J/ApJ/950/163 PANDISC. I. HI model applied to ALFAFA & SPARC gal. (Peng+, 2023)
Parameterized Asymmetric Neutral hydrogen Disk Integrated Spectrum
Characterization (PANDISC).
I. Introduction to a physically motivated HI model.
Peng Bo, Haynes M.P., Ball C.J., Jones M.G.
<Astrophys. J., 950, 163 (2023)>
=2023ApJ...950..163P 2023ApJ...950..163P
ADC_Keywords: H I data; Models; Surveys; Galaxies, radio; Velocity dispersion
Keywords: H I line emission ; Astronomy data analysis ; Galaxy dynamics ;
Galaxy rotation
Abstract:
Modeling the integrated HI spectra of galaxies has been a difficult
task, due to their diverse shapes, but more dynamical information is
waiting to be explored in HI line profiles. Based on simple
assumptions, we construct a physically motivated model for the
integrated HI spectra: Parameterized Asymmetric Neutral hydrogen Disk
Integrated Spectrum Characterization (PANDISC). The model shows great
flexibility in reproducing the diverse HI profiles. We use Monte Carlo
Markov Chain for fitting the model to global HI profiles and produce
statistically robust quantitative results. Comparing with several
samples of HI data available in the literature, we find the
model-fitted widths agree with cataloged velocity widths (e.g., W50)
down to S/N≲6. While dynamical information can only be extracted
reliably from spectra with S/N>8. The model is also shown to be useful
for applications like the baryonic Tully-Fisher relation (BTFR) and
profile-based sample control. By comparing the model parameter vr to
vflat, we uncover how the HI width is affected by the structure of the
rotation curve, following a trend consistent with the difference in
the BTFR slope. We also select a sample of spectra with broad
wing-like features suggestive of a population of galaxies with unusual
gas dynamics. The PANDISC model bears both promise and limitations for
potential use beyond HI lines. Further application on the whole
ALFALFA sample will enable us to perform large-scale ensemble studies
of the HI properties and dynamics in nearby galaxies.
Description:
In this paper, we introduce PANDISC (Parameterized Asymmetric Neutral
hydrogen Disk Integrated Spectrum Characterization), a physically
motivated parameterized model for global HI line profiles. The model
is based on simple physical assumptions that combine an asymmetric
corotating disk component with a Gaussian component. The model
consists of seven parameters, with five of them controlling the shape
of the profile (vr, k, vsigma, vg and r) and the other two setting the
line center (vc) and total flux (F). See Section 2.
The Arecibo Legacy Fast ALFA Survey (ALFALFA; Haynes+ 2018, J/ApJ/861/49)
produced a final catalog (a.100) of ∼31500 extragalactic HI detections
in the local Universe. For the purposes of demonstration and
application, we selected subsamples from the ALFALFA data and
literature as described in Section 3.1.
To demonstrate the application of the line model in Baryonic
Tully-Fisher relation (BTFR) and sample control, we applied it on the
galaxy sample selected in Papastergis+ (2016A&A...593A..39P 2016A&A...593A..39P).
To test the physical assumption of the "projected corotating velocity"
for vr, we apply the line model to the galaxies with rotation curves
presented in Lelli+ (2016, J/AJ/152/157), hereafter known as the SPARC
galaxies. We did an extensive literature search for integrated,
single-dish HI observations for the SPARC sample. A total of 158
galaxies with HI spectra available were cross-matched with sample.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table2.dat 266 911 Model fitted parameter of all the spectra used in
the paper
--------------------------------------------------------------------------------
See also:
II/276 : The SDSS Photometric Catalog, Release 5 (Adelman-McCarthy+, 2007)
VIII/77 : HI spectral properties of galaxies (Springob+, 2005)
J/A+A/3/765 : HI synthesis observations in UMa cluster (Verheijen+, 2001)
J/AJ/128/16 : HIPASS Bright Galaxy Catalog (Koribalski+, 2004)
J/ApJ/669/821 : CO Tully-Fisher relation for host galaxies of QSOs (Ho+, 2007)
J/AJ/138/1938 : EDD: All Digital HI profile catalog (Courtois+, 2009)
J/A+A/532/A117 : AMIGA VIII. Flux ratio asymmetry parameter (Espada+, 2011)
J/AJ/152/157 : Mass models for 175 disk galaxies with SPARC (Lelli+, 2016)
J/ApJ/861/49 : ALFALFA extragalactic HI source catalog (Haynes+, 2018)
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 16 A16 --- Name Common name used in paper and figure set
(UGC for numbers<12943 or AGC numbers)
18- 38 A21 --- AName Alternative name in sample or reference paper
40- 53 A14 --- Sample Sample in paper using the spectrum (1)
55- 60 A6 --- Ref Reference of integrated HI spectra (2)
62- 67 F6.2 km/s vr [5/451] Model projected co-rotating velocity
69- 74 F6.2 km/s E_vr [0.01/104] Upper uncertainty in vr
76- 81 F6.2 km/s e_vr [0.01/218] Lower uncertainty in vr
83- 88 F6.3 --- k [-0.64/0.64] Model gradient of angular
density
90- 94 F5.3 --- E_k [0/1.2] Upper uncertainty in k
96- 100 F5.3 --- e_k [0/0.9] Lower uncertainty in k
102- 106 F5.2 km/s vsigma [3/25] Model velocity dispersion
108- 112 F5.2 km/s E_vsigma [0/15.3] Upper uncertainty in vsigma
114- 118 F5.2 km/s e_vsigma [0/13] Lower uncertainty in vsigma
120- 125 F6.2 km/s vg [8.5/200] Model std of gaussian component
127- 132 F6.2 km/s E_vg [0.01/105] Upper uncertainty in vg
134- 139 F6.2 km/s e_vg [0/104] Lower uncertainty in vg
141- 145 F5.3 --- r [0.059/1] Model disk flux fraction
147- 151 F5.3 --- E_r [0/0.54] Upper uncertainty in r
153- 157 F5.3 --- e_r [0/0.72] Lower uncertainty in r
159- 165 F7.2 Jy.km/s F [0.3/2055] Model integrated line flux
167- 171 F5.2 Jy.km/s E_F [0.05/13.3] Upper uncertainty in F
173- 177 F5.2 Jy.km/s e_F [0.04/13.2] Lower uncertainty in F
179- 186 F8.2 km/s vc [-233/17469] Model heliocentric velocity of
line center
188- 192 F5.2 km/s E_vc [0.01/92] Upper uncertainty in vc
194- 199 F6.2 km/s e_vc [0.01/132] Lower uncertainty in vc
201- 206 F6.2 km/s W50mod [21/937] Model based W50 (the width at 50% of
the peak flux)
208- 213 F6.2 km/s E_W50mod [0.02/192] Upper uncertainty in W50mod
215- 220 F6.2 km/s e_W50mod [0.02/179] Lower uncertainty in W50mod
222- 227 F6.1 --- SRD [3/3650] Model based Square Root Deviance
229- 234 F6.2 --- rmsBl [1.2/118] RMS of blank (line-free) channels
236- 241 F6.2 --- rmsRes [1.3/630] RMS of residual spectrum
243- 246 A4 --- Fqual Flagged by low model fitting quality
248- 251 A4 --- Fasym Flagged by asymmetry
253- 256 A4 --- Fdisk Flagged by low disk fit quality
258- 261 A4 --- FW50 Flagged by W50 to W50_model>2σ
discrepancy
263- 266 A4 --- FBroad Flagged as a broad wing candidate
--------------------------------------------------------------------------------
Note (1): Sample as follows:
P16 = Papastergis+ 2016A&A...593A..39P 2016A&A...593A..39P (93+4 occurrences)
SPARC = Lelli+ 2016, J/AJ/152/157 (135+23 occurrences)
demo = ALFALFA demonstration sample (296+4 occurrences)
high S/N = ALFALFA high S/N sample (356+31 occurrences)
Note (2): Reference codes as follows:
a.100 = ALFALFA; Haynes+ 2018, J/ApJ/861/49 (804 occurrences)
S05 = Springob+ 2005, VIII/77 (56 occurrences)
T88 = Tifft & Cocke, 1988ApJS...67....1T 1988ApJS...67....1T (27 occurrences)
EDD = Courtois+ 2009, J/AJ/138/1938 (11 occurrences)
HIPASS = Koribalski+ 2004, J/AJ/128/16 (10 occurrences)
R80 = Rots A.H. 1980A&AS...41..189R 1980A&AS...41..189R (3 occurrences)
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 19-Aug-2025