil C stocks remains unclear. Variability in the existing data is typically explained by soil and climate factors with little consideration given to land use and management history. To address this knowledge gap, we classified the current and historical land cover and measured SOC stocks under different land cover, soil group and slope type in the humid zone of south-west Rwanda. It was observed that SOC levels were best explained by contemporary land cover types, and not by soil group, conversion history or slope position, although the latter factors explained partly the variation within annual crop land cover type. Lack of the influence of land use history on SOC stocks suggests that after conversion to a new land use/land cover, SOC stocks reached a new equilibrium within the timestep that was observed (25 years). For conversion to annual crops, SOC stocks reach a new equilibrium at about 2.5 % SOC concentration which is below the proposed soil fertility threshold of 3 % SOC content in the Eastern and central African region. SOC stock declined under transitions from banana-coffee to annual crop by 5 % or under transitions from natural forest to degraded forest by 21 % and increased for transitions from annual crops to plantation forest by 193 %. Forest clearing for agricultural use resulted in a loss of 72 %. Assuming steady states, the data can also be used to make inferences about SOC changes as a result of land cover changes. We recommend that SOC stocks should be reported by land cover type rather than by soil groups which masks local land cover and landscape differences. This study addresses a critical issue on sustainable management of SOC in the tropics and global carbon cycle given that it is performed in a part of the world that has high land cover dynamics while at the same time lacks data on land cover changes and SOC dynamics.