Skip to content

1. High-Value Datasets (HVD) and Open Radar Data (ORD): Compliance with the HVD scope and requirements

Under Commission Implementing Regulation (EU) 2023/138, weather radar data are explicitly designated as High-Value Datasets (HVDs) within the meteorological thematic category. This means that public-sector weather radar data and products must be made openly available under harmonised technical and legal conditions to maximise reuse, innovation, and cross-border applications .

For radar data providers (e.g. National Meteorological Services (NMSs)), this is not just about openness — it sets concrete requirements on formats, timeliness, access mechanisms, metadata, and licensing. Data must be free of charge, provided via APIs (with bulk download options), and licensed under open, machine-readable terms (e.g., Creative Commons BY 4.0).

What radar data and products must be provided as HVDs?

Scope

Radar datasets must include:

  • Data per radar station within a Member State
  • National radar composites, where available

Required radar variables

The regulation explicitly lists the following radar products and parameters (modified terminology):

  • Radar reflectivity factor
  • Dual - polarisation variables
  • Precipitation rate and accumulated precipitation
  • Radial velocity
  • Echo tops

This implies that both single-site and composite radar products, including polarimetric information, fall under the HVD obligation.

Technical requirements for radar data publication

1. Update Frequency & Timeliness

Radar data must be made available:

  • With updates every 5 minutes, or at the shortest available interval

2. Access mechanisms

Radar data must be provided through:

  • APIs (Application Programming Interfaces), and
  • Bulk download

This ensures both:

  • automated, real-time access, and
  • efficient retrieval of large historical datasets

3. Data formats

Radar data must be published in open, machine-readable formats, specifically including:

  • HDF5
  • BUFR (old data format for the archived data)

4. Metadata & documentation

Radar datasets must be accompanied by:

  • Complete metadata, published online in an open, widely used machine-readable format

  • Public documentation describing:

  • data structure

  • semantics
  • variables and conventions used

Licensing

Radar HVDs must be made available under:

  • Creative Commons BY 4.0, or
  • an equivalent or less restrictive open licence

This allows free and unrestricted reuse, including commercial use, provided attribution is given .

Cost

High-Value Datasets must be provided free of charge, unless a temporary exemption applies under the Open Data Directive (not typical for meteorological services).

Compliance of ORD API data and product supply with the HVD scope and requirements?

The datasets in ORD are:

1. EUMETNET OPERA single-site radar data

  • 24-hour rolling cache, with a long-term archive (2012–, TBD)
  • Formats: BUFR (older) and ODIM HDF5 (recent)
  • License: CC BY 4.0 (with exceptions noted in metadata)
  • Currently mainly includes:
    • DBZH (Doppler-filtered horizontal reflectivity factor, with national QC applied)
    • TH (unfiltered horizontal reflectivity factor)
    • VRADH (horizontal radial velocity)

:heavy_exclamation_mark: Noting ORD API supply differences to HVD * Dual-polarisation variables: Core dual-polarisation radar variables (e.g. correlation coefficient, differential reflectivity, differential phase shift, specific differential phase) are currently not exchanged within the EUMETNET OPERA data flow. As a result, these variables are not available via the ORD API, even though they are listed in the HVD scope. The inclusion of these variables in the OPERA data exchange — and consequently in the ORD API — is planned from 2027 onwards (timeline to be confirmed). * Data quality and processing: The applied quality control and clutter removal procedures for radar data exchanged through EUMETNET OPERA depend on national processing chains. Therefore, the level and type of data cleaning may vary between countries. * Extended and value-added products: Additional radar variables and more highly processed products may be available through the national data services of individual National Meteorological Services (NMSs), beyond what is provided via the ORD API.

2. European OPERA composite products

  • Products: maximum reflectivity, instantaneous rain rate, 1-hour accumulation, real-time (2011 - coming later)
  • Formats: ODIM HDF5 and cloud-optimized GeoTIFF (TBD)
  • License: CC BY 4.0

:heavy_exclamation_mark: Noting ORD API supply differences to HVD * OPERA composite: OPERA composites are not strickly speaking HVDs, because the Implementation Act is targeted to national public-sector weather radar data and products.

3. National radar products

  • Examples of national composites, rain rate composites, accumulations, echo tops
  • Access: via links to national interfaces (24h access of links)
  • Formats: ODIM HDF5 or cloud-optimized GeoTIFF

:heavy_exclamation_mark: Noting ORD API supply differences to HVD * Radar products: This service is planned to serve the NMSs to fulfill their HVDs requirement. Some NMSs prefer to use their national interface for the data supply.

4. ODIM (OPERA Data Information Model)

High-Value Datasets (HVDs) under the EU Open Data Directive are required to include specific metadata to ensure that data are Findable, Accessible, Interoperable, and Reusable (FAIR). The ORD initiative aims to meet these FAIR requirements.

:heavy_exclamation_mark: Noting ORD API supply differences to HVD To ensure consistent naming, structure, and data representation, the ORD API follows the OPERA Data Information Model (ODIM), which is an internationally recognised and widely adopted standard in radar meteorology.

Clarifying the terminology in HVDs

Some radar-related terms used in Sections 3.1 and 3.2 of the Annex to the Implementation Act differ from the terminology commonly applied in radar meteorology. The differences are briefly clarified below using terminology aligned with current radar meteorology practice, based on the AMS Glossary of Meteorology (https://glossary.ametsoc.org/)

Reflectivity

The basic radar variable is actually called radar reflectivity factor, which is stated in units of mm^6^ m^-3^ or provided also as dBZ. The radar reflectivity factor is proportional to the reflectivity which again has dimensions of area per unit volume (e.g., cm^2^m^-3^, or, more commonly, cm^-1^ or m^-1^), however reflectivity is not given as observable of the weather radar.

Backscatter

By definition backscatter is the scattering of radiant energy into the hemisphere of space bounded by a plane normal to the direction of the incident radiation and lying on the same side as the incident ray. Presumably here it is meant to describe the backscatter coefficient, i.e., the physical quantity used to describe the backscattering process, are m^−1^ sr^−1^ (per meter and steradian). This is a measure of the reflective strength of a radar target, however also here, backscatter coefficient is not given directly as a radar observable, but radar reflectivity factor is the used quantity. Therefore, it is assumed that the two first terms of the key attributes list are describing the same radar observable.

Polarisation

Polarisation is general term describing the correlation between two orthogonal components of its electric (or, equivalently, magnetic) field for a transverse electromagnetic wave. With respect to weather radar, here it is probably meant, the radar variables that can be measured with dual-polarisation radars. Dual-polarisation weather radars measure typically with two linear polarisations, orthogonally horisontal and vertical directions. The basic and operationally measured dual-pol radar variables are differential reflectivity (Zdr), propagation differential phase shift (ϕ~dp~), and co-polar correlation coefficient (ρ~hv~), but there are also other dual-pol variables. However, it is assumed here that in the list of key attributes should not go to this much of detail, but the term the basic dual-polarisation variables could be used to describe these observables.

Precipitation

Weather radar doesn’t measure precipitation directly, however the instantaneous precipitation rate (mm/h) or rain rate (mm/h) can be derived using retrieval algorithm from the measured radar reflectivity factor. Therefore, it is suggested to be described this quantity either as the basic radar product in general terms or as instantaneous precipitation rate (mm/h) or rain rate (mm/h).

Wind

Weather radar doesn’t measure wind directly, however Doppler radars can observe radial velocity component from the hydrometeors or other scatterers in the atmosphere, also known as Doppler velocity. Interpretation of the Doppler velocity depends on the viewing geometry and the kind of target. Clear-air echoes are assumed to move with the wind, so that the Doppler velocity measured at a given location in the atmosphere is equal to the radial component of the wind at that location. Precipitation falls relative to the air, so that the Doppler velocity of a precipitation target is assumed to be the sum of the radial velocity component of the precipitation terminal fall velocity and the radial component of the air motion. To simplify the interpretation, in the list of key attributes the term radial velocity should be used instead of wind. There is also a radar product called wind profile, but this is not necessarily computed in all Meteorological Services in Europe and computational algorithms differ nationally.

Echo tops

Echo top is a product based on radar observations and used often in detection of convective storms. Basically, an echo top is a top of an area of precipitation indicated by the radar. This product is created either from a single or multiple radar volume observations. The definition is depended on the used thresholds to define the cloud top as well as from the wavelength and sensitivity of the radar system. There exist several algorithms how this product can be computed. This is not a direct radar observable, nor it is created in all Meteorological Services in Europe and due to the before-mentioned reasons, this quantity is not comparable between the different member states. Hence, it is suggested to remove this from the list of key attributes.