Project aims

Development and improvement of the central prediction and evaluation system

Project structure

WP1.1: Coordinating and evaluating ensemble decadal hindcasts and forecasts
WP1.2: Performing decadal hindcasts, forecasts, and post-processing
WP1.3: Pre-operational ensemble prediction system infrastructure
WP1.4: Developing the central evaluation system

Tasks of the project

Similar to the first MiKlip phase, the scientific synthesis of MiKlip II has both prediction research and infrastructure components. The decadal climate prediction system is improved by incorporating the research progress across all of MiKlip II. A central step is the doubling of the model resolution of the MPI Earth System Modell (MPI-ESM) from LR/MR in MiKlip to HR in MiKlip II. Two complete sets of hindcasts and forecasts will be performed during the development stages 4 and 5 (DS4, DS5). Hindcasts and forecasts are planned to contribute to the WCRP CMIP6.

The central evaluation system is improved by incorporation of evaluation and verification procedures developed by individual projects across MiKlip II. The evaluation system is designed to serve well-defined front ends for operational hindcast and forecast evaluation.

Both the MPI-M and the DWD are working toward replacing their current models to ones based on the new ICON framework developed jointly between them. DWD has just switched to ICON in an atmosphere-only version as its operational weather-prediction model, and it will be crucial for efficient use of infrastructure that, once the DWD takes over decadal prediction operationally, the MiKlip system is also based on ICON. The MPI-M will switch to ICON as soon as this is scientifically feasible; first test simulations with a coupled model in a realistic setting were planned for the summer of 2015. Hence, MiKlip II must prepare for the situation that the MPI-ESM1 will no longer be supported by the MPI-M. Together this means that preparing for the use of ICON must be part of preparing the prediction system for operational use.


M1.1: Implementation of module recommendations from DS3 MiKlip
M1.2: Provision of a pre-operational set of hindcasts and forecasts with the global prediction system
M1.3: Analysis of a pre-operational set of hindcasts and forecasts with the global prediction system
M1.4: Development of visualization front ends for pre-operational hindcasts and forecasts for the regional and global prediction systems
M1.5: Implementation of nudging in to ICON-ocean and ocean-only assimilation
M1.6: Implementation of module recommendations from DS4
M1.7: Provision of operational set of hindcasts and forecasts with the global prediction system
M1.8: Analysis of a pre-operational set of hindcasts and forecasts with the global prediction system
M1.9: Application of front-end for operational hindcasts and forecasts of the global prediction system
M1.10: Synthesis of CES verification plug-ins of Module E for operational use
M1.11: Coupled ICON assimilation with ocean-only nudging and hindcas

Correlation of surface temperature from the pre-operational hindcasts with observations (HadCRUT4). Red indicates a positive correlation.

Progress so far

MiKlip and MPI-M have developed a model version of MPI Earth System Model (MPI-ESM1) with higher resolution (HR). The model is used to provide a pre-operational set of hindcasts of 5 ensemble members with CMIP6 forcing for DS4. This system is currently analysed with the central evaluation system (Fig. 1).

The Central Evaluation System (CES - www-miklip.dkrz.de) has been extended by the forecast and hindcast frontend to give a brief overview to the MiKlip prediction systems and its development stages. The new result-browser gives the scientists a better way of connecting each other and find others' research.

The CES and its forecast relevant plug-ins are already prepared for the transition to the German Weather Service. The last pieces are put together for the last and official update of the MiKlip-CES at DWD.


HadCRUT4: Morice, C. P., J. J. Kennedy, N. A. Rayner, and P. D. Jones (2012), Quantifying uncertainties in global and regional temperature change using an ensemble of observational estimates: The HadCRUT4 dataset, J. Geophys. Res., 117, D08101, doi:10.1029/2011JD017187


Max-Planck-Institute for Meteorology
Dr. Sebastian Hettrich
Prof. Dr. Jochem Marotzke
Dr. Wolfgang Müller
Dr. Holger Pohlmann

Freie Universität Berlin
Christopher Kadow
Sebastian Illing
Prof. Dr. Ulrich Cubasch
Prof. Dr. Uwe Ulbrich
Prof. Dr. Henning Rust

On skillful decadal predictions of the subpolar North Atlantic

2019 - Meteorologische Zeitschrift Vol. 28 No. 5 (2019), p. 417 - 428

Höschel, I. | Illing, S., Grieger, J., Ulbrich, U., Cubasch, U.

Realistic Quasi‐Biennial Oscillation Variability in Historical and Decadal Hindcast Simulations Using CMIP6 Forcing

2019 - Geophysical Research Letters, 46.

Pohlmann, H. | Müller, W.A., Bittner, M., Hettrich, S., Modali, K., Pankatz, K., Marotzke, J.

Decadal Predictions of the Probability of Occurrence for Warm Summer Temperature Extremes

2019 - Geophys. Res. Lett.

Borchert, L.F. | Pohlmann, H., Baehr, J., Neddermann, N.-C., Suarez-Gutierrez, L., Müller, W.A.

Forecast-oriented assessment of decadal hindcast skill for North Atlantic SST

2019 - Geophysical Research Letters, 46, 11444-11454

Borchert, L.F. | Düsterhus, A., Brune, S., Müller, W.A., Baehr, J.

Skill and added value of the MiKlip regional decadal prediction system for temperature over Europe

2019 - Tellus A: Dynamic Meteorology and Oceanography, 71:1, 1-19

Feldmann, H. | Pinto, J.G., Laube, N., Uhlig, M., Moemken, J., Pasternack, A., Früh, B., Pohlmann, H., Kottmeier, C.