Copernicus Climate Change Service (C3S) integrates multiple seasonal forecast models of climate variables with multiple ensemble realizations. Assessing the risks of natural hazards with high impacts on human and natural systems and providing actionable services at the local scale require high-resolution predictions. We implement the AI-based approach proposed by Heidari et al. (2023) to address such needs and reach a kilometer scale. While downscaling seasonal forecasts, it is crucial to transfer the full range of the uncertainties given by the ensembles.This study assesses how uncertainty is transferred by an AI-based downscaling approach. Quantile-based metrics are here used to measure the ensemble variability between seasonal forecasts and their downscaled products. On the other side, quantile-based metrics can also give an alternative description of the ensemble variabilities, which could replace the raw ensemble members in the downscaling process. In this study, the AI-downscaling system is tested by inputting (a) raw ensemble members and (b) quantile-based metrics. Transferred uncertainty and downscaling accuracy are then evaluated to develop and implement an optimal downscaling approach with hazard-dependent inputs being selected at regional and local scales. Heidari F., Lin Q., Espitia Sarmiento E.F., Toreti A., and Xoplaki E. (2023): A deep learning technique to realistically bias correct and downscale seasonal forecast ensembles of climate variables towards the development of an AI-based early warning system, EGU 2023 abstract

Regional climate models, due to their systematic biases, are not usable for impact assessment and policy-relevant applications. It is common to post-process the regional model outputs with appropriate bias correction methodologies to provide reliable climate change information. We apply a distribution-based, trend-preserving quantile mapping procedure to bias correct the projections of daily precipitation and temperature from an ensemble of 5 RCMs driven by 5 GCMs, each at a resolution of 0.11°, chosen from the EURO-CORDEX initiative. The gridded observations from the German Weather Service, DWD-HYRAS, has been used as a reference for the bias correction. The impact of the bias correction is found to be more pronounced on precipitation than on temperature, as the precipitation biases are larger. The models are wetter and underestimate (overestimate) the daily maximum (minimum) temperature. The correction method eliminates large parts of these biases and maps the distributions of both the variables well with that of observations. The bias adjustment also leads to the narrowing down of the uncertainties in the projected changes of both the variables. The decomposition of total variance into model uncertainty and internal variability suggests that the bias correction acts mostly on the former component. The internal variability component does not seem, however, to undergo considerable changes following the bias correction. Due to the reduction of the uncertainty, we find a slight improvement in the signal-to-noise ratio in the projections.

<p>The interactions and feedbacks between climate stress and social systems are currently the focus of interest for the scientific community and the general public. Understanding how paleo societies responded to extreme climate conditions is important for gaining insight into current and future climate concerns. The East Mediterranean (EM) and the Nile River basin (NR) are ideal areas for scientific and historical studies and modelling experiments due to the abundance of proxy and historical data. The 6<sup>th</sup> century AD is of particular interest from both a historical and scientific perspective, as it coincides with a period of prosperity for the Eastern Byzantine Empire and political stability, but which also experienced a plague pandemic and significant climate variability in parallel or as a result of a major cluster of volcanic eruptions. To investigate these events and the climate variability in the 6th century in more detail, a transient paleo-simulation is carried out with the appropriately adjusted regional climate model COSMO-CLM (COSMO 5.0 clm16). The regional climate model is driven by the global MPI-ESM-LR at 0.44° for the last 2500 years. The state-of-the-art external forcings of the CMIP6 compliant Earth System Model comprise of volcanic (stratospheric aerosol optical depth), orbital (eccentricity, obliquity, longitude), solar (irradiance), land-use (leaf area index and plant coverage) and greenhouse-gas (CO2 equivalent) changes, implemented in the regional climate model. The simulated temperature and precipitation will be compared with those of other CMIP6 models, and proxy records. This research will provide a comprehensive interpretation of the regional climate and its impacts during the 6th century AD in the Mediterranean.</p> <p> </p> <p>Reference</p> <p>Jungclaus, J. H., Bard, E., Baroni, M., Braconnot, P., Cao, J., Chini, L. P., Egorova, T., Evans, M., González-Rouco, J. F., Goosse, H., Hurtt, G. C., Joos, F., Kaplan, J. O., Khodri, M., Klein Goldewijk, K., Krivova, N., LeGrande, A. N., Lorenz, S. J., Luterbacher, J., Man, W., Maycock, A. C., Meinshausen, M., Moberg, A., Muscheler, R., Nehrbass-Ahles, C., Otto-Bliesner, B. I., Phipps, S. J., Pongratz, J., Rozanov, E., Schmidt, G. A., Schmidt, H., Schmutz, W., Schurer, A., Shapiro, A. I., Sigl, M., Smerdon, J. E., Solanki, S. K., Timmreck, C., Toohey, M., Usoskin, I. G., Wagner, S., Wu, C.-J., Yeo, K. L., Zanchettin, D., Zhang, Q., and Zorita, E.: The PMIP4 contribution to CMIP6 – Part 3: The last millennium, scientific objective, and experimental design for the PMIP4 <em>past1000</em> simulations, Geoscientific Model Development, 10, 4005–4033, https://doi.org/10.5194/gmd-10-4005-2017, 2017.</p>

Abstract. Understanding the past climate at regional scale, the impact of natural variability and sensitivity by studying the underlying dynamics and processes, can provide a point of reference for future climate conditions under anthropogenic forcing. The Eastern Mediterranean (EM) and Nile River basin (NR) regions are of particular interest for the study of past climate due to their location under the influence of major atmospheric teleconnections. We developed a high-resolution regional model for paleoclimate applications, COSMO-CLM, by integrating all external forcings and conducted a transient simulation from 500 BCE to 1850 CE. Principal Component Analysis (PCA) was applied for winter/summer precipitation and temperature to validate the model set up and showed very good agreement between simulated and observational/reanalysis data. Further, 400–362 BCE and 1800–1850 CE have been selected for the comparison of the mean climate conditions of the early Roman period (ERP) and pre-industrial times (PI). The comparison of temperature and precipitation suggests comparable mean climatic conditions with spatial differences in terms of variability within the study regions. Over the Eastern Mediterranean (EM), ERP is wetter and warmer in both winter and summer compared to PI, with higher variability in temperature and precipitation in summer than in winter. In the Nile River basin (NR), ERP summers were wetter and more variable compared to PI. The ERP over NR is warmer by approximately 0.5 °C in winter and cooler by 0.5 °C in summer, with low variability in winter and high variability in summer compared to PI. The relevant large-scale circulation of the two periods shows consistent spatial structures with the corresponding precipitation/temperature EOF patterns, albeit with varying amplitudes. The 2500 years transient simulation sheds light to the paleoclimate conditions and relevant atmospheric circulation as well as processes of periods of interest in complex areas with detailed output and comprehensive forcing allowing for better representation of the regional climate variability and change. Comparison of simulated output with proxy records, reconstructions and detailed studies of specific events, e.g., volcanic eruptions, can help to capture the spatiotemporal extent of these events and their impact on climate variability and change, in addition to providing insights into their impact on societal change and human history.

The project DAKI-FWS (Data and AI-supported Early Warning System to stabilize the German Economy), funded by the Federal Ministry of Economic Affairs and Climate Action (Germany), develops an innovative early warning system with a seasonal time horizon to protect and support lives, jobs, land and infrastructure. High-skilled, innovative time and space-dependent bias correction and high resolution downscaling artificial intelligence approaches, such as deep learning and reinforcement learning techniques, are designed and implemented on ensemble seasonal forecast data. A fundamental challenge in bias correction is to preserve climate trends and plausible representation of the physical properties (variables) of the climate data.Thus in this work, a trend preserving AI-based correction approach is implemented. The high quality bias-corrected data can be introduced into the various climate-related practical applications of the overall project, such as the detection of extreme events but also evolution of pandemics or subtropical/tropical diseases and hydrological models. State-of-the-art AI techniques are applied not only for preprocessing and preparation of the climate and sectoral data but also for the analysis and post-processing phases. Weather and climate extremes, such as heatwaves, storms and droughts, and concurrent extremes are identified from the large pool of meteorological and climatological reference datasets, seasonal forecasts as well as event lists. Such a comprehensive early warning system with seasonal horizon that contributes to the estimation of the outbreak and development of climate and health crises and supports disaster management and risk reduction and mitigation, does not yet exist for Germany, illustrating the importance and potential of this work.

Daily precipitation and temperature simulated by regional climate models carry large systematic biases owing to multiple factors including inadequate model resolution and limitations in the parameterization of important processes. Reduction of these biases is a crucial process in rendering the model information more reliable for climate change and hydrological assessments. We present an evaluative study of bias correction of daily precipitation and temperature from an ensemble of regional climate models from the EUR-11 CORDEX domain (CLMCOM-CCLM4, GERICS-REMO15, SMHI-RCA4, DMI-HIRHAM5, and CanRCM4 driven by MPI-ESM). This is an important milestone within a larger framework of the RegiKlim consortium towards generating high-resolution bias corrected and statistically downscaled fields for providing useful climate information in specific areas in Germany. A quantile delta mapping (QDM) approach is applied to adjust the biases in the distribution characteristics of precipitation and temperature. The delta factor, derived from the ratio of the projected value of a given quantile to that of the present value, is applied to the standard transfer function so that the modelled climate change signal can be preserved. High-resolution (0.1°) gridded dataset from the German Weather Service, DWD-HYRAS, is used as the reference for bias correcting the variables. The impact of the bias adjustment on important parameters such as the number and frequency of wet/dry and cold/hot spells are quantified. The response of the quantile mapping method to the seasonal variations in the dominant driving processes is further investigated.

The project DAKI-FWS (BMWi joint-project “Data and AI-supported early warning system to stabilise the German economy”; German: “Daten- und KI-gestütztes Frühwarnsystem zur Stabilisierung der deutschen Wirtschaft”) develops an early warning system (EWS) to strengthen economic resilience in Germany. The EWS enables better characterization of the development and course of pandemics or hazardous climate extreme events and can thus protect and support lives, jobs, land and infrastructures. The weather and climate modules of the DAKI-FWS use state-of-the-art seasonal forecasts for Germany and apply innovative AI-approaches to prepare very high spatial resolution simulations. These are used for the climate-related practical applications of the project, such as pandemics or subtropical/tropical diseases, and contribute to the estimation of the outbreak and evolution of health crises. Further, the weather modules of the EWS objectively identify weather and climate extremes, such as heat waves, storms and droughts, as well as compound extremes from a large pool of key data sets. The innovative project work is complemented by the development and AI-enhancement of the European Flood Awareness System model, LISFLOOD, and forecasting system for Germany at very high spatial resolution. The model combined with the high-end output of the seasonal forecast prepares high-resolution, accurate flood risk assessment. The final output of the EWS and hazard maps not only support adaptation, but they also increase preparedness providing a time horizon of several months ahead, thus increasing the resilience of economic sectors to impacts of the ongoing anthropogenic climate change. The weather and climate modules of the EWS provide economic, political, and administrative decision-makers and the general public with evidence on the probability of occurrence, intensity and spatial and temporal extent of extreme events as well as with critical information during a disaster.

This study compares the performance of three bias correction (BC) techniques in adjusting simulated precipitation estimates over Germany. The BC techniques are the multivariate quantile delta mapping (MQDM) where the grids are used as variables to incorporate the spatial dependency structure of precipitation in the bias correction; empirical quantile mapping (EQM) and, the linear scaling (LS) approach. Several metrics that include first to fourth moments and extremes characterized by the frequency of heavy wet days and return periods during boreal summer were applied to score the performance of the BC techniques. Our results indicate a strong dependency of the relative performances of the BC techniques on the choice of the regional climate model (RCM), the region, the season, and the metrics of interest. Hence, each BC technique has relative strengths and weaknesses. The LS approach performs well in adjusting the first moment but tends to fall short for higher moments and extreme precipitation during boreal summer. Depending on the season, the region and the RCM considered, there is a trade-off between the relative performances of the EQM and the MQDM in adjusting the simulated precipitation biases. However, the MQDM performs well across all considered metrics. Overall, the MQDM outperforms the EQM in improving the higher moments and in capturing the observed return level of extreme summer precipitation, averaged over Germany.

How did climatic and environmental variability and stress affect past societies in an area of increasing relevance for contemporary planning and policy concerns? The Eastern Mediterranean (EM) and the Nile river basin (Nile) bear a long history of human social dynamics, making it a suitable area for exploring potential interactions between climate variability, extreme events, environmental changes and society over a variety of time scales. The areas contain abundant natural and human-historical archives that preserve information on the climate conditions and impacts on humans and ecosystems covering the past centuries to millennia. So far, the links between climate and societies are examined mainly from the proxy records or the derived paleoclimatic reconstruction perspectives, without addressing the detail of the processes and underlying dynamics that offer the regional climate model simulations. In order to improve our understanding of past climate in the EM and Nile at the regional scale, we developed a spatially high resolved fully-forced paleoclimate version of the COSMO-CLM running over the past 2500 years. All forcings used for the driving ESM, namely volcanic (stratospheric aerosol optical depth), orbital (eccentricity, obliquity, precession), solar (irradiance), land-use and greenhouse-gas changes are implemented to COSMO 5.0-clm16 (see Hartmann et al. for more details). As a starting point for exploring the relationship between climate and society over the last 2500 years, we compared the mean climate conditions (2m temperature and precipitation) of two periods that are 2400 years apart, namely BCE 400-362 and 1980-2018 CE. Overall, the results show that summer temperatures differ by up to 3 degrees between the two periods. In particular, over the tropics, the temperature differences are largest. Precipitation changes vary within the study area and the climate regimes covered. We will further analyze the dynamics and climate variability of the area over the two periods to explore more details of regional and local climate change.

The climate of the last 2500 years is documented in natural (speleothems, tree rings, sediments and pollen) and human-historical archives. Proxy records and subsequent climate reconstructions can be subject to a considerable amount of uncertainty, as the proxies can only capture a fraction of the entire variability. Climate model simulations can contribute to the interpretation of variations observed in the paleoclimate data and better understanding of dynamics, mechanisms and procedures. The state-of-the-art simulations following the CMIP6-protocol are highly resolved in time but still present a rather coarse horizontal resolution (200 km or more) to adequately address regional paleoclimate questions/hypotheses. Dynamical downscaling can close the gap between the regional archives and the coarsely resolved Earth System Models (ESMs). Using regional climate models to downscale ESM output requires a consistent implementation of the climate forcings in the regional model used also for the driving ESM. State-of-the-art and CMIP6 compliant reconstructions of volcanic (stratospheric aerosol optical depth), orbital (eccentricity, obliquity, precession), solar (irradiance), land-use and greenhouse-gas changes used for the MPI-ESM are therefore implemented in the regional climate model COSMO-CLM (CCLM, COSMO 5.0 clm16). The functionality of each implemented forcing is tested separately and in combination for the period (1255-1265) that covers the Samalas volcanic eruption of 1257. The orbital forcing is found to have the largest impact in general and the volcanic forcing has a strong but short-lasting effect after the eruption. The other climate forcings only show very small impact in the chosen period. At the moment, a transient CCLM simulation with all forcings implemented with a horizontal resolution of 50 km is running for the last 2500 years in the Eastern Mediterranean, the Middle East and the Nile River basin.