The combined impacts of land use change and climate change on soil organic carbon stocks in the Ethiopian highlands

Abstract

Land Use Change (LUC), especially deforestation in tropical regions, significantly contributes to global anthropogenic greenhouse gas (GHG) emissions. Here, we address potential combined impacts of LUC and Climate Change (CC) on Soil Organic Carbon (SOC) stocks in the Ethiopian highlands. The soil model Q was employed to predict SOC stocks for various combinations of LUC and CC scenarios until the year 2100. Four reference scenarios (cropland, bushland, natural forest, and Eucalyptus plantations under contemporary climatic conditions) were evaluated against reported measurements of SOC stocks. We studied impacts of six common LUC scenarios, including deforestation and planting Eucalyptus, on SOC stocks under contemporary and future climates. To assess the impact of CC, effects of elevated temperature (mean annual temperature + 2.6 °C) together with three litterfall scenarios (no change in litterfall, a 5% reduction and 22% increase, designated CC0, CCd, and CCi, respectively) were considered to test potential vegetation responses to increases in temperature and atmospheric CO2 concentrations. Most of the tested combinations of LUC and CC led to losses of SOC stocks. Losses were most severe, both relatively and absolutely, in the deforestation scenarios: up to 30% was lost if natural forest was converted to cropland and temperature increased (under the CC0 scenario). Gains in SOC stocks of 4–19% were modelled when sparse vegetation was converted to more dense vegetation like Eucalyptus plantation with substantially increased litterfall (the CCi scenario). Elevated temperature accelerated decomposition rates, leading to circa 8% losses of SOC stocks.We conclude that effects of LUC and CC on SOC stocks are additive and changes in litterfall caused by LUC determine which has the largest impact. Hence, deforestation is the biggest threat to SOC stocks in the Ethiopian highlands, and stocks in sparse vegetation systems like cropland and bushland are more sensitive to CC0 than LUC. We recommend conservation of natural forests and longer rotation periods for Eucalyptus plantations to preserve SOC stocks.Finally, we suggest that use of the Q model is a viable option for national reporting changes in SOC stocks at Tier 3 within the LULUCF sector to the United Nations Framework Convention on Climate Change (UNFCCC) as it is widely applicable and robust, although it only requires input data on a few generally available variables.