The AWI Earth System Model (AWI-ESM)

Earth System Components and Feedbacks

The AWI Earth System Model (AWI-ESM) is an extension of the AWI-CM (Sidorenko et al., 2015) for earth system modelling. It is suitable for application at background climates where, among other earth system components, vegetation patterns differ from the modern state. To this end the land surface scheme is enabled to compute vegetation dynamics and the related climate feedbacks. Furthermore, we constantly extend the model suite with further dynamic representations of components of the coupled Earth-Climate system, as, e.g., the ice sheets, water isotopes and the radiocarbon cycle.

The modular framework

The AWI-ESM consists of the atmospheric component ECHAM6, the ocean-sea ice component FESOM, inland ice model PISM, carbon models JSBACH for land and REcoM in the ocean, and in the future the solid earth model VILMA. Our model strategy is a result of the work on building a modular framework (, which ensures that developments can be reused in other contexts even after already announced end of life of different model components (e.g. ECHAM). In the framework of projects, the AWI-ESM is extended with dynamic representations of further earth system components that were before not available in the modelling framework.

AWI-ESM-1 is based on the finite element formulation of the Finite Element Sea ice-Ocean Model (FESOM), version 1.4, and AWI-ESM-2 is based on AWI-CM with the finite-volume formulation of FESOM, (FESOM2.0; Danilov et al. 2017; Sidorenko et al., 2019). One advantage of the FESOM2.0 finite volume approach in the coupled model system is strongly increased simulation throughput in comparison to the finite element approach employed by FESOM1.4 (Koldunov et al., 2019). This characteristic enables the study of time scales in paleoclimate to an extent that has not been possible based on the AWI-ESM-1.

From the viewpoint of general model structure and physics, the AWI-ESM benefits from the expertise in developing and employing AWI-CM. As in AWI-CM, the sub-models FESOM and ECHAM6 are coupled via the OASIS3-MCT coupler (Fig. 1). Improvements implemented in either AWI-CM or AWI-ESM will also be available in the respective other model version due to the source code repository being shared between both model variants. Consequently, advantages of AWI-CM over traditional modelling approaches with conventional model grids, outlined on the AWI-CM website, also apply to AWI-ESM.

Long-term future and Paleoclimates 

 The motivation to use AWI-ESM is manyfold.

  1. We need high spatial resolution in the ocean in dynamically interesting regions, like coasts, high latitudes, tropical wave guide, and even ice shelf cavities (e.g. Ionita et al., 2016; Timmermann and Goeller, 2017; Sein et al., 2018; Naughten et al., 2018). Recent developments have improved the computational efficiency and scalability of unstructured-mesh approaches on high-performance computing systems considerably (Danilov et al., 2017; Koldunov et al., 2019). Given the present high-performance computing power, and the efficient and parallelized model code of FESOM, it is now possible to use our innovative model framework for transient simulations.
  2. Secondly, building upon recent analyses shedding light on the processes that have determined the large-scale ocean circulation and climate feedbacks for the Holocene and deglacial climate (Lohmann et al. 2013; Zhang et al., 2014, 2017; Wassenburg et al., 2016), we need a model set-up with locally high-resolution meshes aiming to analyse the connection between the reconstructed signals and past climate events. For proper data-model comparisons we need a model where the data are critical. Since the overwhelming part of marine records is retrieved from continental margins or ocean ridges, a high spatial model resolution is needed around the relevant locations, which makes traditional ocean-climate models with their uniform meshes impractical. Building upon an unstructured mesh approach it is possible to zoom into regions of interest, while keeping the resolution sufficiently low in other areas towards retaining the ability to run the model climate system into quasi-equilibrium. Recently, we have shown that a multi-scale concept is applicable to paleoclimate modelling (Shi and Lohmann, 2016; Shi et al., 2020).
  3. On long time scales, such as glacial-interglacial, we can study the marine carbon-isotope record by means of the sophisticated marine biogeochemistry model REcoM (e.g., Schourup-Kristensen et al., 2014, 2018). In contrast to most other marine carbon cycle models, REcoM does not rely on fixed Redfield ratios for organic soft tissue. Instead, the ratios of C:N and C:Chl in phytoplankton are calculated as a response to light, temperature and nutrient supply, which allows for assessing potential shifts in marine autotroph stochiometry. REcoM considers dissolved inorganic carbon and alkalinity for the carbonate system, the macronutrients dissolved inorganic nitrogen (DIN), and silicic acid as well as the trace metal iron. REcoM2 has two phytoplankton classes, nanophytoplankton (with implicit representation of calcifiers) and diatoms. The intracellular stoichiometry of C:N:Si:Chl pools for diatoms and C:N:CaCO3:Chl pools for nanophytoplankton is allowed to respond dynamically to changes in environmental conditions. The intracellular iron pool is a function of the intracellular nitrogen concentration (fixed Fe:N), as iron is physiologically linked to enzyme formation, especially in the photosynthetic electron transport chain. Dead organic matter is transferred to detritus by aggregation and grazing, the latter by one zooplankton class. The sinking and advection of detritus is represented explicitly.
  4. Finally, we incorporate isotope modules into the ESM components which makes a direct comparison with paleoclimate data possible. Examples are water isotopes in ECHAM6 and radiocarbon isotopes in FESOM2 (e.g., Cauquoin et al., 2019; Lohmann et al., 2020).

The modular structure of the ESM tools ensures that developments can be reused in several contexts. In the future, the AWI-ESM will be extended with further earth system components and modules that were before not available in the modelling framework.

The Paleoclimate Dynamics' model version

For paleo and long-term future climate simulations the focus with the AWI-ESM is mostly on lower resolutions than employed in typical applications with the AWI-CM. The current standard resolution of AWI-ESM is "LR" (Fig. 2), where the ECHAM6 is set up based on the T63 grid (1.88x1.88°), and where FESOM employs the COREII mesh (~127,000 nodes for a modern land sea mask) or one of the paleo-derivates of COREII. Other meshes may be used for specific applications.

AWI-ESM-1 and AWI-ESM-2 include the land surface scheme JSBACH with interactive vegetation dynamics. This ensures that climate and vegetation are consistent with each other, which is a precondition for realistic simulation of climates that are characterized by vegetation distribution that differs from the modern state.

Recently, the cryosphere is included into the modelling framework by means of the Parallel Ice Sheet Model PISM (version 1.2). This flavour of AWI-ESM improves climate simulations at long time scales by supplying dynamic representations of Greenland and Antarctic Ice Sheets (Gierz et al., 2020; Ackermann et al., 2020). It also enables consideration of further ice sheets (e.g. Laurentide), that played a role during glacial-interglacial cycles of the Pleistocene. Availability of the explicit ice sheet - climate dynamics removes the need to implement past (or future) extent and geometry of ice sheets into a climate simulation via fixed boundary conditions; it also enables the representation of ice sheet - climate feedbacks. Important for realistic simulation of ice sheets is consideration of detailed knowledge of the lower boundary condition at the base of the ice sheet. To this end we have developed dedicated data sets for North America, Greenland, and surrounding areas (Gowan et al., 2019). Our most recent coupling with ice sheets uses the diurnal energy balance approach (Krebs-Kanzow et al., 2018a,b).

Beyond application in the framework of glacial-interglacial cycles (Fig. 3), the AWI-ESM will enrich the toolbox to study climate dynamics over multiple time scales - covering periods, where plate tectonics are of relevance, to the Anthropocene, where man-made impact on climate is the predominant driver (Fig. 4). Consequently, application of the AWI-ESM is suitable for climate research across time scales.

Nomenclature of AWI-ESM versions:

AWI-ESM-1.1-LR: as AWI-CM-1-1-LR, but with interactive vegetation

AWI-ESM-2.1-LR: as AWI-CM-2-1-LR, but with interactive vegetation

AWI-ESM-1.2-LR: as AWI-ESM-1.1-LR, but with interactive ice sheets

AWI-ESM-2.2-LR: as AWI-ESM-2.1-LR, but with interactive ice sheets

Participating in Model Intercomparison Projects

We currently apply the AWI-ESM-1.1-LR and AWI-ESM-2.1-LR in the framework of CMIP6 and PMIP4. The model is employed to study the dynamics of past climate states, in particular those of the Holocene (Shi et al., 2020; Brierley et al., 2020), the Last Glacial Maximum (21 thousand years Before Present (kyr BP); Lohmann et al., 2020; Kageyama et al., 2020a), and the Last Interglacial (127 kyr BP; Otto-Bliesner et al., 2020; Kageyama et al., 2020b). Beyond the scientific interest, this effort also aims at evaluating the model beyond the recent observational period. While highly resolved, but short term, instrumental records of weather and climate are available for grading the model's performance during recent decades, the AWI-ESM also must be tested in a paleoclimatic framework towards evaluating the accuracy of its projections for future climate states that differ from the modern one. One of the many research topics in the framework of our contribution to PMIP4 is the study of the dependency of internal variability on the background climate state (for example ENSO; Brown et al., 2020).

Data prepared for MIPs is available via the Earth System Grid Federation ( The following data sets are available at the end of 2020:

  1.     Danek, C., Shi, X., Stepanek, C. Yang, H., Barbi, D., Hegewald, J., Lohmann, G. (2020). AWI-ESM1.1LR model output prepared for CMIP6. Earth System Grid Federation. .
  2.     Danek, C., Shi, X., Stepanek, C. Yang, H., Barbi, D., Hegewald, J., Lohmann, G. (2020). AWI-ESM1.1LR model output prepared for CMIP6 PI Control. Earth System Grid Federation.
  3.     Danek, C., Shi, X., Stepanek, C. Yang, H., Barbi, D., Hegewald, J., Lohmann, G. (2020). AWI-ESM1.1LR model output prepared for CMIP6 historical. Earth System Grid Federation.
  4.     Shi, X., Yang, H., Danek, C., Lohmann, G. (2020). AWI-ESM1.1LR model output prepared for CMIP6 PMIP. Earth System Grid Federation.
  5.     Shi, X., Yang, H., Danek, C., Lohmann, G. (2020). AWI AWI-ESM1.1LR model output prepared for CMIP6 PMIP lgm. Earth System Grid Federation.
  6.     Shi, X., Yang, H., Danek, C., Lohmann, G. (2020). AWI AWI-ESM1.1LR model output prepared for CMIP6 PMIP lig127k. Earth System Grid Federation.
  7.     Shi, X., Yang, H., Danek, C., Lohmann, G. (2020). AWI AWI-ESM1.1LR model output prepared for CMIP6 PMIP midHolocene. Earth System Grid Federation.

Current Projects

We are involved into several research projects. Here, we mention one major project, that covers several aspects of Earth System Model development.

Our goal in the BMBF funded project PalMod ( is to understand and quantify feedback between climate components during glacial-interglacial cycles. A fundamental question is how to obtain a suitable glacial climate state and how to simulate deglaciation and sea level rise that followed. We investigate mechanisms for abrupt climate changes, test the influence of deglacial meltwater on ocean circulation, and study the stability of ice shelves and ice sheets. In PalMod, different models are used to study the dependence of abrupt climate and carbon changes on solar radiation and greenhouse gases in different configurations. At AWI we work with the AWI-ESM. Our model strategy is a result of the work on building a modular framework (, which ensures that PalMod developments can be reused in other contexts. Another focus is on data-model comparisons.

Code availability

The source code of the FESOM2 model is available to the public via GitHub. The ECHAM6 model is distributed by the Max Planck Institute for Meteorology in Hamburg and must be requested directly from there. 


Ackermann, L., Gierz, P., and Lohmann, G., 2020. AMOC recovery in a multi-centennial scenario using a coupled atmosphere-ocean-ice sheet model. Geophysical Research Letters, e2019GL086810.

Brierley, C. M., Zhao, A., Harrison, S. P., Braconnot, P., Williams, C. J. R., Thornalley, D. J. R., Shi, X., Peterschmitt, J.-Y., Ohgaito, R., Kaufman, D. S., Kageyama, M., Hargreaves, J. C., Erb, M. P., Emile-Geay, J., D'Agostino, R., Chandan, D., Carré, M., Bartlein, P., Zheng, W., Zhang, Z., Zhang, Q., Yang, H., Volodin, E. M., Tomas, R. A., Routson, C., Peltier, W. R., Otto-Bliesner, B., Morozova, P. A., McKay, N. P., Lohmann, G., Legrande, A. N., Guo, C., Cao, J., Brady, E., Annan, J. D., and Abe-Ouchi, A., 2020. Large-scale features and evaluation of the PMIP4-CMIP6 midHolocene simulations. Climate of the Past, 16, 1847–1872,, 2020.

Brown, J. R., Brierley, C. M., An, S.-I., Guarino, M.-V., Stevenson, S., Williams, C. J. R., Zhang, Q., Zhao, A., Braconnot, P., Brady, E. C., Chandan, D., D'Agostino, R., Guo, C., LeGrande, A. N., Lohmann, G., Morozova, P. A., Ohgaito, R., O'ishi, R., Otto-Bliesner, B., Peltier, W. R., Shi, X., Sime, L., Volodin, E. M., Zhang, Z., and Zheng, W., 2020. Comparison of past and future simulations of ENSO in CMIP5/PMIP3 and CMIP6/PMIP4 models. Climate of the Past, 16, 1777–1805,, 2020.

Butzin, M., Heaton, T. J., Köhler, P., and Lohmann, G., 2020. A short note on marine reservoir age simulations used in IntCal20. Radiocarbon, 1-7.

Cauquoin, A., Werner, M., and Lohmann, G., 2019. Water isotopes – climate relationships for the mid-Holocene and preindustrial period simulated with an isotope-enabled version of MPI-ESM. Climate of the Past15, pp.1913-1937.

Danilov, S., Sidorenko, D. Wang, Q., and Jung, T., 2017. The Finite-volumE Sea ice–Ocean Model (FESOM2). Geoscientific Model Development10, pp.765-789.

Gierz, P., Ackermann, L., Rodehacke, C. B., Krebs-Kanzow, U., Stepanek, C., Barbi, D., and Lohmann, G., 2020: Simulating interactive ice sheets in the multi-resolution AWI-ESM 1.2: A case study using SCOPE 1.0, Geoscientific Model Development Discussions,, in review.

Gowan, E. J., Niu, L., Knorr, G., and Lohmann, G., 2019. Geology datasets in North America, Greenland and surrounding areas for use with ice sheet models.  Earth System Science Data, 11,  pp.375-391.

Hinck, S., Gowan, E. J., and Lohmann, G. 2020. LakeCC: a tool for efficiently identifying lake basins with application to paleo-geographic reconstructions of North America.  Journal of Quaternary Science, 35(3), pp.422-432.

Ionita, M., Scholz, P., Lohmann, G., Dima, M., and Prange, M., 2016. Linkages between atmospheric blocking, sea ice export through Fram Strait and the Atlantic Meridional Overturning Circulation. Scientific Reports6(32881).

Kageyama, M., Harrison, S. P., Kapsch, M.-L., Löfverström, M., Lora, J. M., Mikolajewicz, U., Sherriff-Tadano, S., Vadsaria, T., Abe-Ouchi, A., Bouttes, N., Chandan, D., LeGrande, A. N., Lhardy, F., Lohmann, G., Morozova, P. A., Ohgaito, R., Peltier, W. R., Quiquet, A., Roche, D. M., Shi, X., Schmittner, A., Tierney, J. E., and Volodin, E., 2020a. The PMIP4-CMIP6 Last Glacial Maximum experiments: preliminary results and comparison with the PMIP3-CMIP5 simulations. Climate of the Past Discussions, in review.

Kageyama, M., Sime, L. C., Sicard, M., Guarino, M.-V., de Vernal, A., Schroeder, D., Stein, R., Malmierca-Vallet, I., Abe-Ouchi, A., Bitz, C., Braconnot, P., Brady, E., Chamberlain, M. A., Feltham, D., Guo, C., Lohmann, G., Meissner, K., Menviel, L., Morozova, P., Nisancioglu, K. H., Otto-Bliesner, B., O'ishi, R., Sherriff-Tadano, S., Stroeve, J., Shi, X., Sun, B., Volodin, E., Yeung, N., Zhang, Q., Zhang, Z., and Ziehn, T., 2020b. A multi-model CMIP6 study of Arctic sea ice at 127 ka: Sea ice data compilation and model differences. Climate of the Past Discussions, accepted.

Koldunov, N. V., Aizinger, V., Rakowsky, N., Scholz, P., Sidorenko, D., Danilov, S., and Jung, T., 2019. Scalability and some optimization of the Finite-volumE Sea ice-Ocean Model, Version 2.0 (FESOM2). Geoscientific Model Development12, pp.3991-4012.

Krebs-Kanzow, U., Gierz, P., and Lohmann, G., 2018a. Estimating Greenland surface melt is hampered by melt induced dampening of temperature variability. Journal of Glaciology64(244), pp.227-235.

Krebs-Kanzow, U., Gierz, P., and Lohmann, G., 2018b. Brief communication: An ice surface melt scheme including the diurnal cycle of solar radiation. The Cryosphere12, pp.3923-3930.

Lohmann, G., Pfeiffer, M., Laepple, T., Leduc, G., and Kim, J.-H., 2013. A model-data comparison of the Holocene global sea surface temperature evolution. Climate of the Past9, pp.1807-1839.

Lohmann, G., Butzin, M., Eissner, N., Shi, X., and Stepanek, C., 2020. Abrupt climate and weather changes across timescales.  Paleoceanography and Paleoclimatology35,  e2019PA003782.

Naughten, K. A., Meissner, K. J., Galton-Fenzi, B. K., England, M. H., Timmermann, R., and Hellmer, H. H., 2018. Future Projections of Antarctic Ice Shelf Melting Based on CMIP5 Scenarios. Journal of Climate31, pp.5243-5261.

Otto-Bliesner, B. L., Brady, E. C., Zhao, A., Brierley, C., Axford, Y., Capron, E., Govin, A., Hoffman, J., Isaacs, E., Kageyama, M., Scussolini, P., Tzedakis, P. C., Williams, C., Wolff, E., Abe-Ouchi, A., Braconnot, P., Ramos Buarque, S., Cao, J., de Vernal, A., Guarino, M. V., Guo, C., LeGrande, A. N., Lohmann, G., Meissner, K., Menviel, L., Nisancioglu, K., O'ishi, R., Salas Y Melia, D., Shi, X., Sicard, M., Sime, L., Tomas, R., Volodin, E., Yeung, N., Zhang, Q., Zhang, Z., and Zheng, W., 2020. Large-scale features of Last Interglacial climate: Results from evaluating the lig127k simulations for CMIP6-PMIP4. Climate of the Past Discussions, accepted.

Renoult, M., Annan, J. D., Hargreaves, J. C., Sagoo, N., Flynn, C., Kapsch, M.-L., Li, Q., Lohmann, G., Mikolajewicz, U., Ohgaito, R., Shi, X., Zhang, Q., and Mauritsen, T. 2020. A Bayesian framework for emergent constraints: case studies of climate sensitivity with PMIP. Climate of the Past, 16, 1715–1735,, 2020.

Schourup-Kristensen, V., Sidorenko, D., Wolf-Gladrow, D. A., and Völker, C., 2014. A skill assessment of the biogeochemical model REcoM2 coupled to the Finite Element Sea Ice-Ocean Model (FESOM 1.3). Geoscientific Model Development7(6), pp.2769-2802.

Schourup-Kristensen, V., Wekerle, C., Wolf-Gladrow, D., and Völker, C., 2018. Arctic Ocean biogeochemistry in the high resolution FESOM 1.4-REcoM2 model. Progress in Oceanography168, pp.65-81.

Sein, D. V., Koldunov, N. V., Danilov, S., Sidorenko, D., Wekerle, C., Cabos, W., Rackow T., Scholz P., Semmler T., Wang Q., and Jung T., 2018. The relative influence of atmospheric and oceanic model resolution on the circulation of the North Atlantic Ocean in a coupled climate model. Journal in Advances in Modeling Earth Systems10(8),  pp.2026-2041.

Shi, X. and Lohmann, G., 2016. Simulated response of the mid-Holocene Atlantic Meridional Overturning Circulation in ECHAM6-FESOM/MPIOM. Journal of Geophysical Research - Oceans121(8), pp.6444-6469.

Shi, X., Lohmann, G., Sidorenko, D., and Yang, H., 2020. Early-Holocene simulations using different forcings and resolutions in AWI-ESM. The Holocene30(7), pp.996-1015.

Sidorenko, D., Rackow, T., Jung, T., Semmler, T., Barbi ,D., Danilov, S., Dethloff ,K., Dorn, W., Fieg, K., Goessling, H. F., Handorf, D., Harig, S.,  Hiller, W.,  Juricke, S., Losch, M.,  Schröter, J.,  Sein,  D. V., and Wang, Q., 2015. Towards multi-resolution global climate modeling with ECHAM6–FESOM. Part I: model formulation and mean climate. Climate Dynamics44(3-4), pp.757-780.

Sidorenko, D., Goessling, H. F., Koldunov, N., Scholz, P., Danilov, S., Barbi ,D., Cabos, W., Gurses, O., Harig, S., Hinrichs, C., Juricke, S., Lohmann, G., Losch, M., Mu, L., Rackow, T., Rakowsky, N., Sein, D., Semmler, T., Shi, X., Stepanek, C., Streffing, J., Wang, Q., Wekerle, C., Yang, H., and Jung. T., 2019. Evaluation of FESOM2.0 coupled to ECHAM6.3: Preindustrial and HighResMIP simulations. Journal of Advances in Modeling Earth Systems11, pp.3794-3815.

Timmermann, R. and Goeller, S., 2017. Response to Filchner–Ronne Ice Shelf cavity warming in a coupled ocean–ice sheet model – Part 1: The ocean perspective. Ocean Science13, pp.765-776.

Wassenburg, J. A., Dietrich, S., Fietzke, J., Fohlmeister, J., Jochum, K. P., Scholz, D., Richter, D. K., Sabaoui, A., Spötl, C., Lohmann, G., Andreae, M. O., and Immenhauser, A., 2016. Major reorganization of the North Atlantic Oscillation during Early Holocene deglaciation. Nature Geoscience9, pp.602-605.

Yang, H., Lohmann, G., Krebs-Kanzow, U., Ionita, M., Shi, X., Sidorenko, D., Gong, X., Chen, X., and Gowan, E. J., 2020. Poleward shift of the major ocean gyres detected in a warming climate. Geophysical Research Letters47, e2019GL085868.

Yang, H., Lohmann, G., Shi, X., Gowan, E. J., and Liu, J., and Wang, Q., 2020. Tropical expansion driven by poleward advancing subtropical fronts.  Journal of Geophysical Research - Atmospheres125, e2020JD033158.

Zhang, X., Lohmann, G., Knorr, G., and Purcell, C., 2014. Abrupt glacial climate shifts controlled by ice sheet changes. Nature512, pp.290-294.

Zhang, X., Knorr, G., Lohmann, G., and Barker, S., 2017. Abrupt North Atlantic circulation changes in response to gradual CO2 forcing in a glacial climate state. Nature Geosciences10, pp.518-523.