ALBERT

Description: Mittelfristige Klimaprognose (MiKlip), an 8-yr German national research project on decadal climate prediction, is organized around a global prediction system comprising the Max Planck Institute Earth System Model (MPI-ESM) together with an initialization procedure and a model evaluation system. This paper summarizes the lessons learned from MiKlip so far; some are purely scientific, others concern strategies and structures of research that target future operational use. Three prediction system generations have been constructed, characterized by alternative initialization strategies; the later generations show a marked improvement in hindcast skill for surface temperature. Hindcast skill is also identified for multiyear-mean European summer surface temperatures, extratropical cyclone tracks, the quasi-biennial oscillation, and ocean carbon uptake, among others. Regionalization maintains or slightly enhances the skill in European surface temperature inherited from the global model and also displays hindcast skill for wind energy output. A new volcano code package permits rapid modification of the predictions in response to a future eruption. MiKlip has demonstrated the efficacy of subjecting a single global prediction system to a major research effort. The benefits of this strategy include the rapid cycling through the prediction system generations, the development of a sophisticated evaluation package usable by all MiKlip researchers, and regional applications of the global predictions. Open research questions include the optimal balance between model resolution and ensemble size, the appropriate method for constructing a prediction ensemble, and the decision between full-field and anomaly initialization. Operational use of the MiKlip system is targeted for the end of the current decade, with a recommended generational cycle of 2–3 years.

Description: Due to its extension, geography and the presence of several underdeveloped or developing economies, the Central Asia domain of the Coordinated Regional Climate Downscaling Experiment (CORDEX) is one of the most vulnerable regions on Earth to the effects of climate changes. Reliable information on potential future changes with high spatial resolution acquire significant importance for the development of effective adaptation and mitigation strategies for the region. In this context, regional climate models (RCMs) play a fundamental role. In this paper, the results of a set of sensitivity experiments with the regional climate model COSMO-CLM version 5.0, for the Central Asia CORDEX domain, are presented. Starting from a reference model setup, general model performance is evaluated for the present day, testing the effects of singular changes in the model physical configuration and their mutual interaction with the simulation of monthly and seasonal values of three variables that are important for impact studies: near-surface temperature, precipitation and diurnal temperature range. The final goal of this study is two-fold: having a general overview of model performance and its uncertainties for the considered region and determining at the same time an optimal model configuration. Results show that the model presents remarkable deficiencies over different areas of the domain. The combined change of the albedo, taking into consideration the ratio of forest fractions, and the soil conductivity, taking into account the ratio of liquid water and ice in the soil, allows one to achieve the best improvements in model performance in terms of climatological means. Importantly, the model seems to be particularly sensitive to those parameterizations that deal with soil and surface features, and that could positively affect the repartition of incoming radiation. The analyses also show that improvements in model performance are not achievable for all domain subregions and variables, and they are the result of a compensation effect in the different cases. The proposed better performing configuration in terms of mean climate leads to similar positive improvements when considering different observational data sets and boundary data employed to force the simulations. On the other hand, due to the large uncertainties in the variability estimates from observations, the use of different boundaries and the model internal variability, it has not been possible to rank the different simulations according to their representation of the monthly variability. This work is the first ever sensitivity study of an RCM for the CORDEX Central Asia domain and its results are of fundamental importance for further model development and for future climate projections over the area.

Description: One of the most prominent asymmetric features of the southern hemispheric (SH) circulation is the split jet over Australia and New Zealand in austral winter. Previous studies have developed indices to detect the degree to which the upper-level midlatitude westerlies are split and investigated the relationship between split events and the low-frequency teleconnection patterns, viz. the Antarctic Oscillation (AAO) and the El Niño–Southern Oscillation (ENSO). As the results were inconsistent, the relationship between the wintertime SH split jet and the climate variability indices remains unresolved and is the focus of this study. Until now, all split indices' definitions were based on the specific region where the split jet is recognizable. We consider the split jet as hemispheric rather than a regional feature and propose a new, hemispherical index that is based on the principal components (PCs) of the zonal wind field for the SH winter. A linear combination of PC2 and PC3 of the anomalous monthly (JAS) zonal wind is used to identify split-jet conditions. In a subsequent correlation analysis, our newly defined PC-based split index (PSI) indicates a strong coherence with the AAO. However, this significant relationship is unstable over the analysis period; during the 1980s, the AAO amplitude was higher than the PSI, and vice versa in the 1990s. It is probable that the PSI, as well as the AAO, underlie low-frequency variability on the decadal to centennial timescales, but the analyzed period is too short to draw these conclusions. A regression analysis with the Multivariate ENSO Index points to a nonlinear relationship between PSI and ENSO; i.e., split jets occur during both strong positive and negative phases of ENSO but rarely under “normal” conditions. The Pacific South American (PSA) patterns, defined as the second and third modes of the geopotential height variability at 500 hPa, correlate poorly with the PSI (rPSA−1 ≈ 0.2 and rPSA-2= 0.06), but significantly with the individual components (PCs) of the PSI, revealing an indirect influence on the SH split-jet variability. Our study suggests that the wintertime SH split jet is strongly associated with the AAO, while ENSO is to a lesser extent connected to the PSI. We conclude that a positive AAO phase, as well as both flavors of ENSO and the PSA-1 pattern produce favorable conditions for a SH split event.

Description: Paleoclimate data assimilation (DA) is a promising technique to systematically combine the information from climate model simulations and proxy records. Here, we investigate the assimilation of tree-ring-width (TRW) chronologies into an atmospheric global climate model using ensemble Kalman filter (EnKF) techniques and a process-based tree-growth forward model as an observation operator. Our results, within a perfect-model experiment setting, indicate that the "online DA" approach did not outperform the "off-line" one, despite its considerable additional implementation complexity. On the other hand, it was observed that the nonlinear response of tree growth to surface temperature and soil moisture does deteriorate the operation of the time-averaged EnKF methodology. Moreover, for the first time we show that this skill loss appears significantly sensitive to the structure of the growth rate function, used to represent the principle of limiting factors (PLF) within the forward model. In general, our experiments showed that the error reduction achieved by assimilating pseudo-TRW chronologies is modulated by the magnitude of the yearly internal variability in the model. This result might help the dendrochronology community to optimize their sampling efforts.

Description: The East Asian summer monsoon (EASM) is an important part of the global climate system and plays a vital role in the Asian climate. Its sub-seasonal-to-seasonal predictability is a long-standing issue within the monsoon scientist community. In this study, we analyse the seasonal (with six months lead time) prediction skill of the EASM rainfall and its associated general circulation in non-initialised and initialised simulations for the years 1979–2005 performed by six prediction systems (i.e., the BCC-CSM1-1, the CanCM4, the GFDL-CM2p1, the HadCM3, the MIROC5 and the MPI-ESM-LR) from the Coupled Model Intercomparison Project phase 5 (CMIP 5). We find that the simulation of the zonal wind is significantly improved in initialised simulations compared to non-initialized simulations. Based on the knowledge that zonal wind indices can be used as potential predictors for the EASM, we selected an EASM index based upon the zonal wind for further analysis. The assessment show that the GFDL-CM2p1 and the MIROC5 add prediction skill in simulating the EASM index with initialisation, the BCC-CSM1-1, the CanCM4, and the MPI-ESM-LR change the skill insignificantly, and the HadCM3 indicates a decreased skill score. The different response to the initialisation can be traced back to the ability of the models to capture the ENSO (El Niño-Southern Oscillation)-EASM coupled mode, particularly the Southern Oscillation-EASM coupled mode. In summary, we find that the GFDL-CM2p1 and the MIROC5 are capable to predict the EASM on a seasonal time-scale after initialisation.

Description: Paleo-proxy observations have been recently used to constrain the climate models through data assimilation (DA). However, both DA and climate models are computationally very expensive. Moreover, in paleo-DA, the assimilation period is usually too long for a dynamical model to follow the previous analysis state and the chaotic behavior of the model becomes dominant. The majority of the recent paleoclimate studies using DA have performed low or intermediate resolution global simulations along with an off-line DA approach. In an off-line DA, the re-initialisation cycle is completely removed after the assimilation step. In this paper, we design a computationally affordable DA to assimilate yearly pseudo and real observations into an ensemble of COSMO-CLM high resolution regional climate model (RCM) simulations over Europe, where the ensemble members slightly differ in boundary and initial conditions. Within a perfect model experiment, the performance of the applied DA scheme is evaluated with respect to its sensitivity to the noise levels of pseudo-observations. It was observed that the injected bias in the pseudo-observations does linearly impact the DA skill. Such experiments can serve as a tool for selection of proxy records, which can potentially reduce the background error when they are assimilated in the model. Additionally, the sensibility of the COSMO-CLM to the boundary conditions is addressed. The geographical regions, where the model exhibits high internal variability are identified. Two sets of experiments are conducted by averaging the observations over summer and winter. The dependency of the DA skill to different seasons is investigated. Furthermore, the effect of the spurious correlations within the observation space is studied and the optimal correlation length, within which the observations are assumed to be correlated, is detected. Finally, the real yearly-averaged observations are assimilated into the RCM and the performance is evaluated against a gridded observation dataset. We conclude that the DA approach is a promising tool for creating high resolution yearly analysis quantities. The affordable DA method can be applied to efficiently improve the climate field reconstruction efforts by combining high resolution paleo-climate simulations and the available proxy observations.