Estimates and determinants of stocks of deep soil carbon in Gabon, Central Africa
| Autor: | Allan R. Bacon, Vincent P. Medjibe, John R. Poulsen, Daniel Richter, Anna M. Wade, Lee J. T. White, Paul R. Heine |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Total organic carbon
Carbon accounting Soil Science Tropics chemistry.chemical_element 04 agricultural and veterinary sciences Soil carbon 010501 environmental sciences Soil type 01 natural sciences Carbon cycle chemistry Deforestation 040103 agronomy & agriculture 0401 agriculture forestry and fisheries Environmental science Physical geography Carbon 0105 earth and related environmental sciences |
| Zdroj: | Geoderma. 341:236-248 |
| ISSN: | 0016-7061 |
| DOI: | 10.1016/j.geoderma.2019.01.004 |
| Popis: | Despite the importance of tropical forest carbon to the global carbon cycle, research on carbon stocks is incomplete in major areas of the tropical world. Nowhere in the tropics is this more the case than in Africa, and especially Central Africa, where carbon stocks are known to be high but a scarcity of data limits understanding of carbon stocks and drivers. In this study, we present the first nation-wide measurements and determinants of soil carbon in Gabon, a nation in Central Africa. We estimated soil carbon to a 2-m depth using a systematic, random design of 59 plots located across Gabon. Soil carbon to a 2-m depth averaged 163 Mg ha−1 with a CV of 61%. These soil carbon stocks accounted for approximately half of the total carbon accumulated in aboveground biomass and soil pools. Nearly a third of soil carbon was stored in the second meter of soil, averaging 58 Mg ha−1 with a CV of 94%. Lithology, soil type, and terrain attributes were found to be significant predictors of cumulative SOC stocks to a 2-m depth. Current protocols of the IPCC are to sample soil carbon from the surface 30 cm, which in this study would underestimate soil carbon by 60% and underestimate ecosystem carbon by 30%. A nonlinear model using a power function predicted cumulative soil carbon stocks in the second meter with an average error of prediction of 3.2 Mg ha−1 (CV = 915%) of measured values. The magnitude and turnover of deep soil carbon in tropical forests needs to be estimated as more countries prioritize carbon accounting and monitoring in response to accelerating land-use change. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect