Contributors: Swapan Mallick, Harald Schyberg, Jelena Bojarova
<style>
.special_indent ul > li > ul {
padding-left: 0;
}
.special_indent ul > li > ul > li > ul {
padding-left: 1.5em;
}
</style> |
One of the original objectives of the CARRA2 reanalysis was to quantify its uncertainty in essential climate variables. To address this, we developed a machine-learning framework based on Denoising Diffusion Probabilistic Models (DDPM-ML) to infer high-resolution situation-dependent uncertainty using paired ensembles from ERA5 (62 km) and CARRA2 (2.5 km). The model is trained in a supervised manner with ERA5 ensemble fields as inputs and corresponding CARRA2 fields as targets for a limited time period (in the first half of 2022), focusing on 2m temperature.
This framework generates spatially coherent uncertainty fields that effectively capture variability related to atmospheric flow and orographic features. By harnessing the probabilistic capabilities inherent in diffusion models, DDPM-ML is able to learn intricate uncertainty structures from ensemble data and reconstruct the cross-scale uncertainty relationships between the coarser-resolution ERA5 ensemble data assimilation (ERA-EDA) and the finer-resolution CARRA2 dataset. Comprehensive details regarding the methodology, input data, and uncertainty quantification procedures are provided in this documentation of the method: Copernicus pan-Arctic Regional Reanalysis (CARRA2): Uncertainty estimation method
Upon learning the uncertainty structures via the DDPM-ML framework, uncertainty estimates can be generated for the deterministic CARRA2 reanalysis. The trained model produces flow-dependent uncertainty fields by downscaling ERA5 ensemble information onto the high-resolution CARRA2 grid and its associated orography. Although applying this approach to the entire 40-year CARRA2 record presents significant computational challenges, we demonstrate the methodology using data for 2019. In the absence of an intrinsic ensemble prediction system for CARRA2, this dataset represents a pragmatic advancement toward providing uncertainty information to the CARRA2 user community.
Uncertainty estimates for 2-meter temperature were produced at 00, 06, 12, and 18 UTC throughout the entirety of 2019 across the CARRA2 domain. This effort resulted in a dataset comprising 1,460 NetCDF files accompanied by corresponding visualizations in image format. All files are accessible via the CARRA2 project’s GitHub repository https://github.com/CARRA2/Uncertainty_Quantification/tree/Sep_2025/UQ_OUTPUT, where they are systematically organized into distinct directories for NetCDF data and image outputs. Each directory is further subdivided into two segments: 2019_PART1 (covering January to June) and 2019_PART2 (covering July to December).
Figure: Illustration of uncertainty quantification for the specific date of January 1, 2019, at 00:00 UTC.
This dataset is provided as an example to demonstrate how uncertainties change in relation to topography, weather conditions, and their variations over space and time. Specifically, the findings emphasize that uncertainty patterns can differ depending on whether the conditions are linked to frontal systems or other meteorological events. While the dataset is useful for those interested in examining these relationships, its relevance is confined to the demonstration period of 2019. This year was chosen as a typical case to illustrate the methodology across a variety of common atmospheric conditions, rather than to offer a full climatological analysis. Therefore, the uncertainty estimates should not be assumed to apply directly to other time periods without additional evaluation. Although the 2019 results may provide some qualitative understanding of uncertainty behavior in CARRA2 in general, they do not represent descriptions for other years. It is also important to recognize that the underlying method has limitations, such as sensitivity to the selected time frame and possible difficulties in capturing all sources of variability.
This document has been produced in the context of the Copernicus Climate Change Service (C3S). The activities leading to these results have been contracted by the European Centre for Medium-Range Weather Forecasts, operator of C3S on behalf of the European Union (Delegation Agreement signed on 11/11/2014 and Contribution Agreement signed on 22/07/2021). All information in this document is provided "as is" and no guarantee or warranty is given that the information is fit for any particular purpose. The users thereof use the information at their sole risk and liability. For the avoidance of all doubt , the European Commission and the European Centre for Medium - Range Weather Forecasts have no liability in respect of this document, which is merely representing the author's view. |