This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| ESMValTool | |
|---|---|
| Name | ESMValTool |
| Developer | European Centre for Medium-Range Weather Forecasts; Royal Netherlands Meteorological Institute; Max Planck Institute for Meteorology |
| Initial release | 2016 |
| Latest release | 2024 |
| Programming language | Python; Fortran; YAML |
| License | Apache License 2.0 |
| Website | ESMValTool project |
ESMValTool
ESMValTool is an open-source analysis and diagnostics tool designed for multi-model evaluation of climate model output, providing reproducible workflows for model intercomparison, detection and attribution, and process evaluation. It integrates data processing, plotting, and statistical evaluation into modular recipes that interoperate with community datasets and computing infrastructures, enabling collaboration among researchers at institutions such as the European Centre for Medium-Range Weather Forecasts, the Max Planck Institute for Meteorology, and national meteorological services like the Royal Netherlands Meteorological Institute.
ESMValTool automates extraction, standardization, and diagnostics of climate model output, producing diagnostics across metrics like bias, trends, and teleconnections while interfacing with major data archives and analysis libraries. It bridges model output from projects such as the Coupled Model Intercomparison Project and observational archives like the Global Precipitation Climatology Project, supporting community workflows used by panels including the Intergovernmental Panel on Climate Change and collaborations between groups like the World Climate Research Programme and the European Union Copernicus Programme.
Development began to address reproducibility and consistency challenges highlighted by multi-model activities tied to the Coupled Model Intercomparison Project phases. Early contributors included teams at the Max Planck Institute for Meteorology, the Royal Netherlands Meteorological Institute, and the European Centre for Medium-Range Weather Forecasts, with coordination among projects such as CMIP5 and CMIP6. Over successive releases the project incorporated community software practices influenced by initiatives like GitHub, the Python Software Foundation, and standards from the World Meteorological Organization. Major development milestones aligned with assessment cycles of the Intergovernmental Panel on Climate Change and with adoption by infrastructure providers including the European Open Science Cloud.
Key components include recipe-driven workflows, preprocessor modules, diagnostic scripts, and plotting backends that draw on libraries and standards from the scientific Python ecosystem. Preprocessors implement operations referenced in community protocols developed by groups such as the Global Carbon Project and interfaces to file conventions used by the Climate and Forecast Metadata Conventions. Diagnostic modules embed algorithms from literature produced at institutes like the University of Oxford, the National Center for Atmospheric Research, and the Geophysical Fluid Dynamics Laboratory. Visualization and output formats connect with tools and services such as Matplotlib, Cartopy, and archive targets used by the Earth System Grid Federation.
ESMValTool supports a wide range of climate, Earth system, and regional models submitted to projects like CMIP6, CMIP5, and regional ensembles coordinated by the Coordinated Regional Climate Downscaling Experiment. It works with observational datasets including those from the Global Precipitation Climatology Project, the Global Historical Climatology Network, satellite missions by NASA, European Space Agency, and reanalysis products such as ERA5 produced at the European Centre for Medium-Range Weather Forecasts. The tool’s mapping between model-variable names and standard variables follows conventions used by the Climate Model Output Rewriter and community vocabularies endorsed by the World Data System.
Users define workflows via YAML recipes that specify variables, model ensembles, time ranges, and diagnostics; these recipes are executed on desktops, high-performance computing centers, or cloud platforms managed by providers like Amazon Web Services or research clouds coordinated by the European Open Science Cloud. Execution leverages workflow engines and parallelization strategies similar to those employed by projects at the National Energy Research Scientific Computing Center and scheduling systems used at supercomputing centers such as Jülich Supercomputing Centre. Typical usage in research groups at institutions like the University of Cambridge, ETH Zurich, and Princeton University supports reproducible figures and tables for publications and assessment reports.
Validation efforts combine unit tests and integration tests with benchmark suites reflecting distributed experiments from consortia such as the World Climate Research Programme and the Climate Modeling Alliance. Performance tuning addresses I/O bottlenecks when accessing archives like the Earth System Grid Federation and optimizes parallel processing strategies used in community codes originating from centers such as the National Center for Atmospheric Research and the European Centre for Medium-Range Weather Forecasts. Benchmark studies published by groups including the Max Planck Institute for Meteorology and the Met Office compare diagnostics across model ensembles and observational references to quantify uncertainty and computational scaling.
Governance combines steering by core institutions like the European Centre for Medium-Range Weather Forecasts, collaborative development on platforms such as GitHub, and community input via workshops organized by organizations like the World Climate Research Programme and the Global Climate Observing System. Contributions come from universities, national meteorological services, and research laboratories including the Met Office, National Oceanic and Atmospheric Administration, Lamont–Doherty Earth Observatory, and the Scripps Institution of Oceanography. Training, documentation, and user support are delivered through tutorials at conferences such as the American Geophysical Union Fall Meeting and summer schools run by institutions like the International Space Science Institute.
Category:Climate software