A-WP1 - A-Coordination2

Project aims

A-Coordination2 is concerned with the coordination of Module A and with performing research within the module. The main objectives of the coordination activity is to ensure close collaborations and linkages between different WPs within Module A, and between Module A and other modules, especially Module D, thereby assuring the immediate relevance of results obtained in Module A for the development of a final MiKlip prediction system. The main goals of A-Coordination2 research activities are:

  • Deliver initial conditions (GECCO2) for MiKlip II, making ocean-only settings practical for a Deutscher Wetterdienst (DWD) operational use.
  • Develop a mode-initialization approach to improve climate predictions by reducing initialization shocks and initializing climate modes (filtering the modes that cannot be simulated by the model and reducing unbalanced states).
  • Investigate initialization procedures involving flux corrections of buoyancy and momentum fluxes to improve the forecast skill in the equatorial Pacific.
  • Quantify the impact of model improvements through parameter estimation on the predictive skill of a decadal prediction system.

Project structure

The objectives of A-Coordination2 are subdivided into three WPs: 

WP1.1: Estimating initial conditions and developing a mode-initialization method.
WP1.2: Testing the flux correction method.
WP1.3: Developing a coupled data assimilation approach.

Tasks of the project

WP1.1: Creating initial conditions through the global GECCO approach will continue, while taking into account all available ocean and sea ice information. Testing a new climate mode initialization method is subdivided into the following tasks: estimating climate modes for MPI-ESM using empirical orthogonal function analysis; filtering out components from the synthesis/reanalysis that do not correspond to climate modes represented by MPI-ESM which may thus lead to unbalanced states; carrying out the assimilation run to produce the initial conditions for the initialized decadal hindcasts that will be performed and compared with the hindcasts from Module D.

Fig.1. Illustration for the climate modes initialization method: an initial state of the ocean for a decadal climate prediction is projected onto the “climates modes” derived from the prediction system. The relevance of different contributions like the grid size or ocean density signature, etc. for climate modes is controlled through weighting coefficients which are applied prior to singular vector decomposition from which the empirical orthogonal functions (EOF) are derived.

WP1.2: For testing the full state initialization with a momentum and buoyancy flux correction scheme, we plan the following: construction of flux correction terms from repeating the historical run with nudging of SST, SSS to the values from reanalyses, following the strategy described by Polkova et al. (2014a,b); carrying out the assimilation run and the initialized hindcasts with employing the diagnosed flux correction terms. Finally, we plan to analyze the predictive skill and make a comparison with the skill for the Prototype hindcasts (GECCO2, full field initialization) using MiKlip Central Evaluation System (CES).     

Fig.2. Illustration on implementing the flux-correction method. Adapted from Sausen et al., 1988.

WP1.3: We will perform assimilation experiments with the CESAM model to test the efficiency of the approach described by Abarbanel et al. (2010) to improve model parameters by assimilating data in coupled climate models. The objective is twofold; first, we will test how the improved consistency of the model climate with the data and with model-consistent initial conditions influences the forecast skill of a coupled climate model on various space and time scales. Second, we aim at producing a coupled synthesis that can provide model-consistent initial conditions for decadal predictions. Although the synthesis may be at first build on a simpler and coarser resolution climate model than what is currently state-of-the-art and used as the MiKlip prediction system, hindcasts initialized with this optimized model fields will provide insight into how model-consistent initialization affects the forecast skill.

Deliverables

The following deliverables are planned for A-Coordination2:

  • Initial conditions (GECCO2) for MiKlip II, making ocean-only settings practical for the operational use in Deutscher Wetterdienst (DWD)
  • Mode-initialization approach and initialization procedure involving flux corrections
  • Coupled data assimilation approach

Progress so far

WP1.1: Updates of GECCO2 are being performed on a regular basis to provide initial conditions for MiKlip II. The updates of GECCO2 are available through http://icdc.zmaw.de/1/daten/reanalysis-ocean/gecco2.html.

We designed a climate modes-initialization method alternative to the traditional anomaly initialization used in MiKlip Baseline-1 and PreOp-LR to tackle the problem of initialization shocks due to model inconsistencies. ORAS4 reanalysis (temperature and salinity anomalies) provides initial conditions which are filtered before use. In this exercise, the anomalies of the non-native ocean reanalysis, which might not be compatible with the dynamics of the prediction system, is filtered out by projecting the reanalysis anomalies on the truncated set of 3D-EOFs estimated from an ensemble of historical simulations. The reconstructed ORAS4 state retains 66% of explained variance as compared to the original product. The variance explained is sensitive to the choice of weighting and normalization during the multivariate EOF analysis. Initialized with the filtered state, hindcasts show a slight (insignificant) reduction of surface temperature skill in the first lead year but a more persistent skill for later lead years (yr2-5). Thus for yr2-5, modes initialization shows improvement of skill for the equatorial and eastern North Pacific as compared to PreOp-LR. However, comparing this skill to those of historical simulations suggests that neither modes-initialization nor PreOp-LR beat the surface temperature skill of the historical simulations in those regions. Currently, prediction skill for other climate variables is investigated. We also consider alternative methods to improve the filtering method.

WP1.3: The coupled data assimilation model that we are using is the coupled adjoint system CESAM (CEN Earth System Assimilation Model, https://www.cen.uni-hamburg.de/en/research/cen-models/cesam.html). Various experiments were performed with different choices of control parameters and different model configurations: low resolution (4 degrees resolution and 15 depth levels in the ocean, and T21 and 10 levels in the atmosphere) and medium resolution (1 degree resolution and 23 levels in the ocean, and T42 and 10 levels in the atmosphere), forward 500yrs long integrations of the climate model and are used for: (i) evaluation of CESAM and the quality of the coupled model forward runs (this analysis focuses on understanding how parameterization of some processes in the atmosphere and ocean affect the simulated mean-state climate and climate variability), and (ii) generation of initial conditions for sensitivities studies of near-surface air temperature over Europe to changes in sea surface temperature in different regions of global ocean (e.g., the North Atlantic Ocean).

References

  • Abarbanel, Henry DI, Mark Kostuk, and William Whartenby, 2010: Data assimilation with regularized nonlinear instabilities. Quarterly Journal of the Royal Meteorological Society, 136.648 : 769-783.
  • Köhl A., 2015: Evaluation of the GECCO2 Ocean Synthesis: Transports of Volume, Heat and Freshwater in the Atlantic. Quart. J. Roy. Meteor. Soc., 141(686), 166-181. doi:10.1002/qj.2347.
  • Polkova I., A. Köhl, and D. Stammer, 2014a: Impact of initialization procedures on the predictive skill of a coupled ocean–atmosphere model. Clim. Dyn., 42, 3151-3169 doi:10.1007/s00382-013-1969.
  • Polkova, I., 2014b: Impact of initialization procedures on the predictive skill of a coupled ocean–atmosphere mode and related mechanisms for predictability, thesis. Reports on Earth System Science, 146.
  • Sausen R., Barthel K. and Hasselmann K,1988: Coupled ocean-atmosphere models with flux correction. Climate Dynamics, 2:145-163.

Contact

Institut für Meereskunde,Universität Hamburg
Prof. Dr. Detlef Stammer
detlef.stammer(at)nospamuni-hamburg.de
+49 40 42838-5052

Predictive Skill for Regional Interannual Steric Sea Level and Mechanisms for Predictability

2015 - J. Climate, Vol. 28(18), pp. 7407-7419

Polkova, I. | A. Köhl, and D. Stammer

Impact of initialization procedures on the predictive skill of a coupled ocean–atmosphere model

2014 - Climate Dynamics, Vol. 42(11), pp. 3151-3169

Polkova, I. | A. Köhl, and D. Stammer

Evaluation of the GECCO2 ocean synthesis: transports of volume, heat and freshwater in the Atlantic

2014 - Quart. J. Roy. Meteor. Soc., Vol. 141 (686), pp. 166–181

Köhl, A.

Testing variational estimation of process parameters and initial conditions of an earth system model

2014 - Tellus A, Vol. 66(2260)

Blessing, S. | T. Kaminski, F. Lunkeit, I. Matei, R. Giering, A. Köhl, M. Scholze, P. Herrmann, K. Fraedrich, and D. Stammer