About the MGCLS–H I database

The MeerKAT Galaxy Cluster Legacy Survey (MGCLS) is a large programme carried out with the 64-dish MeerKAT telescope in South Africa. Between 2018 and 2019, MeerKAT observed 115 galaxy clusters for 6–10 hours each at L-band (≈900–1670 MHz), producing deep, wide-field radio imaging with typical sensitivities of a few μJy beam−1 and angular resolutions of ≈8 arcsec. The first continuum data release (DR1) provides a rich set of products for studies of radio galaxies, diffuse cluster emission and polarisation.

The MGCLS–H I project builds directly on this legacy by exploiting the neutral hydrogen (H I 21-cm) line information embedded in the same visibilities. By reprocessing the original MGCLS measurement sets in spectral-line mode, the project constructs a homogeneous database of H I detections and data products for galaxies in and around MGCLS clusters. This website hosts the first H I data release (H I–DR1).

From MGCLS to the MGCLS–H I database

The MGCLS–H I project reprocesses the original MGCLS visibilities to extract the 21-cm line signal. The available bandwidth and channelisation allow us to trace H I emission over a primary redshift range of 0 < z < 0.11 (with additional sub-bands at higher redshift), covering galaxies in environments from rich clusters to smaller groups.

The survey strategy is to first process the radio-selected cluster sample with z < 0.1, followed by X-ray selected clusters in the same redshift range. For each cluster, a dedicated H I cube is produced over ≈1280–1430 MHz, covering the relevant velocity range around the cluster systemic redshift.

The goal is to create a legacy H I database that enables robust statistical studies across thousands of galaxies and dozens of clusters, with uniform sensitivity, well-documented completeness and consistent data products. The database is hosted at Rhodes University, and this website provides the first public data release (H I–DR1).

The development of the database and associated workflow has been carried out over a dedicated multi-year effort, in close collaboration with MGCLS team members who initiated the early H I exploration, tuning of pipeline parameters and design of the database schema. This release represents the first step in turning the full MGCLS H I information into a long-lived community resource.

Processing workflow and data preparation

The MGCLS–H I data products are derived from the original MeerKAT visibilities using a uniform spectral-line reduction and staging workflow. At a high level:

  • Raw MeerKAT visibilities are ingested from the science archive.
  • Flagging, calibration and self-calibration are applied using a standard spectral-line reduction pipeline, including radio-frequency interference (RFI) mitigation, cross-calibration and iterative self-calibration.
  • Continuum subtraction and line imaging are performed to isolate the H I line signal, producing primary-beam-corrected data cubes on a regular spatial–spectral grid.
  • 3-D source finding and parameter extraction are run on the cubes to identify H I emission in position–position–velocity space and to measure basic source properties.
  • Completeness testing with mock galaxy injection and recovery experiments is used to quantify the H I mass and linewidth ranges over which the survey is statistically complete, as a function of redshift, linewidth and cube depth.
  • Galaxy-level products (H I cubelets, moment maps, velocity fields, line profiles and masks) are cut out and associated with their parent cluster and catalogued in a consistent schema.
  • Quality assessment and staging are performed before products are ingested into the MGCLS–H I database and exposed through this website.

This end-to-end approach ensures that all clusters in the database are reduced with compatible settings and that the resulting catalogues can be used for homogeneous population studies, completeness-limited analyses and detailed follow-up work.

What is included in MGCLS–H I DR1?

The first MGCLS–H I data release focuses on a staged subset of MGCLS clusters for which the H I cubes and catalogues have passed the current quality checks. For each cluster, the database provides:

  • Cluster-level metadata – basic properties such as redshift, dynamical mass proxies, image depth, beam size and H I column-density sensitivity.
  • Primary-beam-corrected H I cubes – full-field spectral cubes centred on the cluster pointing.
  • Galaxy-level catalogues – H I detections with positions, velocities, integrated fluxes, linewidths and a set of basic morphological descriptors.
  • H I cubelets and moment maps – small cutouts around each detection, including total intensity (moment-0), velocity fields (moment-1) and, where appropriate, velocity-dispersion maps (moment-2).
  • Masks and diagnostic products – source masks, signal-to-noise maps and quality flags that allow users to assess the robustness of each detection.
  • Multi-wavelength preview figures – quick-look overlays that combine H I, X-ray and optical information for many fields.

Access, documentation and further reading

All released products in MGCLS–H I DR1 are made available through this web interface. Users can browse clusters and galaxies, preview figures, and download catalogues, cubelets or full cubes for local analysis.

Technical details of the parent MGCLS survey, observing strategy and continuum data products are described in:

A dedicated MGCLS–H I early-release or description paper provides further details of the H I reduction, completeness analysis and database design. A pre-release version of this work is available here: