mmu_apogee_dr17 HATS Catalog Collection

This is the collection of HATS catalogs representing mmu_apogee_dr17.

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_apogee_dr17")
# One-degree cone toward the Galactic bulge.
catalog = lsdb.open_catalog(
    "https://huggingface.co/datasets/hugging-science/mmu_apogee_dr17",
    search_filter=lsdb.ConeSearch(ra=266.4, dec=-29.0, radius_arcsec=3600.0),
)

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_apogee_dr17 — 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)
  • skymap.fits — HEALPix skymap FITS file with row-counts per HEALPix tile of fixed order 10
• mmu_apogee_dr17_10arcs/ — default margin catalog to ensure data completeness in cross-matching, the margin threshold is 10.0 arcseconds
  • ... margin catalog files and directories ...

Catalog metadata

Metadata of the main HATS catalog, excluding margins and indexes:

Number of rows	Number of columns	Number of partitions	Size on disk	HATS Builder
719,832	60	1,820	30.0 GiB	hats-import v0.7.3, hats v0.7.3

Catalog columns

The main HATS catalog contains the following columns:

Name	_healpix_29	spectrum.flux	spectrum.ivar	spectrum.mask	spectrum.lambda	ra	dec	object_id	telescope	field	nvisits	aspcapflag	starflag	snr	radial_velocity	vscatter	teff	teff_err	logg	logg_err	m_h	m_h_err	alpha_m	alpha_m_err	fe_h	fe_h_err	vmicro	vsini	j	h	k	c_fe	n_fe	o_fe	na_fe	mg_fe	al_fe	si_fe	k_fe	ca_fe	ti_fe	v_fe	cr_fe	mn_fe	co_fe	ni_fe	ce_fe	c_fe_err	n_fe_err	o_fe_err	na_fe_err	mg_fe_err	al_fe_err	si_fe_err	k_fe_err	ca_fe_err	ti_fe_err	v_fe_err	cr_fe_err	mn_fe_err	co_fe_err	ni_fe_err	ce_fe_err
Data Type	int64	list[float]	list[float]	list[int32]	list[float]	double	double	string	string	string	int32	int64	int64	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float	float
Nested?	—	spectrum	spectrum	spectrum	spectrum	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—	—
Value count	719,832	5,408,817,648	5,408,817,648	5,408,817,648	5,408,817,648	719,832	719,832	N/A	N/A	N/A	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832	719,832
Example row	3452007789738841051	[1.005, 1.026, 1.024, 1.020, … (7,514 total)]	[1.209e+04, 1.194e+04, 1.225e+04, … (7,514 total)]	[1, 1, 1, 1, … (7,514 total)]	[1.515e+04, 1.515e+04, 1.515e+04, … (7,514 total)]	312.088	-7.30775	'2M20482122-0718278'	'apo25m'	'041-28_MGA'	4	16388	16777216	129.1	16.4536	0.0376212	4962.49	15.748	4.54971	0.0196983	0.18052	0.00649099	0.0071703	0.058809	0.17915	0.00694241	0.366834	1.50945	11.42	10.96	10.884	0.011954	0.060851	-0.0503501	0.0848236	0.0574575	0.060866	0.024755	0.0400929	-0.027554	0.016182	-0.240591	0.00601831	0.0623625	nan	0.05523	nan	0.0169887	0.0253632	0.0204161	0.0686432	0.0137818	0.0211988	0.0145771	0.0271501	0.0124386	0.0339361	0.0647775	0.036684	0.0131672	nan	0.0129922	nan
Minimum value	1.43e+12	N/A	N/A	N/A	N/A	0.000103	-87.22	N/A	N/A	N/A	1	0	0	0.1117	-1042	-0	3088	2.461	-0.4828	0.01032	-2.469	0.003098	-0.7144	0.001972	-2.469	0.003032	0.3	1.5	5.357	4.835	4.384	-2.57	-2.485	-1.998	-2.827	-1.71	-2.566	-1.392	-2.832	-2.744	-2.089	-2.652	-3.24	-3.103	-2.981	-2.931	-2.511	0.005958	0.00681	0.005896	0.01759	0.007364	0.01028	0.008938	0.01106	0.006475	0.01019	0.01937	0.01213	0.005199	0.0127	0.005889	0.02716
Maximum value	3.459e+18	N/A	N/A	N/A	N/A	360	87.61	N/A	N/A	N/A	69	3.3e+12	6.083e+07	4469	1049	1409	1.987e+04	1.128e+04	5.391	6.477	0.9686	4.834	0.9954	19.25	0.9579	1.469	4.8	96.01	18.42	24	16.89	1.422	2.791	2.04	2.847	1.872	2.931	1.827	3.369	2.138	1.66	3.326	3.31	3.088	3.374	2.982	2.913	67.71	1586	9.281	2.402	4.857	0.7003	8.129	1.092	7.991	6.628	1.838	1.906	1.778	2.118	1.736	3.047

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

"Value count" may be different from the total number of rows for nested columns: each nested element is counted as a single value.

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_apogee_dr17 = lsdb.open_catalog("https://huggingface.co/datasets/hugging-science/mmu_apogee_dr17")
other = lsdb.open_catalog("https://huggingface.co/datasets/<org>/<other_catalog>")

crossmatched = mmu_apogee_dr17.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 the APOGEE survey (Apache Point Observatory Galactic Evolution Experiment), part of SDSS-IV, specifically Data Release 17 — the final SDSS-IV release. APOGEE is a near-infrared (H-band, 1.51–1.70 μm), high-resolution (R ≈ 22,500) spectroscopic survey of ∼700,000 stars across the Milky Way, obtained with twin 300-fibre spectrographs on the SDSS 2.5 m telescope at Apache Point (apo25m) and the du Pont 2.5 m at Las Campanas (lco25m).

Data modality The dataset is multimodal and includes combined-visit stellar spectra cropped to the three active-detector regions (7,514 pixels with the inter-chip gaps removed), 2MASS J/H/K photometry, ASPCAP-derived atmospheric parameters (effective temperature, surface gravity, microturbulence, projected rotational velocity, overall metallicity [M/H] and α-enhancement [α/M]), individual elemental abundances [X/Fe] for 16 species (C, N, O, Na, Mg, Al, Si, K, Ca, Ti, V, Cr, Mn, Co, Ni, Ce), heliocentric radial velocities and visit-to-visit scatter, signal-to-noise, and the ASPCAPFLAG / STARFLAG quality bitmasks.

Typical use cases APOGEE DR17 is the standard reference catalogue for Milky Way chemodynamics: chemical tagging of stellar populations, mapping abundance gradients across the disc and bulge, age–metallicity relations via [C/N]-based asteroseismic calibration, characterising stellar streams, and training data-driven spectroscopic models (The Cannon, The Payne, neural-network emulators) and self-supervised foundation models on stellar spectra.

Caveats Each spectrum.flux / ivar / mask / lambda list has 7,514 elements: the full 8,575-pixel combined-detector grid has been cropped to the active blue/green/red chip regions and the inter-chip gaps dropped. Abundance and parameter columns hold NaN where the ASPCAP pipeline did not return a measurement; the per-bit meanings of aspcapflag and starflag are documented at sdss.org/dr17/algorithms/bitmasks. Sky coverage is heavily concentrated along the Galactic plane and bulge (see skymap above) rather than uniform on the sphere.

Citation Users should cite the SDSS DR17 paper, Abdurro'uf et al. 2022, ApJS 259, 35, and acknowledge SDSS-IV/APOGEE-2 (instrument: Wilson et al. 2019; ASPCAP pipeline: García Pérez et al. 2016). Acknowledgements text is provided at sdss.org/collaboration/citing-sdss.