Search

Abstract Five initialization and ensemble generation methods are investigated with respect to their impact on the prediction skill of the German decadal prediction system “Mittelfristige Klimaprognose” (MiKlip). Among the tested methods, three tackle aspects of model‐consistent initialization using the ensemble Kalman filter, the filtered anomaly initialization, and the initialization method by partially coupled spin‐up (MODINI). The remaining two methods alter the ensemble generation: the ensemble dispersion filter corrects each ensemble member with the ensemble mean during model integration. And the bred vectors perturb the climate state using the fastest growing modes. The new methods are compared against the latest MiKlip system in the low‐resolution configuration (Preop‐LR), which uses lagging the climate state by a few days for ensemble generation and nudging toward ocean and atmosphere reanalyses for initialization. Results show that the tested methods provide an added value for the prediction skill as compared to Preop‐LR in that they improve prediction skill over the eastern and central Pacific and different regions in the North Atlantic Ocean. In this respect, the ensemble Kalman filter and filtered anomaly initialization show the most distinct improvements over Preop‐LR for surface temperatures and upper ocean heat content, followed by the bred vectors, the ensemble dispersion filter, and MODINI. However, no single method exists that is superior to the others with respect to all metrics considered. In particular, all methods affect the Atlantic Meridional Overturning Circulation in different ways, both with respect to the basin‐wide long‐term mean and variability and with respect to the temporal evolution at the 26° N latitude.

The long term variability and its predictability of the monthly mean oceanic surface net freshwater fluxes is compared in a set of retrospective predictions. All are using the same model setup, and only differ in the implemented ocean initialisation method and ensemble generation method. The basic aim is to deduce the differences between the initialization/ensemble generation methods in view of the uncertainty of the verifying observational data sets. The analysis will give an approximation of the uncertainties of the net freshwater fluxes, which up to now appear to be one of the most uncertain products in observational data and model outputs. All ensemble generation methods are implemented into the MPI-ESM earth system model in the framework of the ongoing MiKlip project (www.fona-miklip.de). Hindcast experiments are initialised annually between 2000–2004, and from each start year 8 ensemble members are run for 10 years forward. Four different ensemble generation methods are compared: (i) a method based on the Anomaly Transform method in which the initial oceanic perturbations represent orthogonal and balanced anomaly structures in space and time and between the variables taken from a control run, (ii) one-day-lagged ocean states from the MPI-ESM-LR Baseline 1 system, (iii) one-day-lagged ocean and atmospheric states with preceding full-field nudging to re-analysis in the atmospheric and anomaly nudging in the oceanic component of the system – the Baseline MPI-ESM-LR system, (iv) an Ensemble Kalman Filter (EnKF) implemented into oceanic part of MPI-ESM, assimilating monthly subsurface oceanic temperature and salinity using the Parallel Data Assimilation Framework and full-field nudging in the atmosphere. The hindcasts are evaluated probabilistically using freshwater flux data set from NCEP-R2. On the global scale the physically motivated methods (i) and (iv) provide probabilistic hindcasts to some extent higher correlation and reliability than the lagged initialization methods (ii)/(iii) despite the large uncertainties in the verifying observations and in the simulations. We suggest similar approaches for further evaluations of other variables of decadal hindcasts systems.

Five initialization and ensemble generation methods are investigated with respect to their impact on the prediction skill of the German decadal prediction system "Mittelfristige Klimaprognose" (MiKlip). Among the tested methods, three tackle aspects of model‐consistent initialization using the ensemble Kalman filter, the filtered anomaly initialization, and the initialization method by partially coupled spin‐up (MODINI). The remaining two methods alter the ensemble generation: the ensemble dispersion filter corrects each ensemble member with the ensemble mean during model integration. And the bred vectors perturb the climate state using the fastest growing modes. The new methods are compared against the latest MiKlip system in the low‐resolution configuration (Preop‐LR), which uses lagging the climate state by a few days for ensemble generation and nudging toward ocean and atmosphere reanalyses for initialization. Results show that the tested methods provide an added value for the prediction skill as compared to Preop‐LR in that they improve prediction skill over the eastern and central Pacific and different regions in the North Atlantic Ocean. In this respect, the ensemble Kalman filter and filtered anomaly initialization show the most distinct improvements over Preop‐LR for surface temperatures and upper ocean heat content, followed by the bred vectors, the ensemble dispersion filter, and MODINI. However, no single method exists that is superior to the others with respect to all metrics considered. In particular, all methods affect the Atlantic Meridional Overturning Circulation in different ways, both with respect to the basin‐wide long‐term mean and variability and with respect to the temporal evolution at the 26° N latitude. Key Points: Five initialization and ensemble generation methods are tested with respect to their impact on the skill of a decadal prediction systemResults show that the tested methods provide an added value for the prediction skill as compared to the reference prediction systemThe study deals with dynamical consistency during initialization and ocean initial state uncertainty [ABSTRACT FROM AUTHOR]