The UCC is maintained by Gabriel I Perren. The latest version of the full database (including all the identified members) can be found in its Zenodo repository.
Please contact me for any comments/suggestions, or open a Github issue. If you found the UCC useful for your research, please reference its original article Perren et al. (2023). You can use the following text:
“This research has made use of the Unified Cluster Catalogue (UCC)~\cite{Perren_2023}”
@ARTICLE{Perren_2023,
author = {{Perren}, Gabriel I. and {Pera}, Mar{\'\i}a S. and {Navone}, Hugo D. and {V{\'a}zquez}, Rub{\'e}n A.},
title = "{The Unified Cluster Catalogue: towards a comprehensive and homogeneous data base of stellar clusters}",
journal = {\mnras},
keywords = {methods: data analysis, catalogues, open clusters and associations: general, Astrophysics - Astrophysics of Galaxies},
year = 2023,
month = dec,
volume = {526},
number = {3},
pages = {4107-4119},
doi = {10.1093/mnras/stad2826},
adsurl = {https://ui.adsabs.harvard.edu/abs/2023MNRAS.526.4107P}
}
Table of Contents
- What is the UCC?
- What objects are included in the UCC?
- How are member stars selected?
- How are the core values estimated?
- What is the C3 parameter?
- What is the UTI parameter?
- How is the duplicate probability estimated?
- How are objects flagged as likely not real?
- About the Search page
- About the Overview section
- About the Data section
- About the Visualization section
What is the UCC?
The acronym UCC stands for Unified Cluster Catalogue. It is the largest catalogue of open clusters in existence. It consists of comprehensive and homogeneous data for an ever expanding number of entries, taken from the latest published articles combined with data from the Gaia survey.
What objects are included in the UCC?
The UCC lists any object that was catalogued as an open cluster in the literature. This object might be classified differently in other articles (e.g.: moving group, association, etc.) but it will remain listed in the UCC because at least one article at some point indicated that it was an open cluster.
The UCC is regularly updated to include new research. If your article is not listed in the database, you can contact me with the details and I will add it as soon as possible.
How are member stars selected?
Membership is obtained through the fastMP method described in Sect. 3 of
Perren et al. (2023). The fastMP membership estimation method has been
incorporated into the ASteCA package (see details here).
If the number of stars identified with P>50% (P: membership probability) is less than 25, then the 25 stars with the highest membership probabilities are selected as members.
How are the core values estimated?
The core radius is estimated as the radius where the stellar density drops to half of the central density. Starting from the King surface-density profile,
\[\Sigma(R)=k\left[\frac{1}{\sqrt{1+(R/r_c)^2}}-\frac{1}{\sqrt{1+(r_t/r_c)^2}}\right]^2\]take the limit \(r_t \to \infty\)
\[\Sigma(R)=\frac{k}{1+(R/r_c)^2}\]With \(\Sigma_0=\Sigma(0)=k\), we obtain
\[\frac{\Sigma(R)}{\Sigma_0}=\frac{1}{1+(R/r_c)^2}\]evaluating at \(R=r_c\)
\[\Sigma(r_c)=\frac{1}{2}\Sigma_0\]The core stellar density is obtained as the number of members within the core radius divided by the area of the core.
Notice that both values are approximations since the actual core radius should be estimated by fitting a King or Plummer profile to the observed stellar density distribution.
The conversion to parsec is done using the parallax of the cluster, estimated as the median parallax of the identified members. There is a hard limit imposed to both values: the core radius is capped at 10 pc, and the density at 250 stars/pc^2. This is done to avoid outliers that can arise from very sparse clusters.
What is the C3 parameter?
The C3 parameter is the combined C1 and C2 classes, described in Sect. 4.3 of
Perren et al. (2023) where the UCC was initially introduced. The C1 and C2
classes can be described as:
C1: A density-based metric that quantifies the contrast between the spatial
distribution of cluster member stars and that of the surrounding field stars within
the five-dimensional parameter space defined by celestial coordinates,
proper motions, and parallax.
C2: A photometric metric that estimates the likelihood that the observed stellar
sequence of the candidate members is statistically indistinguishable from a
sequence randomly drawn from the field star population.
Each one takes values [A, B, C, D] where A is best and D is worst.
What is the UTI parameter?
The UTI (UCC Trust Index) is a measure of the reliability of the cluster detection,
ranging from 0 (worst) to 1 (best). It is calculated based on factors such as the number
of members, stellar density, the C3 parameter, the presence of the object
in the literature, and the probability of the object being a duplicate of a previous
entry. It is estimated via the relation:
where the \(C\) factors have values in the [0, 1] range (1 is best) representing normalized estimates of:
- \(C_N\): number of members (0=very few members, 1=many members)
- \(C_{dens}\): stellar density in pc^2 (0=very sparse object, 1=dense object)
- \(C_{C3}\): C3 parameter (0=DD class, 1=AA class)
- \(C_{lit}\): presence in literature (0=rarely mentioned in the literature, 1=frequently mentioned in the literature)
- \(C_{dup}\): likelihood of uniqueness (0=very likely a duplicate entry, 1=not a duplicate entry)
How is the duplicate probability estimated?
To estimate the probability of an object being a duplicate of a previous entry, we simply find the overlap between members. If two objects share a significant number of members, it is likely that they are the same object catalogued twice under different names.
The probability of being a duplicate is calculated as:
\[P_{dup} = max(\mathrm{shared\_members\_percent}) / 100\]where shared_members_percent is the percentage of common members between objects
(since a given object can be compared to many others, we take the maximum value found).
The object presented earlier in the literature is considered the original, while the
later one is considered the duplicate.
The P_dup value is equivalent to 1 - C_dup, where C_dup is the factor used
in the UTI calculation (see What is the UTI parameter?).
How are objects flagged as likely not real?
Objects are flagged as likely not real (or non-clusters) when they meet the following conditions:
- C_dup > 0.75 (not a duplicate)
- C_lit < 0.3 (rarely mentioned in the literature)
- UTI < 0.25 (low UTI parameter)
These dubious candidates are identified in tables by their names appearing in red. The Search page allows you to hide these objects by selecting the Hide likely non-clusters option.
About the Search page
The Search page allows the user to filter the UCC database by the stored fundamental parameter values:
- Dist: distance [pc]
- Av: visual absorption [mag]
- DAv: differential visual absorption [mag]
- Age: age [Myr]
- FeH: metallicity [dex]
- Mass: total cluster mass in [M⊙]
- Bfrac: total binary fraction
- BSS: blue stragglers (total)
Note: The value for each parameter is estimated as the median of all values compiled from the UCC literature.
Users can also filter the search results by:
- Nm: estimated number of members (stars with membership probability >50%)
- Pdup: probability of the object being a duplicate of another object
- UTI: UCC Trust Index
The Hide likely non-clusters toggle filters out objects flagged as non-clusters (e.g., asterisms, moving groups, or data artifacts). If not filtered, these objects are highlighted with their names in red within the results table.
Search Modes
Select between three modes using the button in the lower-left:
- Names: Search by specific object identifiers
- Equatorial: Search by central
[RA, Dec]coordinates - Galactic: Search by central
[Lon, Lat]coordinates
The resulting table displays all objects matching the query or located within the specified spatial proximity, up to a maximum of Nmax rows.
Interactive Map
The page includes a Map displaying the spatial distribution of results in a
[X, Y] Galactic projection. The interface supports interactive zooming and allows
users to click on individual objects to navigate to their dedicated pages.
Data Export
To download the complete UCC catalogue, visit the. Zenodo repository.
About the Overview section
The Overview section consists of two tabs: Summary and Comments. These are automatically generated for each object, based on the available data in the UCC.
Summary
Provides a brief natural language overview of the object’s characteristics and fundamental parameters. Warnings and useful badges are also included here.
Comments
Highlights any notable features found in the literature.
About the Data section
This section features two tabs displaying astrometric data and fundamental parameters for the included objects.
Astrometry
The UCC values are estimated from its identified members, the remaining values are extracted from the included literature.
Parameters
The UCC provides for each entry a table of fundamental parameters, when available. These parameters are extracted from the included literature and their median is estimated to provide a single representative value for each parameter (shown in the UCC row).
The columns DAv, Bfr, BSS correspond to differential extinction, binary fraction, and blue stragglers, respectively. BSS values can be listed as fractions or integers.
In cases where multiple articles provide values for the same parameter, these sometimes need to be transformed to maintain homogeneity. The transformations are as follows:
Metallicity
Metallicity is shown as [Fe/H]. We use the Bressan et al. (2012)
z_sun=0.0152 coefficient to transform from z values as:
Age
Ages are given in [Myr]. When ages are provided in log10(age/yr), we apply:
Absorption / Extinction
The UCC lists Av absorption. Transform coefficients are taken from
Wang & Chen (2019), Table 3.
All approximate coefficients can be estimated for example using the dust_extinction package.
About the Visualization section
There are three tabs with different plots in this section.
Members
The plots show the selected members in the UCC and also in the articles Hunt & Reffert (2023) and Cantat-Gaudin et el. (2020), when available.
Cluster region
The cluster region is shown in this tab in an interactive plot that allows exploring the spatial distribution of other clusters around the cluster. This plot uses the inverse of the median parallax of the UCC estimated members as the distance estimator.
Galactocentric position
The Sun and the Galactic center are represented by the yellow star and the black X, respectively. The spiral arms are taken from Momany et al (2006). The (X_GC, Y_GC, Z_GC) values are estimated applying a -0.02 parallax zero-point offset position of -0.02 [mas] (taken as a reasonable average, see e.g. Ding et al. 2025). The minimum accepted parallax value is 0.035 (~29 Kpc). Similar to the cluster region plot, the distance is estimated as the inverse of the median parallax of the UCC estimated members.
Radial velocities are used when available and set to 0.0 Km/s when they are not.