ar
X

iv
:1

21
1.

29
35

v1
  [

as
tr

o-
ph

.S
R

] 
 1

3 
N

ov
 2

01
2

Mon. Not. R. Astron. Soc. 000, 1–?? (2002) Printed 14 November 2012 (MN LATEX style file v2.2)

A comprehensive study of NGC 2023 with XMM-Newton

and Spitzer ⋆

M. A. López-Garćıa1†, J. López-Santiago1, J. F. Albacete-Colombo2,

P. G. Pérez-González1,3 and E. de Castro1
1Departamento de Astrof́ısica y Ciencias de la Atmósfera, Universidad Complutense de Madrid, E-28040 Madrid, Spain
2Centro Universitario Regional Zona Atlántica (CURZA) Universidad Nacional del COMAHUE, Monseñor Esandi y Ayacucho (8500),
Viedma (Rio Negro), Argentina.
3Associate Astronomer at Stward Observatory, The University of Arizona.

Accepted .... Received ...

ABSTRACT

Nearby star-forming regions are ideal laboratories to study high-energy emission
of different stellar populations, from very massive stars to brown dwarfs. NGC 2023 is
a reflection nebula situated to the south of the Flame Nebula (NGC 2024) and at the
edge of the H ii region IC 434, which also contains the Horsehead Nebula (Barnard 33).
NGC 2023, NGC 2024, Barnard 33 and the surroundings of the O-type supergiant star
ζ Ori constitute the south part of the Orion B molecular complex. In this work, we
present a comprehensive study of X-ray emitters in the region of NGC 2023 and its
surroundings. We combine optical and infrared data to determine physical properties
(mass, temperature, luminosity, presence of accretion disks) of the stars detected in
an XMM-Newton observation. This study has allowed us to analyze spectral energy
distribution of these stars for the first time and determine their evolutionary stage.
Properties of the X-ray emitting plasma of these stars are compared to those found
in other nearby star-forming regions. The results indicate that the stars that are
being formed in this region have characteristics, in terms of physical properties and
luminosity function, similar to those found in the Taurus-Auriga molecular complex.

Key words: (Galaxy:) open clusters and associations: general – stars: pre-main-
sequence – stars: coronae – X-rays: stars

1 INTRODUCTION

Star-forming regions are the best laboratories to study the
different physical processes that give rise to X-ray emission.
X-ray photons are little absorbed by interstellar material
and young stars show higher levels of X-ray emission than
main-sequence stars. High-energy radiation plays an impor-
tant role in the development and evolution of stellar pro-
toplanetary disks (Bally et al. 2000; Scally & Clarke et al.
2001) and planetary atmospheres (e.g. Penz et al. 2008) and
contributes to the disruption of star formation by causing

⋆ This publication makes use of data products from the Two Mi-
cron All Sky Survey, which is a joint project of the University
of Massachusetts and the Infrared Processing and Analysis Cen-
ter/California Institute of Technology, funded by the National
Aeronautics and Space Administration and the National Science
Foundation. This research has also made use of the Centre de
Donnes Astronomiques de Strasbourg (CDS) tool Aladin.
† E-mail: mal@astrax.fis.ucm.es

molecular material blow out. Outflows, jets and winds are
the main mechanisms to modify the molecular cloud en-
vironment, except for supernova explosions. They displace
cloud material but also trigger star formation.

The Orion giant molecular complex (see Fig. A1) con-
sists of two main large structures, usually referred to as
Orion A and B, and a number of less massive filaments (see
Genzel & Stutzki et al. 1989, for a detailed description). The
Orion A molecular cloud contains the Orion Nebula Cluster
(ONC) which is the closest OB association to the sun (∼ 450
pc). Because of its proximity, the ONC is one of the most
visited targets for numerous studies, including the relation
between cloud kinematics and star formation (see Hartman
& Burkert et al. 2007, and references therein), protostellar
and circumstellar disk characteristics (e.g. Fang et al. 2009;
Ingleby et al. 2009; Wilson et al. 2011) and the properties of
the X-ray emission of young stellar objects and T Tauri stars
(see Preibisch et al. 2005, and the rest of the articles of the
Chandra Orion Ultradeep Project (COUP) in the special is-
sue of the ApJS). An extensive study of Orion A combining

http://arxiv.org/abs/1211.2935v1


2 M. A. López-Garćıa et al.

data from the XMM-Newton and Spitzer missions is just be-
ing carried out (see preliminary results in Wolk et al. 2010a;
Pillitteri et al. 2010).

Equally interesting for studies on stellar formation is
the Orion B molecular cloud. The complex extends roughly
south-north and contains the Horsehead Nebula, NGC 2023,
NGC 2024, and several other reflection nebulae, such as
NGC 2068 and NGC 2071. The southern part of Orion B
(L 1630) borders the large H ii region IC 434, which is ex-
panding into the molecular cloud. The interface between the
molecular cloud and the H ii region (IC 434) is seen as a
bright ridge of glowing gas with the Horsehead Nebula and
several smaller pillars (Mookerjea et al. 2009). The Horse-
head Nebula points directly towards the multiple system σ
Ori, which is ionizing IC 434. At a distance of 350–450 pc,
the Horsehead Nebula is the closest pillar to the Sun and
represents an ideal laboratory to study the emergence and
evolution of X-ray emission in very young stars, their X-ray
properties and the influence of X-ray emission and its vari-
ability on the heating and evolution of proto-planetary disks
(see infrared studies by Bowler et al. 2009; Mookerjea et al.
2009).

NGC 2023 is located about 15 arcmin to the north-
east of the pillar. In contrast to NGC 2024, that contains
an embedded stellar cluster with more than 200 members
and several protostar candidates (Haisch et al. 2000; Skin-
ner et al. 2003), NGC 2023 shows low stellar density. Some
radio clumps have been detected using the VLA (Anglada
& Rodŕıguez et al. 2002; Reipurth et al. 2004). One of them,
the source NGC 2023 MM 1, is a confirmed (very cold) Class
0 source driving a large and well collimated molecular out-
flow (Sandell et al. 1999). Three Herbig-Haro objects were
identified by Malin et al. (1987): [MOW87] HH1, HH2, and
HH3. Mookerjea et al. (2009) associated HH3 with the clas-
sical T Tauri star V615 Ori, but the association of HH1 and
HH2 with a detected source is more uncertain. The authors
associated HH2 to the mid-infrared source MIR-51, while
HH1 could be associated to the radio source NGC 2023
MM3, the mid-infrared source MIR-62 or even the near-
infrared source NIR-15, merely at 40 arcsec from HH1.

The age of NGC 2023 is not well established. Alcala et
al. (2000) derived ages in the range 0.5 < τage < 7.0 Myr
for stars in the L 1641 and L 1630 clouds with an X-ray
counterpart. For NGC 2024, Eisner & Carpenter (2003) de-
termined an age of 0.3 Myr from the disk frequency, while
Ali et al (1998) proposed 0.5 Myr as the age of this cluster,
based in infrared photometry. The coexistence of both radio
clumps and stars without a protoplanetary disk in the re-
gion of NGC 2023 (e.g. Mookerjea et al. 2009) suggests that
star formation has taken place during several million years,
although this hypothesis should be tested robustly.

In this work, we present for the first time a compre-
hensive study of the X-ray sources in the star-forming re-
gion NGC 2023 and its surroundings. We give general X-ray
emission characteristics of the sources (temperature, column
density and emission measure) and compare them with the
results obtained for the nearby regions Taurus (Güdel et
al. 2007) and Orion Nebula Cluster (ONC; Preibisch et al.
2005). We complete a multi-wavelength study of the X-ray
detected sources using data from the literature. In addition,
archival observations from the Spitzer infrared observatory
are analyzed. We then classify the X-ray sources into classes

Table 1. NGC 2023 XMM-Newton observation details on each
EPIC detector.

PN MOS1 MOS2

Exp. time (ks) 25.09 29.13 29.12
Effective exp. time (ks) 13.60 15.00 15.80

Filter Medium Thin Medium
Obs. mode Extented Full Frame Full Frame Full Frame

Figure 1. Lightcurve of the whole EPIC-PN detector at E >
10 keV. The dashed line indicates the recommended value for
excluding flaring high background by the SAS manual. Shaded
regions are those showing high background flaring and, therefore,
are those discarded for our analysis.

I, II and III objects according to their infrared characteris-
tics.

2 X-RAY OBSERVATION AND DATA

REDUCTION

The XMM-Newton observation of the NGC 2023 reflection
nebula (ID 0112640201) was part of an observing program
aimed at studying deeply embedded stellar clusters in Orion
B. The observation was performed as a single, 30 ks long
exposure on 2002 during satellite revolution 419. The EPIC
(European Photon Imaging Camera) was used in full-frame
mode as primary instrument. The observation was centered
at α = 05h41m47.20s and δ = 02d16m37.0s. Table 1 summa-
rizes the information of the parameters used for each EPIC
detector.

The data reduction was done through theXMM-Newton

Science Analysis System (SAS) software, version 10.0. Im-
ages and spectra were produced with the standard SAS
tools. We rejected high background periods from the anal-
ysis of the lightcurve of the whole detector (independently
for EPIC-MOS and PN) and selected time periods with con-
stant, low X-ray emission at E > 10 keV (Fig. 1). Bad events
and noise were also rejected to create the final (GTIs; good
time intervals) event tables. From our analysis, we conclude
that the observation was largely contaminated by high X-ray
emission background periods (usually attributed to protons
and other relativistic particles produced during solar flares).
After our processing, the useful exposure time was reduced
to 13.6 ks for PN, 15 ks for MOS1 and 15.8 ks for MOS2.
Hereafter, we use the cleaned event lists.

In Fig. 2, we show a mosaic of PN and MOS images
of NGC 2023 in the energy band 0.3–8.0 keV, created with
the task emosaic. We used the SAS task edetect chain to re-



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 3

Figure 2. Composite EPIC (MOS+PN) image of the XMM-
Newton FoV in the 0.3–8.0 keV energy band.

veal sources at three different energy bands (0.3–1.2, 1.2–2.5,
2.5–4.5 keV) in PN, MOS1 and MOS2 separately. A second
stage consisted of using the 0.3-8.0 keV energy band to con-
firm the first detections. A total of 50 sources were detected.
After a careful inspection that includes a cross-match with
optical and infrared catalogs (see Section 3), we discarded
14 of them because they were multiple or spurious detec-
tions, typically at the detector border or close to a bright
source. The list of the X-ray sources is given in appendix
(Table A1). We have removed those sources considered as
spurious detections during the optical and infrared analyses
(Section 3) from the tables, although we have maintained
the original numeration for convenience. X-ray sources of
Yamauchi et al. (2000) are identified in the table last col-
umn as [YKK2000] followed by the ID number given by the
authors.

3 INFRARED PROPERTIES OF X-RAY

SOURCES

3.1 Spitzer

The region of NGC 2023 and its surroundings present a
large amount of gas and dust that produces high opti-
cal and near-infrared absorption (see Bowler et al. 2009).
The absorption by the gas and dust is less dramatic in the
mid-infrared, in particular in the Spitzer-IRAC bands and
in radio-wavelengths. Mookerjea et al. (2009) used archive
Spitzer observations of this region and sub-millimetric ob-
servations performed by themselves with the Submillimeter
Common User Bolometric Array (SCUBA; Holland et al.
1999) to create an infrared census of young stellar objects
and protostars. Mookerjea et al. (2009) analyzed only part
of the Spitzer -IRAC observations, close to the Horsehead.
The region they selected overlaps only partially with the
XMM-Newton observation, that is centered on NGC 2023.

We have reanalyzed the Spitzer data covering the XMM-

Newton field of view to look for infrared counterparts of all
the X-ray detected sources.

For our work, we used the observations with aorkey

numbers 8773120 and 8773632. Two collections of images
are present in those observations for each IRAC channel, cor-
responding to two different configurations with frame-time
0.6 and 10.4 seconds, respectively. We treated both datasets
separately. The regular IRAC pipeline of MOPEX (version
18.4.9) was used for the reduction and the mosaic creation.
At the end of the reduction process, we had two mosaics
for each IRAC channel and observation, for a total of four
mosaics per channel. Photometric analysis was performed in
each mosaic independently.

It was not the aim of this work to obtain IRAC fluxes
for all the sources in the field. We determined infrared
fluxes only for the IR counterparts of the X-ray sources
(see Fig. A2). To derive the fluxes, we used the daophot

package included in IDL, which is an adaptation of the For-
tran code also used in IRAF1. Aperture photometry with
three different extraction radii (2, 3 and 4 arcsec) was per-
formed. For the background (sky) subtraction, we used an
annulus with inner and outer radii 20 and 30 arcsec, respec-
tively. The mosaics performed by MOPEX are in units of
MJy/sr. We transformed them to mJy and applied an aper-
ture correction. Aperture corrections were estimated from
semi-empirical point response functions (PRFs)2 given in
the Spitzer Manual. Their values are given in Table 2 for
each channel. The uncertainties include the effects of typical
World Coordinate System (WCS) random alignment errors
(always less than 1 pixel; see Pérez-González et al. 2008). For
each source, we determined magnitudes in AB system and
Vega system. The latter were determined using the flux to
magnitude conversion of the GLIMPSE3. For each infrared
source, fluxes determined with extraction radii of two, three
and four arcsec are very similar except for the two most
brilliant IRAC sources, that show differences of up to 10%
in the four channels. The same statement is applicable to
the comparison between observations with frame times 0.6
and 12 seconds. In the latter case, the differences arise from
the fact that those sources are very close to the saturation
regime for long exposures. To avoid problems with those
sources, we decided to use only the images with shorter expo-
sure times. The Spitzer photometry for our sources is given
in Table A3. We note that Src. 24 was not cross-identified
with an unique IRAC counterpart because of the proximity
of several infrared sources that contaminate the photometric
measures. In the tables, we give photometry of the closest
star to this Src. 24 for completeness. However, those values
are not reliable and we did not used them in our study.

Figure 3 shows a color-color diagram of the four IRAC
channels for those of our sources with a counterpart in the
four bands. We have marked the typical location of infrared
class I, II and III objects and the locii of broad-line and
narrow-line active galactic nuclei as shown by Stern et al.
(2005). Source ID numbers for class I and class II objects are

1 IRAF: http://iraf.noao.edu/
2 “Spitzer pixels are relatively big relative to point sources, so
MOPEX fits with Point Response Functions (PRFs), which take
into account intra-pixel sensitivities.”
3 GLIMPSE: http://www.astro.wisc.edu/sirtf/



4 M. A. López-Garćıa et al.

Table 2. Aperture correction factors for several different extrac-
tion radii.

IRAC channel
Aperture 3.6µm 4.5µm 5.4µm 8.0µm
(arcsec) (mag) (mag) (mag) (mag)

1.5 0.59 ± 0.04 0.63 ± 0.04 0.83 ± 0.04 0.94 ± 0.06
2.0 0.32 ± 0.03 0.36 ± 0.03 0.53 ± 0.02 0.65 ± 0.03
3.0 0.13 ± 0.03 0.14 ± 0.02 0.22 ± 0.02 0.36 ± 0.02
4.0 0.03 ± 0.02 0.03 ± 0.02 0.05 ± 0.02 0.21 ± 0.03

Figure 3. IRAC color-color diagram of the X-ray sources of NGC
2023 and its surroundings. The locus of broad-line and narrow-
line active galactic nuclei is marked. Extinction vector is also
overplotted. The cross symbol at the left bottom corner indicates
the typical error for sources in our sample. Squares are sources
classified as class I objects. Dots are class II objects while aster-
isks are stars classified as main-sequence or class III objects. The
extinction vector is AK = 5 mag.

overplotted. The cross at the right bottom corner represents
the mean error bar in both colors. The typical error in the
colors of bright sources is 0.1 mag. Faint sources show error
bars with values as large as 1.2 mag in some cases.

The main concentration of sources is detected upon
class III, that are usually identified with weak-line T Tauri
and main sequence stars (asterisks in Figure 3). Five sources
are situated inside the boundaries of the class II type objects
and another two are very close to those boundaries. Stellar
sources with infrared class II objects nature are typically
identified with classical T Tauri stars. In Figure 3, we have
plotted them as filled circles. Finally, two sources have col-
ors typical of class I objects (squares in the figure). Stellar
sources with such infrared properties use to be related to
protostars and T Tauri stars with circumstellar envelopes.
A note of caution must be given here. Very absorbed class
II objects may present IRAC colors typical of class I ob-
jects. Therefore, the two sources in Fig. 3 classified as class
I objects may be absorbed class II stars. In fact, one of the
sources has been previously classified as a class II star in the
literature. This is the case of the object detected by Mook-
erjea et al. (2009) with Spitzer and SCUBA (named VLA 3

Figure 4. 2MASS color-color diagram of the X-ray sources of
NGC 2023 and its surroundings. The dwarf stars main sequence
is plotted as a continuous line. The extinction vector is AV = 3
mag. The dotted lines correspond to the dwarfs main sequence
with AV = 0.5, 1.0 and 1.5 mag. Symbols are as in Fig. 3.

and MIR 46 in their paper, source number 7 in our work,
Src 7).

3.2 Other photometric data from the literature

We complemented our study by cross-correlating our sample
against different catalogs. In particular, we used the second
version of the TASS Mark IV photometric catalog (Droege
et al. 2006, 2007) for V magnitudes, DENIS database 3rd
release (DENIS Consortium 2005) for I magnitudes, the
2MASS database (Skrutskie et al. 2006) for near-infrared,
and theWISE preliminary data release (Cutri et al. 2012) for
mid-infrared. The cross-correlation was performed through
the specific Aladin Java tool (Bonnarel et al. 2000). We used
a search radius of 5 arcsec to prevent the loss of coincidences
due to the XMM-Newton positional uncertainties. We finally
cross-matched our sample with the list of stars of Mookerjea
et al. (2009).

Tables A2 and A3 summarize the photometric data for
each one of our sources. We note that 13 out of the 50 sources
detected in X-rays have no counterpart in any of the catalogs
mentioned above. We considered them as spurious X-ray de-
tections and, therefore, we removed them from the tables.
We maintained Src. 23 because it is situated at 11 arcsec
from an optical source. We cannot discard this source as a
real detection. The remaining sources have a counterpart in
DENIS and/or 2MASS, except for source 7 (VLA3 in Mook-
erjea et al. 2009, see Section 3.1) and Src 12 (only Spitzer

photometry). In Fig. 4 we plot the 2MASS color-color dia-
gram for these sources. The main sequence for unabsorbed
(dwarf) stars is overplotted as a continuous line. The shaded
region in the figure is the locus of the active galactic nu-
clei (AGN) determined by us using data from Kouzuma &
Yamaoka (2010). The position of most of the stars in the



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 5

Figure 5. I−J - J−[3.6] color-color diagram of the X-ray sources
of NGC 2023. The continuous line represents the ZAMS and the
dashed line is the 1 Myr isochrone (Baraffe et al. 1998). The
shaded region is the locus of quasars (adapted from Barrado et al.
2011). Symbols are as in Fig. 3. The extinction vector is AV = 1.5
mag.

color-color diagram indicates the high interstellar extinction
in this star-forming region. As in Fig. 3, we identified class
II objects with its X-ray source ID number.

We estimated the value of the extinction (AV) for each
star, from the 2MASS color-color diagram (Fig. 4). Our re-
sults are given in Table A2. They will be discussed and com-
pared with those obtained from SED fitting in Section 3.4.

Barrado et al. (2011) showed that the I − J vs J−[3.6]
color-color diagram can be used to reject extragalactic ob-
jects from X-ray selected samples, as active galactic nuclei
are well separated from stars. Figure 5 shows the same dia-
gram for our 32 sources with a DENIS counterpart. Src 10
and Src 28 are situated close to the boundaries of extra-
galactic sources, but they have been clearly identified with
stars in the optical images (in fact Src 10 is V615 Ori). Only
Src 18 and Src 35 may be absorbed quasars, according to
the direction of the extinction vector (In Tables A2 and A3
they are identified as possible AGN).

3.3 Age of the NGC 2023 members

It was mentioned in Section 1 that the age of the star-
forming region around NGC 2023 is not well known. There
are very few works about this issue in the literature (see
Meyer et al. 2008, for a review). Alcala et al. (2000) gave a
range of ages for the stars belonging to the Orion Molecular
Complex, but there is no particular study on each individ-
ual region. Ages determined for these stars are in the range
0.5–7.0 Myr.

In Fig. 6, we show evolutionary tracks from Baraffe et
al. (1998), with an extension to stellar masses over 1.6 M⊙

from Siess et al. (2000), in a 2MASS color-magnitud dia-
gram. We assumed a mean distance of 400 pc for the region.

Figure 6. Ks - J − Ks color-magnitud diagram of the X-ray
sources of NGC 2023. The lines are evolutionary tracks from
Baraffe et al. (1998) and Siess et al. (2000), with the ZAMS,
7, 3 and 1 Myr isochrones and several mass tracks (d = 400 pc).
Symbols are as in Fig. 3. The extinction vector is AV = 5 mag.

The isochrones are 1, 3 and 7 Myr and the ZAMS. We also
plot tracks for masses ranging from 0.2 to 1.4 M⊙. Although
there is a concentration of low mass stars around the 7 Myr
isochrone, the density of isochrones in that region of the
diagram makes difficult to assure this is the age for the clus-
ter. In fact, the most massive stars seem to follow younger
isochrones. Therefore, we only may affirm the age of the
stars in NGC 2023 is in the approximate range 1− 7 Myr.

3.4 SED fitting

To investigate in more detail the nature of our X-ray sources,
we performed SED fitting using the photometric data we
compiled for each source (Tables A2 and A3). We took full
advantage of the grid of young stellar object radiation trans-
fer models of Robitaille et al. (2007). In the model, we per-
mitted the extinction to vary only from zero to the maximum
value given by the 2MASS color-color diagram (Fig. 4). Ex-
tinction and distance are the only parameters external to
the model.

Fitting was performed for the 32 sources in our sample
with a DENIS counterpart, to include at least one optical
band. This should improve spectral type determination. The
blue spectrum of a Class II is usually affected by accretion,
as it produces an excess in the blue and ultraviolet spectral
range. In Figs. A3 and A4, we show the best fit for those
32 sources. In some cases, there is more than one model
that fits well our data. For those cases, we overplot them
as grey lines. We have assumed as good models those in
which (χ2

− χ2
best)/χ

2
best 6 0.1. In most of those cases, it is

not possible to discard any of the fits. We would need some
point in the sub-millimeter spectral range (see discussion in
Section 5).

The classification into the different infrared classes



6 M. A. López-Garćıa et al.

given in Tables A2 and A3 was performed according to the
results of the SED fitting, using the parameter α, defined as
the slope of the SED beyond 2.2 µm (see Robitaille et al.
2007). We used every point of the SED for this purpose, join-
ing WISE and Spitzer data. A review on the classification
method can be found in Stahler & Palla (2005). We classi-
fied as class III objects those sources for which an absorbed
stellar model is enough to fit the observed SED (α 6 −1.5).
Only Src. 14 may show a very thin circumstellar disk (see
Fig. A3). Nevertheless, the data fits well also to a stellar
photosphere. Class II objects (−1.5 < α 6 0.0) show an
obvious excess at the IRAC bands (see Fig. A4). For those
sources, any attempt of fitting a stellar model without disk
was unsuccessful. Src. 13 and Src. 17 show characteristics
of both class I and II objects. Their SED first decreases be-
tween 1.0 and 5.4 µm, but then increases again. We have
classified them as class I/II objects. Finally, there are sev-
eral sources that have SEDs typical of class III objects but
with an excess at wavelengths longer than 10 µm. We have
classified them as class II/III objects. Src 42 is situated in
the IRAC color-color diagram (Fig. 3) in a position that is
usually occupied by stars with transition disks. Our classi-
fication into infrared classes using the SED coincides very
well with that made first using the IRAC color-color dia-
gram, except for Src 44, the SED of which shows an excess
above 10 µm but that in the color-color diagram is situated
inside the boundaries of class III objects.

Finally, we note that the SED fitting tool of Robitaille et
al. (2007) gives an age for each star. This value is determined
through isochrone fitting, so its result depends strongly on
the model chosen. Robitaille et al. (2007) use the Siess et
al. (2000) pre-main sequence evolutionary models that dif-
fer slightly from the Baraffe et al. (1998) models. The tool
also tends to fit low-age isochrones even when a simple pho-
tosphere model is suitable for the observed SED. Therefore,
the results of the fitting tool should be used with some cau-
tion.

Ages determined using the Robitaille et al. (2007) fit-
ting tool are in the range 5–9 Myr for most of the stars,
what may agree with the results presented in Section 3.3.
Nevertheless, for stars showing larger infrared excesses, ages
are in teh range 0.5–2 Myr.

3.5 Extinction, stellar masses, temperatures and

luminosities

Table A4 summarizes the results of the SED fitting for the
32 sources with an optical counterpart (see Section 3.4). For
sources with more than one good fit, we give only parameters
for the best fit. For class II objects, the accretion rate and
evolutionary stage are also given. To account for the evolu-
tionary stage of young stellar objects, Robitaille et al. (2007)
defined a parameter based on physical properties of the cir-
cumstellar envelops, such as mass and accretion rate from
it. The authors distinguished between stars with significant
infalling envelops (Stage 0/I objects), stars with optically
thick disks and thin envelopes (Stage II objects), and stars
with optically thin disks (Stage III objects). We have deter-
mined this stage for our sources and included it in Table A4.
We observe that this classification coincides very well with
that given by the IRAC color-color diagram or the slope of
the SED. Class I objects present properties of Stage I ob-

jects while class II objects are classified as Stage III objects
following this method.

We previously mentioned that parameters listed in Ta-
ble A4 are those given by the model. In fact, the extinction
(AV) and the distance are parameters external to the radia-
tive model (see Robitaille et al. 2007). In addition, stellar
mass and radius are determined from evolutionary tracks
(Robitaille 2008). In essence, the parameters derived from
the model are the temperature and luminosity of the star.

With respect to the interstellar extinction, the values
coming from the SED fitting tool are quite similar to those
determined with the 2MASS color-color diagram if one takes
error bars into account for stars showing low extinction. For
the remaining stars, there is a trend of AV determined using
the color-color diagram to be higher than that coming from
the SED fitting tool. The median of the ratio between AV

determined with both methods is 1.35, or 1.5 if only stars
with AV > 1 are considered. These stars are mainly those
for which the extinction determined with both methods do
not coincide even when error bars are taken into account.

4 X-RAY SPECTRAL ANALYSIS

In Section 2, we showed that the X-ray observation of the
NGC 2023 nebula was affected by high background periods
(see Fig. 1). The highly variable background prevented us
to perform a study about variability of the detected sources.
Therefore, our study was limited to the spectral analysis.
The X-ray spectrum of sources with more than approxi-
mately 100 net counts was compared (using fitting proce-
dures) to hot plasma models using XSPEC (Arnaud et al.
1996, 2004). The software first load the data previously ex-
tracted using specific SAS tools (source and background
spectra, response matrix and ancillary file) and then gen-
erates the plasma model. Two hot (diffuse) plasma models
are available in XSPEC: mekal (Mewe et al. 1985, 1986;
Kaastra et al. 1992; Liedahl et al. 1995) and apec (Smith
et al. 2001a). To be consistent with our previous works
(e.g. López-Santiago & Caballero 2008; López-Santiago et
al. 2010a), we chose the latter for the present analysis. The
apec uses the atomic data contained in the Astrophysical
Plasma Emission Database (APED; Smith et al. 2001b).

Due to the limitations of our observation in terms of
counts – because of the time lost due to the high background
periods – we used only 1T or 2T -models for the fits. The re-
sults of the spectral fitting are given in Table 3 for those
stars with accurate fits. For several stars, the value of the
abundance obtained during the fit was unconstrained. In
those cases, we fixed the abundance to Z/Z⊙ = 0.2. Al-
though some authors have used other values in their works
(e.g. Stelzer et al. 2004; Robrade & Schmitt et al. 2007; Gia-
rdino et al. 2007a), in other stars of this star-forming region
we find always a value of Z/Z⊙ very close to 0.2. We used χ2

Statistics to obtain the goodness of our fits. Energy chan-
nels were then grouped to contain at least 15 counts in each
energy bin.



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 7

Table 3. Results of the X-ray spectral fitting to hot plasma models for stars in NGC 2023. Errors are at 90% confidence level.

ID NH kT1 kT2 EM1/EM2 Z χ2 fX
a logLX

b

(×1022 cm−2) (keV) (keV) (Z⊙) (d.o.f) (×10−13 erg cm−2 s−1) (erg s−1)

3 0.00+0.01
−0.00 0.62+0.09

−0.11 1.26+0.21
−0.13 0.70 0.23+0.05

−0.05 1.29 (276) 3.95 29.2

10 0.50+0.05
−0.04 2.49+0.34

−0.52 ... ... 0.33+0.22
−0.16 1.29 (140) 12.20 31.3

17 0.22+0.44
−0.08 1.24+1.62

−0.53 ... ... = 0.2 1.27 (54) 0.57 30.3

20 0.07+0.05
−0.05 0.78+0.16

−0.80 1.64+2.32
−2.32 2.59 0.13+0.13

−0.03 0.88 (78) 2.30 30.5

22 0.05+0.06
−0.04 0.34+0.07

−0.04 1.21+0.17
−0.20 1.29 0.18+0.10

−0.06 0.83 (120) 3.30 30.5

27 0.41+0.33
−0.21 18.3+60.0

−13.7 ... ... = 0.2 0.51 (10) 1.38 30.0

28 0.44+0.07
−0.05 0.28+0.05

−0.04 1.24+0.04
−0.05 1.02 0.16+0.03

−0.03 1.10 (602) 33.41 31.9

34 0.40+0.10
−0.08 2.73+1.08

−0.58 ... ... = 0.2 1.26 (59) 6.99 30.9

36 0.24+0.16
−0.10 2.14+1.76

−0.81 ... ... = 0.2 0.72 (61) 2.17 30.4

41 0.08+0.03
−0.02 2.36+0.55

−0.61 ... ... 0.15+0.14
−0.10 1.18 (271) 11.53 30.6

50 0.22+0.08
−0.08 2.00+1.86

−0.62 ... ... = 0.2 1.33 (65) 3.41 30.2

(a)Unabsorbed flux in the energy band [0.3-8.0] keV. (b) Unabsorbed X-ray luminosity in the energy band [0.3-8.0] keV. We assumed
d = 450 pc (the distance to the ONC) for all the sources.

5 DISCUSSION

Spectral fitting was performed for the eleven sources in our
sample with at least 500 counts (combining PN and MOS
detectors). The results are shown in Table 3.

The position in the color-color diagram of Fig. 5 of the
eleven spectroscopically sources studied in this work indi-
cates they are indeed stars. According to that diagram, only
Src 18 and Src 35 may be (absorbed) extragalactic sources.
General X-ray parameters obtained for these eleven stars
are very similar to those observed in stars of the same class
in other star-forming regions (e.g. Getman et 2005; Güdel
et al. 2007). Among them, three are classified as Class II
from their SEDs (see Tables A2 and A3.). They are Src 10
(V615 Ori), Src 17 and Src 28. Another two stars likely have
transition disks (Src 36 and Src 41). The remainder are clas-
sified as Class III or main-sequence stars.

Coronal temperatures are consistent with being young
T Tauri stars (e.g. Güdel et al. 2007). Only Src 27 shows an
extremely high temperature, although it is unconstrained
(this is the source with the lowest number of counts we
could fit). Unabsorbed X-ray fluxes in the energy band [0.3–
8.0] keV were determined for the eleven sources for which a
spectral fitting was performed and then transformed to lumi-
nosities using the distances obtained during the SED fitting
procedure (Table A4). We note that for HD 37805 (Src 3),
the distance given by the SED fitting tool is slightly under-
estimated (Hipparcos distance for HD 37805 is d = 82 ± 7
pc). However, this is the only star for which we could find a
value for its distance in the literature. Therefore, an inter-
polation of this result to the remaining stars should not be
done. In fact, the distance obtained with this procedure for
V615 Ori (Src 10, d = 371 pc) is consistent with some results
found in the literature: there are hints that the Horsehead
Nebula is a few parsecs closer to us than σ Orionis, that is
situated at approximately 385 pc (see Caballero & Solano
2008, and references therein). V615 Ori must be closer than
the Horsehead Nebula as it is observed at the base of the
nebula, in front of it. Therefore, the use of the distances

obtained from the SED fitting tool seems to be a good op-
tion, undoubtedly better than using the same distance for
all the sources. Observed X-ray luminosities are indicative of
the shallowness of this observation. Typical X-ray luminosi-
ties of young M dwarfs are in the range LX ∼ 1028.5 − 1030

erg s−1 (e.g. Preibisch & Feigelson 2005). In this observa-
tion, the completeness limit is fX ∼ 5×10−14 erg cm−2 s−1,
what corresponds to LX ∼ 1030 erg s−1 at a distance of 400
pc. Stelzer et al. (2012) pointed out that there is a trend of a
mass dependence of the X-ray emission level throughout the
various evolutionary stages of young stellar objects. Com-
paring the luminosities of class II and class III objects in
our sample, we do not find such differences, but our sample
is too small to achieve robust results.

With the aim of comparing general results of our X-ray
analysis of NGC 2023 with other star-forming regions, we
plotted the X-ray luminosity cumulative distribution func-
tion of our sample, the Orion Nebula Cluster and the Taurus
Molecular Complex together (see Fig. 7). The data of those
regions were taken from Getman et (2005) and Güdel et al.
(2007), respectively. For each star-forming region, we plot
the whole sample accessible in those works with no distinc-
tion between infrared classes. For NGC 2023, we discarded
HD 37805 (Src 3), that is likely a field star, and those stars
for which we did not fitted a model to their SEDs (Src 7,
Src 12, Src 16 and Src 24). For the stars for which a spec-
tral analysis with XSPEC could not be performed, we de-
termined X-ray fluxes from the observed count-rate in the
following way. We used the subsample of eleven stars in Ta-
ble 3 for which X-ray spectral fitting was done to determine
a conversion factor between observed count-rate and (ab-
sorbed) flux. This conversion factor was determined for each
star and each EPIC detector (PN, MOS1 and MOS2) inde-
pendently. Then, we performed a linear regression to each
one of the three datasets. The relationships obtained by us
were:

fX [MOS1] = (0.5± 0.3) + (0.056 ± 0.005) · CR



8 M. A. López-Garćıa et al.

Figure 7. Cumulative distribution function of logLX in

NGC 2023 (continuous line, black), Taurus (dashed-
dotted line, blue) and in the Orion Nebula Cluster
(ONC). The ONC is plotted twice, one for the whole
sample (dashed line, green) and the other with stars with
masses less than 3 M⊙, plotted as a dotted (red) line.

fX [MOS2] = (0.2± 0.4) + (0.057± 0.005) · CR

fX [PN ] = (1.5± 0.2) + (0.003± 0.005) · CR

Fluxes determined with these three relationships are
compatible. Table A1 gives the flux values determined for
our sample from their count-rates. We include in that table
the eleven stars of Table 3, for completeness. For them, the
values of the fluxes are those determined from the spectral
fitting. As for the eleven stars for which we could perform
spectral fitting (see above), fluxes were transformed into lu-
minosities using the distances obtained from the SED fitting
(Table A4).

The comparison of the X-ray luminosity function of the
different star-forming regions was made only in the range
of completeness of the XMM-Newton observation of NGC
2023 (logLX[erg s

−1] > 29.8). Known non-members of the
ONC identified by Getman et (2005) were discarded for
this study. From Fig. 7, NGC 2023 seems to be more sim-
ilar to the Taurus molecular complex. In both cases, there
is a lack of stars brighter than LX ∼ 1031 erg s−1, while
the ONC have stars with X-ray luminosities up to 1031.5−32

erg s−1. This result may be explained as differences in the
stellar population in each star-forming region. While the
ONC have massive stars, they are not present in Taurus
and NGC 2023. We have explored this hypothesis by con-
structing a new COUP sample with stars less massive than
3 M⊙ (similar to the mass range in NGC 2023). In Fig. 7,
this new sample is plotted as a dotted (red) line. The fig-
ure suggests that the COUP sample restricted to the same
mass range is more similar to NGC 2023 than the sample
without mass restriction. We note here that the ages of the
Taurus and ONC samples are not equal and that we could

not determine robustly the age of NGC 2023. Therefore, this
result must be used with caution.

In Fig. 8, we overplot our sample of NGC 2023 sources
over the COUP data (Getman et 2005). For NGC 2023,
we used the masses and bolometric luminosities given in Ta-
ble A4 and X-ray luminosities determined from the distances
given in Table A4 and fluxes of Table A1. All the NGC 2023
low-mass stars (M < 1M⊙) are inside the boundaries of
the stars of the COUP. Only the more massive stars in our
sample (as determined by the SED fitting tool of Robitaille
et al. 2007) have very low values of logLX/Lbol (out of the
limits of Fig. 8, see Table A4). There are three types of
sources with very low values of logLX/Lbol in our sample:
(1) stars showing large infrared excess, for which the SED fit-
ting tool gives very high luminosities, (2) intermediate-mass
pre-main or main-sequence stars (A and F spectral types),
and (3) Src. 18, that may be an AGN instead of a star (see
Section 3.2 and Fig. 5). The SED fitting procedure derives
bolometric luminosities and temperatures directly and then
determine masses using evolutionary tracks (see Robitaille
2008). For stars with large infrared excesses, to obtain a
robust value for their luminosity may be difficult since it
depends on the accurate determination of the disk (and en-
velope in some cases) parameters. Therefore, in those stars
of our sample, Lbol may be overestimated.

For the two stars classified as infrared class III ob-
jects (i.e. stars without an accretion disk) their temperature
may be also overestimated. Nevertheless, the star HD 37805
(Src. 3) that is one of these two pre-main or main-sequence
stars with a very low value of the ratio logLX/Lbol, it is a
known A star (e.g. Mookerjea et al. 2009), already detected
in X-rays with ASCA (Yamauchi et al. 2000). Therefore, the
presence of a low-mass companion responsible for the X-ray
emission cannot be discarded in these three intermediate-
mass stars in our sample. The other class III object with
low logLX/Lbol is Src 33 (MIR 83).

6 SUMMARY AND CONCLUSSIONS

We have carried out a comprehensive analysis of the X-ray
and infrared properties of 36 young stellar objects in the
NGC 2023 star-forming region and its surroundings for the
first time. The spectral energy distribution (SED) of 90%
of this sample (32 out of 36 sources) has been analyzed
using the star-disk models of Robitaille et al. (2007). We
have identified only two AGN candidate among the 36 X-
ray sources. A classification on different infrared classes has
been made according to the position of the stars in infrared
color-color diagrams and to their SED. The study of the
SEDs has allowed us to determine several stellar parameters
such as masses, radii and luminosities, and other physical
parameters such as extinctions and distances.

The X-ray properties of NGC 2023 have been compared
with those of other star-forming regions with different char-
acteristics in terms of stellar population. Although the val-
ues of temperature and column density determined for our
sample of young stars are similar to those observed in stars
of the Orion Nebula Cluster and Taurus, the (X-ray lu-
minosity) cumulative distribution function of members of
NGC 2023 is more similar to that of the Taurus molecu-
lar complex. A possible explanation is the different stellar



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 9

Figure 8. logLX/Lbol as a function of stellar mass for the stars

of the COUP catalog (small crosses) and our sample of NGC 2023
stars. Squares are stars classified as class III in Tables A2 and A3.
Filled circles are infrared class II objects. Triangles are class II/III
objects.

populations in each region. While in the ONC there is a
number of massive stars, they are not present in Taurus and
NGC 2023.

ACKNOWLEDGMENTS

M.A.L-G. and J.L-S. acknowledge support by the Spanish
Ministerio de Ciencia e Innovación under grant AYA2008-
06423-C03-03. J.F.A.C. is researcher of the CONICET
and acknowledges support by grant PICT 2007-02177 (Se-
cyT). PGP-G acknowledges support from the Spanish Pro-
grama Nacional de Astronomı́a y Astrof́ısica under grants
AYA2009-10368 and AYA2009-07723-E. We would like to
acknowledge the anonymous referee for a fruitful discussion
about the results of this work. His/Her comments on the
text content have allowed us to notoriously improve this
manuscript.

REFERENCES

Alcalá, J. M., Covino, E., Torres, G., et al. 2000, A&A,
353, 186

Ali, B., Sellgren, K., Depoy, D. L., et al. 1998, Cool Stars,
Stellar Systems, and the Sun, 154, 1663

Anglada, G., & Rodŕıguez, L. F. 2000, Rev. Mexicana As-
tron. Astrofis., 38, 13

Arnaud, K. 1996, XSPEC: The First Ten Years. In ASP
Conf. Ser. 101, eds. George H. Jacoby and Jeannette
Barnes, 17

Arnaud, K. 2004, XSPEC: Progress and Plans. In the
American Astronomical Society, HEAD meeting 8, 16.29

Bally, J., O’Dell, C. R., & McCaughrean, M. J. 2000, AJ,
119, 2919

Baraffe, I., Chabrier, G., Allard, F., & Hauschildt, P. H.
1998, A&A, 337, 403

Barrado, D., Stelzer, B., Morales-Calderón, M., et al. 2011,
A&A, 526, A21

Bonnarel, F., et al. 2000, A&AS, 143, 33
Bowler, B. P., Waller, W. H., Megeath, S. T., Patten, B. M.,
& Tamura, M. 2009, AJ, 137, 3685

Caballero, J. A., & Solano, E. 2008, A&A, 485, 931
Cash, W. 1979, ApJ, 228, 939
Cutri, R. M., et al. 2012, VizieR Online Data Catalog, 2307,
0

DENIS Consortium 2005, VizieR Online Data Catalog,
2263, 0

Droege, T. F., Richmond, M. W., Sallman, M. P., & Crea-
ger, R. P. 2006, PASP, 118, 1666

Droege, T., F., Richmond, M., W., & Sallman, M. 2007,
VizieR Online Data Catalog, 2271, 0

Eisner, J. A., & Carpenter, J. M. 2003, ApJ, 598, 1341
Fang, M., van Boekel, R., Wang, W., Carmona, A., Sicilia-
Aguilar, A., & Henning, T. 2009, A&A, 504, 461

Genzel, R., & Stutzki, J. 1989, ARA&A, 27, 41
Getman, K. V., Flaccomio, E., Broos, P. S., et al. 2005,
ApJS, 160, 319

Giardino, G., Favata, F., Micela, G., Sciortino, S., & Win-
ston, E. 2007a, A&A, 463, 275

Giardino, G., Favata, F., Pillitteri, I., et al. 2007b, A&A,
475, 891

Güdel, M., Briggs, K. R., Arzner, K., et al. 2007, A&A,
468, 353

Hartmann, L., & Burkert, A. 2007,ApJ, 654, 988
Haisch, K. E., Jr., Lada, E. A., & Lada, C. J. 2000, AJ,
120, 1396

Holland, W. S., et al. 1999, MNRAS, 303, 659
Ingleby, L., et al. 2009, ApJ, 703, L137
Kaastra, J.S. 1992, An X-Ray Spectral Code for Optically
Thin Plasmas (Internal SRON-Leiden Report, updated
version 2.0)

Kouzuma, S., & Yamaoka, H. 2010, A&A, 509, A64
Liedahl, D. A., Osterheld, A. L., & Goldstein, W. H. 1995,
ApJ, 438, L115

López-Santiago, J., & Caballero, J. A. 2008, A&A, 491, 961
López-Santiago, J., Albacete Colombo, J. F., & López-
Garćıa, M. A. 2010, A&A, 524, A97

Malin, D. F., Ogura, K., & Walsh, J. R. 1987, MNRAS,
227, 361

Mewe, R., Gronenschild, E. H. B. M., & van den Oord,
G. H. J. 1985, A&AS, 62, 197

Mewe, R., Lemen, J. R., & van den Oord, G. H. J. 1986,
A&AS, 65, 511

Meyer, M. R., Calvet, N., & Hillenbrand, L. A. 1997, AJ,
114, 288

Meyer, M. R., Flaherty, K., Levine, J. L., et al. 2008, Hand-
book of Star Forming Regions, Volume I, 662

Mookerjea, B., Sandell, G., Jarrett, T. H., & McMullin,
J. P. 2009, A&A, 507, 1485

Penz, T., Micela, G., & Lammer, H. 2007, A&A, 477, 309
Pérez-González, P. G., Rieke, G. H., Villar, V., et al. 2008,
ApJ, 675, 234

Pickles, A. J. 1998, PAPS, 110, 863
Pillitteri, I., Sciortino, S., Flaccomio, E., et al. 2010, A&A,



10 M. A. López-Garćıa et al.

519, A34
Pillitteri, I., Wolk, S. J., Allen, L., Megeath, S. T., Guter-
muth, R. A., & Gutermuth 2010, IAU Symposium, 266,
509

Pizzolato, N., Maggio, A., Micela, G., Sciortino, S., & Ven-
tura, P. 2003, A&A, 397, 147

Prisinzano, L., Micela, G., Flaccomio, E., et al. 2008, ApJ,
677, 401

Preibisch, T., & Feigelson, E. D. 2005, ApJS, 160, 390
Preibisch, T., Kim, Y.-C., Favata, F., et al. 2005, ApJS,
160, 401

Reipurth, B., Rodŕıguez, L. F., Anglada, G., & Bally, J.
2004, AJj, 127, 1736

Robitaille, T. P., Whitney, B. A., Indebetouw, R., Wood,
K., & Denzmore, P. 2006, ApJS, 167, 256

Robitaille, T. P., Whitney, B. A., Indebetouw, R., & Wood,
K. 2007, ApJS, 169, 328

Robitaille, T. P. 2008, Massive Star Formation: Observa-
tions Confront Theory, 387, 290

Robrade, J., & Schmitt, J. H. M. M. 2007, A&A, 461, 669
Sandell, G., et al. 1999 ApJ, 519, 236
Scally, A., & Clarke, C. 2001, MNRAS, 325, 449
Sciortino, S. 2008, Astronomische Nachrichten, 329, 214
Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593
Skinner, S., Gagné, M., & Belzer, E. 2003, ApJ, 598, 375
Skrutskie, M. F., et al. 2006, AJ, 131, 1163
Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Ray-
mond, J. C. 2001a, ApJ, 556, L91

Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Ray-
mond, J. C. 2001b, ASP Conf. Ser. 247: Spectroscopic
Challenges of Photoionized Plasmas, 247, 161

Stahler, S. W., & Palla, F. 2005, The Formation of Stars,
by Steven W. Stahler, Francesco Palla, pp. 865. ISBN 3-
527-40559-3. Wiley-VCH , January 2005.

Stelzer, B., Micela, G., & Neuhäuser, R. 2004, A&A, 423,
1029

Stelzer, B., Preibisch, T., Alexander, F., et al. 2012, A&A,
537, A135

Stern, D., Eisenhardt, P., Gorjian, V., et al. 2005, APJ,
631, 163

Telleschi, A., Güdel, M., Briggs, K. R., Audard, M., &
Palla, F. 2007, A&A, 468, 425

Testa, P., Drake, J. J., Ercolano, B., et al. 2008, ApJL, 675,
L97

Tsujimoto, M., Feigelson, E. D., Grosso, N., et al. 2005,
ApJS, 160, 503

Wall, J. V., & Jenkins, C. R. 2003, Princeton Series in
Astrophysics

Wilson, T. L., Muders, D., Dumke, M., Henkel, C., &
Kawamura, J. H. 2011, ApJ, 728, 61

Wolk, S. J., Pillitteri, I., Megeath, S. T., Allen, L., & con-
sortium, S. 2010, Bulletin of the American Astronomical
Society, 42, 683

Yamauchi, S., Kamimura, R., & Koyama, K. 2000, PASJ,
52, 1087

APPENDIX A: TABLES AND FIGURES



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 11

Table A1. X-ray parameters from the SAS multitask edetect chain.

Src ID α δ rate [pn] rate [mos1] rate [mos2] Obs. fluxa HR1 HR2 Remarksb

(deg) (deg) (ks−1) (ks−1) (ks−1) (×10−13 erg cm−2 s−1)

1 05 40 54.2 -02 19 04.8 ... 8 ± 1 10 ± 2 0.79 ± 0.44 -0.55 -0.49
3 05 41 02.6 -02 18 18.0 310 ± 8 67 ± 5 87 ± 4 3.95 ± 0.10 -0.72 -0.80 [YKK2000] A5
4 05 41 03.8 -02 21 39.6 12 ± 2 ... 6 ± 1 0.97 ± 0.42 -0.56 -0.31
5 05 41 06.7 -02 23 52.8 13 ± 2 5 ± 1 4 ± 1 0.98 ± 0.42 -0.71 -1.00
6 05 41 11.5 -02 27 25.2 ... 2 ± 1 1 ± 1 0.22 ± 0.44 ... ...
7 05 41 21.6 -02 11 09.6 8 ± 1 ... ... 0.93 ± 0.41 -0.53 -0.15 [YKK2000] A6
10 05 41 24.7 -02 22 37.2 203 ± 6 79 ± 4 77 ± 3 5.84 ± 1.28 0.24 -0.49 [YKK2000] A7
12 05 41 25.7 -02 23 06.0 3 ± 1 ... ... 0.87 ± 0.41 -0.45 -0.34
13 05 41 28.3 -02 13 58.8 4 ± 1 ... ... 0.88 ± 0.41 -0.70 -1.00
14 05 41 28.3 -02 19 44.4 3 ± 1 ... ... 0.87 ± 0.41 -0.78 -1.00
15 05 41 31.9 -02 10 19.2 5 ± 1 ... ... 0.89 ± 0.41 -0.73 -1.00
16 05 41 32.6 -02 13 19.2 3 ± 1 ... ... 0.87 ± 0.41 -1.00 -0.08
17 05 41 34.8 -02 17 24.0 17 ± 2 11 ± 1 9 ± 1 1.03 ± 0.15 -0.20 -0.89
18 05 41 36.5 -02 16 48.0 ... 1 ± 1 1 ± 1 0.22 ± 0.44 -0.63 -0.42
20 05 41 43.0 -02 10 26.4 100 ± 4 26 ± 2 27 ± 2 1.81 ± 0.22 -0.50 -0.70 [YKK2000] A9
21 05 41 43.2 -02 17 38.4 2 ± 1 ... ... 0.86 ± 0.41 -0.95 -1.00
22 05 41 43.2 -02 03 54.0 161 ± 7 51 ± 4 48 ± 4 2.34 ± 2.30 -0.72 -0.91 [YKK2000] A10
23 05 41 43.7 -02 10 22.8 ... 4 ± 1 ... 0.72 ± 0.39 -0.79 -1.00 opt. source at 11′′

24 05 41 44.2 -02 16 08.4 3 ± 1 ... ... 0.87 ± 0.41 -0.95 -0.03
27 05 41 45.6 -02 24 18.0 14 ± 2 6 ± 1 5 ± 1 1.29 ± 0.38 0.57 -0.34
28 05 41 47.0 -02 16 37.2 477 ± 7 166 ± 4 159 ± 4 8.32 ± 0.87 -0.18 -0.68 [YKK2000] A11
31 05 41 48.2 -02 10 40.8 7 ± 1 2 ± 1 ... 0.92 ± 0.41 -0.79 -0.66
33 05 41 51.1 -02 29 52.8 ... 15 ± 2 10 ± 2 0.73 ± 0.44 -0.01 -0.70
34 05 41 52.6 -02 03 50.4 131 ± 6 50 ± 4 35 ± 4 4.10 ± 1.01 0.30 -0.61 [YKK2000] A12
35 05 41 55.7 -02 23 42.0 14 ± 2 3 ± 1 ... 1.26 ± 0.47 0.16 -0.77
36 05 42 09.1 -02 08 20.4 62 ± 6 16 ± 2 20 ± 2 1.48 ± 0.99 -0.05 -0.55
38 05 42 10.1 -02 10 22.8 7 ± 1 ... ... 0.92 ± 0.41 -0.91 -1.00
41 05 42 12.5 -02 05 09.6 517 ± 13 ... 145 ± 7 9.90 ± 0.93 -0.36 -0.53
42 05 42 16.1 -02 06 46.8 72 ± 5 27 ± 3 18 ± 2 1.57 ± 0.74 -0.71 -0.61 [YKK2000] A13
43 05 42 16.8 -02 06 39.6 12 ± 3 6 ± 2 ... 0.97 ± 0.42 -0.89 -0.43 [YKK2000] A13
44 05 42 18.5 -02 19 44.4 4 ± 1 ... ... 0.88 ± 0.41 -0.95 -0.58
45 05 42 19.7 -02 15 32.4 ... 1 ± 1 3 ± 1 0.34 ± 0.44 ... ...
46 05 42 20.2 -02 11 27.6 9 ± 2 3 ± 1 ... 0.94 ± 0.41 -0.46 -1.00
47 05 42 21.1 -02 15 18.0 96 ± 4 34 ± 3 34 ± 2 1.77 ± 0.91 -0.06 -0.68 [YKK2000] A14?
48 05 42 21.4 -02 07 44.4 50 ± 4 20 ± 3 19 ± 2 1.36 ± 0.59 -0.38 -0.92
49 05 42 33.8 -02 09 46.8 9 ± 2 ... ... 0.94 ± 0.41 ... ...
50 05 42 36.7 -02 20 38.4 96 ± 4 25 ± 3 35 ± 3 1.78 ± 1.58 -0.08 -0.60

(a) Flux converted from observed count-rates as explained in Section 5; (b) As in Tables A2 and A3, we have removed sources
considered as spurious. [YKK2000] is for Yamauchi et al. (2000).



12 M. A. López-Garćıa et al.

Table A2. Optical and near-infrared magnitudes for the sources detected in the XMM-Newton observation.

Src. ID Name V I J H Ks AV IR-Class† and
(mag) (mag) (mag) (mag) (mag) (mag) other remarks

1 ... 14.06 ± 0.03 12.81 ± 0.03 12.11 ± 0.02 11.93 ± 0.02 1.0 ± 0.5 III
3 HD 37805/MIR 31 7.53 ± 0.05 7.23 ± 0.05 6.95 ± 0.03 6.87 ± 0.05 6.76 ± 0.02 0.0 ± 0.5 III
4 ... 14.48 ± 0.03 12.86 ± 0.03 12.24 ± 0.03 11.95 ± 0.02 0.7 ± 0.7 III
5 ... 14.56 ± 0.03 13.03 ± 0.03 12.45 ± 0.03 12.21 ± 0.02 0.0 ± 0.5 III
6 ... 17.31 ± 0.11 13.70 ± 0.08 ... 11.34 ± 0.08 ... III
7 VLA 3/MIR 46 ... ... ... ... ... ... I
10 V615 Ori/MIR 52 13.11 ± 0.24 10.99 ± 0.17 9.10 ± 0.02 8.07 ± 0.04 7.39 ± 0.02 12.2 ± 3.5 II
12 ... ... ... ... ... ... -
13 ... 14.21 ± 0.03 12.09 ± 0.02 11.12 ± 0.02 10.73 ± 0.02 6.6 ± 1.5 I/II
14 MIR 55 12.83 ± 0.25 12.03 ± 0.09 11.34 ± 0.02 10.94 ± 0.02 10.89 ± 0.02 0.0 ± 0.6 III
15 ... 15.74 ± 0.05 14.23 ± 0.03 13.60 ± 0.03 13.33 ± 0.04 0.7 ± 0.5 III
16 NIR 17 ... ... 17.29 ± 0.23 15.55 ± 0.10 14.47 ± 0.08 23.8 ± 6.0 III?
17 MIR 60 ... 14.27 ± 0.03 12.28 ± 0.03 11.42 ± 0.02 10.97 ± 0.02 1.0 ± 0.5 I/II
18 MIR 62 ... 14.75 ± 0.03 11.73 ± 0.03 10.05 ± 0.02 8.83 ± 0.02 8.5 ± 1.5 II (AGN?)
20 MIR 71 13.58 ± 0.39 12.28 ± 0.10 11.25 ± 0.02 10.63 ± 0.02 10.49 ± 0.02 0.0 ± 0.5 III
21 WB 57/NIR 24 ... 14.57 ± 0.03 13.03 ± 0.03 12.35 ± 0.02 12.11 ± 0.03 1.0 ± 0.3 III
22 13.13 ± 0.24 11.83 ± 0.10 10.93 ± 0.03 10.32 ± 0.02 10.16 ± 0.02 0.0 ± 0.8 III
23 ... ... ... ... ... ... opt. source at 11′′

24 W 218/NIR 25 ... ... 13.38 ± 0.04 12.51 ± 0.05 11.93 ± 0.04 ... not reliable‡

27 MIR 76 ... 16.90 ± 0.11 13.19 ± 0.03 11.52 ± 0.02 10.80 ± 0.02 14.8 ± 1.5 III
28 MIR 80 12.39 ± 0.20 10.64 ± 0.13 9.04 ± 0.02 8.14 ± 0.04 7.55 ± 0.03 1.0 ± 1.0 II
31 MIR 81 11.93 ± 0.14 11.30 ± 0.07 10.89 ± 0.03 10.62 ± 0.02 10.58 ± 0.02 0.5 ± 0.5 II/III
33 MIR 83 ... 10.00 ± 0.22 8.90 ± 0.02 8.60 ± 0.02 8.45 ± 0.02 2.9 ± 0.5 III
34 V621 Ori ... 12.63 ± 0.02 10.57 ± 0.03 9.51 ± 0.02 9.11 ± 0.02 7.5 ± 0.7 III
35 MIR 86 ... 15.76 ± 0.07 12.85 ± 0.02 11.66 ± 0.02 11.05 ± 0.02 5.8 ± 0.5 II (AGN?)
36 ... 12.74 ± 0.03 11.07 ± 0.02 10.15 ± 0.02 9.87 ± 0.02 1.7 ± 0.5 II/III
38 12.60 ± 0.20 11.88 ± 0.08 11.31 ± 0.02 10.97 ± 0.02 10.93 ± 0.02 0.0 ± 1.5 III
41 ... 13.36 ± 0.03 11.84 ± 0.02 11.15 ± 0.02 10.92 ± 0.02 1.0 ± 0.5 II/III
42 V622 Ori 12.96 ± 0.23 11.75 ± 0.12 10.95 ± 0.02 10.43 ± 0.02 10.31 ± 0.02 1.5 ± 2.0 III
43 ... 13.76 ± 0.03 12.21 ± 0.02 11.51 ± 0.03 11.21 ± 0.02 0.0 ± 1.5 II/III
44 ... 14.29 ± 0.04 12.79 ± 0.03 12.06 ± 0.02 11.85 ± 0.03 1.5 ± 0.7 II/III

45 V623 Ori ... 14.10 ± 0.04 12.80 ± 0.08 ... 11.53 ± 0.10 ... II/III
46 ... 15.54 ± 0.06 13.73 ± 0.02 13.17 ± 0.03 12.92 ± 0.03 0.5 ± 0.6 II
47 13.03 ± 0.18 11.28 ± 0.14 9.40 ± 0.02 8.68 ± 0.04 8.39 ± 0.02 0.5 ± 0.5 III
48 ... 13.67 ± 0.03 12.28 ± 0.02 11.62 ± 0.02 11.37 ± 0.02 1.0 ± 0.6 III
49 ... 14.48 ± 0.04 12.57 ± 0.02 12.00 ± 0.03 11.67 ± 0.02 0.0 ± 0.7 III
50 ... 13.32 ± 0.03 11.56 ± 0.03 10.68 ± 0.02 10.39 ± 0.02 4.5 ± 1.0 III

† Infrared object class from the SED and Spitzer color-color diagrams and remarks for particular sources. We have removed spurious
sources (2, 8, 9, 11, 19, 25, 26, 29, 30, 32, 37, 39, 40). ‡ The photometry of Src. 24 is not reliable due to confusion of sources in the
region.



A
co
m
p
reh

en
sive

stu
d
y
o
f
N
G
C

2
0
2
3
w
ith

X
M
M
-N

ew
to
n
a
n
d
S
p
itzer

1
3

Table A3. Infrared photometric data for the sources detected in the XMM-Newton observation. W is for WISE, S is for Spitzer.

Src. ID Name FW
3.4 (mJy) FS

3.6 (mJy) FS
4.5 (mJy) FW

4.6 (mJy) FS
5.4 (mJy) FS

8.0(mJy) FW
12 (mJy) FW

22 (mJy) FS
24 (mJy) IR class†/Remarks

1 5.9 ± 0.4 5.3 ± 0.3 3.5 ± 0.3 3.5 ± 0.3 1.6 ± 0.6 1.0 ± 0.1 ... ... ... III

3 HD 37805/MIR 31 ... 518.2 ± 5.2 336.5 ± 3.5 ... 219.7 ± 2.5 149.8 ± 2.0 ... ... 55.4 ± 0.2 III

4 5.5 ± 0.4 5.1 ± 0.3 4.6 ± 0.3 3.8 ± 0.3 ... ... ... ... ... III

5 ... 3.6 ± 0.3 3.1 ± 0.3 ... ... ... ... ... ... III

6 13.1 ± 0.5 ... 8.2 ± 0.3 ... ... ... ... ... ... III

7 VLA 3/MIR 46 ... 3.0 ± 0.2 4.6 ± 0.3 ... 6.5 ± 0.5 13.6 ± 0.6 ... ... 23.4 ± 0.2 I

10 V615 Ori/MIR 52 998.9 ± 36.3 931.9 ± 9.0 995.2 ± 9.6 1468.9 ± 48.3 1020.1 ± 9.9 1207.5 ± 11.7 1215.3 ± 32.9 1396.89 ± 55.5 1067.0 ± 0.5 II

12 ... 0.1 ± 0.2 0.7 ± 0.3 ... 2.2 ± 0.7 6.1 ± 1.0 ... ... ... -

13 23.6 ± 1.3 20.9 ± 0.5 11.8 ± 0.4 16.1 ± 1.2 27.3 ± 1.0 ... ... ... .. I/II

14 MIR 55 14.0 ± 0.4 12.4 ± 0.4 7.4 ± 0.3 7.5 ± 0.2 9.3 ± 0.6 4.5 ± 0.4 ... ... ... III

15 1.5 ± 0.1 1.1 ± 0.2 1.3 ± 0.2 1.0 ± 0.1 ... ... ... ... ... III

16 NIR 17 ... ... ... ... ... ... ... ... ... III?

17 MIR 60 21.1 ± 0.8 20.7 ± 0.5 15.8 ± 0.5 19.59 ± 0.7 40.4 ± 2.1 62.5 ± 2.4 156.1 ± 18.4 559.7 ± 79.3 ... I/II

18 MIR 62 508.1 ± 21.9 595.6 ± 6.0 686.7 ± 6.8 700.5 ± 20.8 764.1 ± 10.0 957.6 ± 21.8 ... 2560.7 ± 162.5 1063.0 ± 1.0 II (AGN?)

20 MIR 71 ... 19.0 ± 0.5 12.4 ± 0.4 ... 7.4 ± 0.6 4.9 ± 0.6 ... ... ... III

21 WB 57/NIR 24 ... 5.6 ± 0.3 3.0 ± 0.3 ... ... ... ... ... ... III

22 28.7 ± 0.7 25.4 ± 0.6 ... 16.24 ± 0.4 11.6 ± 0.6 ... ... ... ... III

23 ... ... ... ... ... ... ... ... ... opt. source at 11′′

24 W 218/NIR 25 ... 295.4 ± 3.2 501.3 ± 5.1 ... 706.4 ± 7.3 1161.2 ± 12.6 ... ... ... not reliable‡

27 MIR 76 19.7 ± 0.5 20.3 ± 0.5 13.9 ± 0.4 14.7 ± 0.4 13.2 ± 0.7 6.0 ± 0.3 ... ... ... III

28 MIR 80 811.2 ± 30.8 785.9 ± 7.7 755.0 ± 7.4 944.0 ± 27.2 719.1 ± 7.2 789.3 ± 8.7 827.8 ± 21.1 785.5 ± 33.2 458.7 ± 0.3 II

31 MIR 81 19.1 ± 0.5 14.9 ± 0.4 10.3 ± 0.4 10.37 ± 0.3 5.3 ± 0.6 ... 4.3 ± 1.6 ... ... II/III

33 MIR 83 160.5 ± 4.3 134.1 ± 1.7 89.7 ± 1.2 91.1 ± 2.3 65.9 ± 1.2 36.0 ± 0.7 11.6 ± 1.5 ... 24.4 ± 0.2 III

34 V621 Ori 83.8 ± 2.2 74.7 ± 1.1 ... 50.8 ± 1.3 38.2 ± 0.8 ... 12.4 ± 0.5 ... ... III

35 MIR 86 22.6 ± 0.5 27.2 ± 0.6 27.1 ± 0.6 24.1 ± 0.6 30.5 ± 0.8 38.9 ± 0.8 58.3 ± 2.2 154.9 ± 9.6 ... II (AGN?)

36 39.8 ± 1.1 33.2 ± 0.6 21.0 ± 0.5 22.1 ± 0.5 16.1 ± 0.6 11.6 ± 0.5 12.8 ± 1.0 17.9 ± 4.5 ... II/III

38 13.9 ± 0.4 11.0 ± 0.4 7.8 ± 0.3 7.6 ± 0.2 8.1 ± 0.6 3.5 ± 0.3 ... ... ... III

41 15.4 ± 0.4 13.3 ± 0.4 ... 9.4 ± 0.2 6.9 ± 0.7 ... 3.4 ± 0.6 ... ... II/III

42 V622 Ori 24.5 ± 0.7 20.4 ± 0.5 13.0 ± 0.4 13.4 ± 0.3 6.3 ± 0.6 6.8 ± 0.3 4.8 ± 0.9 ... .. III

43 13.2 ± 0.4 11.1 ± 0.4 9.2 ± 0.4 9.6 ± 0.2 8.7 ± 0.6 7.5 ± 0.4 4.6 ± 0.1 ... ... II/III

44 6.3 ± 0.2 5.1 ± 0.3 3.7 ± 0.3 3.9 ± 0.1 3.4 ± 0.5 2.1 ± 0.3 2.6 ± 0.8 5.7 ± 1.6 ± ... II/III

45 V623 Ori 9.1 ± 0.2 ... ... 8.8 ± 0.2 ... ... 7.6 ± 0.9 13.7 ± 4.2 ... II/III

46 2.5 ± 0.1 2.3 ± 0.2 2.0 ± 0.2 1.8 ± 0.1 0.9 ± 0.5 ... 2.8 ± 0.7 9.9 ± 2.7 ± ... II

47 149.9 ± 4.0 131.1 ± 1.6 79.7 ± 1.1 85.4 ± 2.1 61.0 ± 1.1 35.6 ± 0.7 19.8 ± 1.1 9.6 ± 3.2 ... III

48 8.8 ± 0.2 9.5 ± 0.3 6.0 ± 0.3 5.6 ± 0.2 3.1 ± 0.6 ... ... ... ... III

49 7.4 ± 0.2 7.9 ± 0.3 5.3 ± 0.3 5.1 ± 0.1 3.0 ± 0.5 1.8 ± 0.3 ... ... ... III

50 24.8 ± 0.7 22.1 ± 0.5 15.3 ± 0.5 14.6 ± 0.4 11.9 ± 0.6 5.3 ± 0.4 ... ... ... III

† Infrared object class from the SED and Spitzer color-color diagrams and remarks for particular sources. We have removed spurious sources (2, 8, 9, 11, 19, 25, 26, 29, 30, 32, 37, 39, 40). ‡ The photometry of

Src. 24 is not reliable due to confusion of sources in the region.



14 M. A. López-Garćıa et al.

Table A4. Main results from the use of the fitting tool of Robitaille et al. (2007) with the NGC 2023 X-ray detected sources.

Src. ID Name AV Distance M⋆ R⋆ T⋆ Lbol logLX/Lbol Ṁ Stagea IR Classb

(mag) (pc) (M⊙) (R⊙) (K) (L⊙) (M⊙/yr)

1 0.05 398 0.48 1.02 3735 0.19 -2.68 ... ... III
3 HD 37805/MIR 31 0.10 56 1.45 1.77 7500 8.53 -13.02 ... ... III
4 1.11 299 0.35 0.77 3550 0.09 -3.41 ... ... III
5 0.25 398 0.38 0.52 3650 0.01 -3.08 ... ... III

6 8.81 630 0.66 0.76 4672 0.27 -3.59 ... ... III
10 V615 Ori/ MIR 52 5.19 371 2.96 4.39 7902 67.4 -69.89 1.82× 10−9 III II
13 1.90 500 1.09 10.14 4037 24.6 -27.52 1.03× 10−8 I I/II
14 MIR 55 0.00 575 1.60 1.88 5084 2.11 -3.37 ... ... III
15 0.24 479 0.29 0.76 3399 0.07 -3.78 ... ... III
17 MIR 60 1.19 575 0.48 5.40 3692 4.87 -8.36 2.15× 10−9 I I/II
18 MIR 62 5.33 457 3.42 2.63 11725 118 -121.6 4.51× 10−9 III II
20 MIR 71 0.76 347 1.31 1.46 4809 1.02 -3.18 ... ... III
21 NIR 24 0.42 245 0.24 0.72 3366 0.06 -2.57 ... ... III
22 0.34 288 1.35 1.52 4886 1.18 -3.29 ... ... III
27 MIR 76 7.65 275 0.35 1.92 3500 0.63 -3.30 ... ... III
28 MIR 80 3.11 436 2.96 4.39 7902 67.4 -69.09 1.82× 10−9 III II
31 MIR 81 0.88 616 1.85 1.53 7978 9.32 -12.03 1.70× 10−14 III II/III
33 MIR 83 2.15 280 1.93 2.78 6835 15.1 -4.92 ... ... III
34 V621 Ori 4.30 309 2.05 2.96 4828 4.27 -3.54 ... ... III
35 MIR 86 3.87 550 0.22 2.86 3119 0.93 -3.61 1.07× 10−7 I II
36 2.88 309 1.60 1.91 4847 1.82 -3.61 2.70× 10−13 III II/III
38 0.07 199 0.33 0.47 3500 0.01 -3.71 ... ... III
41 0.55 166 0.35 0.77 3567 0.09 -3.08 2.54× 10−10 III II/III
42 V622 Ori 0.11 407 1.52 1.83 4759 1.54 -3.27 1.13× 10−13 III III
43 0.37 182 0.24 0.82 3343 0.08 -3.84 1.53× 10−11 III II/III
44 0.51 257 0.31 0.29 3556 0.02 -3.52 7.88× 10−12 III II/III
45 V623 Ori 0.13 309 0.26 0.27 3401 0.01 -3.76 1.09× 10−8 III II/III
46 0.07 500 0.13 0.15 3052 0.002 -2.89 3.07× 10−11 III II
47 1.49 148 0.73 0.61 4331 0.13 -3.80 3.00× 10−14 III III
48 0.09 275 0.36 1.00 3548 1.43 -3.64 ... ... III
49 0.29 398 0.17 0.16 3042 0.20 -2.63 ... ... III
50 1.37 190 0.46 1.11 3706 0.21 -3.02 ... ... III

(a) Evolutionary stage parameter from Robitaille et al. (2007). (b) Infrared class based on the SEDs (Tables A2 and A3).



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 15

Figure A1. IRAS 100µm mosaic of the complex molecular system Orion A, Orion B and Monoceros. CO map contours are overplotted
as continuous lines. The light open circle in the center of the image represents the position and field of view of the XMM-Newton
observation.



16 M. A. López-Garćıa et al.

Figure A2. Spitzer-IRAC channel 1 (3.6 µm) image of the NGC 2023 nebula and its surroundings. The Horsehead Nebula is visible
at the right, bottom corner. X-ray sources detected in the XMM-Newton observation are marked and numbered in magenta. To avoid
confusion, the numbers of the likely spurious X-ray sources identified with numbers 11 and 12 (probable multi-detection of Src 10), 23
and 25 (multi-detection of Src 20) and 29 (multi-detection of Src 28) have been removed (see Section 3).



A comprehensive study of NGC 2023 with XMM-Newton and Spitzer 17

A3.1: Source 1 A3.2: Source 3 A3.3: Source 4 A3.4: Source 5

A3.5: Source 6 A3.6: Source 14 A3.7: Source 15 A3.8: Source 20

A3.9: Source 21 A3.10: Source 22 A3.11: Source 27 A3.12: Source 33

A3.13: Source 34 A3.14: Source 38 A3.15: Source 42 A3.16: Source 47

A3.17: Source 48 A3.18: Source 49 A3.19: Source 50

Figure A3. Spectral energy distribution of objects classified as class III in Table A3. The solid black lines indicate the best-fitting model
using Robitaille et al. (2007). Gray lines are other models with low χ2 values similar to that of the best-fitting model ((χ2 −χ2

best
)/χ2 6

10%). The dashed lines represent the stellar photosphere for the accepted model.



18 M. A. López-Garćıa et al.

A4.1: Source 10 A4.2: Source 13 A4.3: Source 17 A4.4: Source 18

A4.5: Source 28 A4.6: Source 31 A4.7: Source 35 A4.8: Source 36

A4.9: Source 41 A4.10: Source 43 A4.11: Source 44 A4.12: Source 45

A4.13: Source 46

Figure A4. Same as Fig. A3 for stars classified as class I/II, II and II/III in Table A3. The dashed lines are the unabsorbed stellar SED.