Abstract. Soil is currently thought to be a sink for carbon; however, the response of this sink to increasing levels of atmospheric carbon dioxide and climate change is uncertain. In this study, we analyzed soil organic carbon (SOC) changes from 11 Earth system models (ESMs) contributing simulations to the Coupled Model Intercomparison Project Phase 5 (CMIP5). We used a reduced complexity model based on temperature and moisture sensitivities to analyze the drivers of SOC change for the historical and high radiative forcing (RCP 8.5) scenarios between 1850 and 2100. ESM estimates of SOC changed over the 21st century (2090–2099 minus 1997–2006) ranging from a loss of 72 Pg C to a gain of 253 Pg C with a multi-model mean gain of 65 Pg C. Many ESMs simulated large changes in high-latitude SOC that ranged from losses of 37 Pg C to gains of 146 Pg C with a multi-model mean gain of 39 Pg C across tundra and boreal biomes. All ESMs showed cumulative increases in global NPP (11 to 59%) and decreases in SOC turnover times (15 to 28%) over the 21st century. Most of the model-to-model variation in SOC change was explained by initial SOC stocks combined with the relative changes in soil inputs and decomposition rates (R2 = 0.89, p < 0.01). Between models, increases in decomposition rate were well explained by a combination of initial decomposition rate, ESM-specific Q10-factors, and changes in soil temperature (R2 = 0.80, p < 0.01). All SOC changes depended on sustained increases in NPP with global change (primarily driven by increasing CO2). Many ESMs simulated large accumulations of SOC in high-latitude biomes that are not consistent with empirical studies. Most ESMs poorly represented permafrost dynamics and omitted potential constraints on SOC storage, such as priming effects, nutrient availability, mineral surface stabilization, and aggregate formation. Future models that represent these constraints are likely to estimate smaller increases in SOC storage over the 21st century.

Abstract. We present results from an ensemble of eight climate models, each of which has carried out simulations of the early Eocene climate optimum (EECO, ∼ 50 million years ago). These simulations have been carried out in the framework of the Deep-Time Model Intercomparison Project (DeepMIP; http://www.deepmip.org, last access: 10 January 2021); thus, all models have been configured with the same paleogeographic and vegetation boundary conditions. The results indicate that these non-CO2 boundary conditions contribute between 3 and 5 ∘C to Eocene warmth. Compared with results from previous studies, the DeepMIP simulations generally show a reduced spread of the global mean surface temperature response across the ensemble for a given atmospheric CO2 concentration as well as an increased climate sensitivity on average. An energy balance analysis of the model ensemble indicates that global mean warming in the Eocene compared with the preindustrial period mostly arises from decreases in emissivity due to the elevated CO2 concentration (and associated water vapour and long-wave cloud feedbacks), whereas the reduction in the Eocene in terms of the meridional temperature gradient is primarily due to emissivity and albedo changes owing to the non-CO2 boundary conditions (i.e. the removal of the Antarctic ice sheet and changes in vegetation). Three of the models (the Community Earth System Model, CESM; the Geophysical Fluid Dynamics Laboratory, GFDL, model; and the Norwegian Earth System Model, NorESM) show results that are consistent with the proxies in terms of the global mean temperature, meridional SST gradient, and CO2, without prescribing changes to model parameters. In addition, many of the models agree well with the first-order spatial patterns in the SST proxies. However, at a more regional scale, the models lack skill. In particular, the modelled anomalies are substantially lower than those indicated by the proxies in the southwest Pacific; here, modelled continental surface air temperature anomalies are more consistent with surface air temperature proxies, implying a possible inconsistency between marine and terrestrial temperatures in either the proxies or models in this region. Our aim is that the documentation of the large-scale features and model–data comparison presented herein will pave the way to further studies that explore aspects of the model simulations in more detail, for example the ocean circulation, hydrological cycle, and modes of variability, and encourage sensitivity studies to aspects such as paleogeography, orbital configuration, and aerosols.

Abstract. The Last Glacial Maximum (LGM, 21,000 years ago) is one of the suite of paleoclimate simulations included in the current phase of the Coupled Model Intercomparison Project (CMIP6). It is an interval when insolation was similar to present, but global ice volume was at a maximum, eustatic sea level was at or close to a minimum, greenhouse gas concentrations were lower, atmospheric aerosol loadings were higher than today, and vegetation and land-surface characteristics were different from today. The LGM has been a focus for the Paleoclimate Modelling Intercomparison Project (PMIP) since its inception, and thus many of the problems that might be associated with simulating such a radically different climate are well documented. The LGM state provides an ideal case study for evaluating climate model performance because the changes in forcing and temperature between the LGM and pre-industrial are of the same order of magnitude as those projected for the end of the 21st century. Thus, the CMIP6 LGM experiment could provide additional information that can be used to constrain estimates of climate sensitivity. The design of the Tier 1 LGM experiment (lgm) includes an assessment of uncertainties in boundary conditions, in particular through the use of different reconstructions of the ice sheets and of the change in dust forcing. Additional sensitivity experiments have been designed to quantify feedbacks associated with land-surface changes and aerosol loadings, and to isolate the role of individual forcings. Model analysis and evaluation will capitalise on the relative abundance of palaeoenvironmental observations and quantitative climate reconstructions already available for the LGM.