mmu_manga HATS Catalog Collection

This is the collection of HATS catalogs representing mmu_manga.

This dataset is part of the Multimodal Universe, a large-scale collection of multimodal astronomical data. For full details, see the paper: The Multimodal Universe: Enabling Large-Scale Machine Learning with 100TBs of Astronomical Scientific Data.

Access the catalog

We recommend the use of the LSDB Python framework to access HATS catalogs. LSDB can be installed via pip install lsdb or conda install conda-forge::lsdb, see more details in the docs. The following code provides a minimal example of opening this catalog:

import lsdb

# Full sky coverage.
catalog = lsdb.open_catalog("https://huggingface.co/datasets/hugging-science/mmu_manga")

Each catalog in this collection is represented as a separate Apache Parquet dataset and can be accessed with a variety of tools, including pandas, pyarrow, dask, Spark, DuckDB.

File structure

This catalog is represented by the following files and directories:

• collection.properties — textual metadata file describing the HATS collection of catalogs
• mmu_manga — main HATS catalog directory
  • dataset/ — Apache Parquet dataset directory for the main catalog
    • ... parquet metadata and data files in sub directories ...
  • hats.properties — textual metadata file describing the main HATS catalog
  • partition_info.csv — CSV file with a list of catalog HEALPix tiles (catalog partitions)

Catalog metadata

Metadata of the main HATS catalog:

Number of rows	Number of columns	Number of partitions	Size on disk	HATS Builder
10,735	41	3,705	1.05 GiB	hats-import v0.9.0, hats v0.9.0

Catalog columns

The main HATS catalog contains the following columns:

Name	_healpix_29	z	spaxel_size	spaxel_size_units	spaxels.flux	spaxels.ivar	spaxels.mask	spaxels.lsf	spaxels.lambda	spaxels.x	spaxels.y	spaxels.spaxel_idx	spaxels.flux_units	spaxels.lambda_units	spaxels.skycoo_x	spaxels.skycoo_y	spaxels.ellcoo_r	spaxels.ellcoo_rre	spaxels.ellcoo_rkpc	spaxels.ellcoo_theta	spaxels.skycoo_units	spaxels.ellcoo_r_units	spaxels.ellcoo_rre_units	spaxels.ellcoo_rkpc_units	spaxels.ellcoo_theta_units	images.filter	images.flux	images.flux_units	images.psf	images.psf_units	images.scale	images.scale_units	ra	dec	maps.group	maps.label	maps.flux	maps.ivar	maps.mask	maps.array_units	object_id
Data Type	int64	double	double	string	fixed_size_list<element: fixed_size_list<element: float>[4563]>[1]	fixed_size_list<element: fixed_size_list<element: float>[4563]>[1]	fixed_size_list<element: fixed_size_list<element: int32>[4563]>[1]	fixed_size_list<element: fixed_size_list<element: float>[4563]>[1]	fixed_size_list<element: fixed_size_list<element: float>[4563]>[1]	int64	int64	int64	string	string	float	float	float	float	float	float	string	string	string	string	string	string	fixed_size_list<element: fixed_size_list<element: float>[96]>[96]	string	fixed_size_list<element: fixed_size_list<element: float>[96]>[96]	string	float	string	double	double	string	string	fixed_size_list<element: fixed_size_list<element: float>[96]>[96]	fixed_size_list<element: fixed_size_list<element: float>[96]>[96]	fixed_size_list<element: fixed_size_list<element: float>[96]>[96]	string	string
Nested?	—	—	—	—	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	spaxels	images	images	images	images	images	images	images	—	—	maps	maps	maps	maps	maps	maps	—

"Nested" indicates whether the column is stored as a nested field inside another "struct" column.

Crossmatch with another catalog

HATS catalogs can be efficiently crossmatched using LSDB, which leverages the HEALPix partitioning to avoid loading the full datasets into memory:

import lsdb

mmu_manga = lsdb.open_catalog("https://huggingface.co/datasets/hugging-science/mmu_manga")
other = lsdb.open_catalog("https://huggingface.co/datasets/<org>/<other_catalog>")

crossmatched = mmu_manga.crossmatch(other, radius_arcsec=1.0)
print(crossmatched)

See the LSDB documentation for more details on crossmatching and other operations.

Dataset-specific context

Original survey This dataset is based on MaNGA (Mapping Nearby Galaxies at Apache Point Observatory), one of the three core programs of SDSS-IV. MaNGA used fiber-bundle integral field units (IFUs) on the SDSS 2.5m telescope to obtain spatially resolved optical spectroscopy for a sample of roughly 10,000 nearby galaxies, selected to give a roughly flat distribution in stellar mass.

Data modality Each row represents one galaxy and bundles together its integral field spectroscopy and imaging: a spaxels array giving, for every spatial element of the IFU footprint, the calibrated flux, inverse variance, mask, line-spread function, and wavelength solution across ~4563 spectral elements, along with its sky and elliptical (deprojected) coordinates; an images array with multi-band (4-filter) broadband image cutouts and PSF images on a 96×96 pixel grid; and a maps array of derived 96×96 pixel-grid quantities (e.g. kinematic and emission-line maps) grouped by analysis type. Basic catalog-level quantities (ra, dec, z, spaxel_size) are also included.

Typical use cases Spatially-resolved spectroscopic data of this kind has been used for machine learning applications such as estimating galaxy properties (e.g. star formation rate, metallicity, stellar population parameters) from spectra, spectral classification, and learning representations of resolved galaxy structure.

Caveats The dataset includes only objects with successfully reduced MaNGA data products. No further scientific processing is applied beyond the standard pipeline reduction, so spaxels affected by foreground stars, low signal-to-noise, or instrumental artifacts should be filtered using the provided mask arrays.

Citation When using this dataset in scientific publications, users should cite the MaNGA survey papers (Bundy et al. 2015) and the relevant SDSS-IV/SDSS-V data release paper, and follow the official SDSS acknowledgment guidelines: https://www.sdss.org/collaboration/citing-sdss/