HARMONIE

HARMONIE
Name	HARMONIE
Programming language	Fortran, C, Python
Operating system	Unix-like

HARMONIE

HARMONIE is a high-resolution numerical weather prediction system used for short-range forecasting and research, developed through collaborations among European meteorological services and research institutes. It serves operational centers and academic groups, integrating models, data assimilation, and post-processing pipelines to support forecasting for aviation, hydrology, and emergency management. The system has been adopted by national services and projects across Scandinavia, the British Isles, the Netherlands, and continental Europe.

Overview

HARMONIE originated from collaborations linking European Centre for Medium-Range Weather Forecasts initiatives with national services such as Météo-France, Royal Netherlands Meteorological Institute, Met Éireann, Danish Meteorological Institute, Met Office, and SMHI. It is part of a family of limited-area models alongside ALADIN, AROME, and IFS-derived systems used by Norwegian Meteorological Institute partners and project consortia like HIRLAM and ALADIN Consortium. HARMONIE combines dynamical cores, parameterization suites, and coupling modules influenced by developments at ECMWF, Met Office UK, and research groups from Université de Toulouse and KNMI. Operational deployments often interface with forecasting tools from EUMETSAT, Copernicus, and national warning systems.

Development and Technical Architecture

The software stack traces contributions from research teams at Météo-France Research and Development, ECMWF, KNMI, and university groups including University of Reading, Uppsala University, and University of Helsinki. HARMONIE uses a nonhydrostatic dynamical core similar to approaches in AROME and exploits numerical schemes promoted by ECMWF and Met Office Hadley Centre research. Implementations use languages common in numerical modeling such as Fortran 90, C, and Python for scripting and workflow integration, relying on libraries from NetCDF, MPI, and OpenMP ecosystems. The model couples to post-processing and visualization tools developed by partners like KNMI and Météo-France, and is deployed on high-performance computing platforms from vendors including Cray Inc., IBM, and cloud systems used by European Space Agency projects.

Data Assimilation and Model Physics

HARMONIE integrates data assimilation methods influenced by variational and ensemble techniques developed at ECMWF, Met Office, and research centers such as INRIA and LMD (Laboratoire de Météorologie Dynamique). Observational inputs include profiles and radiances from satellite missions like Metop, MSG (satellite), and Sentinel-3 coordinated by EUMETSAT and ESA, as well as conventional reports from Synoptic station networks, radiosonde launches by national services including Météo-France and Danish Meteorological Institute, and remote sensing from Doppler radar networks managed by UK Met Office and KNMI. Physics parameterizations draw on schemes tested at Météo-France, SMHI, and Met Éireann for microphysics, boundary layer, and convection, and include land-surface coupling modules comparable to those used in SURFEX and hydrology interfaces developed with groups at Rijkswaterstaat and Deltares.

Operational Use and Applications

National meteorological services deploy HARMONIE for real-time forecasts supporting aviation services at airports like Schiphol Airport and Copenhagen Airport, coastal warnings coordinated with agencies such as Fisheries and Oceans-style services, and hydrometeorological forecasting partnered with Deltares and regional emergency planners in Scandinavia and Ireland. Research applications include convection-permitting simulations used in studies led by universities like University of Oxford, University of Leeds, and Université Grenoble Alpes for convective initiation and urban meteorology. HARMONIE outputs feed decision-support systems tied to Copernicus Emergency Management Service products, EUMETNET coordination activities, and aviation safety tools used by Eurocontrol.

Validation and Performance

Verification frameworks for HARMONIE utilize metrics and standards from ECMWF, World Meteorological Organization, and research groups at Met Office and KNMI, employing deterministic and probabilistic verification against observations from Synoptic station networks, radiosonde datasets archived by NCAR, and satellite validation from EUMETSAT projects. Performance assessments consider ensemble spread tested in intercomparison campaigns involving HARMONIE-AROME, ALADIN, and COSMO consortia, with computational benchmarks run on infrastructure from PRACE and national supercomputing centers such as CSC (Finland) and KNMI facilities. Skill scores reported in peer-reviewed studies by teams at Météo-France and SMHI address precipitation timing, wind gusts, and temperature biases versus reference analyses from ECMWF and Met Office operational products.

Community, Governance, and Licensing

Development governance is coordinated by consortia including HIRLAM and ALADIN Consortium member services and research institutes such as Météo-France, KNMI, Danish Meteorological Institute, and Met Éireann. Community contributions come from academic partners at Uppsala University, University of Reading, University of Helsinki, and Université Victor Segalen Bordeaux 2, with working groups organized under networks like EUMETNET and project frameworks funded by Horizon 2020 and national research councils. Licensing and dissemination arrangements reflect agreements among participating services and are aligned with practices used in open modeling collaborations involving ECMWF and Copernicus partners. Implementation, training, and user support are delivered through workshops hosted by Météo-France, KNMI, and university partners, while collaborative research outputs are published in journals associated with American Meteorological Society, Royal Meteorological Society, and European geoscience conferences.

Category:Numerical weather prediction