Data-mining analysis of the global distribution of soil carbon in observational databases and Earth system models

Future climate change will dramatically change the carbon balance in the soil, and this change will affect the terrestrial carbon stock and the climate itself. Earth system models (ESMs) are used to understand the current climate and to project future climate conditions, but the soil organic carbon (SOC) stock simulated by ESMs and those of observational databases are not well correlated when the two are compared at fine grid scales. However, the specific key processes and factors, as well as the relationships among these factors that govern the SOC stock, remain unclear; the inclusion of such missing information would improve the agreement between modeled and observational data. In this study, we sought to identify the influential factors that govern global SOC distribution in observational databases, as well as those simulated by ESMs. We used a data-mining (machine-learning) (boosted regression trees – BRT) scheme to identify the factors affecting the SOC stock. We applied BRT scheme to three observational databases and 15 ESM outputs from the fifth phase of the Coupled Model Intercomparison Project (CMIP5) and examined the effects of 13 variables/factors categorized into five groups (climate, soil property, topography, vegetation, and land-use history). Globally, the contributions of mean annual temperature, clay content, carbon-to-nitrogen (CN) ratio, wetland ratio, and land cover were high in observational databases, whereas the contributions of the mean annual temperature, land cover, and net primary productivity (NPP) were predominant in the SOC distribution in ESMs. A comparison of the influential factors at a global scale revealed that the most distinct differences between the SOCs from the observational databases and ESMs were the low clay content and CN ratio contributions, and the high NPP contribution in the ESMs. The results of this study will aid in identifying the causes of the current mismatches between observational SOC databases and ESM outputs and improve the modeling of terrestrial carbon dynamics in ESMs. This study also reveals how a data-mining algorithm can be used to assess model outputs.

[1]  A. Pitman,et al.  Response of microbial decomposition to spin-up explains CMIP5 soil carbon range until 2100 , 2014 .

[2]  D. Jenkinson,et al.  RothC-26.3 - A Model for the turnover of carbon in soil , 1996 .

[3]  J. Randerson,et al.  Causes of variation in soil carbon simulations from CMIP5 Earth system models and comparison with observations , 2012 .

[4]  G. Collatz,et al.  ISLSCP II Global Gridded Soil Characteristics , 2011 .

[5]  N. Batjes,et al.  Total carbon and nitrogen in the soils of the world , 1996 .

[6]  R. B. Jackson,et al.  THE VERTICAL DISTRIBUTION OF SOIL ORGANIC CARBON AND ITS RELATION TO CLIMATE AND VEGETATION , 2000 .

[7]  Wolfgang Lucht,et al.  Quantifying uncertainties in soil carbon responses to changes in global mean temperature and precipitation , 2014 .

[8]  Ben Bond-Lamberty,et al.  Temperature-associated increases in the global soil respiration record , 2010, Nature.

[9]  Pierre Friedlingstein,et al.  Controls on terrestrial carbon feedbacks by productivity versus turnover in the CMIP5 Earth System Models , 2015 .

[10]  Maurizio Santoro,et al.  Global covariation of carbon turnover times with climate in terrestrial ecosystems , 2014, Nature.

[11]  T. Morishita,et al.  Increasing trends of soil greenhouse gas fluxes in Japanese forests from 1980 to 2009 , 2011, Scientific reports.

[12]  G. Bonan,et al.  Evaluating soil biogeochemistry parameterizations in Earth system models with observations , 2014 .

[13]  J. Canadell,et al.  Soil organic carbon pools in the northern circumpolar permafrost region , 2009 .

[14]  A. Pitman,et al.  Examining soil carbon uncertainty in a global model: response of microbial decomposition to temperature, moisture and nutrient limitation , 2013 .

[15]  Pierre Friedlingstein,et al.  Uncertainties in CMIP5 Climate Projections due to Carbon Cycle Feedbacks , 2014 .

[16]  Global Soil Data Task,et al.  Global Gridded Surfaces of Selected Soil Characteristics (IGBP-DIS) , 2000 .

[17]  Markus Reichstein,et al.  Modeling the vertical soil organic matter profile using Bayesian parameter estimation , 2012 .

[18]  K. Denef,et al.  The Microbial Efficiency‐Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter? , 2013, Global change biology.

[19]  J. Six,et al.  Soil carbon storage controlled by interactions between geochemistry and climate , 2015 .

[20]  Jianyang Xia,et al.  Evaluation and improvement of a global land model against soil carbon data using a Bayesian Markov chain Monte Carlo method , 2014 .

[21]  Atul K. Jain,et al.  Global patterns and controls of soil organic carbon dynamics as simulated by multiple terrestrial biosphere models: Current status and future directions , 2015, Global biogeochemical cycles.

[22]  Dominique Arrouays,et al.  Spatial distribution of soil organic carbon stocks in France , 2010 .

[23]  Thomas H. Painter,et al.  Climatic, edaphic, and biotic controls over storage and turnover of carbon in soils , 1994 .

[24]  G. J. Collatz,et al.  ISLSCP II Global Primary Production Data Initiative Gridded NPP Data , 2011 .

[25]  S. Zaehle,et al.  Terrestrial nitrogen–carbon cycle interactions at the global scale , 2013, Philosophical Transactions of the Royal Society B: Biological Sciences.

[26]  Bart R. Johnson,et al.  Quantifying global soil carbon losses in response to warming , 2016, Nature.

[27]  J. Randerson,et al.  Changes in soil organic carbon storage predicted by Earth system models during the 21st century , 2013 .

[28]  L. Vesterdal,et al.  Tree species is the major factor explaining C:N ratios in European forest soils , 2014 .

[29]  Alan H. Strahler,et al.  ISLSCP II MODIS (Collection 4) IGBP Land Cover, 2000-2001 , 2010 .

[30]  R. Hiederer,et al.  Global distribution of soil organic carbon – Part 1: Masses and frequency distributions of SOC stocks for the tropics, permafrost regions, wetlands, and the world , 2015 .

[31]  Amilcare Porporato,et al.  Soil carbon and nitrogen mineralization: Theory and models across scales , 2009 .

[32]  P. Dirmeyer,et al.  Trends in Land–Atmosphere Interactions from CMIP5 Simulations , 2013 .

[33]  R. Schnur,et al.  Climate-carbon cycle feedback analysis: Results from the C , 2006 .

[34]  Nuno Carvalhais,et al.  Implications of the carbon cycle steady state assumption for biogeochemical modeling performance and inverse parameter retrieval , 2008 .

[35]  J Elith,et al.  A working guide to boosted regression trees. , 2008, The Journal of animal ecology.

[36]  G. Collatz,et al.  ISLSCP II Historical Croplands Cover, 1700-1992 , 2010 .

[37]  M. Lomas,et al.  Decomposing uncertainties in the future terrestrial carbon budget associated with emission scenarios, climate projections, and ecosystem simulations using the ISI-MIP results , 2014 .

[38]  J. Liski,et al.  Leaf litter decomposition-Estimates of global variability based on Yasso07 model , 2009, 0906.0886.

[39]  Auroop R. Ganguly,et al.  Regional and seasonal intercomparison of CMIP3 and CMIP5 climate model ensembles for temperature and precipitation , 2014, Climate Dynamics.

[40]  E. Chang,et al.  CMIP5 multimodel ensemble projection of storm track change under global warming , 2012 .

[41]  Dale W. Johnson,et al.  Humus buildup in boreal forests : effects of litter fall and its N concentration , 2001 .

[42]  Michael Obersteiner,et al.  Nutrient availability as the key regulator of global forest carbon balance , 2014 .

[43]  J. Lamarque,et al.  Nitrogen Availability Reduces CMIP5 Projections of Twenty-First-Century Land Carbon Uptake* , 2015 .

[44]  Eric A. Davidson,et al.  Gas diffusivity and production of CO2 in deep soils of the eastern Amazon , 1995 .

[45]  Corinne Le Quéré,et al.  Climate Change 2013: The Physical Science Basis , 2013 .

[46]  J. Liski,et al.  Carbon and decomposition model Yasso for forest soils , 2005 .

[47]  William R. Wieder,et al.  Global soil carbon projections are improved by modelling microbial processes , 2013 .

[48]  Yujie He,et al.  Explicitly representing soil microbial processes in Earth system models , 2015 .

[49]  Benjamin L Turner,et al.  Mycorrhiza-mediated competition between plants and decomposers drives soil carbon storage , 2014, Nature.

[50]  H. Komatsu,et al.  Relationships between soil CO2 concentration and CO2 production, temperature, water content, and gas diffusivity: implications for field studies through sensitivity analyses , 2006, Journal of Forest Research.

[51]  Russell K. Monson,et al.  Terrestrial Carbon Cycle: Climate Relations in Eight CMIP5 Earth System Models , 2013 .

[52]  J. Liski,et al.  Leaf litter decomposition — Estimates of global variability based on Yasso 07 model , 2009 .

[53]  Nuno Carvalhais,et al.  Global spatiotemporal distribution of soil respiration modeled using a global database , 2015 .

[54]  P. Döll,et al.  Development and validation of a global database of lakes, reservoirs and wetlands , 2004 .

[55]  Roland Hiederer,et al.  Global soil carbon: understanding and managing the largest terrestrial carbon pool , 2014 .

[56]  J. Canadell,et al.  The Northern Circumpolar Soil Carbon Database: spatially distributed datasets of soil coverage and soil carbon storage in the northern permafrost regions , 2012 .

[57]  Matthew J. Smith,et al.  Responses of two nonlinear microbial models to warming and increased carbon input , 2015 .

[58]  J. Randerson,et al.  Changes in soil organic carbon storage predicted by Earth system models during the 21 st century , 2013 .

[59]  Bart Muys,et al.  Predicting forest site productivity in temperate lowland from forest floor, soil and litterfall characteristics using boosted regression trees , 2011, Plant and Soil.

[60]  P. Cox,et al.  1 EVALUATING THE LAND AND OCEAN COMPONENTS OF THE GLOBAL 1 CARBON CYCLE IN THE CMIP 5 EARTH SYSTEM MODELS 2 3 4 , 2022 .

[61]  K.,et al.  Carbon–Concentration and Carbon–Climate Feedbacks in CMIP5 Earth System Models , 2012 .

[62]  C. Sierra,et al.  A general mathematical framework for representing soil organic matter dynamics , 2015 .

[63]  E. Davidson,et al.  Sensitivity of decomposition rates of soil organic matter with respect to simultaneous changes in temperature and moisture , 2015 .

[64]  Marc L. Imhoff,et al.  Exploring global patterns of net primary production carbon supply and demand using satellite observations and statistical data , 2006 .

[65]  Yujie He,et al.  Toward more realistic projections of soil carbon dynamics by Earth system models , 2016 .

[66]  R. Betts,et al.  Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model , 2000, Nature.

[67]  Markus Reichstein,et al.  Soils apart from equilibrium – consequences for soil carbon balance modelling , 2006 .

[68]  A. Lehtonen,et al.  Underestimation of boreal soil carbon stocks by mathematical soil carbon models linked to soil nutrient status , 2016 .

[69]  K. Tachiiri,et al.  Modeling in Earth system science up to and beyond IPCC AR5 , 2014, Progress in Earth and Planetary Science.

[70]  Pete Smith,et al.  Integrating plant–soil interactions into global carbon cycle models , 2009 .

[71]  Roland Hiederer,et al.  Global Soil Organic Carbon Estimates and the Harmonized World Soil Database , 2011 .

[72]  Steve Easterbrook,et al.  The software architecture of climate models: a graphical comparison of CMIP5 and EMICAR5 configurations , 2015 .

[73]  W. Parton,et al.  Analysis of factors controlling soil organic matter levels in Great Plains grasslands , 1987 .

[74]  Matthew J. Smith,et al.  Oscillatory behavior of two nonlinear microbial models of soil carbon decomposition , 2013 .

[75]  Yiqi Luo,et al.  Scale-Dependent Performance of CMIP5 Earth System Models in Simulating Terrestrial Vegetation Carbon* , 2015 .