Development and prospects of the regional MiKlip decadal prediction system over Europe: predictive skill, added value of regionalization, and ensemble size dependency

Abstract

Abstract. The current state of development and the prospects of the regional MiKlip decadal prediction system for Europe are analysed. The MiKlip regional system consists of two 10-member hindcast ensembles computed with the global coupled model MPI-ESM-LR downscaled for the European region with COSMO-CLM to a horizontal resolution of 0.22∘ (∼25 km). Prediction skills are computed for temperature, precipitation, and wind speed using E-OBS and an ERA-Interim-driven COSMO-CLM simulation as verification datasets. Focus is given to the eight European PRUDENCE regions and to lead years 1–5 after initialization. Evidence of the general potential for regional decadal predictability for all three variables is provided. For example, the initialized hindcasts outperform the uninitialized historical runs for some key regions in Europe, particularly in southern Europe. However, forecast skill is not detected in all cases, but it depends on the variable, the region, and the hindcast generation. A comparison of the downscaled hindcasts with the global MPI-ESM-LR runs reveals that the MiKlip prediction system may distinctly benefit from regionalization, in particular for parts of southern Europe and for Scandinavia. The forecast accuracy of the MiKlip ensemble is systematically enhanced when the ensemble size is increased stepwise, and 10 members is found to be suitable for decadal predictions. This result is valid for all variables and European regions in both the global and regional MiKlip ensemble. The present results are encouraging for the development of a regional decadal prediction system.

The regional MiKlip decadal prediction system for Europe: Hindcast skill for extremes and user‐oriented variables

“…• As expected, results are more mixed for precipitationbased parameters: The spatial skill patterns are more heterogeneous and the skill scores are generally lower (mean ACC for Europe up to 0.3; Figure 6). This is in line with findings by Mieruch et al (2014) and Reyers et al (2019) for mean precipitation in a sub-sample of the full-regional MiKlip ensemble. The selected indices (R10mm and RM7P95) address more extreme precipitation events and might thus be useful to users, as such events may, for example, trigger floods in large river catchments.…”

“…Thus, we aim to identify potential fields of application for regional decadal predictions. Thereby, our work extends previous regional studies like Feldmann et al (2019) and Reyers et al (2019). Methods and datasets are described in Section 2.…”

“…Most global decadal prediction systems provide a high hindcast skill for the AMV Index as well as North Atlantic sea surface-and 2 m-temperatures up to 9 years ahead (Doblas-Reyes et al, 2013;Bellucci et al, 2015). A strong potential for skilful decadal predictions has also been identified for Europe (Feldmann et al, 2019;Reyers et al, 2019). For example, Ghosh et al (2017) provided evidence that an increased heat flux due to higher sea surface temperatures in the North Atlantic is triggered by the positive phase of the AMV.…”

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

“…Therefore, the regional as well as the global ensemble consists of 5800 simulation years. This provides a much larger sample size compared to previous studies of regional decadal predictions (Mieruch et Reyers et al, 2019). This larger sample size here allows for more robust skill estimates in this work, compared to the aforementioned studies.…”

“…The goal of this module is to estimate the predictive skill and added value of high-resolution decadal predictions at the regional scale. First analyses of the regionalized MiKlip hindcasts by Mieruch et al (2014), Reyers et al (2019) and Moemken et al (2016) indicated skill and added value of downscaling and initialization for parts of Europe. In particular, they could show that the dynamical downscaling may reduce the bias and increase the reliability of the hindcasts.…”