juliensimon/gswlc-galaxy-properties

---
license: cc-by-4.0
pretty_name: "GSWLC-2 Galaxy Properties"
language:
- en
description: "659K galaxies with stellar masses, star formation rates, and dust attenuation from UV+optical+IR SED fitting of GALEX, SDSS, and WISE photometry."
task_categories:
- tabular-classification
- tabular-regression
tags:
- space
- galaxies
- stellar-mass
- star-formation
- sdss
- galex
- wise
- astronomy
- open-data
- tabular-data
size_categories:
- 100K<n<1M
configs:
- config_name: default
data_files:
- split: train
path: data/gswlc_galaxy_properties.parquet
default: true
---

# GSWLC-2 Galaxy Properties

*Part of the [Astronomy Datasets](https://huggingface.co/collections/juliensimon/astronomy-datasets-69c24caf2f17e36128946743) collection on Hugging Face.*

**659,229** galaxies with physical properties derived from UV-to-infrared spectral energy distribution (SED) fitting. GSWLC-2 (GALEX-SDSS-WISE Legacy Catalog 2) combines ultraviolet photometry from GALEX, optical photometry from SDSS, and mid-infrared photometry from WISE to estimate stellar masses, star formation rates, and dust attenuation for galaxies at redshifts 0.01 < z < 0.30.

## Dataset description

The GSWLC is the definitive catalog for physical properties of low-redshift galaxies, covering ~90% of the SDSS spectroscopic footprint. Version 2 (Salim et al. 2018) incorporates WISE mid-IR photometry to better constrain dust-obscured star formation. The "X" variant (GSWLC-X2) is the master catalog that selects the deepest available UV observation for each galaxy from the A (shallow), M (medium), and D (deep) sub-catalogs.

Physical properties are derived using the CIGALE SED fitting code with Bayesian estimation of stellar mass, star formation rate, and dust attenuation.

## Quick stats

- **659,229** galaxies in the catalog
- **650,599** with valid stellar mass estimates
- **650,599** with valid SFR estimates
- **340,258** classified as star-forming (log sSFR > -11)
- **310,341** classified as quiescent
- Median stellar mass: **10^{10.81}** solar masses
- Median redshift: **0.1039**

## Schema

| Column | Type | Description |
|--------|------|-------------|
| `objid` | int64 | SDSS photometric object ID |
| `glxid` | int64 | GALEX photometric ID (null if no UV match) |
| `plate` | int64 | SDSS spectroscopic plate number |
| `mjd` | int64 | SDSS spectroscopic plate date (MJD) |
| `fiber_id` | int64 | SDSS spectroscopic fiber ID |
| `ra` | float64 | Right Ascension (J2000, degrees) |
| `dec` | float64 | Declination (J2000, degrees) |
| `redshift` | float64 | Spectroscopic redshift from SDSS |
| `chi2_r` | float64 | Reduced chi-squared of SED fit |
| `log_mstar` | float64 | Log stellar mass (solar masses) |
| `log_mstar_err` | float64 | Error on log stellar mass |
| `log_sfr_sed` | float64 | Log UV/optical SFR (solar masses/yr) |
| `log_sfr_sed_err` | float64 | Error on log SFR |
| `a_fuv` | float64 | Dust attenuation in rest-frame FUV (mag) |
| `a_fuv_err` | float64 | Error on A_FUV |
| `a_b` | float64 | Dust attenuation in rest-frame B band (mag) |
| `a_b_err` | float64 | Error on A_B |
| `a_v` | float64 | Dust attenuation in rest-frame V band (mag) |
| `a_v_err` | float64 | Error on A_V |
| `flag_sed` | int64 | SED fitting flag (0=OK, 1=broad-line, 2=chi2>30, 5=missing photometry) |
| `uv_survey` | int64 | UV survey depth (1=shallow/A, 2=medium/M, 3=deep/D) |
| `flag_uv` | int64 | UV detection flag (0=none, 1=FUV only, 2=NUV only, 3=both) |
| `flag_midir` | int64 | Mid-IR flag (0=none, 1=12um, 2=22um, 5=AGN-corrected) |
| `flag_mgs` | int64 | SDSS Main Galaxy Sample flag (0=no, 1=yes) |
| `log_ssfr` | float64 | Derived: log specific SFR (log SFR - log M*, yr^-1) |
| `is_star_forming` | bool | Derived: log sSFR > -11 |
| `uv_survey_name` | string | Derived: human-readable UV survey name |

## Usage

```python
from datasets import load_dataset

ds = load_dataset("juliensimon/gswlc-galaxy-properties", split="train")
df = ds.to_pandas()

# Star-forming galaxies
sf = df[df["is_star_forming"]]

# Massive quiescent galaxies
massive_quiescent = df[(df["log_mstar"] > 11) & (~df["is_star_forming"])]

# Star formation main sequence
import matplotlib.pyplot as plt
valid = df[df["log_sfr_sed"].notna() & df["log_mstar"].notna()]
plt.hexbin(valid["log_mstar"], valid["log_sfr_sed"], gridsize=100, mincnt=1)
plt.xlabel("log M* (Msun)")
plt.ylabel("log SFR (Msun/yr)")
plt.title("Star Formation Main Sequence")

# Dusty galaxies (high FUV attenuation)
dusty = df[df["a_fuv"] > 3.0]

# Cross-match with SDSS using objid
```

## Data source

[GSWLC-2](https://salims.pages.iu.edu/gswlc/) — Salim et al. (2016, 2018).

- Salim et al. (2016), "GALEX-SDSS-WISE Legacy Catalog (GSWLC): Star Formation Rates, Stellar Masses, and Dust Attenuations of 700,000 Low-Redshift Galaxies", *ApJS*, 227, 2. [arXiv:1610.00712](https://arxiv.org/abs/1610.00712)
- Salim et al. (2018), "Dust Attenuation Curves in the Local Universe: Demographics and New Laws for Star-forming Galaxies and High-redshift Analogs", *ApJ*, 859, 11. [arXiv:1804.05850](https://arxiv.org/abs/1804.05850)

## Related datasets

- [galaxy-zoo-2-morphology](https://huggingface.co/datasets/juliensimon/galaxy-zoo-2-morphology) — Galaxy Zoo 2 visual morphological classifications
- [open-ngc](https://huggingface.co/datasets/juliensimon/open-ngc) — NGC/IC galaxy and nebula catalog

## Pipeline

Source code: [juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)

## Citation

```bibtex
@dataset{gswlc_galaxy_properties,
author = {Simon, Julien},
title = {GSWLC-2 Galaxy Properties},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/gswlc-galaxy-properties},
note = {Based on GSWLC-2 data (Salim et al. 2016, ApJS 227, 2; Salim et al. 2018, ApJ 859, 11)}
}
```

## License

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)