We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Causes of variation in soil carbon predictions from CMIP5 Earth system models and comparison with observations.
- Authors
Todd-Brown, K. E. O.; Randerson, J. T.; Post, W. M.; Hoffman, F. M.; Tarnocai, C.; Schuur, E. A. G.; Allison, S. D.
- Abstract
Stocks of soil organic carbon represent a large component of the carbon cycle that may participate in climate change feedbacks, particularly on decadal and century scales. For Earth system models (ESMs), the ability to accurately represent the global distribution of existing soil carbon stocks is a prerequisite for predicting future carbon-climate feedbacks. We compared soil carbon predictions from 16 ESMs to empirical data from the Harmonized World Soil Database (HWSD) and Northern Circumpolar Soil Carbon Database (NCSCD). Model estimates of global soil carbon stocks ranged from 510 to 3050 Pg C, compared to an estimate of 890-1660 Pg C from the HWSD. Model predictions for the high latitudes fell between 60 and 800 Pg C, compared to 380-620 Pg C from the NCSCD and 290 Pg C from the HWSD. This 5.3-fold variation in global soil carbon across models compared to a 3.4-fold variation in net primary productivity (NPP) and a 3.8-fold variation in global soil carbon turnover times. The spatial distribution of soil carbon predicted by the ESMs was not well correlated with the HWSD (Pearson's correlations < 0.4, RMSE 9.4 to 22.8 kgCm-2), although model-data agreement generally improved at the biome scale. There was poor agreement between the HWSD and NCSCD datasets in northern latitudes (Pearson's correlation=0.33), indicating uncertainty in empirical estimates of soil carbon. We found that a reduced complexity model dependent on NPP and soil temperature explained most of the spatial variation in soil carbon predicted by most ESMs (R² values between 0.73 and 0.93). This result suggests that differences in soil carbon predictions between ESMs are driven primarily by differences in predicted NPP and the parameterization of soil carbon responses to NPP and temperature not by structural differences between the models. Future work should focus on accurately representing these driving variables and modifying model structure to include additional processes.
- Subjects
CARBON in soils; CARBON cycle; CLIMATE change; COMPARATIVE studies; PREDICTION theory; MATHEMATICAL models; EMPIRICAL research
- Publication
Biogeosciences Discussions, 2012, Vol 9, Issue 10, p14437
- ISSN
1810-6277
- Publication type
Article
- DOI
10.5194/bgd-9-14437-2012