Developments in the MPI-M Earth System Model Version 1.2 (MPI-ESM1.2) and Its Response to Increasing CO2
Observation and Context
Complex climate models contain hundreds of thousands of lines of computer code, making subtle programming errors easy to miss. In earlier versions of the Max Planck Institute Earth System Model (MPI-ESM), hidden coding flaws in atmospheric physics caused artificial energy leaks, while older soil and marine biology schemes struggled to match real-world carbon and nitrogen data. Furthermore, an intermediate update displayed an unrealistically high climate sensitivity of around 7 K due to tropical low-level cloud feedbacks, which would have grossly exaggerated observed 20th-century warming. Scientists required a corrected, physically consistent, and computationally faster model for international CMIP6 climate assessments.
Hypothesis
If developers resolve internal coding errors violating energy conservation, refine parameterizations for clouds, aerosols, soils, and ocean biogeochemistry, and explicitly tune cloud processes to match the historical instrumental warming record, then MPI-ESM1.2 will maintain strict physical energy conservation, improve present-day environmental fidelity, and produce realistic climate responses under rising carbon dioxide concentrations.
Experiment and Methodology
Researchers updated and evaluated the coupled model components of MPI-ESM1.2:
- Atmosphere (ECHAM6.3): Corrected three major compensating coding errors in convective mass fluxes, detrainment, and turbulent heat diffusion that had caused localized energy budget errors of up to several W/m2. Replaced the aerosol scheme with the observationally constrained MACv2-SP simple plume parameterization (including Twomey indirect cloud effects) and introduced PSrad radiative transfer and an improved marine stratocumulus scheme.
- Land (JSBACH3.2): Swapped the simple soil bucket model for a multilayer soil hydrology scheme, replaced unconstrained decomposition with the observation-based YASSO model, and added a process-based wildfire model (SPITFIRE) and terrestrial nitrogen cycling.
- Ocean and Carbon Cycle (MPIOM1.6 & HAMOCC6): Fixed a vertical viscosity coding error reducing sea surface temperature biases, added prognostic nitrogen-fixing cyanobacteria, and refined particle settling.
- Testing: Ran long pre-industrial control simulations, historical integrations (1850–2014), and abrupt 2×, 4×, 8×, and 16× CO2 experiments run for 1,000 years each to analyze climate sensitivity and feedback behavior.
Results and Data
- Energy Conservation & Biases: Column-level energy budget errors from atmospheric physics were eliminated in ECHAM6.3 (reduced to 0.00 W/m2 in offline tests). The ocean vertical viscosity fix significantly lowered sea surface temperature biases in upwelling zones and the Southern Ocean.
- Ecosystem Improvements: The YASSO decomposition model removed spurious soil carbon peaks in drylands, matching global soil databases. Prognostic cyanobacteria confined marine nitrogen fixation realistically to warm waters between 40°S and 40°N (82 Tg N/yr in the LR configuration).
- Historical Tuning & Sensitivity: By increasing shallow convective entrainment tenfold, developers lowered the model’s equilibrium climate sensitivity (ECS) to a doubling of CO2 from ~7 K down to 2.77 K (or 2.83 K using standard linear regression methods), matching historical 20th-century warming trends.
- Nonlinear Warming Response: Despite eliminating energy leakages, the model retained a rising climate sensitivity under strong forcing: ECS increased from 2.77 K (2×CO2) to 3.6 K (4×CO2), 4.9 K (8×CO2), and nearly 10 K (16×CO2). This was driven by strengthening water vapor and cloud feedbacks and declining planetary emissivity.
Conclusion and Climate Impact
The hypothesis was supported: fixing internal code errors and upgrading process parameterizations produced a stable, energy-conserving Earth System Model with superior simulation fidelity. Crucially, the discovery that climate sensitivity accelerates under high greenhouse gas concentrations—increasingly driven by cloud and moisture feedbacks as planetary emissivity drops—demonstrates that high-emission scenarios risk triggering much stronger warming than simple linear projections suggest. MPI-ESM1.2 provides a robust, reliable tool to support CMIP6 and guide global climate policy.
Full Citation
Mauritsen, T., Bader, J., Becker, T., Behrens, J., Bittner, M., Brokopf, R., Brovkin, V., Claussen, M., Crueger, T., Esch, M., Fast, I., Fiedler, S., Fläschner, D., Gayler, V., Giorgetta, M., Goll, D. S., Haak, H., Hagemann, S., Hedemann, C., … Roeckner, E. (2019). Developments in the MPI-M Earth System Model version 1.2 (MPI-ESM1.2) and its response to increasing CO2. Journal of Advances in Modeling Earth Systems, 11(4), 998–1038. https://doi.org/10.1029/2018MS001400