Mission Overview

Physics at High Angular resolution in Nearby GalaxieS - Catalogs ("PHANGS-CAT")

NOTE: For HST imaging products, see the PHANGS-HST HLSP page.

Primary Investigator: Janice C. Lee

HLSP Authors: Sinan Deger, Kirsten Larson, Janice C. Lee, Jimmy Lilly, Dave Thilker, Brad Whitmore

Released: 2022-02-08

Updated: 2022-02-08

Primary Reference(s): Lee et al. 2022Thilker et al. 2022

DOI: 10.17909/jray-9798

Citations: See ADS Statistics

Read Me

Map showing locations of Class 1 and Class 2 clusters, color-coded by age.
Location of Class 1 and 2 Clusters In NGC 1566

From Thilker et al. (2022), map of machine learning classified clusters in NGC 1566, color coded by age.  Spiral arms are indicated.

Clusters in NGC 3627 overlaid on an image of the galaxy.
NGC 3627 Clusters and Associations

Visualization of multi-scale associations and compact clusters in NGC 3627, detected using the methods of Thilker et al. (2022) and Larson et al. (2022, in prep.)

Plot of log(Mass) vs. log(Age) for clusters in NGC 1566.
Age-Mass Diagram for NGC 1566

From Thilker et al. (2022), age-mass diagram for Class 1 and 2 clusters in NGC 1566 confirmed with machine learning or human classification.

UB-VI-M_20 three-dimensional plot of an isolated cluster and compact association sample.
UB-VI-M_20 Diagram From Deger et al 2022

From Deger et al. (2022), The UB-VI-M_{20} diagram of a bright, isolated cluster and compact association sample.

Four-panel plot showing evolutionary tracks of synthetic population models compared to Class 1/2/3 structures.
Evolutionary Tracks Compared To Class 1/2/3 Structure

From Deger et al. (2022), NUVB-VI (top) and UB-VI (bottom) evolutionary tracks of various synthetic stellar population models compared to class 1/2/3 structures.

Plots showing age-mass and age-reddening for NGC 3351 compared to associations of Class 1/2/3 sources.
Age-Mass and Age-Reddening Diagrams For NGC 3351 From 16-pc-scale Associations

From Larson et al. (2022, in prep.), age-mass and age-reddening diagrams for NGC 3351 comparing 16 pc scale associations to LEGUS Class 1, 2, 3 sources.

Associations at different scales overlaid on a CO(2-1) map from ALMA for NGC 1566.
Associations In NGC 1566 at Three Scales Compared To ALMA CO(2-1) Emission

From Larson et al. (2022, in prep.), multi-scale associations in a subsection of NGC 1566 (16, 32, 64 pc scale) compared to the distribution of CO(2-1) emission mapped by ALMA.

Color-color diagrams for Class 1 and 2 clusters for machine learning compared to human classifications.
Color-Color Diagrams Of Class 1 and 2 Clusters From Human vs. Machine Learning

From Whitmore et al. (2021), color-color diagrams for Class 1 and 2 clusters obtained via human classification (PHANGS, LEGUS, upper panels) and machine learning classification (ResNet and VGG by PHANGS, lower panels).  The plot demonstrates machine learning classification outcomes are similar to those of human-based efforts.


The PHANGS program is building the first dataset to enable the multi-phase, multi-scale study of star formation across nearby spiral galaxies, by combining Atacama Large Millimeter/submillimeterArray (ALMA) CO(2-1) mapping, Very Large Telescope/Multi Unit Spectroscopic Explorer (VLT/MUSE) optical spectroscopy, and Hubble Space Telescope (HST) UV-optical imaging.  Here, the team provides catalog products from the PHANGS survey, the HST portion of which is obtaining five band NUV-U-B-V-I imaging of the disks of 38 spiral galaxies at distances of 4-23 Mpc, and parallel V and I band imaging of their halos, to provide a census of tens of thousands of compact star clusters and associations.  The combination of HST, ALMA, and VLT/MUSE observations will yield an unprecedented joint catalog of the observed and physical properties of ~100,000 star clusters, associations, HII regions, and molecular clouds.

The PHANGS-CAT products (Version 1) provide compact cluster (CLUST) catalogs and multi-scale association (ASSOC) catalogs for five galaxies observed with HST as part of the PHANGS-HST project.  The corresponding HST imaging products are available as the PHANGS-HST High Level Science Products. The team provides two kinds of compact cluster catalogs (CLUST products): human classified and machine learning (ML) classified. Only class 1 (symmetric, centrally concentrated, radial profile extended relative to point source), and class 2 (asymmetric, centrally concentrated, radial profile extended relative to point source) clusters are included. The multi-scale stellar associations (ASSOC products) are determined using both the HST NUV (F275W) and V-band (F555W) to independently delineate the stellar associations in each galaxy. The associations are calculated at three or four physical scale levels depending on the distance of the galaxy. The physical scale levels of the associations correspond to (8pc, 16pc, 32pc,64pc), or (16pc, 32pc, 64pc) for more distant galaxies.  A flowchart of the entire cluster and multi-scale association processing pipeline (from Lee et al. 2022) illustrates the entire process.

Complementary PHANGS Data From ALMA and MUSE

30 April 2021: Associated ALMA and MUSE data are now publicly available though the PHANGS CANFAR site or https://sites.google.com/view/phangs/home/data.

DR1 - Image Release 1 (Jan. 11, 2021)

Primary Source: Visit the PHANGS-HST HLSP page for references and documentation.

Provides image products for observations in the NUV-U-B-V-I bands for 29 out of 38 PHANGS-HST galaxies.

DR2 - Image Release 2 (August 09, 2021)

Primary Source: Visit the PHANGS-HST HLSP page for references and documentation.

Completes the image product release for the full PHANGS-HST sample.  It also provides convolutional neural network (CNN) models trained with deep transfer learning, for the classification of star clusters.

DR3 - Catalog Release 1 (February 08, 2022)

A set of catalogs for compact clusters (CLUST) and a separate set of catalogs for multi-scale associations for five galaxies (NGC 1433, NGC 1566, NGC 1559, NGC 3351, NGC 3627) are being released as part of DR3. These are described at different levels of detail in Lee et al. 2022, Whitmore et al. 2021, Thilker et al. 2022, and Larson et al. 2022 (in prep). Published papers using these datasets include Wei et al. 2020, Turner et al. 2021Whitmore et al. 2021, Lee et al. 2022, Deger et al. 2022, and Thilker et al. 2022.

Data Products

The catalog files have the following naming convention:



  • <ins> = Name of instrument, either "wfc3" or "acs-wfc3" if both instruments used in the catalog
  • <target> = Name of galaxy, e.g., "ngc1566"
  • <filter> = "multi", since all the catalogs include data from several filters
  • <version> = "v1" for version 1 of the catalogs
  • <type> = Type of file:
    • For CLUSTER files:
      • one of "cluster-human-class12.fits" or "cluster-ml-class12.fits"
    • For ASSOC files:
      • At the galaxy level:
        • "assoc-brightstarmask.reg" = circular regions used to eliminate objects around bright stars
      • At the association level:
        • "assoc-<selection band>-<scalepc>-main.fits" = main catalog
        • "assoc-<selection band>-<scalepc>-idmask.fits" = segmentation map
        • "assoc-<selection band>-<scalepc>-peaks.reg" = position of the peak flux in each region
        • "assoc-<selection band>-<scalepc>-region.reg" = DS9 region file showing outline of the region
      • where:
        • <selection band> = wavelength band used for selection, one of "nuvselect" (F275W) or "vselect" (F555W)
        • <scalepc> = the size scale in parsecs used to form associations (one of "ws8pc", "ws16pc", "ws32pc", or "ws64pc", where "ws" stands for "watershed method"

Data file types:


human-classified cluster catalog, includes class 1 and class 2 clusters


machine-learning-classified cluster catalog, includes class 1 and class 2 clusters


bright star mask region file


multi-scale associaton catalog


segmentation map, values correspond to "ID" in the main catalog


region file containing peak flux positions


region file containing outline of the region

Data Access

See the PHANGS-HST HLSP page for images and more information about individual galaxies.

Download tar files containing every CLUSTER or ASSOC product across all available galaxies in the table below.  A copy of the README files are included in each tar file, but are also linked on their own for direct access in the table below.  The README files explain the catalog columns in more detail, and also include information about the galactic extinction tables, aperture corrections, and assumed distances.


ASSOC Bundle

Catalogs: hlsp_phangs-cat_hst_wfc3_all_multi_v1_cc-cats.tar.gz

Catalogs: hlsp_phangs-cat_hst_wfc3_all_multi_v1_assoc-cats.tar.gz

README: hlsp_phangs-cat_hst_wfc3_all_multi_v1_compact-cluster-readme.txt

README: hlsp_phangs-cat_hst_wfc3_all_multi_v1_assoc-readme.txt


Download tar files containing every CLUSTER or ASSOC product for a specific galaxy using the table below.


Distance (Mpc)


NGC 1433

8.3 hlsp_phangs-cat_hst_wfc3_ngc1433_multi_v1_cc-cats.tar.gz hlsp_phangs-cat_hst_wfc3_ngc1433_multi_v1_assoc-cats.tar.gz

NGC 1559

19.4 hlsp_phangs-cat_hst_wfc3_ngc1559_multi_v1_cc-cats.tar.gz hlsp_phangs-cat_hst_wfc3_ngc1559_multi_v1_assoc-cats.tar.gz

NGC 1566

18.0 hlsp_phangs-cat_hst_wfc3_ngc1566_multi_v1_cc-cats.tar.gz hlsp_phangs-cat_hst_wfc3_ngc1566_multi_v1_assoc-cats.tar.gz

NGC 3351

10.0 hlsp_phangs-cat_hst_wfc3_ngc3351_multi_v1_cc-cats.tar.gz hlsp_phangs-cat_hst_wfc3_ngc3351_multi_v1_assoc-cats.tar.gz

NGC 3627

10.1 hlsp_phangs-cat_hst_wfc3_ngc3627_multi_v1_cc-cats.tar.gz hlsp_phangs-cat_hst_wfc3_ngc3627_multi_v1_assoc-cats.tar.gz


Information about the source catalogs below may be of interest (this information is also available in the CLUSTER catalog README file.)






















8.3 29.6 24.1 -5.5 90 101 25.2 -4.4 209 132


19.4 31.4 23.5 -7.9 419 301 25.7 -5.8 659 494


18.0 31.3 23.5 -7.8 379 271 25.2 -6.1 796 398


10.0 30.0 24.0 -6.0 137 166 25.4 -4.6 439 246


10.1 30.0 22.5 -7.5 462 312 24.9 -5.2 1725 820



Please remember to cite the appropriate paper(s) below and the DOI if you use these data in a published work. 

Note: These HLSP data products are licensed for use under CC BY 4.0.