Available via license: CC BY 4.0
Content may be subject to copyright.
MNRAS 503, 292–311 (2021) doi:10.1093/mnras/stab388
Advance Access publication 2021 February 11
High-contrast and resolution near-infrared photometry of the core of R136
Zeinab Khorrami ,1‹Maud Langlois,2Paul C. Clark ,1Farrokh Vakili,3,4Anne S. M. Buckner,5
Marta Gonzalez,6Paul Crowther,7Richard W¨
unsch,8Jan Palouˇ
s,8Stuart Lumsden9and Estelle Moraux6
1School of Physics and Astronomy, Cardiff University, The Parade, Cardiff CF24 3AA, UK
2Universite de Lyon, Universite Lyon 1, CNRS, CRAL UMR5574, F-69561, Saint-Genis Laval, France
3Universite Cote d’Azur, OCA, CNRS, F-06304, Lagrange, France
4Department of Physics, Shahid Beheshti University, G.C., Tehran, Iran
5School of Physics and Astronomy, University of Exeter, Stocker Road, Exeter EX4 4QL, UK
6Universite Grenoble Alpes, CNRS, IPAG, F-38000 Grenoble, France
7Department of Physics and Astronomy, Hounsfield Road, University of Sheffield, Sheffield S3 7RH, UK
8Astronomical Institute of the Czech Academy of Sciences, Boˇ
cn´
ıII 1401/1a, CZ-141 00 Praha 4, Czech Republic
9School of Physics and Astronomy, University of Leeds, Leeds LS2 9JT, UK
Accepted 2021 February 5. Received 2021 February 5; in original form 2020 November 17
ABSTRACT
We present the sharpest and deepest near-infrared photometric analysis of the core of R136, a newly formed massive star cluster
at the centre of the 30 Doradus star-forming region in the Large Magellanic Cloud. We used the extreme adaptive optics of
the SPHERE focal instrument implemented on the ESO Very Large Telescope and operated in its IRDIS imaging mode for the
second time with longer exposure time in the H and K filters. Our aim was to (i) increase the number of resolved sources in
the core of R136, and (ii) to compare with the first epoch to classify the properties of the detected common sources between
the two epochs. Within the field of view (FOV) of 10.8 ×12.1 (2.7pc×3.0 pc), we detected 1499 sources in both H and K
filters, for which 76 per cent of these sources have visual companions closer than 0.2 . The larger number of detected sources
enabled us to better sample the mass function (MF). The MF slopes are estimated at ages of 1, 1.5, and 2 Myr, at different radii,
and for different mass ranges. The MF slopes for the mass range of 10–300 Mare about 0.3 dex steeper than the mass range
of 3–300 M, for the whole FOV and different radii. Comparing the JHK colours of 790 sources common in between the two
epochs, 67 per cent of detected sources in the outer region (r>3 ) are not consistent with evolutionary models at 1–2 Myr
and with extinctions similar to the average cluster value, suggesting an origin from ongoing star formation within 30 Doradus,
unrelated to R136.
Key words: instrumentation: adaptive optics – stars: luminosity function, mass function – stars: massive – open clusters and
associations: individual: R136.
1 INTRODUCTION
R136 is a very massive young star cluster that lies at the centre
of the Tarantula nebula in the Large Magellanic Cloud (LMC).
Hosting the most massive stars known in the Local Universe
(Crowther et al. 2010,2016), R136 provides a unique opportunity
to observationally study the formation of massive stars and clusters
in the earliest stages of their evolution. Our understanding of the
true nature of R136 has significantly improved with increasing
telescope resolution. The combination of photometry (Hunter, Shaya
& Holtzman 1995; Brandl et al. 1996; Andersen et al. 2009;De
Marchi et al. 2011; Cignoni et al. 2015), ultraviolet spectrometry
[STIS/MAMA, Crowther et al. (2016)], and visible [HST/FOS,
Massey&Hunter(1998)] and near-infrared (NIR) observations
(VLT/SINFONI, Schnurr et al. 2009) has resulted in more constraints
on the R136 stellar population and its most luminous stars.
E-mail: KhorramiZ@cardiff.ac.uk
In 2015, R136 was observed for the first time in the NIR by
the second-generation Spectro-Polarimetric High-contrast Exoplanet
Research1(SPHERE, Beuzit et al. 2019) instrument of the Very Large
Telescope (VLT). Thanks to SPHERE’s high-contrast and extreme
adaptive optics (XAO), the sharpest images from the central region
of R136 were recorded in J and K. For the first time, more than
1000 sources were detected in these bandpasses within the small
field of view (FOV) of IRDIS (10.9 ×12.3) covering almost
2.7 ×3.1 pc of R136 core (Khorrami et al. 2017). The SPHERE
data on R136 core were used to partially resolve and study the initial
mass function (IMF) covering a mass range of 3–300 Mat ages
of 1 and 1.5 Myr. The density in the core of R136 (r<1.4 pc)
was estimated and extrapolated in 3D for larger radii (up to 6pc).
Even at high angular resolution in the NIR, the stars in the core are
still unresolved due to crowding and central concentration of bright
sources. Using evolutionary models of stars more massive than 40
solar masses, Bestenlehner et al. (2020) find that the IMF of massive
1https://www.eso.org/sci/facilities/paranal/instruments/sphere.html
C
2021 The Author(s).
Published by Oxford University Press on behalf of Royal Astronomical Society. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium,
provided the original work is properly cited.
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 293
Tab le 1 . Exposure time log of VLT/SPHERE observations on R136.
Obs. date Filter Single/Total λcen(nm) λ (nm)
Exposure(s)
2015-09-22 BB-J 4.0/1200 1245 235
2015-09-22 BB-K 4.0/1200 2182 294
2018-10-10 BB-K 4.0/4352 2182 294
2018-11-14 BB-H 2.0/4352 1626 286
stars in R136 is top heavy with a power-law exponent γ=2.0 ±0.3.
They also estimate the age of R136 between 1 and 2 Myr with a
median age of around 1.6 Myr.
This paper presents the second epoch observation of R136 in the
NIR by VLT/SPHERE. These observations, obtained with longer
exposures in different NIR filters, aimed first to increase the number
of resolved sources in the core of R136, and second to enable
comparison with the first epoch to classify the property of the detected
common sources between these two epochs. We divided this paper as
follows. In Section 2, we explain in detail the data and the observing
conditions. We then present the photometric analysis in Section 3
and the method we use to correct for completeness in Section 4.
In Section 5, we describe the details of extinction estimation. On
the basis of the results, we investigate the stellar mass functions
(MFs) and the density in Section 6. In Section 7, we compare the
photometric analysis between the two epochs and between the two
sets of filters. We conclude by summarizing our results in Section 8.
2 OBSERVATIONS
We obtained the data via ESO Time Observation run (0102.D-0271)
to image R136 using the ‘classical imaging’ mode of IRDIS Langlois
et al. (2014) in H- and K-bandpass filters. For our purpose, we used
the same spectral band split into the two IRDIS channels to correct for
residual detector artefacts such as hot pixels and uncorrelated detector
noise among other instrumental effects. We maximized the image
sharpness of the total exposure by discarding the single frames with
poorer Strehl ratio (SR) and by correcting ‘a posteriori’ the residual
tip-tilt image motion on each short exposure before combining them.
Observations were performed in October/November (K/H) 2018,
achieving high dynamical range and high angular resolution imaging
in both the H and K bandpass, over a FOV of 10.8 ×12.1 (2.7pc×
3.0 pc), centred on the core of the cluster. The natural seeing was
0.69 ±0.10 arcsec in K and 0.58 ±0.05 arcsec in H during the
observations. The night was rated as ‘Clear’ meaning that less than
10 per cent of the sky (above 30◦elevation) was covered by clouds,
and the transparency variations were less than 10 per cent during the
exposures.
Our data consist of 1088/544 frames of 2.0/4.0 s exposures taken
with the IRDIS broad-band H and K filters (BB-H, BB-K). Table 1
shows the exposure time log of the two epoch observations and the
filters information. Fig. 1shows the final reduced images in the H
(left) and the K (right). These images are the deepest and sharpest
images taken from the core of R136 so far. The Wolf–Rayet (WR) star
R136a1 was used for guiding the AO loop of SPHERE confirming
the high level of performances even for faint guide stars that surpass
both NACO and MAD performances. Fig. 2compares the highest
resolution available images from the very central part of R136 taken
by HST (WFPC2byHunteretal.1995)inleft(topinV,bottominH),
VLT/MAD (Campbell et al. 2010) in the middle (top in K, bottom
in H), and IRDIS in right (top in K, bottom in H). These images
show the effect of better contrast, pixel-sampling, sensitivity, and
AO correction on resolving the stellar sources in the most crowded
part of R136.
The range of airmass during these observations was 1.52 –1.45
in K, and 1.70–1.55 in the H. A log of observations is provided in
Tab le 2. For comparison, we include the observing log for our first
epoch data sets taken in 2015.
We used the SPHERE pipeline package to correct for dark
(current), flat (fielding), (spatial) distortion, bad pixels, and thermal
background due to instrument and sky (in K). Furthermore, and
in order to reach the highest sensitivity and the largest number of
detectable sources, additional corrections were carried out on the
images. Based on a Gaussian fit using selected stars, we estimated and
corrected the subpixel image drifts before combining the individual
images. This allowed us to correct for residual tip-tilt errors with
an accuracy of a few mas. The final step is performed to combine
the parallel images on the left and right sides of the detector, after
the anamorphism correction. Some uncorrected background leaks
persist in our final K images due to thermal background fluctuations,
which are stronger in K than in H.
3 PHOTOMETRY
To analyse the final images, we applied the same method/tools as on
first epoch data (Khorrami et al. 2017). For the photometry, we used
the Starfinder package implemented in Interactive Data Language
(IDL, Diolaiti et al. 2000). Starfinder is designed for the analysis of
AO images of crowded fields, like the Galactic Centre, for instance,
as in Pugliese et al. (2002). This method determines the empirical
local point spread function (PSF) from several isolated sources in the
image and uses this PSF to extract other stellar sources across the
FOV.
For the present study, four and seven well isolated stars (within
0.47 ×0.47 ) were used to extract the PSF in the K and H,
respectively. The FOV in K is more crowded than in H (see Fig. 1), so
finding the bright isolated stars (within 0.47 ×0.47 )inKisharder
than in H. That is why we limit our selection of stars to be used for PSF
extraction to four in K. Fig. A3 shows the separation with the closest
neighbour for each star in H and K. The extracted PSFs were used as
an input for stellar sources detection by Starfinder. The full width at
half-maximum (FWHM) of the extracted PSFs are 58.8 and 63.7 mas
in H and K, with 1658 and 2528 sources detected, respectively. We
stopped the source extraction after obtaining a minimum correlation
coefficient of 65 per cent and 80 per cent, in K and H, between the
extracted star with the locally determined PSF according to Starfinder
procedures. In H, we put a higher threshold on minimum correlation
coefficient than in K, because the AO correcting efficiency in H
degrades faster as a function of distance from R136a1 – which is the
reference star for the AO loop – than in K. The isoplanatic angle in
H is smaller than in K, so at larger radii from R136a1, the PSF is not
centrosymmetric. Indeed, stars with higher correlation coefficients,
i.e. more similarity to the PSF, represent higher reliability on their
photometry estimation.
In addition to the correlation coefficient criterion, we applied the
limit of standard deviation from the sky brightness (σsky) for stopping
the extraction of sources by Starfinder, i.e. the local PSF maximum
value must exceed 2σsky over the background. Fig. 3shows the
signal-to-noise ratio (SNR) of 1658 and 2528 detected sources in H
(purple circles) and K (green squares), respectively. Common sources
(1499) between H and K are shown as filled circles/squares. The solid
horizontal line shows the SNR =2.0 where some of the detected
sources have SNR less than this value. These stars are located further
from the central region where the local σsky is smaller than the σsky of
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
294 Z. Khorrami et al.
Figure 1. R136 core images taken by VLT/SPHERE/IRDIS in the H (left) and the K (right). FOV is 10.8 ×12.1.
Figure 2. Comparison of R136 central images at different wavelengths with the highest available angular resolution telescopes. Top, left to right: HST/WFPC2
in V , VLT/MAD in K, and VLT/SPHERE in K. Bottom, left to right: HST/NICMOS/NIC1, VLT/MAD, and VLT/SPHERE in the H. The identification of bright
sources is shown in Fig. A2.
the whole FOV. 76 per cent of these sources have visual companions
closer than 0.2 , which is much higher than the value found in
the first epoch in 2015 data. Fig. 4shows the histogram of closest
neighbour (visual companion) separation from each source versus its
distance from the centre of R136, in 2015 (top) and 2018 (bottom)
data. Comparing the histogram of closest companions between two
epochs, 85 per cent of the new detected sources in 2018 have visual
companions with separation less than 0.2 .
The SR in H and K is determined as 0.71 ±0.05 and 0.83 ±0.03,
respectively, on average within 5 arcsec from the core. These
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 295
Tab le 2 . Observing condition for R136 data in 2015 and 2018.
Data SR Seeing(”) Airmass Nphot
2015-J 0.40 ±0.05 0.63 ±0.1 1.54–1.59 1110
2015-K 0.75 ±0.03 0.63 ±0.1 1.61–1.67 1059
2018-H 0.71 ±0.05 0.58 ±0.05 1.70–1.55 1658
2018-K 0.83 ±0.03 0.69 ±0.10 1.52–1.45 2528
Figure 3. Signal-to-noise ratio of detected sources in H (purple circles) and
K (green squares) versus their magnitudes. Filled dots are common sources
between H and K data. The solid horizontal line shows the SNR =2.0.
Figure 4. Histogram of the separation of the close detected sources versus
their distance from the core of R136. Top: 818 common sources between J
and K data in 2015. Bottom: 1499 common sources between H and K data in
2018.
estimates are assessed from the SPARTA files recorded during
SPHERE runs simultaneously with the AO-corrected images of
R136. They are based on the slope measurements of the Shack–
Hartmann wavefront sensor in Sphere AO for eXoplanet Observation
(SAXO) by extrapolating the phase variance deduced from the
reconstruction of SAXO open-loop data using deformable mirror,
tip-tilt voltages, and wavefront sensor closed-loop measures (Fusco
et al. 2004). The method has been proved quite robust for FOV
smaller than the anisoplanetism in the past for differential imaging
by NACO (Maire et al. 2014). We are therefore quite confident on
our photometry corrected for the SR effect in H and K. To convert
stellar fluxes to magnitudes, we set the zero-point magnitude to
the instrumental zero-point (one ADU/s in H and K, are 25.2 and
24.8 magnitudes, respectively). To double check our calibration,
we compared H and K magnitudes of 26 sources (located r≥3
from R136a1) common between our catalogue and the VLT/MAD
(Campbell et al. 2010)inHandK.
2We cannot compare our zero-
points with the MAD/H and K magnitudes in the central 2.8 region
because of the completeness issue and the very low AO correction
(SR ∼15–30 per cent) of their data. Fig. 2shows the completeness
and contrast problem of the previous data, leading to an inability to
distinguish certain sources. By way of example R136a6 (H19) and
H263– are very close in terms of brightness (flux ratio of about
90 per cent) and location (separation of about 70 mas). These bright
O-type stars are most easily detectable in our data compared with
others (Figs 2and A2). Cases like these close bright stars, bring
confusion even in the HST spectroscopic analysis (Bestenlehner et al.
2020).
The K magnitude of R136a1 and R136a2 (11.15 mag and
11.43 mag) increased about 0.08 and 0.11 mag, compared to their K
mag in the first epoch (11.07 mag and 11.32 mag). The brightness of
R136a3 and R136b in K (11.45 mag and 11.67 mag) is consistent with
the first epoch K data (11.44 mag and 11.66 mag). R136c in the K
(11.31 mag) is 0.17 mag brighter than in 2015 (11.48 mag), making
this source the second brightest in our K catalogue. In previous
studies, R136c shows indications of binarity based on its strong
X-ray emission (Portegies Zwart, Pooley & Lewin 2002; Townsley
et al. 2006; Guerrero & Chu 2008) and possible low-amplitude RV
variations (Schnurr et al. 2009).
Comparing the brightness of common detected sources within two
epochs, the K magnitude of 56 per cent of the sources has changed
more than twice their photometric errors.4To make sure that this
is not due to Starfinder photometry analysis (which is based on
the reference input PSF), we have used DAOPHOT algorithm for this
analysis as well. The variation of the K magnitude of these sources
still remains.5The K mag of about 67 per cent of the sources located
outside of the core (r>3 ) has changed more than twice their
photometric errors. Since the observations were made in different
nights and the AO correction in the outer part of the FOV is not as
good as its centre, this inconsistency is probably an observational
effect; otherwise, it is originated from physical reasons like variables
or multiple systems with periods less than 3 yr.
4 COMPLETENESS
We produced 6.2 ×104artificial stars (500 per magnitude) in H and
K, in order to determine the completeness as a function of magnitude.
These artificial stars are created using the same PSF as we used for
source extraction process. They are added to the original image one
by one (500 times per magnitude) and the same photometric tools
and criteria to determine how often they can be recovered from the
source extraction process are used. These tests were performed one
star at a time to avoid the artificial stars from affecting one another.
Fig. 5shows the completeness values as a function of magnitude in H
(purple dots) and K (green squares). The completeness is 80 per cent
at K=20 mag and H=20.5 mag.
2See the comparison between two catalogues in Fig. A7.
3Identification from Hunter et al. (1995)–seeFig.A2 to check the position
of these stars in the core.
4See Fig. A4 to compare the K mag of these sources within two epochs.
5See Fig. A5 to compare the photometric analysis using Starfinder and
DAOPHOT.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
296 Z. Khorrami et al.
Figure 5. Average value of completeness values for the whole FOV as a
function of magnitude using 62 000 artificial stars in H (purple dots) and K
(green squares).
The completeness value in each magnitude is the average value
for detection of 500 artificial stars that are distributed randomly in
the FOV. But the completeness varies across the FOV depending
on how crowded and bright that region is. Using these 6.2 ×104
artificial stars, we created completeness maps covering magnitude
range of 17–22 (five maps), both in H and in K. Fig. 6shows two
examples of completeness map in the magnitude of 19–20 both in H
(top) and in K (bottom). The average value of completeness is about
90 per cent both in H and in K (Fig. 5), but one can notice that the
completeness in the very central part of the cluster is much lower than
the outskirts. In the K, we can also see the effect on the completeness
of the thermal background noise from the instrument lying in the
arcs located in the far east and west parts of the FOV as seen in
Fig. 6.TheblackplusesinFig.6shows the position of the observed
sources (398 in H and 410 in K) within the magnitude range of 19–
20. This plot clearly shows the effect of completeness on the detected
sources that are used to create MF. To overcome this limitation, we
corrected the MF for all stars fainter than 17 magnitude. Each star
contributes a mass distribution to the MF (see Section 6). When
we added stars with a given mass to the histogram of mass, we
corrected the contribution of each star to the MF (fainter than 17
mag) for completeness using the completeness maps in every H and
K magnitudes. We know that the completeness in a given magnitude
varies across the FOV in both filters, so depending on the position and
magnitude of an observed star, we estimated its completeness using
these maps (see Fig. 6). Since the average value of completeness
at 17 mag is above 95 per cent (above 80 percent at the core), we
stopped the completeness correction for sources brighter than K=
17 mag.
5 EXTINCTION
In order to fit the evolutionary models to our data and estimate the
stellar masses, we need to measure the extinction first. We used two
methods to estimate the extinction for the spectroscopically observed
massive stars in the core of R136.6In both methods, we adopt
the LMC distance modulus (DM) of 18.49 magnitude estimated by
6List of spectroscopically known stars is provided in Table A1. List of stars
brighter than 16 mag in K (138 sources) is provided in Table A2 with their
SPHERE (J, H, and K) and HST (U, V, and B) magnitudes. See Fig. A2 to
check the position of these sources in our data.
Figure 6. Completeness map for magnitude range of 19–20 in H (top) and in
K (bottom). Black pluses shows the position of photometric detected sources
within the given magnitude range. colour shows completeness in percents.
Pietrzy´
nski et al. (2013), which is consistent with the value suggested
by Gibson (2000) for LMC.
Method I: We used the effective temperature (Teff) and luminosity
(logL) of 49 stars by Bestenlehner et al. (2020) in the optical.
The values of these temperatures and luminosities are provided in
Tab le A1. We also chose a grid of isochrones at different ages (from
0.1 up to 10 Myr) with the LMC metallicity (Z=0.006) from the
latest sets of PARSEC evolutionary model7Bressan et al. (2012),
which is a complete theoretical library that includes the latest set
of stellar phases from pre-main sequence to main sequence and
covering stellar masses from 0.1 to 350 M. By fitting the PARSEC
isochrones to the observed stellar parameters of each star (Teff, logL),
we estimated the age and intrinsic colour for each star with an error.
7http://stev.oapd.inaf.it/cgi-bin/cmd, YBC version of bolometric corrections
as in Chen et al. (2019).
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 297
Figure 7. Histogram of extinction in H and K filters from observations using
49 spectroscopically known stars (Bestenlehner et al. 2020)intheFOV.
The extinction in H and K (AHand AK) is estimated by comparing
these intrinsic H and K magnitudes with the observed values from our
catalogue. Fig. 7shows the histogram of the extinction values of those
49 stars. The mean extinction in H and K and the mean colour excess
are AH=0.38 ±0.55, AK=0.22 ±0.57, and E(H−K)=0.11 ±0.21.
These values derived by fitting a Gaussian function on the extinction
distributions, while these distributions are not exactly Gaussian and
symmetric. That is why the mean colour excess is slightly (0.05 mag)
higher than AH−AK. The large errors on these values result from
the errors in the reported stellar parameters (Teff and log L) from the
spectroscopic analysis in the optical (see Table A1).
Method II: Using the stellar parameters from Bestenlehner et al.
(2020), we adopt the closest stellar atmosphere model of that star.
We used the grids of TLUSTY8atmosphere models (Hubeny &
Lanz 1995) for O-type (Lanz & Hubeny 2003) stars with the LMC
metallicity (Z=1/2Z). The apparent magnitudes of these stars in
the SPHERE/IRDIS H and K filters are calculated by MYOSOTIS9
(Khorrami et al. 2019). We adopt AF555Wfor these sources from
Crowther et al. (2016) as an input for MYOSOTIS to estimate the
H and K magnitudes of these stars using the synthetic extinction
curves10 from Draine (2003a,2003b,2003c), Li & Draine (2001),
and Weingartner & Draine (2001)forRV=4.0. The simulated H and
K magnitudes of 55 per cent (85 per cent) of these sources (except
for three WRs) are less than 0.1 (0.2) magnitudes different from the
observed values. The simulated extinction and colour excess of these
49 stars in H and K are AH=0.46 ±0.05, AK=0.33 ±0.05, and E(H
−K)=0.13 ±0.02, shown in Fig. 8. One can compare the result
of these simulations with observation in Fig. 9(top), where black
circles are 49 spectroscopically known stars from Bestenlehner et al.
(2020) and black squares are their simulated magnitudes.
The colour excess, E(H−K), estimated by two methods is
consistent, and the extinction values, AHand AK,inMethod II are
0.1 mag higher than in Method I. We adopt the extinctions of AH=
0.45 and AK=0.35 magnitudes, which are consistent with the two
analyses (within the error bars) and De Marchi & Panagia (2014).
The colour excess of E(H−K)=0.1 is also consistent with Tatton
et al. (2013) and with the one used in previous studies by Campbell
et al. (2010).
8http://nova.astro.umd.edu/Tlusty2002/tlusty-frames-cloudy.html
9Star cluster simulator tool https://github.com/zkhorrami/MYOSOTIS
10www.astro.princeton.edu/∼draine/dust/dustmix.html
Figure 8. Histogram of extinction in H and K filters modelled based on
observed stellar parameters of Teff, logL, and logg from Bestenlehner et al.
(2020)andAF555Wfrom Crowther et al. (2016).
6 MASS FUNCTION AND CORE DENSITY
Fig. 9shows the colour magnitude diagram (CMD) of 1499 detected
sources common between the H and K data (top) for the full FOV in
left (N=1499), for inner r<3 region (N=627) in middle, and
for outer r>3 region (N=872) in right.11 Red, yellow, and blue
solid lines are PARSEC isochrones at DM =18.49 with AK=0.35
and E(H−K)=0.1 at the ages of 1, 1.5, and 2 Myr, respectively.
We note that there is a large scatter in the CMD, especially for the
lower part of the main-sequence and pre-main-sequence sources,
which are located (detected) in the outer region. A significant scatter
in CMD and colour–colour diagrams of 30 Doradus has previously
been reported in the visible and NIR (Hunter et al. 1995; Andersen
et al. 2009; De Marchi et al. 2011; Cignoni et al. 2015). The scatter
is likely due to a combination of observational confusion (affected
mainly by the visual multiple systems or variables), photometric
errors, differential extinction, a possible age spread, colour excess
due to warm circumstellar matter, and ongoing star formation within
30 Doradus. Pre-main-sequence stars are often associated with
circumstellar discs and outflows, which will introduce additional
extinction for the clusters low-mass content. Brandl et al. (1996)
found that the extinction varies significantly from star to star within
the cluster, with the range of 1–2 mag. The HST observations also
reveal the presence of considerable differential extinction across the
30 Doradus region. De Marchi et al. (2011) quantified the total
extinction towards massive main-sequence stars younger than 3 Myr
to be in the range of 1.3 <AV<2.1. Fig. 9shows that the evolutionary
models (isochrones) cannot be fitted for these scattered sources. We
prefer to consider an (larger) error of 0.5 mag on the extinction to
estimate the stellar mass range for each star at a given age. We
estimated the stellar masses just for common sources between H and
K data (1499 sources) using both their H and K magnitudes fitted to
PARSEC isochrones at three different ages: 1, 1.5, and 2 Myr.
The MF is plotted in Fig. 10 considering a Gaussian distribution for
each stellar mass. We convert the photometric/extinction uncertainty
into a mass (Gaussian) distribution. This Gaussian uncertainty in
the mass of each star is accounted for, when constructing the MF.
11In the first epoch data, in J data, the PSF is not centrosymmetric at large
distances from R136a1, starting typically around 3 (approximately half of
FOV radius). For the sake of comparison of two epochs, we choose r=3
for analysing detected sources in/out of this radius.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
298 Z. Khorrami et al.
Figure 9. Top: CMD of detected sources common between the H and K data for the full FOV in left (N=1499), for central r<3 region (N=627) in middle,
and for outer r>3 region (N=872) in right. Red, yellow, and blue solid lines are PARSEC isochrones at DM =18.49 with AK=0.35 and E(H−K)=0.1
at the ages of 1, 1.5, and 2 Myr, respectively. The black circles are the 49 spectroscopically known stars from Bestenlehner et al. (2020) and the black squares
represent the simulated magnitudes for these 49 sources (explained in Section 6: Method II). Bottom plots are same as top plots but for 818 sources detected in
J and K in our first epoch observation (Khorrami et al. 2017).
Each star fainter than K=17 (about 13.4 Mat 2 Myr) is corrected
for completeness, according to its brightness and its location in the
FOV (see Section 4 for more information). Fig. 10 shows the MF
at three different ages (1, 1.5, and 2 Myr) for the whole FOV (top),
central r≤3 region (middle), and outer r>3 region (bottom). The
MF slope values in these regions, at different ages and mass ranges,
are provided in Table 3, both for the completeness-corrected (CC)
and not corrected MF (NC). The MF slopes are calculated for the
minimum mass of 3 M(about K=19.8 at 2 Myr) where the average
value of completeness is above 85 percent (Fig. 5) but the difference
between CC and NC is not negligible for the inner region of the
cluster where the completeness reaches to 30 per cent (see Fig. 6)
and this affects the MF slope for the whole FOV (top values in
Tab le 3). The slopes for the mass range of 3–300 Mfor the inner
region are flatter than those for the outer region and consequently
for the whole FOV. This can be explained by completeness since
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 299
Figure 10. Mass function at 1, 1.5, and 2 Myr for the whole FOV (top),
central r<3 region (middle), and outer region r>3 (bottom). Grey
symbols are the values without completeness correction. See Table 3for the
MF slopes values.
the CC and the NC are not compatible. But for the mass range of
10–300 Mwhere the data are complete (see Figs 5and A1)and
the MFs are not affected by completeness (compare CC and NC
in Table 3), the slope values in the inner region are slightly flatter
than those in the outer region and consequently than the whole FOV.
Comparing the stellar population in the inner region with the outer
region in CMD (Fig. 9), detected sources are compatible with the
evolutionary models in the inner region [Fig. 9(top middle)], but for
the outer region, detected sources differ from the isochrones. About
50 per cent of the high-mass stars are located in the right side of the
main sequence, far from the evolutionary models. If these population
truly belong to R136, then they have K bandpass excess emission due
Tab le 3 . CC is the slope of the completeness-corrected MF and NC is the
not-corrected one. The stellar masses are estimated using parsec isochrones
with DM =18.49, AK=0.35, and E(H−K)=0.1 at different ages (first
column) and mass ranges (second column) for the whole FOV (top), and
central r<3 region (middle), and outer r>3 region (bottom). Nis the
number of detected sources and NCis the number of stars fainter than K=
17 mag that are corrected for completeness according to their position on
the image and their brightness (see Section 4).
Age (Myr) Mass range (M)CC NC
All, N=1499, NC=1226
1.0 3–300 −0.93 ±0.08 −0.89 ±0.09
1.5 3–300 −0.98 ±0.09 −0.94 ±0.10
2.0 3–300 −1.04 ±0.10 −0.99 ±0.11
1.0 10–300 −1.17 ±0.02 −1.16 ±0.03
1.5 10–300 −1.26 ±0.01 −1.25 ±0.01
2.0 10–300 −1.34 ±0.03 −1.33 ±0.03
r<3 ,N=627 , NC=444
1.0 3–300 −0.80 ±0.10 −0.74 ±0.12
1.5 3–300 −0.83 ±0.11 −0.76 ±0.14
2.0 3–300 −0.87 ±0.14 −0.79 ±0.16
1.0 10–300 −1.12 ±0.06 −1.10 ±0.06
1.5 10–300 −1.19 ±0.03 −1.17 ±0.03
2.0 10–300 −1.26 ±0.03 −1.24 ±0.03
r>3 ,N=872 , NC=782
1.0 3–300 −1.14 ±0.08 −1.13 ±0.08
1.5 3–300 −1.24 ±0.08 −1.22 ±0.08
2.0 3–300 −1.32 ±0.09 −1.30 ±0.10
1.0 10–300 −1.31 ±0.08 −1.31 ±0.08
1.5 10–300 −1.47 ±0.06 −1.47 ±0.05
2.0 10–300 −1.60 ±0.05 −1.59 ±0.04
to hot dust (K-excess sources). The MF slopes in the outer region are
complete for both the mass ranges and the slope values are steeper
than the inner region but might be the effect of the existence of these
K-excess sources.
The MF slopes for the massive part (M>10 M) in the inner
region are in line with Bestenlehner et al. (2020)(γ∼2.0 ±0.3)
and for the whole, FOV is consistent with Schneider et al. (2018)
(γ∼1.90+0.37
−0.26) for the 30 Doradus region, considering the large error
bars from these studies. Massey & Hunter (1998) found the MF slope
of −1.4 <<−1.3 for the mass range of 15–120 M,whichis
consistent with the slope for the whole FOV at 2 Myr.
To check the presence of mass segregation, we used the minimum
spanning tree (MST) algorithm by Allison et al. (2009) comparing
the length of the MST connecting massive stars to that connecting
randomly selected stars. Mass-segregation ratio (MSR, equation 1 in
Allison et al. 2009) defined as the average random path length of the
MST and that of the massive stars. NMST is the number of selected
massive stars and randomly selected stars. Fig. 11 shows MSR
calculated for different NMST subsets. MSR reaches to a constant
value when NMST is larger than the total number of massive stars (m
>10 M) in the FOV, which is about 339 using 1.5-Myr isochrones.
The bottom plot in Fig. 11 shows the MSR calculated within different
radii of the cluster, centred at R136a1. The numbers within parenthe-
ses denote the total number of stars more massive than 10 M,which
are chosen for NMST. The colour shows the ratio of number of massive
stars (m>10 M) to the total number of stars (down to 1 M)
within each radius. This plot shows the variation of MSR locally for
different sample of NMST.MSR is less than 1.2 within the whole FOV
and for NMST >350, and it is larger than 1.0 in all the conditions.
The values of are point to a moderate but significant (above 2
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
300 Z. Khorrami et al.
Figure 11. Top: the mass-segregation ratio (MSR) calculated for different
NMST subsets. Bottom: MSR calculated within different radii of the cluster,
centred at R136a1. The numbers within parentheses denote the total number
of stars more massive than 10 M, which are chosen for NMST. The colour
shows the ratio of number of massive stars (m>10 M) to the total number
of stars within each radius.
sigma) degree of mass segregation. This must be taken with caution,
considering the spatial distribution of completeness (Fig. A1). The
fact that, globally, we detect less low-mass stars towards the centre
of the cluster can have an effect on the mass segregation ratio. If
more low-mass stars are injected in the centre of the cluster, the size
of the N most massive stars would still be the same, but the average
size of a group of N stars within the complete sample is expected to
decrease. The values of MSR would also decrease in that case, but
quantifying this is out of the scope of this work.
Still, these MF slopes in our study are limited to the resolution
of the instrument (55 mas in K) and in future, using higher angular
resolution data, we may resolve binaries and low-mass stars, which
affects the shape of MF.
Using the stellar masses estimated at the age of 1, 1.5, and 2 Myr,
which are corrected12 for the completeness (for 1226 sources), we
plot the 2D (projected) mass density at a given radius, i.e. the
mass between rand r+dr divided by the corresponding area [ρ,
Fig. 12 (top)]. We used an Elson–Fall–Freeman profile Elson, Fall
&Freeman(1987)tofitρin the core of R136 (equation 1) up to the
maximum radius of Rmax =1.32pc from R136a1.
ρ[Mpc−2]=ρ0
(1 +r2
a2)γ+1
2
(1)
We estimated the central mass density of ρ0=(3.80+1.57
−1.11)×
104[Mpc−2], ρ0=(2.51+0.72
−0.56)×104[Mpc−2], and ρ0=
(2.24+0.58
−0.46)×104[Mpc−2], at the ages of 1, 1.5, and 2 Myr,
respectively. The central density at 1 Myr is higher than the other
ages because most of the massive stars are located at the central part
12The reported values are consistent for ones derived without completeness
correction. To compare these plots w/o completeness at 1.5 Myr, see Fig. A6.
Figure 12. Top: Projected mass density (ρ)profileofR136inIRDISFOV
centred on R136a1, using 1499 stars in H–K. The stellar masses are estimated
at the age of 1 (purple squares), 1.5 (green circles), and 2 Myr (red triangles)
with extinction values of AK=(0.35 ±0.5) mag, E(H−K)=(0.1 ±0.1) mag
in H and K. Equation (1) is used to fit the purple, green, and red solid lines
to the data at 1, 1.5, and 2 Myr, respectively. Centre: Surface densities ()
within a given radius. Bottom: Total stellar masses (Mtot ) within a given
radius. Solid blue lines show the average value of surface density (=2.7×
103Mpc−2) and total mass (Mtot =1.5×104M)uptoRma x =1.3pc.
of the cluster [see Fig. 9(middle)], and for these stars, the stellar
mass estimation is higher at lower ages.
The total mass of the cluster (Mtot) and the cluster’s surface density
(), which is the projected mass density within a given radius, i.e.
all the mass between 0 and rdivided by the area of the corresponding
circle, are shown in Fig. 12.and Mtot become consistent at different
ages for larger radii (starting at 0.3 pc up to 1.3 pc), where lots of pre-
main-sequence and low-mass stars are located/detected. The surface
density of R136 at R=0.3pc (and Rmax =1.3pc) reaches to 1.4+0.7
0.4×
104[Mpc−2](and2.7+1.1
0.6×103[Mpc−2]), and the total mass
down to 1.0 ±0.1 Mis 4.0+1.7
−1.2×103M(and 1.5+0.5
−0.4×104M),
respectively.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 301
The densities and total estimated masses at 1 Myr are slightly
higher than the values reported in the first epoch [compare Fig. 12 to
fig.13inKhorramietal.(2017)].
Fitted γand aparameters (in equation 1) vary from 0.8 to 0.85
and 0.14 to 0.19, respectively, depending on the selected age.
Comparing these values to the previous ones estimated for 818
stars in the 2015 data in J–K data (Khorrami et al. 2017), ρ0and
aare consistent within the given errors to their previous values of
ρ0=(3.89+1.60
−1.14)×104[Mpc−2]anda=0.17 ±0.05, but γis
decreased by about 0.4.
7 JHK COLOURS
Among 818 sources detected in J and K in 2015 data, and 1499
sources detected in H and K in 2018, we could detect 790 sources
common in total (J, H, and K). 30 per cent of the newly detected
sources in 2018 had enough SNR to be detected in the K data in 2015
(they are brighter than the faintest star in the 2015 K catalogue), but
these sources were not listed in the J since the AO correction for
J data is not as good as in K. These sources were located either in
the central part of the cluster where AO halo (uncorrected part of
the stellar signal with the FWHM of seeing) was large leading to
decreasing their SNR or in the outer part of the cluster where the
PSF was distorted [see photometric selection criteria in Khorrami
et al. (2017)] due to anisoplanatism.
For these 790 common sources between two epochs, we have
plotted colour–colour diagram in (J–K) versus (H–K) for 413 stars
in the central region r<3 (Fig. 13 top) and 377 sources in the outer
region r>3 [Fig. 13 (bottom)]. The black circles in Fig. 13 show 49
stars studied by Bestenlehner et al. (2020) in the centre of R136, and
the black solid line is the PARSEC isochrone at 2 Myr. These plots
show that the detected sources in the inner region of R136 are more
consistent with the evolutionary models than the sources in the outer
region of the cluster. Comparing the CMD in H–K and J–K (Fig. 9)
for inner and outer regions, the dispersion in the colour of the detected
sources in the outer region is higher than in the inner region. In the first
epoch, 48 per cent of the detected sources in J and K were in the outer
region, but in the second epoch, this number increases to 58 per cent.
Only about 30 per cent (70 per cent) of the new detected sources in the
second epoch (H–K) are located in the inner (outer) region. This can
be explained by the effect of incompleteness in the core, so that even
with the longer exposure time and better observing condition (e.g.
SR >70 per cent), central region remains too bright for low-mass
MS stars to be detected. The completeness in H and K, for the central
region of R136, is less than 50 per cent (20 per cent) for the magnitude
range of 19–20 (20–21) in our completeness maps (Fig. A1).
8 CONCLUSIONS
We presented a new photometric analysis of the core of R136 using
the second epoch data from VLT/SPHERE instrument in the near-IR.
We observed R136 in H and K filters in better atmospheric observing
conditions and longer exposure time than the first epoch. This enabled
us to detect twice as many sources (in H and K) as we have detected
in the first epoch (in J and K) in the FOV of IRDIS (10.8 ×12.1)
covering almost 2.7 ×3.0 pc of R136 core. Among 1658 and 2528
detected sources in H and K, respectively, we found 1499 common
sources between these two sets of data, where 76 per cent of these
sources have visual companion closer than 0.2, which is higher than
the value found in the first epoch in 2015 data (Khorrami et al. 2017).
About 71 per cent of the newly detected sources are located in the
outer region (r>3 )ofthecluster.
Figure 13. Colour–colour diagram for the common sources between J, H,
and K, within two epochs J–K (2015) and H–K (2018), for the central region
r<3 (top) and outer region r>3 (bottom). The colour indicates the K
magnitude of detected sources and the black circles represent 49 stars studied
by Bestenlehner et al. (2020) listed in Table A1. The black solid line is the
PARSEC isochrone at 2 Myr with E(H−K)=0.1 and E(J−K)=0.25. The
right-hand side of the curve represents pre-main-sequence part of isochrone,
followed by the main-sequence locus on the straight line on the centre and
the post-main-sequence part on the top.
Using the stellar parameters (Teff, log g, and logL) of 49 stars
studied spectroscopically by Bestenlehner et al. (2020) in the optical,
PARSEC isochrones, TLUSTY SEDs (Hubeny & Lanz 1995;Lanz&
Hubeny 2003), and synthetic extinction curves from Draine (2003a,
2003b,2003c), Li & Draine (2001), and Weingartner & Draine
(2001), we estimated the extinction at DM of 18.49 magnitude, using
two methods (Section 6). We adopt the extinctions of AH=0.45 and
AK=0.35 magnitudes, which are consistent with the two methods
(within the error bars) and De Marchi & Panagia (2014). The colour
excess of E(H−K)=0.1 is also consistent with Tatton et al. (2013)
and with the one used in previous studies by Campbell et al. (2010).
Consequently, the stellar masses were calculated at the ages of 1.0,
1.5, and 2.0 Myr.
The MF slope for 1, 1.5, and 2 Myr isochrone at the inner (r<3 )
and outer regions (r>3 ) of the cluster is estimated and shown in
Tab le 3. One can check the effect of incompleteness by comparing
the MF slopes before and after the completeness correction, shown as
NC and CC, respectively, in Table 3. The completeness-corrected
MF slopes for the whole FOV (1Myr =−0.93 ±0.08, 1.5Myr
=−0.98 ±0.09) are consistent with the values derived from the
photometric analysis of the first epoch data (Khorrami et al. 2017)
for the mass range of 3–300 M(1Myr =−0.90 ±0.13, 1.5Myr =
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
302 Z. Khorrami et al.
−0.98 ±0.18) and are closer to the Salpeter value (Salpeter 1955)
for the high mass range of 10–300 M(1.5Myr =−1.26 ±0.01,
2Myr =−1.34 ±0.03). The MF slopes for the mass range of 10–
300 Mare about 0.3 dex steeper than the mass range of 3–300 M,
for the whole FOV and for different radii. The MF slopes in the
inner region are shallower than in the outer region for different mass
ranges. In Fig. 11,MSR with a value (about 2 sigma) above 1 shows
a degree of mass segregation. Considering the spatial distribution of
completeness (Fig. A1), both MF slope in the core and MSR would
decrease, if the number of low-mass stars increases in the centre
where the completeness is very low.
We corrected the MF for completeness for sources fainter than
17 mag (1226 stars in the whole FOV and 444/782 stars in inner/outer
region). Still these values are low limits to the steepness due to
incompleteness and central concentration of bright stars.
The surface density of R136 at R=0.3 pc (and Rmax =1.3 pc)
reaches to 1.4+0.7
0.4×104[Mpc−2](and2.7+1.1
0.6×103[Mpc−2]),
and the total mass down to 1.0 ±0.1 Mis 4.0+1.7
−1.2×103M(and
1.5+0.5
−0.4×104M), respectively. The densities and total estimated
masses at 1 Myr are slightly higher than the values reported in the
first epoch [compare Fig. 12 to fig. 13 in Khorrami et al. (2017)].
Comparing data with the first epoch ones, we could detect 790
sources common in total (J, H, and K) and the majority (67 per cent)
of detected sources in the outer region (r>3 ) are not consistent
with the evolutionary models at 1−2 Myr and with extinction similar
to the average cluster value, suggesting an ongoing star formation
within 30 Doradus. A significant scatter in the CMD (Fig. 9)and
colour–colour diagram (Fig. 13) is originated mainly from the lower
part of the main-sequence and pre-main-sequence sources, located
(detected) at the outer region. The observed scatter in CMD and
colour–colour diagrams of 30 Doradus has previously been reported
in the visible and NIR (Hunter et al. 1995; Andersen et al. 2009;
De Marchi et al. 2011; Cignoni et al. 2015). The scatter is likely
due to a combination of observational confusion (affected mainly
by the visual multiple systems or variables), photometric errors,
differential extinction, and a possible age spread. In addition, pre-
main-sequence stars are often associated with circumstellar discs and
outflows, which will introduce additional extinction for the cluster’s
low-mass content. Brandl et al. (1996) found that the extinction varies
significantly from star to star within the cluster, with the range of 1–2
mag. The HST observations also reveal the presence of considerable
differential extinction across the 30 Doradus region. De Marchi et al.
(2011) quantified the total extinction towards massive main-sequence
stars younger than 3 Myr to be in the range of 1.3 <AV<2.1.
This motivates us to observe this cluster again in J and K in future
with even longer exposure time, so the number of common sources
in three filters (J, H, and K) increases.
ACKNOWLEDGEMENTS
The Star Form Mapper project has received funding from the
European Union’s Horizon 2020 research and innovation program
under grant agreement no. 687528. Zeinab Khorrami acknowledges
the support of a STFC Consolidated Grant (ST/K00926/1). This
work has made use of the SPHERE Data Centre, jointly operated
by OSUG/IPAG (Grenoble), PYTHEAS/LAM/CeSAM (Marseille),
OCA/Lagrange (Nice), Observatoire de Paris/LESIA (Paris), and
Observatoire de Lyon (OSUL/CRAL). This work is supported by
the French National Programmes (PNPS). Anne Buckner is funded
by the European Research Council H2020-EU.1.1 ICYBOB project
(grant no. 818940). Richard W¨
unsch and Jan Palouˇ
s acknowledge
support by the Czech Science Foundation project no. 19-15008S and
by the institutional project RVO:67985815.
Based on observations made with ESO Telescopes at the La Silla
Paranal Observatory, under programme ID 0102.D0271.
DATA AVAILABILITY
All data are incorporated into the article and its online supplementary
material.
REFERENCES
Allison R. J., Goodwin S. P., Parker R. J., Portegies Zwart S. F., de Grijs R.,
Kouwenhoven M. B. N., 2009, MNRAS, 395, 1449
Andersen M., Zinnecker H., Moneti A., McCaughrean M. J., Brandl B.,
Brandner W., Meylan G., Hunter D., 2009, ApJ, 707, 1347
Bestenlehner J. M. et al., 2014, A&A, 570, A38
Bestenlehner J. M. et al., 2020, MNRAS, 499, 1918
Beuzit J. L. et al., 2019, A&A, 631, A155
Brandl B. et al., 1996, ApJ, 466, 254
Bressan A., Marigo P., Girardi L., Salasnich B., Dal Cero C., Rubele S.,
Nanni A., 2012, MNRAS, 427, 127
Campbell M. A., Evans C. J., Mackey A. D., Gieles M., Alves J., Ascenso J.,
Bastian N., Longmore A. J., 2010, MNRAS, 405, 421
Chen Y. et al., 2019, A&A, 632, A105
Cignoni M. et al., 2015, ApJ, 811, 76
Crowther P. A., Dessart L., 1998, MNRAS, 296, 622
Crowther P. A., Schnurr O., Hirschi R., Yusof N., Parker R. J., Goodwin S.
P., Kassim H. A., 2010, MNRAS, 408, 731
Crowther P. A. et al., 2016, MNRAS, 458, 624
De Marchi G., Panagia N., 2014, MNRAS, 445, 93
De Marchi G. et al., 2011, ApJ, 739, 27
Diolaiti E., Bendinelli O., Bonaccini D., Close L., Currie D., Parmeggiani G.,
2000, A&AS, 147, 335
Draine B. T., 2003a, ARA&A, 41, 241
Draine B. T., 2003b, ApJ, 598, 1017
Draine B. T., 2003c, ApJ, 598, 1026
Elson R. A. W., Fall S. M., Freeman K. C., 1987, ApJ, 323, 54
Fusco T. et al., 2004, J. Opt. A: Pure Appl. Opt., 6, 585
Gibson B. K., 2000, MemSAI, 71, 693
Guerrero M. A., Chu Y.-H., 2008, ApJS, 177, 216
Hubeny I., Lanz T., 1995, ApJ, 439, 875
Hunter D. A., Shaya E. J., Holtzman J. A., 1995, ApJ, 448, 179
Khorrami Z. et al., 2017, A&A, 602, A56
Khorrami Z. et al., 2019, MNRAS, 485, 3124
Langlois M. et al., 2014, in Ramsay S. K., McLean I. S., Takami H., eds, Proc.
SPIE Conf. Ser. Vol. 9147, Ground-based and Airborne Instrumentation
for Astronomy V. SPIE, Bellingham, p. 91471R
Lanz T., Hubeny I., 2003, ApJS, 147, 225
Li A., Draine B. T., 2001, ApJ, 554, 778
Maire A.-L. et al., 2014, A&A, 566, A126
Massey P., Hunter D. A., 1998, ApJ, 493, 180
Pietrzy´
nski G. et al., 2013, NAT, 495, 76
Portegies Zwart S. F., Pooley D., Lewin W. H. G., 2002, ApJ, 574, 762
Pugliese G., Christou J., Koehler R., Drummond J., 2002, BAAS, 34, 55.05
Salpeter E. E., 1955, ApJ, 121, 161
Schneider F. R. N. et al., 2018, A&A, 618, A73
Schnurr O., Chen´
e A. -N., Casoli J., Moffat A. F. J., St-Louis N., 2009,
MNRAS, 397, 2049
Tatton B. L. et al., 2013, A&A, 554, A33
Townsley L. K., Broos P. S., Feigelson E. D., Garmire G. P., Getman K. V.,
2006, AJ, 131, 2164
Weigelt G., Baier G., 1985, A&A, 150, L18
Weingartner J. C., Draine B. T., 2001, ApJ, 548, 296
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 303
SUPPORTING INFORMATION
Supplementary data are available at MNRAS online.
Table A 3 . Sample catalogue of detected sources in H and K data in
2018.
Please note: Oxford University Press is not responsible for the content
or functionality of any supporting materials supplied by the authors.
Any queries (other than missing material) should be directed to the
corresponding author for the article.
APPENDIX A: ADDITIONAL TABLE AND
FIGURES
Figure A1. Completeness maps in K (top) and H (bottom), used for correcting stellar masses for plotting MF. Colour range (0:100). Stars show the location of
the observed detected sources within the given magnitude range for each map. If a star with a given magnitude is located in the yellow area, it will be detected
100 per cent and it has difficulty to be detected if it is located in the dark purple.
Figure A2. Stars brighter than 16 mag (14 mag) in K data shown in blue circles (yellow), with HSH95 identification written on top of each source. See Table A2
for their IRIDS (J, H, and K) and HST (U, V, and I) magnitudes. Left: Whole FOV, Right: Same as left but zoomed in the core to avoid confusion. Background
image is the SPHERE/IRDIS/K in 2018.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
304 Z. Khorrami et al.
Figure A3. Separation from the nearest neighbour for each detected sources in H (purple circles) and K (green squares) versus their magnitudes.
Figure A4. Difference between the K magnitude of the sources within two epochs versus their K magnitude in 2018. The colour shows the correlation between
the detected source and the input PSF. The higher the value, the more similar the shape of PSF is to the reference input PSF.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 305
Figure A5. Comparing the photometry by Starfinder (purple crosses) with DAOPHOT (red pluses). Top show the CMD in H–K in 2018 data (left) and J–K in
2015 (right). Bottom is the comparison between the K magnitude of common sources between two epochs.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
306 Z. Khorrami et al.
Figure A6. Comparing the effect of completeness correction on the density
of the cluster at 1.5 Myr. Same as Fig. 12 for 1.5 Myr. Black stars are the
completeness-corrected values.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 307
Figure A7. Comparing H–K CMD between IRDIS (this work, red crosses) and MAD [Campbell et al. (2010), black pluses]. Top shows the MAD data within
the same FOV of IRDIS (2.8 excluded), and bottom shows the whole MAD data covering F1-F2-F3. Yellow and blue circles show the 28 common sources
used for adjusting zero-points. The number of common sources between IRDIS FOV and MAD catalogue is larger than 26, but since both our catalogues have
pix position for the location of stars, we needed to identify stars by eye in each image/catalogue.
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
308 Z. Khorrami et al.
Tab l e A 1 . List of brightest spectroscopically known stars in the core of R136 studied in optical. Teff,logL/L
, and log g, and spectral types are taken from
Bestenlehner et al. 2020 (Table 1). Ones noted in brackets, [MH98], [CD98], [C10], [B14], and [C16] are from Massey & Hunter (1998), Crowther & Dessart
(1998), Crowther et al. (2010), Bestenlehner et al. (2014), and Crowther et al. (2016), respectively. HSH95, WB85, and IDKare the stars’ identification in Hunter
et al. (1995), Weigelt & Baier (1985), and SPHERE/IRDIS/K catalogue in 2018 (this work), respectively. K and H are the SPHERE IRDIS magnitudes and J15
is the SPHERE IRDIS magnitude in our first epoch data (Khorrami et al. 2017). ris the distance from r136a1 in our K data. The stellar initial masses (Mini)and
ages are estimated by fitting grids of PARSEC evolutionary models (0.1–10 Myr) to the Teff and log L/L.
HSH95 IDKK(±0.01) H(±0.01) J15(±0.13) rT
eff log L/Llog g Mini Age Spectral
(WB85) (mag) (mag) (mag) (arcsec) (kK) (M)(Myr) type
3 (a1) 1 11.15 11.29 11.33 0.00 46.0 ±2.5 6.79+0.10
−0.10 4.0 207+142
−28 2.4+0.1
−1.5WN5h [CD98]
10 (c) 2 11.31 11.70 11.78 3.35 42.0 ±2.0 [B14] 6.58 ±0.10 [B14] – – – WN5h [C10]
5 (a2) 3 11.43 11.70 11.55 0.11 50.0 ±2.5 6.75+0.10
−0.10 4.0 202+147
−26 2.4+0.1
−1.8WN5h [CD98]
6 (a3) 4 11.45 11.73 11.77 0.47 50.0 ±2.5 6.63+0.10
−0.10 4.0 200+50
−45 0.9+1.7
−0.3WN5h [CD98]
9 (b) 5 11.67 11.90 11.67 2.06 35.0 ±2.5 6.34+0.12
−0.10 3.3 102+36
−23 3.0+0.4
−1.1O4If/WN8 [C16]
20 (a5) 6 12.79 12.76 12.75 0.29 47.0 ±3.3 6.29+0.10
−0.09 4.0 120+19
−45 1.3+2.2
−0.2O2I(n)f∗
19 (a6) 11 13.17 13.30 13.28 0.71 53.0 ±3.5 6.27+0.09
−0.09 4.1 121+12
−46 0.7+2.8
−0.6O2I(n)f∗p
36 8 13.05 13.05 12.86 1.49 52.0 ±3.4 6.33+0.12
−0.10 4.1 130+19
−50 0.8+2.6
−0.7O2If∗
21 (a4) 9 13.14 13.06 13.00 0.41 48.0 ±5.8 6.24+0.18
−0.18 4.1 110+39
−51 1.4+2.7
−1.2O3V((f∗))(n)
24 (a7) 13 13.35 13.20 12.95 0.36 49.0 ±5.5 6.25+0.18
−0.17 4.2 112+37
−48 1.3+2.6
−1.2O3III(f∗) [MH98]
46 15 13.42 13.32 13.11 1.69 49.0 ±6.0 6.16+0.18
−0.17 4.2 100+30
−46 1.2+3.2
−1.1O2-3III(f∗)
47 17 13.51 13.36 13.17 1.72 47.0 ±7.0 6.09+0.22
−0.21 4.0 90+30
−36 1.4+3.0
−1.3O2V((f∗))
31 18 13.60 13.80 13.69 1.57 48.0 ±5.0 6.01+0.16
−0.16 4.0 80+20
−26 1.4+3.0
−1.3O2V((f∗))
48 19 13.68 13.62 13.41 1.19 49.0 ±7.2 6.05+0.21
−0.20 4.1 83+36
−29 1.5+2.9
−1.4O2-3III(f∗)
45 20 13.69 13.74 13.37 2.39 42.0 ±5.0 5.84+0.17
−0.16 3.9 59+15
−13 2.5+1.9
−1.0O4:Vz
30 21 13.73 13.69 13.62 0.61 37.0 ±3.5 5.68+0.14
−0.14 3.9 45+10
−73.6+0.8
−0.7O6.5Vz
40 23 13.86 13.72 13.61 1.76 45.0 ±5.6 5.88+0.18
−0.18 3.9 65+20
−15 2.0+2.4
−1.9O3V
65 27 13.94 13.97 13.67 2.21 42.0 ±5.2 5.74+0.17
−0.16 4.0 51+15
−11 2.8+1.0
−1.9O4 V [C16]
50 28 13.97 13.98 13.82 0.72 42.0 ±3.0 5.71+0.11
−0.11 3.8 50+7
−72.7+0.7
−0.7O3-4 V((f∗))
64 29 14.01 13.86 13.69 2.39 40.0 ±5.1 5.69+0.18
−0.17 3.9 47+14
−10 3.0+1.3
−1.3O4-5V
58 33 14.02 13.91 13.78 0.58 50.0 ±5.9 5.94+0.16
−0.16 4.1 75+22
−21 1.0+3.4
−0.9O2-3V
35 34 14.04 14.03 14.04 0.87 44.0 ±5.6 5.74+0.18
−0.18 4.0 53+0
−02.3+0.0
−0.0O3V
55 36 14.06 14.28 14.15 1.77 47.0 ±5.0 5.76+0.15
−0.15 3.9 57+17
−12 1.5+1.3
−1.4O2V((f∗))z
49 37 14.08 13.96 13.94 1.64 48.0 ±12.0 5.89+0.37
−0.37 4.2 68+51
−30 1.5+3.0
−1.4O3V [MH98]
52 39 14.17 14.28 14.10 1.09 44.0 ±4.8 5.67+0.16
−0.16 4.0 50+11
−10 2.2+1.3
−2.1O3-4Vz
70 40 14.31 14.16 14.06 0.62 47.0 ±6.0 5.78+0.18
−0.18 4.2 60+20
−15 1.4+1.6
−1.3O5Vz
62 41 14.36 14.41 14.28 0.63 49.0 ±6.2 5.75+0.17
−0.17 4.0 60+15
−15 0.8+1.8
−0.7O2-3V
69 42 14.45 14.49 14.29 1.06 41.0 ±4.6 5.51+0.16
−0.16 4.1 40+10
−83.0+1.4
−2.8O4-5Vz
66 43 14.45 14.44 14.31 0.46 46.0 ±6.6 5.64+0.21
−0.21 4.1 50+20
−13 1.7+1.8
−1.6O2V-III(f∗)
73 44 14.46 14.44 14.24 0.98 33.0 ±3.6 5.27+0.14
−0.14 4.3 27+4
−45.6+2.0
−1.1O9.7-B0V
71 46 14.58 14.88 14.80 1.96 48.0 ±8.0 5.56+0.23
−0.23 3.9 50+15
−16 0.2+3.2
−0.1O2-3V((f∗))
86 47 14.63 14.82 14.80 0.43 41.0 ±5.0 5.26+0.16
−0.16 3.8 31+10
−53.1+1.9
−3.0O5:V
78 49 14.67 14.76 14.76 0.94 48.0 ±8.0 5.60+0.24
−0.24 4.2 50+20
−15 0.9+2.5
−0.8O4:V
80 54 14.77 14.83 14.71 0.84 35.0 ±3.8 5.15+0.15
−0.15 3.8 25+4
−45.3+1.9
−2.3O8V
75 55 14.79 14.82 14.84 1.53 39.0 ±6.9 5.29+0.22
−0.22 4.3 31+11
−83.7+2.9
−3.6O6V
90 59 14.83 14.86 14.88 0.75 40.0 ±3.7 5.32+0.13
−0.13 4.1 32+7
−53.4+1.4
−2.9O4:V:
94 60 14.84 14.92 14.69 1.26 48.0 ±8.2 5.52+0.23
−0.23 4.2 47+12
−15 0.3+3.1
−0.2O4-5Vz
129 61 14.84 14.79 14.43 0.97 37.0 ±8.2 4.37+0.26
−0.26 4.0 – – –
92 64 14.90 14.94 14.96 1.02 39.0 ±4.0 5.26+0.14
−0.14 4.0 30+6
−43.9+1.5
−3.2O6Vz
112 67 14.96 14.80 14.61 1.04 36.0 ±6.0 5.21+0.19
−0.19 4.3 27+8
−65.0+2.7
−4.6O7-9Vz
114 72 15.21 15.38 15.33 1.51 44.0 ±6.8 5.25+0.21
−0.21 4.2 35+8
−90.7+3.7
−0.6O5-6V
143 75 15.24 15.14 14.93 1.03 39.0 ±6.0 5.18+0.20
−0.20 4.2 28+9
−73.3+3.2
−3.2O8-9 V-III
121 81 15.31 15.63 15.50 1.98 34.0 ±4.8 4.86+0.16
−0.16 4.2 20+5
−36.3+3.5
−5.7O9.5V
116 82 15.32 15.60 15.86 1.75 34.0 ±6.1 4.84+0.16
−0.16 3.7 20+5
−35.9+3.5
−5.8O7V
141 87 15.40 15.46 15.33 0.75 32.0 ±6.0 4.79+0.21
−0.21 3.6 18+5
−27.8+2.2
−7.7O5-6V [C16]
135 90 15.44 15.28 15.17 1.06 33.0 ±4.9 4.89+0.17
−0.17 4.0 20+5
−36.9+3.1
−6.3B
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136 Core 309
Tab l e A 1 –continued
HSH95 IDKK(±0.01) H(±0.01) J15(±0.13) rT
eff log L/Llog g Mini Age Spectral
(WB85) (mag) (mag) (mag) (arcsec) (kK) (M)(Myr) type
132 91 15.45 15.35 15.38 1.45 39.0 ±5.8 5.05+0.20
−0.20 4.0 26+8
−63.1+3.5
−3.0O7:V
159 92 15.47 15.62 15.51 0.32 36.0 ±8.9 4.93+0.28
−0.28 4.3 – – –
120 93 15.48 16.05 16.11 0.80 37.0 ±6.8 4.81+0.22
−0.22 4.3 – – –
123 95 15.49 15.83 15.88 2.36 41.0 ±6.5 5.01+0.22
−0.22 4.1 28+7
−80.4+5.4
−0.3O6V
134 99 15.54 15.91 15.90 2.21 36.0 ±4.8 4.81+0.17
−0.17 4.0 20+4
−43.9+4.4
−3.8O7Vz
162 101 15.54 15.78 15.98 0.42 37.0 ±13.0 4.87+0.39
−0.39 4.3 – – –
108 102 15.55 15.57 15.71 1.55 43.0 ±7.6 5.04+0.24
−0.24 4.2 28+6
−70.7+4.6
−0.6OVn
139 105 15.56 15.78 15.72 0.40 38.0 ±5.1 4.90+0.17
−0.17 4.0 – – –
173 121 15.71 15.63 15.70 1.51 30.0 ±10.0 4.65+0.33
−0.33 4.3 16+9
−29.4+0.6
−9.3O9+V [C16]
Tab l e A 2 . List of detected sources in the centre of R136 brighter than 16 mag in K. IDKis the star’s identification
in our K catalogue. K(2018) and H(2018) are the K and H magnitudes of stars in the second epoch data in 2018.
K(2015) and J(2015) are the K and J magnitudes of stars in the first epoch data in 2015. HSH95 and WB85 are the
stars’ identification in Hunter et al. (1995) and Weigelt & Baier (1985), followed by their U, V, and I magnitudes
in the HST/WFPC2 filters. Last column is V–K (F555W-K(2018)). See Fig. A2 for their positions in the image.
IDKK H K J HSH95 U V I (V–K)
(2018) (2018) (2015) (2015) (WB85) (F336W) (F555W) (F814W)
1 11.15 11.29 11.07 11.33 3 (a1) 11.56 12.84 12.18 1.69
2 11.31 11.98 11.48 11.78 10 (c) 12.52 13.47 12.71 2.16
3 11.43 11.70 11.32 11.55 5 (a2) 11.94 12.96 12.48 1.53
4 11.45 11.73 11.44 11.77 6 (a3) 11.86 13.01 12.46 1.56
5 11.67 11.90 11.66 11.67 9 (b) 12.29 13.32 12.76 1.65
6 12.79 12.76 12.73 12.75 20 (a5) 12.77 13.93 13.54 1.14
7 13.04 13.18 13.05 13.24 26 12.89 14.19 13.64 1.15
8 13.05 13.05 12.96 12.86 36 13.36 14.49 13.98 1.44
9 13.14 13.06 13.09 13.00 21 (a4) 12.81 13.96 13.63 0.82
10 13.17 13.73 13.36 13.52 57 14.03 14.87 14.23 1.70
11 13.18 13.30 13.19 13.28 19 (a6) 12.86 13.92 13.72 0.74
12 13.33 13.50 13.45 13.37 17 13.00 13.78 13.89 0.45
13 13.35 13.20 13.05 12.95 24 (a7) 12.84 14.06 13.66 0.71
14 13.41 13.27 13.27 13.18 27 (a8) 12.93 14.22 13.86 0.81
15 13.42 13.32 13.29 13.11 46 13.64 14.73 14.26 1.31
16 13.49 13.82 13.58 13.37 59 13.81 14.88 14.35 1.39
17 13.51 13.36 13.32 13.17 47 13.64 14.75 14.31 1.24
18 13.60 13.80 13.62 13.69 31 13.14 14.41 14.08 0.81
19 13.68 13.62 13.58 13.41 48 13.61 14.78 14.37 1.10
20 13.69 13.74 13.63 13.37 45 13.51 14.68 14.31 0.99
21 13.73 13.69 13.65 13.62 30 13.09 14.32 14.02 0.59
22 13.83 14.41 13.84 14.33 124 999.99 15.97 15.46 2.14
23 13.86 13.72 13.72 13.61 40 13.32 14.60 14.35 0.74
24 13.87 13.77 13.78 13.70 39 13.31 14.59 14.31 0.72
25 13.92 14.08 13.99 14.23 33 13.07 14.43 14.19 0.51
26 13.93 14.04 13.84 13.73 102 14.75 15.70 15.10 1.77
27 13.94 13.97 13.85 13.67 65 14.03 15.18 14.49 1.24
28 13.97 13.98 13.96 13.82 50 13.62 14.81 14.48 0.84
29 14.01 13.86 13.84 13.69 64 14.01 15.12 14.68 1.11
30 14.01 14.10 13.93 14.07 111 14.94 15.77 15.20 1.76
31 14.01 14.11 13.96 13.63 68 14.13 15.22 14.76 1.21
32 14.01 15.05 14.68 14.94 38 13.28 14.57 14.11 0.56
33 14.02 13.91 13.91 13.78 58 13.65 14.88 14.55 0.86
34 14.04 14.03 14.00 14.04 35 13.35 14.48 14.27 0.44
35 14.05 14.18 14.11 14.30 42 13.35 14.63 14.58 0.58
36 14.06 14.28 14.15 14.15 55 13.62 14.86 14.53 0.80
37 14.08 13.96 13.98 13.94 49 13.55 14.80 14.53 0.72
38 14.16 15.05 14.64 15.30 29 12.89 14.30 14.03 0.14
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
310 Z. Khorrami et al.
Tab l e A 2 –continued
IDKK H K J HSH95 U V I (V–K)
(2018) (2018) (2015) (2015) (WB85) (F336W) (F555W) (F814W)
39 14.17 14.28 14.20 14.10 52 13.52 14.82 14.55 0.65
40 14.31 14.16 14.20 14.06 70 13.93 15.23 14.80 0.92
41 14.36 14.41 14.35 14.28 62 13.72 15.02 14.75 0.66
42 14.45 14.49 14.39 14.29 69 14.02 15.22 15.03 0.77
43 14.45 14.44 14.42 14.31 66 13.95 15.19 14.88 0.74
44 14.46 14.44 14.38 14.24 73 14.16 15.27 15.01 0.81
45 14.50 15.48 14.84 15.29 262 16.08 16.92 16.18 2.42
46 14.58 14.88 14.70 14.80 71 14.01 15.23 15.00 0.65
47 14.63 14.82 14.80 14.80 86 14.23 15.49 15.19 0.86
48 14.65 15.10 14.79 14.80 136 15.14 16.06 15.58 1.41
49 14.67 14.76 14.68 14.76 78 14.11 15.34 15.05 0.67
51 14.72 14.68 14.49 14.77 107 14.58 15.74 15.34 1.02
52 14.72 14.69 14.72 14.56 77 13.97 15.30 14.99 0.58
53 14.76 14.70 14.64 14.42 118 14.79 15.89 15.45 1.13
54 14.77 14.83 14.77 14.71 80 14.13 15.35 15.12 0.58
55 14.79 14.82 14.78 14.84 75 13.89 15.28 15.06 0.49
56 14.81 14.70 14.61 14.49 177 15.35 16.31 15.76 1.50
57 14.81 14.80 14.69 14.62 89 14.39 15.52 15.21 0.71
58 14.83 15.18 14.97 14.92 87 14.32 15.50 15.28 0.67
59 14.83 14.86 14.83 14.88 90 14.30 15.55 15.29 0.72
60 14.84 14.92 14.85 14.69 94 14.32 15.60 15.30 0.76
61 14.84 14.79 14.70 14.43 129 14.97 16.03 15.65 1.19
62 14.86 14.82 14.78 14.81 93 14.34 15.60 15.24 0.74
63 14.87 15.22 15.02 14.96 97 14.43 15.61 15.36 0.74
64 14.90 14.94 14.87 14.96 92 14.28 15.59 15.38 0.69
65 14.91 15.15 15.03 15.34 79 13.98 15.35 15.09 0.44
66 14.96 15.41 15.10 15.39 219 15.81 16.68 16.03 1.72
67 14.96 14.80 14.78 14.61 112 14.63 15.77 15.45 0.81
68 14.97 15.84 15.75 17.21 623 18.01 18.41 17.45 3.44
69 15.10 15.09 14.98 14.93 193 15.66 16.51 16.03 1.41
70 15.12 15.61 15.32 15.22 183 15.33 16.35 15.85 1.23
71 15.14 15.83 15.50 16.08 84 14.20 15.48 15.20 0.34
72 15.21 15.38 15.26 15.33 114 14.51 15.82 15.59 0.61
73 15.21 15.28 15.16 14.93 222 15.76 16.70 16.21 1.49
74 15.23 15.25 15.18 15.17 119 15.00 15.92 15.88 0.69
75 15.24 15.14 15.11 14.93 143 14.89 16.10 15.76 0.86
77 15.28 15.33 15.27 15.42 127 14.80 16.02 15.73 0.74
78 15.29 15.42 15.22 15.33 149 14.94 16.14 15.85 0.85
79 15.30 15.38 15.22 15.03 277 16.07 16.97 16.41 1.67
80 15.31 15.31 15.25 15.24 150 14.97 16.16 15.85 0.85
81 15.31 15.63 15.47 15.50 121 14.81 15.96 15.70 0.65
82 15.32 15.60 15.45 15.86 116 14.64 15.85 15.60 0.53
83 15.35 16.50 15.98 16.73 76 13.92 15.29 15.11 −0.06
84 15.38 15.96 15.63 15.60 197 15.49 16.53 16.04 1.15
85 15.38 15.62 15.23 15.70 213 15.70 16.65 16.38 1.27
86 15.40 15.41 15.33 15.35 165 15.14 16.23 15.91 0.83
87 15.40 15.46 15.40 15.33 141 14.74 16.09 15.80 0.69
88 15.40 16.49 15.96 15.97 125 14.84 15.98 15.72 0.58
89 15.43 15.38 15.33 15.08 207 15.72 16.62 16.28 1.19
90 15.44 15.28 15.32 15.17 135 14.89 16.05 15.79 0.61
91 15.45 15.35 15.36 15.38 132 14.77 16.04 15.81 0.59
92 15.47 15.62 15.51 15.51 159 15.07 16.21 15.90 0.74
93 15.48 16.05 15.75 16.11 120 14.76 15.95 15.69 0.47
94 15.48 16.10 15.78 15.85 128 14.91 16.02 15.76 0.54
95 15.49 15.83 15.61 15.88 123 14.76 15.96 15.70 0.47
96 15.49 15.58 15.47 15.31 205 15.49 16.59 16.23 1.10
97 15.50 15.58 15.42 15.21 189 15.35 16.46 16.13 0.96
98 15.51 16.36 15.98 15.97 153 14.99 16.17 15.86 0.66
99 15.54 15.91 15.71 15.90 134 14.83 16.05 15.82 0.51
100 15.54 15.75 15.46 15.77 480 17.40 17.91 17.01 2.37
101 15.54 15.79 15.66 15.98 162 15.04 16.23 15.92 0.69
102 15.55 15.57 15.59 15.71 108 14.65 15.74 15.49 0.19
MNRAS 503, 292–311 (2021)
Downloaded from https://academic.oup.com/mnras/article/503/1/292/6133467 by guest on 14 July 2021
Photometry of R136