MNRAS 532, 2842–2859 (2024) https://doi.org/10.1093/mnras/stae1640 
Advance Access publication 2024 July 3 

MeerKAT obser v ations of starburst galaxies and AGNs within the core of 

XMMXCS J2215.9 −1738 at z = 1.46 

D. Y. Klutse, 1 , 2 ‹ M. Hilton , 2 , 3 ‹ I. Heywood , 4 , 5 , 6 I. Smail , 7 A. M. Swinbank , 7 K. Knowles 5 , 6 

and S. P. Sikhosana 

1 , 2 

1 Astrophysics Research Centre, University of KwaZulu – Natal, Westville Campus, Durban 4041, South Africa 
2 School of Mathematics, Statistics & Computer Science, University of KwaZulu – Natal, Westville Campus, Durban 4041, South Africa 
3 Wits Centre for Astrophysics, School of Physics, University of the Witwatersrand, Private Bag 3, Johannesburg 2050, South Africa 
4 Astrophysics, Department of Physics, University of Oxford, Keble Road, Oxford OX1 3RH, UK 

5 Centre for Radio Astronomy Techniques and Technologies, Department of Physics and Electronics, Rhodes University, P.O. Box 94, Makhanda 6140, South 
Africa 
6 South African Radio Astronomy Observatory, 2 Fir Street, Black River Park, Observatory, Cape Town 7925, South Africa 
7 Centre for Extragalactic Astronomy, Department of Physics, Durham University, South Road, Durham DH1 3LE, UK 

Accepted 2024 July 1. Received 2024 June 26; in original form 2024 April 18 

A B S T R A C T 

We present the first detailed radio study of the galaxy cluster XMMXCS J2215.9 −1738 at z = 1.46 using MeerKAT L -band 

(1.3 GHz) observations. We combine our radio observation with archival optical and infrared data to investigate the star formation 

and active galactic nucleus (AGN) population within R 200 ( R = 0.8 Mpc) of the cluster centre. Using three selection criteria; 
the radio luminosity, the far-infrared radio ratio ( q IR 

), and the mid-infrared colour, we distinguish galaxies with radio emission 

predominantly powered by star formation from that powered by AGNs. We selected 24 cluster members within R 200 in the 
MeerKAT image based on either their photometric or spectroscopic redshift. We classified 12/24 (50 per cent) as galaxies whose 
radio emission is dominated by star-formation activity, 6/24 (25 per cent) as intermediate star-forming galaxies, and 6/24 (25 

per cent) as AGN-dominated galaxies. Using the radio continuum luminosities of the star-forming cluster galaxies, we estimated 

an integrated star formation rate (SFR) value of 1700 ± 330 M � yr −1 within R 200 . We derived a mass-normalized integrated SFR 

value of (570 ± 110) × 10 

−14 yr −1 . This supports pre vious observ ational and theoretical studies that indicated a rapid increase 
in star formation activity within the core of high-redshift clusters. We also show that the high-AGN fraction within the cluster 
core is consistent with previous cluster observations at z > 1.5. 

Key words: galaxies: clusters: general – galaxies: clusters: individual: (XMMXCS J2215.9 −1738) – galaxies: star formation. 

1

G  

o  

d  

e  

t
 

e  

d  

w  

d  

2  

2  

g  

l  

f  

�

b  

S
 

r  

g  

s  

e  

S  

X  

c  

d  

r  

A  

s  

a  

a  

2  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
 I N T RO D U C T I O N  

alaxy clusters provide a powerful means to study the formation
f large-scale structures, the evolution of galaxies, and the thermo-
ynamics of the intergalactic medium. They also complement other
xisting probes of structural growth over cosmological time due to
heir high masses (Kravtsov & Borgani 2012 ). 

Environment plays a major role in galaxy evolution and this is
vident via the relationship of star formation rate to local galaxy
ensity. Star formation rate (SFR) in galaxies at z � 1.0 increases
ith increasing galaxy density up to galaxy group scales, and then
eclines in the denser regions of galaxy clusters (Kodama & Bower
001 ; Lewis et al. 2002 ; G ́omez et al. 2003 ; Blanton & Moustakas
009 ; Cluver et al. 2020 ; Pearson et al. 2021 ). Denser regions within
alaxy clusters at z � 1.0 are mostly populated by elliptical and
enticular galaxies (early-type galaxies) which show minimal star
ormation activity in contrast to low-density regions mostly occupied
 E-mail: dianasmoke1@gmail.com (DYK); matt.hilton@wits.ac.za (MH) 

(  

s  

o  

Published by Oxford University Press on behalf of Royal Astronomical Socie
Commons Attribution License ( https:// creativecommons.org/ licenses/ by/ 4.0/ ), whi
y star-forming spirals (late-type disc galaxies; Dressler et al. 1997 ;
mith et al. 2005 ). 
Sev eral studies hav e suggested a rev erse in this order at higher

edshifts (i.e. in some clusters at z > 1, the fraction of star-forming
alaxies increases with galaxy density) leading to the so-called rever-
al in the star formation density relation (Hayashi et al. 2010 ; Tran
t al. 2010 ; Tadaki et al. 2012 ; Brodwin et al. 2013 ; Santos et al. 2015 ;
mail 2024 ). For example, a study of the z = 1 . 56 galaxy cluster
MMU J1007.4 + 1237 showed strong starburst activity within the

luster core (Fassbender et al. 2011 ). Whilst Wang et al. ( 2016 )
etected nine star-forming cluster galaxies within the central 80 kpc
egion of the X-ray detected cluster CL J1001 + 0220 at z = 2 . 5.
 Sub-millimetre Common User Bolometer Array 2 (SCUBA-2;

ub-millimetre observation) study of the CL 0218.3 −0510 cluster
t z = 1 . 6 by Smail et al. ( 2014 ) showed that active star formation
ctivities take place outside the cluster core, whilst the outcome of a
4- μm (mid-infrared) observation of the same cluster from Tran et al.
 2010 ) showed that this high-redshift cluster was actively forming
tars. In contrast, an 850 μm SCUBA-2 continuum observations
f eight X-ray-detected massive galaxy clusters at z ≈ 0 . 8 − 1 . 6
© 2024 The Author(s). 
ty. This is an Open Access article distributed under the terms of the Creative 
ch permits unrestricted reuse, distribution, and reproduction in any medium, 

provided the original work is properly cited. 

http://orcid.org/0000-0002-8490-8117
http://orcid.org/0000-0001-6864-5057
http://orcid.org/0000-0003-3037-257X
http://orcid.org/0000-0003-1192-5837
http://orcid.org/0000-0002-8452-0825
http://orcid.org/0000-0003-3199-1161
mailto:dianasmoke1@gmail.com
mailto:matt.hilton@wits.ac.za
https://creativecommons.org/licenses/by/4.0/


MeerKAT observations of XCS J2215 2843 

s  

S  

t  

f
e  

(  

fi
 

c  

o  

C  

f
t  

r  

W
t
2

g
(  

w
s
s
c  

i  

m  

×  

b  

m
w

r  

J  

I  

t  

c
=  

v  

h
s
(  

o
 

o
a  

M  

P
e  

a  

b  

h

t  

a  

a
e  

f  

i
c  

r  

(
o  

s  

c

s
(  

c  

b  

b  

d
s

 

g
w
o  

t  

L

s  

J

o
b  

f  

d
l  

I  

g
o
s
l  

m  

i  

W  

c  

a

=  

s  

u

2

T  

o
a

 

a
o  

2

T  

L  

r  

f
d  

−  

d
m
B
c
c  

f  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
tudied by Cooke et al. ( 2019 ) revealed that the mass-normalized
FR for clusters at 1 < z < 1 . 6 is a factor of 1 . 5 ± 0 . 3 lower than

he field galaxies and they do not find any reversal for local star-
ormation-rate-density (SFRD) relation in their study. In a study of 
ight submillimetre (850 μm) galaxy clusters at z = 1 . 6 − 2 . 0, Smail
 2024 ) found a reversal in the local SFRD relation for clusters and
elds at z = 1 . 8. 
There may be a wide range of star formation activity in galaxy

lusters at z > 1. A narrow band imaging surv e y of the [ O II ] emitters
f the CIG J0218.3 −0510 cluster at z = 1 . 62 using the Suprime-
am on Subaru Telescope led to the detection of some very high-star

ormation rate galaxies as well as some quiescent galaxies within 
he cluster core (Tadaki et al. 2012 ). Similarly, the study of a higher
edshift cluster JKCS 041 at z = 1 . 8 (Andreon et al. 2009 ) using the

ide Field Camera 3 of the Hubble Space Telescope showed that 
he cluster core was populated by quiescent galaxies. (Newman et al. 
014 ). 
Studies have shown that galaxy mergers, interactions and active 

alactic nuclei (AGN) activities take place in high-redshift clusters 
Lotz et al. 2013 ; Alberts et al. 2016 ; Krishnan et al. 2017 ) but
hether these processes lead to the enhancement or suppression of 

tar formation has not been conclusively established. The study of 
tar formation and AGNs in 11 spectroscopically selected massive 
lusters at 1.0 < z < 1.75 by Alberts et al. ( 2016 ) showed an increase
n star formation rate, excess AGN activity and an increase in galaxy
erger rate within the clusters at this high redshift. The central 24
24 kpc region within the cluster XDCP J0044.0 −2033 at z � 1.5

y Lepore et al. ( 2022 ) showed that high-star formation and gas-rich
erger-driven nuclear activities take place in the cluster core (i.e. 
ithin ≈ 0.16 Mpc from the cluster centre). 
One well studied cluster to investigate starburst activity at high- 

edshift is XMMXCS J2215.9 −1738 at z = 1 . 46 (subsequently
2215, located at R.A. = 22 h 15 m 58 . s 5 and Dec. = 17 ◦38 ′ 02 ′′ ).
t was one of the distant clusters with a well-developed structure
o be disco v ered at X-ray wav elengths (Stanford et al. 2006 ). The
luster’s velocity dispersion σv within a virial radius R 200 (i.e. R 

 0.8 Mpc) is 720 ± 110 kms −1 (Hilton et al. 2010 ), resulting in a
irial mass of M cl = 3 × 10 14 M �. A SCUBA-2 surv e y found sev eral
ighly star-forming galaxies in the cluster’s core with an integrated 
tar formation rate of ≈1400 M � yr −1 (Salpeter initial mass function 
IMF) assumed, Ma et al. 2015 ). Thus J2215 appears to be an example
f a distant and highly star-forming galaxy cluster. 
Twelve [ O II ] emitting cluster members and four star forming dust-

bscured ultra luminous infrared galaxies (ULIRGs) were detected 
t < 0.25 Mpc from the cluster centre by Hayashi et al. ( 2010 ) and
a et al. ( 2015 ), respectively. A 24- μm Spitzer/Multiband Imaging

hotometer for Spitzer (MIPS) cluster surv e y conducted by Hilton 
t al. ( 2010 ) resulted in the detection of three star-forming galaxies
nd a potential AGN all located at R < 0.25 Mpc, suggesting that
oth starburst and AGN activities take place in the cluster core at
igher redshifts. 
There is also evidence of mergers between cluster galaxies within 

his high-redshift cluster; a surv e y of the core region of J2215 cluster
t z = 1 . 46 using the Atacama Large Millimetre Array (ALMA)
nd the MUSE spectrograph on the Very Large Telescope by Stach 
t al. ( 2017 ) led to the detection of 14-mm sources within ≈0.5 Mpc
rom the cluster centre. The result obtained by Stach et al. ( 2017 )
ndicated that there was an intense star formation within the cluster 
ore and evidence of a likely merger event taking place. The most
ecent study of the J2215 cluster from the ALMA CO J = 2 − 1 line
0.4 arcsec resolution) and 870 μm continuum (0.2 arcsec resolution) 
bservation by Ikeda et al. ( 2022 ) has shown evidence of enhanced
tar formation in the central region of the cluster, with 6 out of 17
luster members observed to be early-stage mergers. 

Observations have also shown that both quiescence galaxies and 
tar-forming systems reside within the cluster core; Hayashi et al. 
 2018 ) detected 12 quiescent and 27 star-forming galaxies within the
ore of the J2215 cluster at z = 1.46 via the study of the ALMA
and 3 data. The study of the J2215 cluster using the KMOS ( K-
and Multi-Object Spectrograph) by Maier et al. ( 2019 ) led to the
etection of apparently slow quenching systems and slightly lower 
tar formation activity within the cluster core. 

As noted abo v e, there hav e been sev eral studies of the J2215
alaxy cluster conducted at infrared and millimetre/submillimetre 
avelengths. Here, we present the first detailed radio observation 
f the J2215 galaxy cluster using the MeerKAT radio telescope. In
his work, we combine archi v al optical and infrared data with new
 -band radio observations from MeerKAT to further investigate the 

tar formation, AGN and merger activities within the core of the
2215 cluster. 

This paper is structured as follows: We describe the MeerKAT 

bservations of the J2215 cluster, the data reduction process and 
riefly describe the optical and infrared archi v al observ ations used
or this work in Section 2 . We describe the MeerKAT source
etection, cluster membership classification scheme, and the radio 
uminosity and continuum star formation rate estimation in Section 3 .
n Section 4 , we show the colour–magnitude relation of the cluster
alaxies and their morphologies. In Section 5 , we characterize 
ur galaxy samples into normal star-forming galaxies, intermediate 
tar-forming galaxies, and AGNs using three indicators; the radio 
uminosity value, the far-infrared radio ratio ( q IR ) value, and the

id-infrared colours. In Section 6 , we compare the AGN activity
n the J2215 cluster core to other higher and low-redshift surv e ys.

e discuss the star formation activity in J2215 and the evolution of
lusters with redshift in Section 7 . We conclude this work by giving
 summary of the entire work done and the prospects in Section 8 . 

We assume a � cold dark matter cosmology with H 0 

 70 km s −1 Mpc −1 , �� 

= 0 . 7, and �m 

= 0 . 3. The AB magnitude
ystem and the Chabrier ( 2003 ) IMF were used throughout this work
nless stated otherwise. 

 OBSERVATI ON  A N D  DATA  R E D U C T I O N  

his work is based on the radio observation of the J2215 cluster
btained from the MeerKAT telescope in combination with other 
rchi v al data. 

Sections 2.1 and 2.2 co v er the details of the MeerKAT observation
nd data processing respectively, while we summarize the properties 
f the archi v al multiwavelength data used in this study in Section 2.3 .

.1 MeerKAT obser v ations 

he J2215 galaxy cluster was observed in May 2019 using the
 -band receivers of the 64-dish MeerKAT telescope. The L -band

eceiv er co v ers a frequenc y range of 856 − 1712 MHz with a central
requency of 1284 MHz. The observation was conducted on two 
ifferent days, from 2019 May 11–12 between the hours of 02:16

10:20 UTC each day. A total of 61 antennas were in operation
uring the observation. The 4096 (4K) channel wideband-coarse 
ode of the Square Kilometre Array Reconfigurable Application 
oard processing nodes correlator was used resulting in 209 kHz 
hannel resolution.The data was recorded in full polarization. The 
orrelator dump time was 8 s, and the data were acquired for all
our polarization products, labelled as XX, XY, YX, and YY. The
MNRAS 532, 2842–2859 (2024) 



2844 D. Y. Klutse et al. 

M

d  

a  

e  

f  

s  

(  

c  

e  

1  

t  

o

2

D  

O  

t  

M  

t  

c  

t  

p  

R  

‘  

a  

m  

d  

S
 

t  

a  

T  

t  

o  

o  

D  

e  

c
 

B  

s  

f  

o  

a  

D  

a  

t  

o  

c  

o  

T
 

f  

M  

E  

o  

s  

−  

t

1

 

c  

S  

V  

w  

(  

d  

t  

=  

c

2

T  

d  

t  

t  

s  

A  

6  

b  

a  

S  

(  

b  

i  

M  

C  

v  

b
 

i  

2  

8  

4  

s  

8
 

o  

t  

J  

i  

l  

i  

I  

s  

p
 

r  

(  

T  

i  

f
 

P  

a  

e  

I  

a

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ata set in measurement set format for the two-day observation is
bout 5.2 TB with an approximate 2.6 TB measurement set obtained
ach day. J1939 −6342 was used as the primary calibrator (i.e.
or flux, delay, and bandpass calibration) while a much closer
ource ( ≈ 13 degrees of the target source position), J2225 −0457
3C446) served as the secondary calibrator for amplitude and phase
alibration. We observed the bandpass calibrator for 10 min after
very two hours and two minutes on the gain calibrator after every
5-min target scan. In total, we spent ≈12 h on the target source and
he total integration time for the entire observation is ≈16 h i.e. 8 h
f total integration each day. 

.2 Data processing 

ata reduction was performed using a semi-automated pipeline,
 XKAT 1 (Heywood 2020 ). O XKAT comprises a set of python scripts

hat incorporates other radio astronomical packages to process the
eerKAT data. The first step in the OXKAT pipeline was to duplicate

he MS (main sequence) and average to 1024 channels to make the
ontinuum data processing easier i.e. faster with shorter computation
ime. The primary calibrator was then re-phased to the correct
osition via the CASA (McMullin et al. 2007 ) task FIXVIS . The known
adio Frequency Interference (RFI) channels documented in the

MeerKAT-Cookbook’ were flagged, other flags were also applied to
ll the fields using the CASA task FLA GD ATA (i.e. manual, clip, quack
ode) while auto-flaggers were applied to only the calibrators. We

erived model data visibilities for the primary calibrator defined by
tev ens–Re ynolds 2016 flux density scale (Partridge et al. 2016 ). 
We derive an intrinsic model for the secondary calibrator based on

he primary calibrator and further derived delay ( K), bandpass ( B),
nd gain ( G ) calibrations from the primary and secondary calibrators.
he gain solutions were then applied to all the calibrators and the

arget. K , B, and G corrections are derived iteratively, with rounds
f residual flagging in between. The calibrated target data was split
ut into individual MS, with the reference calibrated data in the
ATA column of the new MS. We then imaged the radio continuum
mission with WSCLEAN (Offringa et al. 2014 ) in Stokes I and then
onducted one round of phase and then amplitude self-calibration. 

Imaging was performed using a cell size of 1.1 arcsec, and a
riggs weighting of −0 . 3. Due to the presence of a ≈4.6 Jy extended

ource 2MASX J22142575 −1701362 located at ≈ 0.7 degrees away
rom our target (see Fig. 1 ), we performed a peeling operation
n the ‘troublesome source’ in our field. This involved modelling
nd subtracting using CUBICAL (Kenyon et al. 2018 ). We exploit
DFACET (Tasse et al. 2018 ) on the residual visibility with the best

vailable cleaning mask from the previous WSCLEAN run (cropped to
he appropriate DDFACET image size i.e. 10125 × 10125 pixels) to
btain a sky model and then defined the directions that will form the
entres of the tesselated sk y. We deriv ed the gain solutions for each
f the tesselated sky models using KILLMS (Tasse 2014 ; Smirnov &
asse 2015 ). 
Further, we re-ran DDFACET to apply the gain corrections derived

rom the KILLMS run and also the primary beam correction. The
eerKAT L -band primary beam model was generated using the

IDOS software (Asad et al. 2019 ). The primary beam has a diameter
f ≈1.4 ◦. The resulting 1284 MHz image has a synthesized beam
ize of 6 . 01 arcsec × 5 . 26 arcsec with a positional angle, PA of
10.4 degrees and an rms noise level of ≈3.5 μJy beam 

−1 . This is
he most sensitive radio observation of the cluster to date. 
NRAS 532, 2842–2859 (2024) 

 https:// github.com/ IanHeywood/ oxkat 2
We determined the accuracy of the astrometry of our image by
omparing the coordinates of the MeerKAT sources (10 890, see
ection 3.1 ) with its closest counterpart in the JVLA (Karl G. Jansky
ery Large Array) map (298 sources). The processed JVLA map
ith a sensitivity of ≈7.5 μJy beam 

−1 was obtained from Ma et al.
 2015 ). The source extraction was done following the same procedure
escribed in Section 3.1 . The mean offset of the 275 counterparts in
he JVLA map was � RA = ( −0 . 22 arcsec ± 0 . 61 arcsec ) and � Dec
 ( −0 . 11 arcsec ± 0 . 61 arcsec which is insignificant so we did not

orrect the astrometry. 

.3 Archi v al multiwav elength obser v ations 

o study the multiwavelength properties of the radio sources
etected in the cluster we made use of archi v al images, pho-
ometry or catalogues from submillimetre/far-infrared data from
he SCUBA-2 (Ma et al. 2015 ), submillimetre/millimetre ob-
ervations from Atacama Large Millimetre/Sub-millimetre Array,
LMA band 3 (Hayashi et al. 2010 ), ALMA 1.25 mm (Band-
) observation (Stach et al. 2017 ), optical data from the Hub-
le Space Telescope ( HST ), Advanced Camera for Surv e ys, ACS
nd infrared data from the Multi-Object Infrared Camera and
pectrograph, MOIRCS mounted on the 8.2 m Subaru telescope
Hilton et al. 2009 ), the Infrared Array Camera (IRAC) on-
oard the Spitzer Space Telescope (Fazio et al. 2004 ), the mid-
nfrared (24- μm) observations of the cluster obtained using the

IPS (Hilton et al. 2010 ), and the PACS (Photodetecting Array
amera and Spectrometer) onboard the Herschel Space Obser-
atory . 2 A summary of the archi v al observ ations can be found
elow. 
SCUBA-2 450 μm and 850 μm : The J2215 cluster was observed

n band 1 using the SCUBA-2 camera by Ma et al. ( 2015 ) between
013 July and August. The total observation time of the 450 and
50 μm surv e ys was 8 h (i.e. 12 scans and each scan lasted for
0 min). For detailed information on data reduction and imaging,
ee Ma et al. ( 2015 ). The resulting noise level of the 450 and the
50 μm data was 5.4 and 0.63 mJy beam 

−1 , respectively. 
ACS/ HST : The J2215 galaxy cluster was among the 25 clusters

bserved during the HST type Ia Supernovae search observed with
he Wide Field Channel of the ACS mounted on the HST from 2005
uly to 2006 December. The total integration time for the observation
n the i 775 band was 3320 s (resulting in a 5 σ point source magnitude
imit of ≈ 25 . 1 ) whilst a total of 16 935 s exposure time was recorded
n the z 850 band (reaching a deeper magnitude limit of ≈26.0).
nformation on the observation is reported by Dawson et al. ( 2009 ),
ee also Hilton et al. ( 2009 ) for detailed information on the data
rocessing. 
MOIRCS/Subaru : Observation of the J2215 cluster was car-

ied out with the Multi-Object Infrared Camera and Spectrograph
MOIRCS, Ichikawa et al. 2006 ) mounted on the Subaru telescope.
he 5 σ point source limiting magnitude from the J and K s -band

mage was 24. More detailed information on the observation can be
ound in Hilton et al. ( 2009 ). 

IRAC/Spitzer : The IRAC imaging of J2215 was obtained through
rogramme 50 333 using all four channels (Ch1-Ch4: 3.6, 4.5, 5.8,
nd 8.0 μm) on 2008 July 12. The total observation time was 1500 s
xposures. The 5 σ point source magnitude limit obtained from the
RAC catalogue was ≈23 mag (Ch1 and Ch2) and ≈20 mag (Ch3
nd Ch4) (see Hilton et al. 2010 for details). 
 https:// irsa.ipac.caltech.edu/ Missions/ herschel.html 

https://github.com/IanHeywood/oxkat
https://irsa.ipac.caltech.edu/Missions/herschel.html


MeerKAT observations of XCS J2215 2845 

Figure 1. Left panel: an ≈ 70 arcmin × 50 arcmin zoom-in image of the radio continuum image produced from MeerKAT’s L -band observation with the image 
before peeling displayed (rms noise value = 3.9 μJy beam 

−1 ). The very bright ≈4.6 Jy source is visible in the image (indicated with ≈4.0 ′ radius magenta circle) 
and the target source is located at a distance of 0.7 ◦ (indicated with a 2.75 ′ × 2.75 ′ cyan box). Right panel: same as the left panel but the very bright ≈4.6 Jy 
source has been mitigated after the peeling operation. The o v erall image has been well impro v ed after the peeling process (rms noise value = 3.5 μJy beam 

−1 ). 

 

2
s
2  

w  

r
 

c  

b  

w
w
v  

l  

(
 

t  

a
T
0  

t
o  

e
 

t  

A
t
P

w  

i
m
a  

0  

v  

C
o  

o  

≈  

t  

a

i  

N  

a  

J

2

T  

u  

fi  

p
E  

e
c  

b  

l
a  

H

3

3

W
M
(  

o  

t  

u  

M
p
1  

t  

o
 

M
1  

i  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
MIPS/Spitzer obser v ation : The target cluster w as observed on
008 June 21 before the IRAC observation was conducted via the 
ame proposal. The total integration time for the observation was 
7 000 s. The 50 per cent completeness limit of the 24 μm observation
as estimated to be 70 μJy at 5 σ . Detailed information about data

eduction and photometry can be found in Hilton et al. ( 2010 ). 
ALMA (BAND 3) : A CO( J = 2–1) emission line surv e y of the

luster was by carried out by Hayashi et al. ( 2017 ) using ALMA in
and 3 in 2016 May. The total integration time for the observation
as 3.12 h. Data reduction and processing were done using CASA 

ith a standard pipeline. The mosaicked 3D cubes with five different 
elocity resolutions 50, 100, 200, 400, and 600 km s −1 yielded noise
evels of 0.17, 0.12, 0.11, 0.12, and 0.12 mJy beam 

−1 , respectively
see Hayashi et al. 2017 for detailed information). 

ALMA (BAND 6) : The CO( J = 5–4) observations of the core of
he J2215 were carried out by Stach et al. ( 2017 ) using forty-two 12 m
ntennae on 2016 June 19 under the project ID: 2015.1.00575.S. 
he six pointing observations yielded a synthesized beam size of 
 . 66 arcsec × 0 . 47 arcsec (PA = 78 ◦). The rms noise recorded for
he final mosaicked image was 48 μJy beam 

−1 . Detailed information 
n the data acquisition and reduction process can be found in Stach
t al. ( 2017 ). 

PACS Herschel 100 μm (green) and 160 μm (Red) : We obtained
he J2215 PACS 100 and 160 μm data from the Herschel Science
rchive. We conducted aperture photometry at the positions of 

he detected MeerKAT sources (see Section 3.1 below) using the 
HOTOUTILS package (Bradley et al. 2016 ). Fluxes were extracted 
ithin aperture sizes 6 arcsec and 12 arcsec for the 100 and 160 μm

mages, respectively. The noise in the images was estimated using the 
edian absolute deviation of the data in source-free regions using 

pertures of 6 arcsec and 12 arcsec. We obtained a noise level of
.2 and 0.3 mJy at 100 and 160 μm, respectively. Comparing our
alues to that of the catalogue from the Herschel /PACS Point Source
atalogue (HPPSC) via the Herschel User Provided Data Products 2 , 
ur values are consistent with that of the HPPSC (i.e. a median ratio
f ≈1 for both bands). Ho we ver, our v alues are higher by a factor of
3 and 2 for the green and red bands, respectively, with reference

o the values obtained by Ma et al. ( 2015 ) even after conducting
perture photometry using similar aperture sizes used in that work 
.e. 4.2 arcsec and 8.5 arcsec for the green and red bands, respectively.
ote that Ma et al. ( 2015 ) erroneously treat the 100 μm PACS data

s if it were 70 μm data, although no 70 μm PACS observations of
2215 exist in the archive. 

.4 Photometric redshifts 

he photometric redshift catalogue from Hilton et al. ( 2009 ) was
sed in this work. Hilton et al. ( 2009 ) used the EAZY spectral template
tting code of Brammer, van Dokkum & Coppi ( 2008 ) to estimate the
hotometric redshift for the cluster galaxies. A default option within 
AZY was selected to fit a linear combination of all of the spectral
nergy distribution templates to the cluster galaxies available in the 
atalogue within a redshift range of 0 < z < 4. The K−magnitude
ased Bayesian redshift prior used in the code resulted in a maximum
ikelihood redshift estimate z p and this estimated redshift value was 
dopted as the photometric redshift of the J2215 cluster galaxies, see
ilton et al. ( 2009 ) for detailed information. 

 ANALYSI S  

.1 MeerKAT source detection 

e extracted sources with > 4 σ detection significance from the 
eerKAT image using the Python Blob Detector and Source Finder 

 PyBDSF ; Mohan & Rafferty 2015 ) resulting in an initial catalogue
f 10 892 radio sources. We performed a ‘ne gativ e peak analysis test’
o quantify the level of noise contamination similar to the approach
sed by Patil et al. ( 2019 ). This was accomplished by inverting the
eerKAT continuum map and re-running PyBDSF using the same 

arameters we used for the original source extraction. We detected 
54 ne gativ e sources as compared to the > 10 K sources detected in
he original map which corresponds to a false positive detection rate
f ≈1.4 per cent. 
We estimated a 4 σ flux detection limit of 14 μJy from the
eerKAT image which translates to a limiting luminosity value of 

.5 ×10 23 WHz −1 at the cluster redshift of z = 1 . 46. This luminosity
s equi v alent to SFR = 46 M �yr −1 when assuming the Bell ( 2003 )
MNRAS 532, 2842–2859 (2024) 



2846 D. Y. Klutse et al. 

M

T able 1. The MeerKAT -detected cluster galaxies within 0.8 Mpc of the cluster centre. Columns: (1) The ID numbers of each source in the MeerKAT 

L -band image selected as a cluster member; (2) MeerKAT pointing coordinates of each galaxy the units of right ascension (R.A.) are in hours, minutes, 
and seconds; (3) the units of declination (Dec) are in degrees, arcminutes, and arcseconds; (4) the observed flux measured in μJy; (5) radio luminosities 
are measured in WHz −1 ; (6) the Far Infrared Radio Luminosity Ratio; (7) star formation rates measured in M � yr −1 . Note that these estimates will not 
be reliable for galaxies where the radio emission is primarily due to the presence of an AGN; (8) radial distance of sources measured with respect to the 
cluster X-ray position measured in Mpc. The potential radio AGNs are marked with asterisks ( ∗) whilst objects blended in the MeerKAT radio image 
are denoted by the obelisk mark ( † ). 

(1) (2) (3) (4) (5) (6) (7) (8) 
MKT ID R.A. Dec. F 1 . 3 L 1 . 4 q IR SFR R 

(J2000) (J2000) ( μJy) (10 23 WHz −1 ) (M � yr −1 ) (Mpc) 

5242 ∗ 22:16:04.79 −17:37:52.5 102 ± 8 10.6 ± 0.8 2.37 ± 0.25 336 ± 26 0.76 
5307 22:16:03.75 −17:37:10.4 24 ± 7 2.5 ± 0.8 2.53 ± 0.37 81 ± 24 0.77 
5332 ∗ 22:16:03.10 −17:38:39.3 197 ± 8 20.4 ± 0.8 1.74 ± 0.39 650 ± 26 0.64 
5336 22:16:03.26 −17:39:09.0 18 ± 7 1.9 ± 0.7 1.95 ± 1.0 60 ± 23 0.80 
5373 22:16:02.50 −17:37:56.5 40 ± 9 4.2 ± 1.0 1.76 ± 0.35 133 ± 31 0.49 
5442 † 22:16:00.88 −17:38:31.7 77 ± 7 8.0 ± 0.7 2.48 ± 0.22 254 ± 22 0.38 
5443 † 22:16:00.59 −17:38:35.3 61 ± 6 6.3 ± 0.6 2.55 ± 0.22 201 ± 20 0.38 
5469 ∗ 22:16:00.30 −17:37:50.7 174 ± 12 18.1 ± 1.3 1.85 ± 0.42 576 ± 40 0.24 
5470 ∗† 22:15:59.67 −17:37:59.2 125 ± 11 13.0 ± 1.2 2.45 ± 0.35 412 ± 37 0.14 
5492 22:16:00.29 −17:38:58.7 21 ± 8 2.2 ± 0.8 2.19 ± 0.26 70 ± 26 0.52 
5517 † 22:15:59.75 −17:38:17.0 47 ± 7 4.9 ± 0.8 2.48 ± 0.24 154 ± 24 0.19 
5551 † 22:15:59.10 −17:37:41.5 69 ± 11 7.2 ± 1.1 2.45 ± 0.34 228 ± 35 0.19 
5574 ∗ 22:15:58.13 −17:38:18.3 108 ± 10 11.2 ± 1.1 2.21 ±40.41 356 ± 34 0.14 
5618 ∗† 22:15:57.27 −17:37:55.8 140 ± 11 14.5 ± 1.1 2.01 ± 0.33 462 ± 35 0.16 
5619 † 22:15:57.70 −17:37:45.6 66 ± 8 6.9 ± 0.8 2.31 ± 0.37 219 ± 25 0.17 
5642 22:15:57.30 −17:37:07.1 42 ± 9 4.3 ± 1.0 2.18 ± 0.5 137 ± 30 0.49 
5675 22:15:56.95 −17:38:05.9 28 ± 9 2.9 ± 0.9 2.32 ± 0.38 92 ± 30 0.19 
5685 22:15:56.75 −17:37:22.2 29 ± 8 3.0 ± 0.8 2.26 ± 0.59 96 ± 25 0.40 
5766 † 22:15:55.01 −17:38:03.7 73 ± 14 7.6 ± 1.5 2.50 ± 0.45 242 ± 47 0.42 
5790 22:15:54.73 −17:38:13.0 30 ± 8 3.1 ± 0.9 – 99 ± 27 0.46 
5793 22:15:54.65 −17:38:54.8 29 ± 8 3.0 ± 0.8 2.55 ± 0.27 96 ± 27 0.64 
5810 22:15:54.37 −17:37:41.1 23 ± 7 2.4 ± 0.7 – 76 ± 22 0.53 
5837 22:15:53.80 −17:37:32.9 21 ± 7 2.2 ± 0.7 2.49 ± 0.34 71 ± 22 0.62 
5872 22:15:52.86 −17:37:11.8 74 ± 7 7.7 ± 0.7 2.19 ± 0.22 244 ± 23 0.80 

r  

C

3

W  

p  

2  

m  

W  

p  

w  

m  

r  

w
i  

m  

w  

(  

C  

c  

H  

a  

2  

t  

o  

t  

g
 

t  

o  

J  

p  

d  

S

3

W  

C  

l  

m  

o  

B  

b  

C  

e  

r  

o  

a  

l  

l  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
elation between radio continuum luminosity and SFR assuming a
habrier ( 2003 ) IMF. 

.2 XMMXCS J2215.9 −1738 cluster membership classification 

e cross-matched the MeerKAT source catalogue with the archi v al
hotometric (Hilton et al. 2009 ) and spectroscopic (Hilton et al.
010 ) redshift catalogues of J2215 at z = 1.46 using a cross-
atching radius of 6.0 arcsec to determine their cluster membership.
e selected galaxies as cluster members within R 200 based on their

hotometric or spectroscopic redshifts. For the photometric selection,
e adopted a similar redshift cutoff that was used to define the cluster
embership in Hilton et al. ( 2009 ) (i.e. cluster galaxies within the

edshift range 1.27 < z p < 1.65 were assumed to be cluster members)
hilst sources having spectroscopic redshifts = 1.46 ± 3 σ (where σ

s the line-of-sight velocity dispersion) were also selected as cluster
embers. We also cross-matched the MeerKAT source catalogues
ith catalogues obtained from [ O II ] emitters via the narrow band

NB) observation of the J2215 cluster at z = 1.46 using the Suprime-
am on the Subaru Telescope. Two sources were selected from the
ross-match between the NB912 (NB912 � = 9139 Å, � � = 134 Å;
ayashi et al. 2010 , 2011 ), one source was detected using the NB921

nd seven sources using the NB 912 + 921 catalogues (Hayashi et al.
014 ). In total, we detected 24 cluster members within 0.8 Mpc of
he cluster centre (see Table 1 and Fig. 2 ). The median angular size
f the selected sources presented in Table 1 is ≈ 7 ′′ which suggests
NRAS 532, 2842–2859 (2024) 
hat the majority of our selected radio sources are barely resolved
iven a synthesized beam size of ≈ 6 arcsec. 
We further cross-matched these 24 cluster members detected in

he MeerKAT image with catalogues from other archi v al infrared
bservations as described in Section 2.3 . The final sample of the
2215 cluster members in the MeerKAT L -band and their counter-
arts in other archi v al observ ations are summarized in T able 2 . W e
iscuss how the far-infrared and radio luminosities were derived in
ections 3.3 and 3.4 , respectively. 

.3 SED fitting with CIGALE 

e performed spectral energy distribution (SED) fitting using
IGALE V2022.0 (Yang et al. 2022 ) to derive the far-infrared (IR)

uminosity values of the cluster members. We used the following
odels; the delayed star formation history model (see equation 4

f Boquien et al. 2019 ), the single stellar populations model of
ruzual & Charlot ( 2003 ), the nebular emission template adopted
ased on Inoue ( 2011 ), dust attenuation templates based on the
harlot & Fall ( 2000 ) model, dust emission template of Draine
t al. ( 2014 ), AGN models from Stalevski et al. ( 2016 ), and galaxy
adio synchrotron emission based on the radio-infrared correlation
f Helou, Soifer & Rowan-Robinson ( 1985 ). CIGALE V2022.0
lso models the AGN components of galaxies using the radio-
oudness parameter (Ballo et al. 2012 ) and assumes an AGN power-
aw SED within the wavelength range 0.01–100 cm. The fitting



MeerKAT observations of XCS J2215 2847 

Figure 2. A zoom-in of the MeerKAT radio map centred on the cluster members. The positions of the cluster members and the ID numbers of the sources in 
Table 1 are shown with circles and numbers, respectively. The ≈6 arcsec synthesized beam size is displayed in the lower-left corner of the image. 

p  

t
l

t  

A  

(  

t  

M  

d  

c
fi  

v  

S  

t  

w
a
o  

(

w  

e  

fl  

d  

t  

o  

t  

f  

m  

s  

 

V
v  

c  

χ  

r  

d  

SED fitting with CIGALE V2022.0 . 

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
arameters we used in this work are listed in Table B1 . One of
he output parameters derived from the SED fitting was the infrared 
uminosity. 

Comparing the infrared luminosity values derived after fitting 
he models to the observed SEDs (i.e. optical to radio bands, see
ppendix A ) to that of the counterpart values in table 2 of Ma et al.

 2015 ) we observed that the infrared luminosity v alues deri ved in
his work are about a factor of 2.4 higher than that estimated in

a et al. ( 2015 ). We probed further and realized that the J2215
ata was taken using the 100 and the 160 μm filters of the PACS
amera onboard the Herschel Space Telescope and not the 70 μm 

lters presented in Ma et al. ( 2015 ) and therefore the luminosity
alues they obtained after the SED fit are likely to be in error (see
ection 2.3 ). We also fitted the SED models to only the mid-infrared

o the radio data (the exact bands used in Ma et al. 2015 ) to ascertain
hether the inclusion of the optical and near-infrared bands had 

ny impact on the results generated, we obtained a median ratio 
f 1.6 which is still higher than the value obtained in Ma et al.
 2015 ). 
We also compared the MeerKAT flux densities reported in this 
ork to those of the nine detected JVLA sources reported in Ma

t al. ( 2015 ). We find a median flux ratio of 1.34, with the MeerKAT
ux densities being higher on average. Ho we ver, only 6/9 of the
etected JVLA sources reported on in Ma et al. ( 2015 ) are considered
o be cluster members in our work, and as indicated in Table 1 , all
f these are either blended in the MeerKAT image, or are thought
o be AGN, which may explain the higher MeerKAT flux densities
or these sources. We hav e v erified that the MeerKAT flux density
easurements in our catalogue are within 5 per cent on average of

ources detected in the NRAO VLA Sk y Surv e y (Condon et al. 1998 ).
We show in Appendix B examples of SED fitting using CIGALE
2022.0 , the goodness of the fit is measured via the reduced χ2 

alue. The median reduced χ2 value is 1.9 for all galaxies selected as
luster members. In total, 17/24 (71 per cent) sources had a reduced
2 value of less than 3. Whilst 20/24 (83 per cent) sources had

educed χ2 value less than 5. This implies that the observed SED
ata are reasonably well-fitted with the models we selected for the
MNRAS 532, 2842–2859 (2024) 



2848 D. Y. Klutse et al. 

M

Table 2. We compare the sources detected in the MeeKAT L -band image to other archi v al observ ations. Columns: (1) Same as in Table 1 ; (2) source 
ID numbers from the SCUBA-2 observation compiled by Ma et al. ( 2015 ); (3) ACS/ HST (Hilton et al. 2009 ); (4) 24 μm MIPS/Spitzer observation 
(Hilton et al. 2010 ); (5) ALMA (Band 3; Hayashi et al. 2017 ); (6) ALMA (Band 6; Stach et al. 2017 ); (7) KMOS (Chan et al. 2018 ); (8) morphology E 

= elliptical, S0 = lenticular, Sp = spiral, Irr = irregular Hilton et al. (see Section 3.2 of 9 for details); (9) log 10 M ∗[M �]; the logarithmic expression of 
the cluster galaxy stellar mass. 

(1) (2) (3) (4) (5) (6) (7) (8) (9) 
MKT ID M15ID H9ID H10ID HY17ID ST17ID C18 Morph log 10 M ∗[M �] Notes 

5242 ∗ 10 709 – – – – Sp + Irr 11.11 ± 0.06 –
5307 – 460 – – – – Sp + Irr 9.58 ± 0.09 –
5332 ∗ 2 1022 – – – – Sp + Irr 9.72 ± 0.15 –
5336 – 1201 – – – – Sp + Irr 9.61 ± 0.08 –
5373 – 728 – – – – Sp + Irr 10.63 ± 0.11 –
5442 † 3 983 – ALMA.B3.16 – – Sp + Irr 11.02 ± 0.12 NB921 [O II ] 
5443 † – 1004 – – – 575 E 9.81 ± 0.13 –
5469 ∗ – 692 53 ALMA.B3.14 5 1011 E 11.06 ± 0.10 NB912 [O II ] 
5470 ∗† 6 747 – ALMA.B3.06 9 930 Sp + Irr 11.49 ± 0.07 NB921 [O II ] 
5492 – 1130 – ALMA.B3.17 – – Sp + Irr 11.29 ± 0.05 NB912 + NB921 [O II ] 
5517 † – 874 – ALMA.B3.13 12 724 – 11.01 ± 0.09 –
5551 † – 614 – ALMA.B3.11 2 1118 E 11.03 ± 0.05 NB912 + NB921 [O II ] 
5574 ∗ 4 884 899 ALMA.B3.05 13 732 – 9.82 ± 0.16 NB921 [O II ] 
5618 ∗† 13 734 35 ALMA.B3.07 7 940 Sp + Irr 9.64 ± 0.02 NB912 + NB921 [O II ] 
5619 † – 653 – ALMA.B3.09 – 1038 Sp + Irr 11.25 ± 0.11 NB912 + NB921 [O II ] 
5642 – 419 – – – – E 9.48 ± 0.13 NB912 + NB921 [O II ] 
5675 – 786 39 ALMA.B3.12 – 829 Sp + Irr 10.65 ± 0.25 NB912 [O II ] 
5685 – 529 529 – – 1261 Sp + Irr 10.93 ± 0.08 –
5766 † – 777 – – – 798 S0 9.85 ± 0.06 –
5790 – 863 – – – – – – –
5793 – 1118 1118 – – 412 – 11.27 ± 0.08 –
5810 – 658 – – – – – – –
5837 – 571 – – – – – 10.41 ± 0.14 –
5872 – 446 – – – – – 11.52 ± 0.09 –

3

T  

s  

l  

o  

t  

a  

1

L

w  

F  

u  

b

S

w
1  

l  

(  

T  

r  

i  

v

a  

1

4
O

A  

g  

m  

g  

l  

2  

e  

t  

o  

t  

w  

m  

W  

t  

c

4

W  

m  

t  

d  

(  

o  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
.4 Radio luminosities and continuum star formation rates 

he observed 1.3 GHz radio flux F 1 . 3 , obtained from the MeerKAT
urv e y was scaled to the rest frame 1.4 GHz ( F 1 . 4 ) using a power
aw with a slope ( α) of −0.8 (i.e. F 1 . 4 = ( ν1 . 4 /ν1 . 3 ) 

α×F 1 . 3 , this value
f α is assumed for non-thermal radio emissions, Bell 2003 ) and
hen converted the F 1 . 4 into a rest-frame 1.4 GHz luminosity ( L 1 . 4 )
ssuming a spectral index ( α) dependent K -correction factor (Condon
992 ; Karim et al. 2011 ), 

 1 . 4 = 

9 . 52 × 10 12 × F 1 . 4 D 

2 
L 4 π

(1 + z) 1 + α
, (1) 

here, D L is the luminosity distance in Mpc, L 1 . 4 in units of WHz −1 ,
 1 . 4 in units of μJy, and z = 1.46. We converted the L 1 . 4 into SFR
sing the calibration of the Far Infrared Radio Correlation (FIRRC)
y Bell ( 2003 ) scaled to a Chabrier IMF (Karim et al. 2011 ), i.e. 

FR (M � yr −1 ) = 

⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ 

⎪ ⎪ ⎪ ⎪ ⎪ ⎩ 

3 . 18 × 10 −22 L 1 . 4 , L 1 . 4 > L c , 

3 . 18 × 10 −22 L 1 . 4 

0 . 1 + 0 . 9 

(
L 1 . 4 

L c 

)0 . 3 , L 1 . 4 � L c , 
(2) 

here L c is the radio luminosity of an L ∗ galaxy (i.e. 6 . 4 ×
0 21 WHz −1 ). Bell ( 2003 ) sees the need to distinguish between low-
uminous and high-luminous star-forming galaxies since non-thermal
low-luminous) galaxies may have their SFR rate underestimated.
hough the MeerKAT observation mostly exploits the non-thermal

adio regime, our radio luminosities mostly fall abo v e the L c threshold
.e. L 1 . 4 ≥ 10 23 WHz −1 and also fall below radio luminosity ( L rad )
alue of 10 25 WHz −1 where radio emissions at L rad � 10 24 −25 WHz −1 
NRAS 532, 2842–2859 (2024) 
re mostly dominated by radio loud AGNs (Miller, Peacock & Mead
990 ; Yun, Reddy & Condon 2001 ). 

 M O R P H O L O G I E S  A N D  O P T I C A L  C O L O U R S  

F  R A D I O  SOURCES  IN  J 2 2 1 5  

GN host galaxies can be identified via their colours and morpholo-
ies. Previous colour–magnitude relation studies have shown that
ost galaxies found on the red sequence or blue cloud are normal

alaxies (Baldry et al. 2004 ) whilst the majority of the AGNs are
ocated in the green valley (Nandra et al. 2007 ; Schawinski et al.
010 ; Povi ́c et al. 2012 ; Wang et al. 2017 ; Gu et al. 2018 ; Lacerda
t al. 2020 ). The green valley represents a transitional region between
he blue cloud (mostly populated by late-type galaxies having spiral
r irregular morphologies) and the red sequence (populated by early-
ype galaxies having elliptical or lenticular morphologies). In this
ork, we investigated the position of the MeerKAT-detected cluster
embers on the colour–magnitude diagram and their morphologies.
e describe the morphological classification in Section 4.1 and

he position of the morphologically classified radio galaxies on the
olour–magnitude diagram in Section 4.2 . 

.1 Morphological classification 

e used the catalogue of morphologically classified J2215 cluster
embers in Hilton et al. ( 2009 ) for this work. We cross-matched

he catalogue in Hilton et al. ( 2009 ) with that of the MeerKAT-
etected cluster members within a radius of 6 arcsec. In Hilton et al.
 2009 ) morphological classification of the galaxies was performed
n magnitudes brighter than z 850 ≤ 24 via visual inspection by



MeerKAT observations of XCS J2215 2849 

Figure 3. The postage stamp images of the selected members of the J2215 cluster within a radius of 0.8 Mpc of the original X-ray cluster position. The source 
ID number presented in Table 1 is located in the upper left corner of each extracted postage stamp image. The cross marker represents the MeerKAT source 
position. Each set of the postage stamp images was taken from the MeerKAT L -band image (left panel), the JVLA L -band image in A configuration (Ma 
et al. 2015 ) (middle panel) and the Hubble Space Telescope ACS z 850 -band image (right panel image from which the initial classification of the cluster galaxy 
morphology was performed in Hilton et al. 2009 ). The MeerKAT and JVLA postage stamp images are 40 arcsec, whilst the HST image is 3.75 arcsec on a side 
with east at the left. 

fi
t
T
g  

s  

d
c  

s

i  

s  

p  

(
f  

(

l  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ve human classifiers under four different morphological bins i.e. 
he elliptical (E), lenticular (S0), spiral (S), and irregular (Irr). 
hey determined the morphologies using an existing training set of 
alaxies compiled by Postman et al. ( 2005 ) from the morphological
tudies of high-redshift cluster galaxies i.e. q0.8 < z < 1.3. An in-
epth description of the morphological classification of the J2215 
luster galaxies can be found in Hilton et al. ( 2009 ). The postage
tamp images of the selected galaxies from the MeerKAT L -band 
mage listed in Table 1 can be seen in Fig. 3 . Each set of the postage
tamp images was taken from the MeerKAT L -band image (left
anel), the JVLA L -band image in A configuration (Ma et al. 2015 )
middle panel) and the z 850 -band counterpart images (right panel) 
rom which the initial morphologies were determined by Hilton et al.
see section 3.2 of Hilton et al. 2009 ). 

Only 17/24 MeerKAT-detected cluster members had their morpho- 
ogical counterparts in the Hilton et al. ( 2009 ) catalogue, 7/24 did
MNRAS 532, 2842–2859 (2024) 



2850 D. Y. Klutse et al. 

M

Figure 4. Caption as per Fig. 3 . 

n  

d  

s  

w  

1  

s

4

F  

o  

i  

c  

a  

d  

t  

t  

r  

u  

b  

c  

S  

o
 

o  

b  

h  

b  

i  

m  

t
 

3  

w  

t  

e  

g

5

5

T  

t  

i  

f
 

c  

s  

s

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ot hav e an y morphological classification. Out of the 17 MeerKAT-
etected cluster members having morphological classification 12 had
piral/irregular morphology (6/12–SFGs, 2/6–ISFGs, 4/6–AGNs)
hilst five had elliptical/lenticular galaxy morphology (3/12–SFGs,
/6–ISFGs, and 1/6–AGNs). Therefore, the majority of the MeerKAT
ources correspond to host galaxies with late-type morphologies. 

.2 Colour–magnitude relation 

ig. 5 shows the ( z 850 − J ) colour versus magnitude ( J ) diagram
f all cluster members selected in the MeerKAT L -band radio
mage that had counterparts in the optical and infrared catalogue
ompiled by Hilton et al. ( 2009 ). The solid black line shows
 fit to the colour–magnitude relation of the early-type galaxies
etected in Hilton et al. ( 2009 ) i.e. the red sequence. We show
he early-type galaxies (i.e. Elliptical/lenticular–E + S0) and late-
ype galaxies (spiral/irregular–Sp + Irr) with circles and stars,
espectively, whilst sources without morphological classification and
ncertain morphologies are shown as diamond markers. The red,
lue, and black markers signify MeerKAT-detected cluster members
lassified as AGNs, star-forming galaxies (SFGs), and intermediate
FGs (ISFGs), respectively (see Section 5 ). A significant number of
ur sources fall below the red sequence line i.e. in the blue cloud. 
The majority of these sources are blue and faint, ho we ver, three

utliers with source IDs #5332, #5242, and #5872 are blue but very
NRAS 532, 2842–2859 (2024) 
right compared to the other cluster members. IDs #5332 and #5242
ave dual morphologies i.e. spiral/irregular suggesting that they are
lended in the infrared image ( J ) (Hilton et al. 2009 ) and the JVLA
mage (see Figs 3 and 4 ) thus their photometry and estimated SFR
ay not be reliable. We also classified them as radio AGNs based on

he three selection schemes mentioned in Section 5 . 
Also, we detected both early-type and late-type galaxies within

- σ of the red sequence in this work (Fig. 5 ). The elliptical galaxies
ithin the red sequence of J2215 are very bright ( J < 22.5) compared

o other ellipticals outside the red sequence. Is it likely that J2215 is
xperiencing the final episode of star formation within its early-type
alaxies? 

 CLASSI FI CATI ON  O F  R A D I O  S O U R C E S  

.1 Mid-infrared colour–colour criteria 

he mid-infrared photometry provides a robust approach to the iden-
ification of A GNs. A GN SEDs are redder than star-forming galaxies
n the mid-infrared bands where stellar populations dominate the red,
alling portion of the stellar spectra. 

Stern et al. ( 2005 ) found that simple mid-infrared colour criteria
ould robustly separate AGNs from normal star-forming galaxies and
tars (this method could identify > 90 per cent of spectroscopically
elected quasars and Seyfert 1 galaxies). 



MeerKAT observations of XCS J2215 2851 

Figure 5. The z 850 − J colour–magnitude diagram of all the selected cluster 
members detected in the MeerKAT L -band image located within 0.8 Mpc 
of the cluster centre. Elliptical/lenticular (E + S0) galaxies are shown as 
circles; spiral/irregular galaxies (Sp + Irr) with stars; galaxies without any 
morphological classification are marked with a diamond shape (None). The 
filled markers signify the MeerKAT sources classified as AGNs whilst the 
thick and thin unfilled markers represent SFGs and ISFGs, respectively. The 
solid line denotes the red sequence path derived from a fit to the colour–
magnitude relation of the elliptical/lenticular galaxies (early-type) detected 
in Hilton et al. ( 2009 ). The dotted dashed lines show the 3- σ deviations 
abo v e and below the red sequence. It can be seen that the majority of our 
cluster members are located below the red sequence line and 12 out of the 17 
MeerKAT-detected cluster members with morphological classification have 
spiral/irregular morphology ( ≈71 per cent). This may imply that the majority 
of all the MeerKAT-detected cluster galaxies may be actively forming stars 
irrespective of whether they have been classified as AGNs, ISFGs or SFGs. 

Figure 6. The IRAC colour–colour plot of the selected cluster members 
detected with the MeerKAT telescope, with some sources falling within the 
‘AGN zone’ (shaded re gion). Ov erlaid are two evolutionary tracks of spectral 
templates from the library of Polletta et al. ( 2007 ); a starburst galaxy (M82) 
and an active galaxy (Seyfert 2) as they are redshifted from z = 0 (the crosses 
in each template track) to z = 2 (the open end of each track). The majority 
of the cluster members are all located outside the AGN wedge which implies 
that using only the IRAC colour–colour plot as diagnostic tool, the majority 
of the radio emission is mostly powered by star-formation activity rather than 
AGNs. 

 

s
c
o
f  

o
=
c

Figure 7. The distribution of q IR (Far Infrared Radio Luminosity Ratio) 
for all sources in the MeerKAT/FIR sample (hashed histogram). The black 
vertical dashed line denotes the mean q IR value, < q IR > greater than 2 for the 
entire MeerKAT/FIR sample (i.e. < q IR > = 2 . 36 ± 0 . 04). The black vertical 
dot–dashed line represents the threshold below which sources are classified 
as radio excess AGN i.e. q IR = 1.68 adopted from Del Moro et al. ( 2013 ). 
We compared our work with other surv e ys of high-redshift galaxies from the 
ALESS surv e y of the sub-millimetre galaxies in the Extended Chandra Deep 
Field South observation with ALMA 870 μm and JVLA (dotted histogram, 
Swinbank et al. 2014 ) and the GOODS-Herschel (North) field with the VLA, 
24 μm Spitzer/MIPS, Chandra X-ray, and Herschel infrared data (star-filled 
histogram, Del Moro et al. 2013 ). All the MeerKAT/FIR sources fall within 
the radio normal region using only the q IR criterion. 

A  

s  

S
q  

c  

r
n
p  

t  

t

[  

c
T
a
A  

2

5

T  

d
e
i  

(

q

w  

l
u  

S

(
h

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
We adopted the same approach from Stern et al. ( 2005 ). Fig. 6
hows the mid-infrared colour–colour plot of the MeerKAT-detected 
luster members obtained from the IRAC catalogue. We show an 
 v erlay of two non-evolving tracks of spectral templates obtained 
rom the library of Polletta et al. ( 2007 ) co v ering a wav elength range
f 8–1000 μm. The figure shows these tracks as they evolve from z 

 0 to z = 2. The shaded region depicts the area within the colour–
olour space dominated by broad-lined AGNs (Stern et al. 2005 ). 
ny source located outside the ‘AGN wedge’ is likely to be a normal
tar-forming galaxy or another type of AGN apart from quasars and
eyfert 1 (since the mid-infrared colour criteria can identify only 
uasars and Seyfert 1 efficiently; Stern et al. 2005 ). From Fig. 6 , it
an be observed that at least five cluster members fall within this
egion, IDs: #5492, #5442, #5517, #5551, and #5469. Sources with 
o IRAC data were not included in the colour–colour AGN diagnostic 
lot. i.e. 8/24 cluster members did not have IRAC photometry for the
wo longest wavelength IRAC channels (i.e. 5.8 and 8.0 μm) because
he IRAC images at those bands are shallow. 

Due to the large error bars in both the [3.4] − [4.5] and [5.8] −
8.0] colours we can not draw a firm conclusion based on only this
riterion (i.e. the mid-infrared colour–colour AGN diagnostic plot). 
he Far-Infrared Radio Luminosity Ratio, q IR and the FIRRC are 
lternatives to distinguish normal star-forming galaxies from radio 
GN via the so-called ‘radio excess’ approach (Del Moro et al.
013 ). 

.2 Far-Infrared Radio Luminosity Ratio 

he Far Infrared Radio Luminosity Ratio, q IR value is used to
istinguish normal star-forming galaxies from AGNs (Del Moro 
t al. 2013 ; Delvecchio et al. 2021 ; Radcliffe et al. 2021 ). This q IR 
s defined similarly to Condon ( 1992 ), Bell ( 2003 ), Magnelli et al.
 2015 ), Calistro Rivera et al. ( 2017 ), and Algera et al. ( 2020 ) as: 

 IR = log 10 

(
L IR 

3 . 75 × 10 12 

)
− log 10 ( L 1 . 4 ) , (3) 

here L 1 . 4 is the rest-frame radio luminosity. L IR is the total infrared
uminosity obtained from the best-fitting model to the observed SED 

sing galaxy dust emission templates from Draine et al. ( 2014 ), see
ection 5.3 for details. 
Fig. 7 shows the q IR distribution of all the MeerKAT/far-infrared 

FIR) selected cluster members (i.e. hashed histogram) and other 
igh-redshift field galaxies from the VLA/ALMA 870- μm obser- 
MNRAS 532, 2842–2859 (2024) 



2852 D. Y. Klutse et al. 

M

Figure 8. The FIRRC for all the J2215 MeerKAT/FIR galaxies located within 
0.8 Mpc of the cluster centre. It can be seen that all the MeerKAT radio 
normal star-forming galaxies 18/24 (solid star markers) and the radio loud 
AGNs 6/24 (solid square markers) are all located within the radio normal 
region i.e. below the radio excess AGN fit. We compare our work with other 
high-redshift galaxies from the GOODS- North field (Del Moro + 2013 
radio excess star-forming galaxies–plus symbol; Del Moro et al. 2013 ) and 
the ECDFS star-forming galaxies (Swinbank + 2014 SFG–cross marker; 
Swinbank et al. 2014 ). The dashed and dotted–dashed lines represent the 
best-fitting models to the radio excess star-forming galaxies and AGNs from 

Del Moro et al. ( 2013 ), respectively. 

v  

(  

b  

1  

w  

d  

a  

s  

‘
 

i  

g

5

A  

o  

i  

f  

1  

2  

f  

o  

g  

L  

o  

S  

d  

e  

f  

c
 

s  

m  

o  

o  

t  

f

5

W  

i  

s  

p  

o  

I  

1

6

B  

a  

2  

A  

i
 

c  

D  

d  

r  

t  

a  

a
 

s  

l  

G  

o
 

g  

t  

t  

f  

c  

a  

r
 

z  

fi
 

a  

o

7

7

O  

r  

s  

e  

4  

N  

T  

t  

t  

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ations (dotted histogram; Swinbank et al. 2014 ) and VLA/24- μm
star-filled histogram, Del Moro et al. 2013 ). The cutoff/separation
oundary adopted from Del Moro et al. ( 2013 ) was defined at q IR =
.68 (the black vertical dot–dashed line). Sources with q IR > 1.68
ere defined as ‘radio normal’ whilst those below the cutoff were
efined to be ‘radio excess’ sources. The ‘radio normal’ sources
ccording to Del Moro et al. ( 2013 ) are mostly dominated by
tar-forming galaxies and low-luminous radio AGNs, whilst the
radio excess’ sources are most likely to be dominated by AGNs. 

Comparing our work with other high-redshift field galaxy surv e ys
n Fig. 7 , it can be seen that, all the MeerKAT/FIR detected cluster
alaxies fall within the radio normal region. 

.3 Far-infrared radio correlation 

lthough there is a v ast dif ference between the emission mechanism
f radio (synchrotron) and infrared galaxies (dust re-emission), there
s a tight correlation between the radio and FIR luminosities of star-
orming galaxies (Helou et al. 1985 ; Condon, Anderson & Helou
991 ; Yun et al. 2001 ; Bell 2003 ; Appleton et al. 2004 ; Ivison et al.
010 ; Basu et al. 2015 ; Algera et al. 2020 ). The FIRRC has been
ound to be well correlated for several galaxy samples irrespective
f their stellar activities or morphology excluding radio-loud active
alaxies or some other interacting galaxies (Lisenfeld & V ̈olk 2010 ;
isenfeld et al. 2015 ). Previous works have established the validity
f the FIRRC out to z ≈ 2 (Garrett 2002 ; Appleton et al. 2004 ;
argent et al. 2010a , b ; Algera et al. 2020 ), with only a few uncertain
eviations at higher redshifts ( 3 ≤ z ≤ 6 ) (Murphy 2009 ; Seymour
t al. 2009 ; Shen et al. 2022 ). The FIRRC for all the J2215 members
rom the MeerKAT L -band surv e y (this work) and its FIR counterpart
an be seen in Fig. 8 . 

We compare our work with other high-redshift radio excess
tar-forming galaxies from the GOODS-Herschel North field (plus
arker; Del Moro et al. 2013 ) and the ALMA LESS (ALESS) surv e y

f sub-millimetre Extended Chandra Deep Field South (ECDFS)
bservation (cross marker; Swinbank et al. 2014 ) in Fig. 8 . All of
NRAS 532, 2842–2859 (2024) 
he MeerKAT/FIR sources are located within the radio normal star-
orming region. 

.4 Source classification 

e employed three selection criteria to distinguish SFGs from AGNs
.e. sources with L 1 . 4 < 10 24 WHz −1 , q IR values > 1.68 and finally
ources detected outside the IRA C A GN zone were all considered as
otential SFG candidates. Ho we ver, sources that passed at least two
ut of the three selection criteria mentioned earlier are considered
SFGs otherwise, they are selected as AGNs. In total, we classified
2/24 sources as SFGs, 6/24 as ISFGs, and 6/24 as AGNs. 

 A  G N  A  CTI VI TY  WI THI N  J 2 2 1 5  

ased on the three selection criteria described in Section 5 we classify
ll the other non-SFGs as potential AGNs i.e., 6/24 this implies that
5 per cent of the MeerKAT sources that are cluster members are
GN hosts and there is a possibility of a higher AGN activity ongoing

n the core of this high-redshift cluster. 
Alberts et al. ( 2016 ) found that in a sample of ≈250 galaxy

lusters obtained from the IRAC Shallow Cluster Surv e y and IRAC
istant Cluster Surv e y at 0.5 < z < 2, AGN-composite (moderately
ominated AGN fraction) cluster galaxies increased with respect to
edshifts abo v e the field galaxies at 1.0 < z < 1.5. They attributed
he rise in the AGN-composite cluster galaxies at these redshifts to
 conducive cluster environment that allows black hole growth or
ctivities. 

Also, a multiwavelength study of AGNs using three different
election criteria i.e. mid-IR colour, radio luminosity, and X-ray
uminosity of galaxy clusters in the 8 . 5 ◦ × 8 . 5 ◦ Boots field by
alametz et al. ( 2009 ) showed that a higher fraction of AGN is
bserved in cluster centres than in the field at z � 1.0. 
Further, Martini et al. ( 2013 ) studied the AGN evolution of 13

alaxy clusters from the Spitzer/IRAC Shallow Cluster Surv e y and
heir surrounding field galaxies at 1 < z < 1.5. They observed that
here was an order of magnitude increase in the evolution of AGN
raction in clusters from 0 < z < 1.25 than in the field of these
lusters whilst a reverse is seen at lower redshift i.e. AGN fraction is
 factor of ≈6 times higher in the field than in the clusters at lower
edshift. 

High-redshift studies of two proto-clusters and field galaxies at
 > 2 showed an increase in AGN fractions in proto-cluster than in
eld galaxies (Lehmer et al. 2009 ; Digby-North et al. 2010 ). 
Giv en the abo v e e xamples in relation to the high-redshift clusters

nd their AGN fraction, the J2215 galaxy cluster is no different from
ther higher redshift clusters investigated by previous surveys. 

 STAR  F O R M AT I O N  AC TI VI TY  WI THI N  J 2 2 1 5  

.1 Star formation rate versus stellar mass 

bservations suggest a rough correlation between the star formation
ates in galaxies and their stellar mass ( M ∗), the so-called ‘main
equence’ (Brinchmann et al. 2004 ; Elbaz et al. 2007 ; Noeske
t al. 2007 ). This correlation claimed to hold from z = 0 to z =
 (Brinchmann et al. 2004 ; Daddi et al. 2007 ; Elbaz et al. 2007 ;
oeske et al. 2007 ; Pannella et al. 2009 ; Schreiber et al. 2015 ).
his has resulted in the ‘dual mode’ of star formation evolution i.e.

he normal star-forming mode and the starburst mode depending on
he position of the galaxy in the SFR–M ∗ plot although there are



MeerKAT observations of XCS J2215 2853 

Figure 9. The SFR versus M ∗ of the SFGs (solid squares) and the interme- 
diate SFGs (solid triangle) detected in the J2215 MeerKAT image. The solid 
line represents the best-fit to the MS relations derived from observations in 
literature (i.e. equation 28 of Speagle et al. 2014 ) at the cluster redshift ( z = 

1.46). The ±0.3 dex scatter about the adopted MS fit to our data is represented 
as the blue dotted line. The 4- σ radio flux detection limit equi v alent to an 
SFR value of 46 M � yr −1 is represented as the black horizontal dashed line. 

i
2

t
s  

2
a  

w
i  

2
 

w  

w  

o  

o  

i  

F  

f
 

c  

s
o

 

t  

m  

o  

p
s
a  

e  

m
 

(  

g
o  

m  

t
a  

a  

a
o  

h  

m  

r
 

g  

t  

a  

t  

i  

w  

l  

m
w
i  

g

7

W
g  

m
 

a  

t  

r  ∑
 

v  

c
 

m  

e  

h
r
w  

p  

μ  

s
n

C  

a  

v  

e

1  

g  

t
h
S

t  

t

7

W
c
M  

f  

v

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ncreasing doubts about the reality of this distinction (Elbaz et al. 
018 ; Puglisi et al. 2021 ). 
Galaxies found within the ‘main sequence’ region are termed 

he ‘normal star-forming’ galaxies, whilst those abo v e the main 
equence regions are termed the ‘star -b urst’ galaxies (Elbaz et al.
011 ). The ‘normal star-forming’ mode is associated with the gas 
ccretion process (Dekel, Sari & Ceverino 2009 ; Dav ́e et al. 2010 )
hilst the ‘starburst’ mode is often associated with the major merger 

nteractions (Daddi et al. 2010 ; Genzel et al. 2010 ; Rodighiero et al.
011 ). 
In Fig. 9 , we show how the SFR of the cluster galaxy varies

ith the galaxy stellar mass. The stellar masses used in the work
ere obtained by fitting SED models described in Section 3.3 to the
bserved data. The majority of the SFGs and ISFGs are located on
r abo v e the best-fit to the MS relations derived from observations
n literature (black line; Speagle et al. 2014 ). It can also be seen in
ig. 9 that, our SFR detection limit of 46 M � yr −1 lies abo v e the MS
or all but the most massive galaxies in our sample. 

Coogan et al. ( 2018 ) showed that only 2/8 of the Cl J1449 + 0856
luster galaxies at z = 2 were found to reside abo v e the main
equence relation (see fig. 8 of that paper) although the majority 
f the cluster galaxies exhibited starburst-like characteristics. 
A similar trend was also observed by Smith et al. ( 2019 ) during

he study of CLJ1449 at z = 2. They found that most of the cluster
embers lay within the MS region (see fig. 10 of that paper) with

nly a few cluster galaxies located about a factor of 2 or 3 abo v e the
redicted MS region at the cluster redshift. They suggested that the 
tarburst activities within those cluster galaxies could be caused by 
 merger event and this goes to support the claim made by Coogan
t al. ( 2018 ) that the star-formation activity of the same cluster is
ostly merger-driven. 
‘Starb urst’ galaxies ha v e also been observ ed by Hung et al.

 2013 ) via visual morphological analysis of the 2084 COSMOS field
alaxies (Herschel-PACS and SPIRE observation) at z = 0–1.5 they 
bserved that about 50 per cent of the galaxies were undergoing a
erging process and these systems were seen to be deviating from

he main sequence relation supporting the claim that galaxies that 
re located abo v e the MS relation are mostly merging systems. The y
lso observed that some main sequence galaxies ( ≈18 per cent) were
lso undergoing merging events. Is it possible that the majority 
f J2215 cluster galaxies are undergoing a merger episode at this
igh redshift? This is because in Fig. 9 it can be observed that the
ajority of the cluster galaxies fall on or abo v e the main sequence

elation. 
Also, in a CO (2–1) emission line surv e y of the J2215 cluster

alaxies conducted by Hayashi et al. ( 2017 ) they showed that, two of
he cluster galaxies located at R ≈ 0.5 Mpc were undergoing a merger
ctivity (i.e. # ALMA.B3.15 and # ALMA.B3.16 – this is visible as
wo bulges in the intensity maps of the individual cluster galaxies
n Fig. 3 of 14 , the # ALMA.B3.16 counterpart is # 5442 in this
 ork). Again, Ik eda et al. ( 2022 ) classified six out of 17 CO emitters

ocated within ≈0.5 Mpc of the J2215 cluster centre as early-stage
ergers i.e. #ALMA.B3.06, #ALMA.B3.09, and #ALMA.B3.16 
hich correspond to IDs #5470, #5619, and #5442, respectively, 

n this work. Therefore J2215 may be undergoing a high rate of
 alaxy-to-g alaxy merger episode at this high redshift. 

.2 Integrated star formation rate 

e estimated a cluster-wide integrated star formation rate for 12/24 
alaxies selected to be star forming based on the three criteria
entioned in Section 5.4 . 
This resulted in 

∑ 

SFR = 1700 ± 330 M � yr −1 at � 0.8 Mpc
nd 1100 ± 210 at � 0.5 Mpc for 7/24 SFGs. Our measurement of
he integrated SFR is approximately two times higher than the value
eported by Ma et al. ( 2015 ) within 0.8 Mpc of the cluster centre i.e.

 

SFR value of 800 + 360 
−250 M � yr −1 but it is in agreement with the

alue reported by Stach et al. ( 2017 ) (i.e. 840 M � yr −1 within the
entral ≈0.5 Mpc of the J2215 cluster). 

The integrated SFR that we obtain for J2215 is of similar
agnitude to that seen in other clusters at similar redshift. Santos

t al. ( 2014 ) obtained an 
∑ 

SFR value of 780 ± 90 M � yr −1 for a
igh-redshift cluster at 1.62 (i.e. CLG0218.3 −0510) within 0.5 Mpc 
adius. Tran et al. ( 2010 ) also estimated an 

∑ 

SFR of ≈1370 M � yr −1 

ithin < 0.5 Mpc of the same cluster at that redshift emphasizing the
ossibility of an underestimation due to the low sensitivity of the 24-
m image, which implies that there may be a possibility of higher
tar formation activity ongoing in this high-redshift cluster that has 
ot been captured. 
The 0.5 Mpc central region of another high-redshift cluster 

LJ1449 at z = 2 was also seen to be actively forming stars with
n 
∑ 

SFR value of 470 ± 120 M � yr −1 (Smith et al. 2019 ). Similar
alues were also estimated within the core of this cluster by Strazzullo
t al. ( 2018 ) and Coogan et al. ( 2018 ). 

Cooke et al. ( 2019 ) estimated a median 
∑ 

SFR = 750 ±
90 M � yr −1 within 1 Mpc of the cluster centre of eight submillimetre
alaxy clusters at z ≈ 0.8–1.6. This value is also in agreement with
he residual field contamination corrected integrated SFR of eight 
igh-redshift ( z = 1.6–2.0) submillimeter galaxy clusters studied by 
mail ( 2024 ) i.e. a median 

∑ 

SFR = 530 ± 80 M � yr −1 . 
All these high-redshift galaxy cluster studies give an indication 

hat there is indeed an increase in the star formation activity within
he core of some high-redshift clusters at z = 1.5 and beyond. 

.3 Mass–normalized integrated SFR 

e determined the mass-normalized integrated SFR for all the 12/24 
luster members classified as non-AGNs, adopting a cluster mass 
 cl of 3 × 10 14 M � from Ma et al. ( 2015 ) and an integrated star-

ormation rate of 
∑ 

SFR of 1700 ± 330 M � yr −1 . We obtained a
alue of 

∑ 

(SFR)/ M cl = (570 ± 110) × 10 −14 yr −1 . 
MNRAS 532, 2842–2859 (2024) 



2854 D. Y. Klutse et al. 

M

Figure 10. The 
∑ 

(SFR)/ M cl − z relation for low- and high-redshift clusters. 
The solid line depicts the evolutionary model for the number density of star- 
forming ULIRGs radio sources out to z ≈ 1.5 that follows the (1 + z) 7 − z 

relation (Cowie et al. 2004 ; Geach et al. 2006 ). The dashed line represents 
the best-fitting model to the 

∑ 

(SFR)/ M cl of massive galaxy clusters studied 
in Smail ( 2024 ) (i.e. (1 + z) 5 . 5 − z; Webb et al. 2013 ). The dash–dotted 
line denotes the best-fitting model to the observed 

∑ 

(SFR)/ M −z (the 
integrated star formation rate per unit total halo mass) relation for the 
nine cluster sample detected by Popesso et al. ( 2012 ) including a blind 
extrapolation of the best-fitting model beyond the observed cluster redshifts 
(i.e. 0 . 15 < z < 0 . 85). All the power-law models have been normalized at z = 

0.8 and all measurements have been scaled to an equi v alent FIR luminosity 
limit of 10 12 L � where necessary to ensure a uniform comparison with other 
high-redshift clusters studied in Smail ( 2024 ). All three power-law models 
converge at the normalization redshift ( z = 0 . 8) ho we v er be yond that redshift 
the 

∑ 

(SFR)/ M cl increases sharply with high z for the Cowie et al. ( 2004 ) 
and Smail ( 2024 ) model with a relatively uniform pattern for the Popesso 
et al. ( 2012 ) model which deviates by a factor of ≈3 and 2 below the Cowie 
et al. ( 2004 ) and Smail ( 2024 ) model, respectively. The wider deviations from 

the two models at higher redshifts may be due to insufficient data at higher 
redshifts to better represent the best-fitting model beyond the redshift range 
of the nine cluster sample investigated by Popesso et al. ( 2012 ). 

 

f  

C  

a  

o  

2
 

r
z  

r  

l
 

o  

0
 

c  

m  

w  

l
F  

b  

p  

S  

i  

p  

s  

a  

e  

b  

m  

b  

i  

g  

(  

i  

1

8

W  

M  

r  

i  

t  

s  

q  

f  

1  

(
 

i  

m  

m
 

(  

2  

i  

(
 

r  

g  

a
 

c  

h  

e  

i  

h  

a  

A

A

W  

t  

A  

C  

c  

S  

a  

h  

S  

(  

o  

M  

t  

o  

o

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
Previous works have suggested that the critical epoch of star
ormation is at z ≈ 1.4; a study conducted via the IRACS Shallow
luster Surv e y rev ealed that abo v e z = 1 . 4 activ e star formation
ctivities are dominant within the cluster core and the reverse is
bserved below this redshift (e.g., Brodwin et al. 2013 ; Alberts et al.
014 ). 
This trend can be visualized in Fig. 10 where the 

∑ 

(SFR)/ M cl − z

elation for some high-redshift clusters nearly follows the (1 + z) 7 −
 relation. i.e. the evolutionary model for the number density for
adio star-forming galaxies whose luminosity values correspond to
ocal ULIRGs from z ≈ 0–1.5 proposed by Cowie et al. ( 2004 ). 

The black dash–dotted line denotes the best-fitting model to the
bserved 

∑ 

(SFR)/ M cl − z relation for the nine cluster sample ( z =
.15–0.85) studied by Popesso et al. ( 2012 ). 
Comparing our work with previous surv e ys of high-redshift

lusters one can see that J2215 generally follows the ‘higher
ass-normalized integrated SFR at higher redshift’ trend. When
e correct our measurements to a corresponding FIR luminosity

imit of ≥ 10 12 L � and normalize all the power-law models in 
ig. 10 at z = 0.8 following Smail ( 2024 ) the 

∑ 

(SFR)/ M cl value
ecomes a factor of ≈4.4, 1.7, and 3.5 higher than the value
redicted by the Popesso et al. ( 2012 ), Cowie et al. ( 2004 ) and
mail ( 2024 ) relation, respectively. The J2215 cluster at z = 1.46

s in relatively better agreement with the Cowie et al. ( 2004 )
ower law and also consistent with the submillimetre value of the
ame cluster studied by Cooke et al. ( 2019 ). Again, J2215 was
mong the two highest star forming clusters studied by Cooke
NRAS 532, 2842–2859 (2024) 
t al. ( 2019 ). They also suggested that the mild evidence of a
inomial velocity distribution of the cluster (Hilton et al. 2010 )
ay be due to cluster to cluster merger event and that event may

e responsible for the extremely high-star formation rate recorded
n their work. Should this claim be true then J2215 may be under-
oing a ‘twin’ merger event i.e. cluster galaxy to galaxy merger
see Section 7.1 ) and cluster to cluster merger event. This may
mply that J2215 is at the peak of its star-formation phase at z =
.46. 

 SUMMARY  

e have studied star formation and AGN activity within the XM-
XCS J2215.9 −1738 cluster at z = 1.46 using MeerKAT L -band

adio data. We combined the radio data with other archi v al optical and
nfrared data to study the star formation and AGN activities within
he cluster. We use for the first time in the study of the cluster three
election criteria; the radio luminosity, the far-infrared radio ratio
 IR and the mid-infrared colour, to classify sources as normal star-
orming, intermediate star-forming, and AGN galaxies. We classified
2/24 (50 per cent) as SFGs and 6/24 (25 per cent) as ISFGs and 6/24
25 per cent) as radio AGNs. 

We further investigated the evolution of star-forming galaxies
n clusters with redshifts. We achieved this by comparing the
ass-normalized integrated SFR of the MeerKAT-detected cluster
embers with other lower and higher redshift clusters. 
We observed that XMMXCS J2215.9 −1738 is consistent with the

1 + z) 7 − z relation proposed by (Cowie et al. 2004 ; Geach et al.
006 ) and these higher star formation activities occur within clusters
n their youthful ages (high redshift) compared to their older ages
low redshift). 

Finally, we also showed that most of the cluster members that are
adio sources detected by MeerKAT at this high redshift are late-type
alaxies. This also gives an indication that this cluster at z = 1.46 is
ctively star forming. 

This work can be extended to impro v e upon the morphological
lassification and the visual identification of merger events with
igher resolution observations like that of the JWST . We hope to also
mploy a more advanced technique to mitigate the blending effect
n our MeerKAT radio images whilst we anticipate the arri v al of the
igher resolution and more sensitive near future radio telescopes such
s the MeerKAT extension project (MK + ) and the Square Kilometer
rray. 

C K N OW L E D G E M E N T S  

e thank the referee for a number of useful suggestions that impro v ed
his work. DYK acknowledges financial support from the South
frican Radio Astronomy Observatory (SARAO) via the Human
apital Development (HCD) programme. MH acknowledges finan-
ial support from SARAO and the National Research Foundation of
outh Africa. IS and AMS acknowledge STFC (ST/X001075/1). We
cknowledge the use of the University of Kwazulu-Natal UKZNs’
igh-computing facility ( https:// www.acru.ukzn.ac.za/ ∼hippo/ ) and
outh Africa’s National Integrated Cyberinfrastructure System
NICIS) Centre for High performance computing facility (CHPC) for
ur data processing. The primary data used for this work was from the
eerKA T radio telescope. The MeerKA T telescope is operated by

he South African Radio Astronomy Observatory, which is a facility
f the National Research Foundation, an agency of the Department
f Science and Innovation. 

https://www.acru.ukzn.ac.za/~hippo/


MeerKAT observations of XCS J2215 2855 

D

T  

t
A
a

R

A
A
A
A
A
A
B  

B
B  

B
B
B  

B

B
B  

B
B
C
C
C
C
C
C
C
C  

C
C  

C  

D
D
D  

D
D
D
D
D
D
D
E
E
E
F
F
G
G
G
G
G
G

H  

H  

H
H
H
H
H
H
H
H
I  

I
I
I
K
K  

K
K
K
L  

L
L  

L
L
L  

L
M
M
M  

M
M  

 

M
M

M
N
N  

N
O
P
P  

P  

P  

P
P
P
P
P
R  

R

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
ATA  AVAILABILITY  

he MeerKAT continuum data used in this work can be found in
he MeerKAT data archive managed by the South African Radio 
stronomy Observ atory ( https://archi ve.sarao.ac.za/). The data are 

ssociated with proposal ID SCI-20190418-MH-01. 

EFERENCES  

lberts S. et al., 2014, MNRAS , 437, 437 
lberts S. et al., 2016, ApJ , 825, 72 
lgera H. S. B. et al., 2020, ApJ , 903, 138 
ndreon S. , Maughan B., Trinchieri G., Kurk J., 2009, A&A , 507, 147 
ppleton P. et al., 2004, ApJS , 154, 147 
sad K. M. B. et al., 2019, MNRAS, 502, 2970 
aldry I. K. , Glazebrook K., Brinkmann J., Ivezi ́c Ž., Lupton R. H., Nichol

R. C., Szalay A. S., 2004, ApJ , 600, 681 
allo L. , Heras F. J. H., Barcons X., Carrera F. J., 2012, A&A , 545, A66 
asu A. , Wadadekar Y., Beelen A., Singh V., Archana K. N., Sirothia S.,

Ishwara-Chandra C. H., 2015, ApJ , 803, 51 
ell E. F. , 2003, ApJ , 586, 794 
lanton M. R. , Moustakas J., 2009, ARA&A , 47, 159 
oquien M. , Burgarella D., Roehlly Y., Buat V., Ciesla L., Corre D., Inoue

A. K., Salas H., 2019, A&A , 622, A103 
radley L. et al., 2016, Astrophysics Source Code Library, 

record(ascl:1609.011) 
rammer G. B. , van Dokkum P. G., Coppi P., 2008, ApJ , 686, 1503 
rinchmann J. , Charlot S., White S. D. M., Tremonti C., Kauffmann G.,

Heckman T., Brinkmann J., 2004, MNRAS , 351, 1151 
rodwin M. et al., 2013, ApJ , 779, 138 
ruzual G. , Charlot S., 2003, MNRAS , 344, 1000 
alistro Rivera G. et al., 2017, MNRAS , 469, 3468 
habrier G. , 2003, PASP , 115, 763 
han J. C. C. et al., 2018, ApJ , 856, 8 
harlot S. , Fall S. M., 2000, ApJ , 539, 718 
luver M. E. et al., 2020, ApJ , 898, 20 
ondon J. J. , 1992, ARA&A , 30, 575 
ondon J. J. , Anderson M. L., Helou G., 1991, ApJ , 376, 95 
ondon J. J. , Cotton W . D., Greisen E. W ., Yin Q. F., Perley R. A., Taylor G.

B., Broderick J. J., 1998, AJ , 115, 1693 
oogan R. T. et al., 2018, MNRAS , 479, 703 
ooke E. A. , Smail I., Stach S. M., Swinbank A. M., Bower R. G., Chen

C.-C., Koyama Y., Thomson A. P., 2019, MNRAS , 486, 3047 
owie L. L. , Barger A. J., Fomalont E. B., Capak P., 2004, ApJ , 603,

L69 
addi E. et al., 2007, ApJ , 670, 156 
addi E. et al., 2010, ApJ , 714, L118 
av ́e R. , Finlator K., Oppenheimer B. D., Fardal M., Katz N., Kere ̌s D.,

Weinberg D. H., 2010, MNRAS , 404, 1355 
awson K. S. et al., 2009, AJ , 138, 1271 
ekel A. , Sari R., Ceverino D., 2009, ApJ , 703, 785 
el Moro A. et al., 2013, A&A , 549, A59 
elvecchio I. et al., 2021, A&A , 647, A123 
igby-North J. A. et al., 2010, MNRAS , 407, 846 
raine B. T. et al., 2014, ApJ , 780, 172 
ressler A. et al., 1997, ApJ , 490, 577 
lbaz D. et al., 2007, A&A , 468, 33 
lbaz D. et al., 2011, A&A , 533, A119 
lbaz D. et al., 2018, A&A , 616, A110 
assbender R. et al., 2011, A&A , 527, L10 
azio G. G. et al., 2004, ApJS , 154, 10 
alametz A. et al., 2009, ApJ , 694, 1309 
arrett M. A. , 2002, A&A , 384, L19 
each J. E. et al., 2006, ApJ , 649, 661 
enzel R. et al., 2010, MNRAS , 407, 2091 
 ́omez P. L. et al., 2003, ApJ, 584, 210 
u Y. , Fang G., Yuan Q., Cai Z., Wang T., 2018, ApJ , 855, 10 
ayashi M. , K odama T., K oyama Y., Tanaka I., Shimasaku K., Okamura S.,
2010, MNRAS , 402, 1980 

ayashi M. , Kodama T., Koyama Y., Tadaki K.-I., Tanaka I., 2011, MNRAS ,
415, 2670 

ayashi M. et al., 2014, MNRAS , 439, 2571 
ayashi M. et al., 2017, ApJ , 841, L21 
ayashi M. et al., 2018, ApJ , 856, 118 
elou G. , Soifer B. T., Rowan-Robinson M., 1985, ApJ , 298, L7 
eywood I. , 2020, Astrophysics Source Code Library, record(ascl: 2009.003) 
ilton M. et al., 2009, ApJ , 697, 436 
ilton M. et al., 2010, ApJ , 718, 133 
ung C.-L. et al., 2013, ApJ , 778, 129 

chikawa T. et al., 2006, in McLean I. S., Iye M., eds, Proc. SPIE Conf. Ser.,
Vol. 6269, Ground-based and Airborne Instrumentation for Astronomy, 
SPIE, Bellingham, p. 626916 

keda R. et al., 2022, ApJ , 933, 11 
noue A. K. , 2011, MNRAS , 415, 2920 
vison R. J. et al., 2010, A&A , 518, L31 
arim A. et al., 2011, ApJ , 730, 61 
en yon J. S. , Smirno v O. M., Grobler T. L., Perkins S. J., 2018, MNRAS ,

478, 2399 
odama T. , Bower R. G., 2001, MNRAS , 321, 18 
ravtsov A. V. , Borgani S., 2012, ARA&A , 50, 353 
rishnan C. et al., 2017, MNRAS , 470, 2170 
acerda E. A. D. , S ́anchez S. F., Cid Fernandes R., L ́opez-Cob ́a C., Espinosa-

Ponce C., Galbany L., 2020, MNRAS , 492, 3073 
ehmer B. D. et al., 2009, ApJ , 691, 687 
epore M. , Bongiorno A., Tozzi P., Travascio A., Zappacosta L., Merlin E.,

Fassbender R., 2022, A&A , 668, A123 
ewis I. et al., 2002, MNRAS , 334, 673 
isenfeld U. , V ̈olk H. J., 2010, A&A , 524, A27 
isenfeld U. , P ́erez-Torres M., Alberdi A., Colina L., Garcıa-Burillo S.,

Alonso-Herrero A., 2015, The Spanish Square Kilometre Array White 
Book, Sociedad Espa ̃ nola de Astronom ́ıa, p. 101 

otz J. M. et al., 2013, ApJ , 773, 154 
a C. J. et al., 2015, ApJ , 806, 257 
agnelli B. et al., 2015, A&A , 573, A45 
aier C. , Hayashi M., Ziegler B. L., Kodama T., 2019, A&A , 626,

A14 
artini P. et al., 2013, ApJ , 768, 1 
cMullin J. P. , Waters B., Schiebel D., Young W., Golap K., 2007, in Shaw

R. A., Hill F., Bell D. J.eds, ASP Conf. Ser.Vol. 376, Astronomical Data
Analysis Software and Systems XVI. Astron. Soc. Pac., San Francisco, 
p. 127 

iller L. , Peacock J. A., Mead A. R. G., 1990, MNRAS, 244, 207 
ohan N. , Rafferty D., 2015, Astrophysics Source Code Library, 

record(ascl:1502.007) 
urphy E. J. , 2009, ApJ , 706, 482 
andra K. et al., 2007, ApJ , 660, L11 
ewman A. B. , Ellis R. S., Andreon S., Treu T., Raichoor A., Trinchieri G.,

2014, ApJ , 788, 51 
oeske K. G. et al., 2007, ApJ , 660, L43 
ffringa A. R. et al., 2014, MNRAS , 444, 606 
annella M. et al., 2009, ApJ , 698, L116 
 artridge B. , L ́opez-Canie go M., Perle y R. A., Stev ens J., Butler B. J., Rocha

G., Walter B., Zacchei A., 2016, ApJ , 821, 61 
atil P. , Nyland K., Lacy M., Farrah D., Afonso J., Barkhouse W., Surace J.,

2019, ApJ , 871, 109 
earson W. J. , Wang L., Brough S., Holwerda B. W., Hopkins A. M., Lo v eday

J., 2021, A&A , 646, A151 
olletta M. et al., 2007, ApJ , 663, 81 
opesso P. et al., 2012, A&A , 537, A58 
ostman M. et al., 2005, ApJ , 623, 721 
ovi ́c M. et al., 2012, A&A , 541, A118 
uglisi A. et al., 2021, MNRAS , 508, 5217 
adcliffe J. F. , Barthel P. D., Garrett M. A., Beswick R. J., Thomson A. P.,

Muxlow T. W. B., 2021, A&A , 649, L9 
odighiero G. et al., 2011, ApJ , 739, L40 
MNRAS 532, 2842–2859 (2024) 

https://archive.sarao.ac.za/
http://dx.doi.org/10.1093/mnras/stt1897
http://dx.doi.org/10.3847/0004-637X/825/1/72
http://dx.doi.org/10.3847/1538-4357/abb77b
http://dx.doi.org/10.1051/0004-6361/200912299
http://dx.doi.org/10.1086/422425
http://dx.doi.org/10.1086/380092
http://dx.doi.org/10.1051/0004-6361/201117464
http://dx.doi.org/10.1088/0004-637X/803/2/51
http://dx.doi.org/10.1086/367829
http://dx.doi.org/10.1146/annurev-astro-082708-101734
http://dx.doi.org/10.1051/0004-6361/201834156
http://dx.doi.org/10.1086/591786
http://dx.doi.org/10.1111/j.1365-2966.2004.07881.x
http://dx.doi.org/10.1088/0004-637X/779/2/138
http://dx.doi.org/10.1046/j.1365-8711.2003.06897.x
http://dx.doi.org/10.1093/mnras/stx1040
http://dx.doi.org/10.1086/376392
http://dx.doi.org/10.3847/1538-4357/aaadb4
http://dx.doi.org/10.1086/309250
http://dx.doi.org/10.3847/1538-4357/ab9cb8
http://dx.doi.org/10.1146/annurev.aa.30.090192.003043
http://dx.doi.org/10.1086/170258
http://dx.doi.org/10.1086/300337
http://dx.doi.org/10.1093/mnras/sty1446
http://dx.doi.org/10.1093/mnras/stz955
http://dx.doi.org/10.1086/383198
http://dx.doi.org/10.1086/521818
http://dx.doi.org/10.1088/2041-8205/714/1/L118
http://dx.doi.org/10.1111/j.1365-2966.2010.16395.x
http://dx.doi.org/10.1088/0004-6256/138/5/1271
http://dx.doi.org/10.1088/0004-637X/703/1/785
http://dx.doi.org/10.1051/0004-6361/201219880
http://dx.doi.org/10.1051/0004-6361/202039647
http://dx.doi.org/10.1111/j.1365-2966.2010.16977.x
http://dx.doi.org/10.1088/0004-637X/780/2/172
http://dx.doi.org/10.1086/304890
http://dx.doi.org/10.1051/0004-6361:20077525
http://dx.doi.org/10.1051/0004-6361/201117239
http://dx.doi.org/10.1051/0004-6361/201732370
http://dx.doi.org/10.1051/0004-6361/201016169
http://dx.doi.org/10.1086/422843
http://dx.doi.org/10.1088/0004-637X/694/2/1309
http://dx.doi.org/10.1051/0004-6361:20020169
http://dx.doi.org/10.1086/506469
http://dx.doi.org/10.1111/j.1365-2966.2010.16969.x
http://dx.doi.org/10.3847/1538-4357/aaad0b
http://dx.doi.org/10.1111/j.1365-2966.2009.16026.x
http://dx.doi.org/10.1111/j.1365-2966.2011.18892.x
http://dx.doi.org/10.1093/mnras/stu115
http://dx.doi.org/10.3847/2041-8213/aa71ad
http://dx.doi.org/10.3847/1538-4357/aab3e7
http://dx.doi.org/10.1086/184556
http://dx.doi.org/10.1088/0004-637X/697/1/436
http://dx.doi.org/10.1088/0004-637X/718/1/133
http://dx.doi.org/10.1088/0004-637X/778/2/129
http://dx.doi.org/10.3847/1538-4357/ac6cdc
http://dx.doi.org/10.1111/j.1365-2966.2011.18906.x
http://dx.doi.org/10.1051/0004-6361/201014552
http://dx.doi.org/10.1088/0004-637X/730/2/61
http://dx.doi.org/10.1093/mnras/sty1221
http://dx.doi.org/10.1046/j.1365-8711.2001.03981.x
http://dx.doi.org/10.1146/annurev-astro-081811-125502
http://dx.doi.org/10.1093/mnras/stx1315
http://dx.doi.org/10.1093/mnras/staa008
http://dx.doi.org/10.1088/0004-637X/691/1/687
http://dx.doi.org/10.1051/0004-6361/202244078
http://dx.doi.org/10.1046/j.1365-8711.2002.05558.x
http://dx.doi.org/10.1051/0004-6361/201015083
http://dx.doi.org/10.1088/0004-637X/773/2/154
http://dx.doi.org/10.1088/0004-637X/806/2/257
http://dx.doi.org/10.1051/0004-6361/201424937
http://dx.doi.org/10.1051/0004-6361/201935522
http://dx.doi.org/10.1088/0004-637X/768/1/1
http://dx.doi.org/10.1088/0004-637X/706/1/482
http://dx.doi.org/10.1086/517918
http://dx.doi.org/10.1088/0004-637X/788/1/51
http://dx.doi.org/10.1086/517926
http://dx.doi.org/10.1093/mnras/stu1368
http://dx.doi.org/10.1088/0004-637X/698/2/L116
http://dx.doi.org/10.3847/0004-637X/821/1/61
http://dx.doi.org/10.3847/1538-4357/aaf7a4
http://dx.doi.org/10.1051/0004-6361/202039177
http://dx.doi.org/10.1086/518113
http://dx.doi.org/10.1051/0004-6361/201117973
http://dx.doi.org/10.1086/428881
http://dx.doi.org/10.1051/0004-6361/201117314
http://dx.doi.org/10.1093/mnras/stab2914
http://dx.doi.org/10.1051/0004-6361/202140791
http://dx.doi.org/10.1088/2041-8205/739/2/L40


2856 D. Y. Klutse et al. 

M

S
S
S
S
S
S
S  

S
S
S
S
S  

S  

S  

S  

S  

S
S
S
S
T
T
T
T
W
W
W
Y
Y

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024
antos J. S. et al., 2014, MNRAS , 438, 2565 
antos J. S. et al., 2015, MNRAS , 447, L65 
argent M. T. et al., 2010a, ApJS , 186, 341 
argent M. T. et al., 2010b, ApJ , 714, L190 
chawinski K. et al., 2010, ApJ , 711, 284 
chreiber C. et al., 2015, A&A , 575, A74 
eymour N. , Huynh M., Dwelly T., Symeonidis M., Hopkins A., McHardy

I., Page M., Rieke G., 2009, MNRAS , 398, 1573 
hen L. et al., 2022, ApJ , 935, 177 
mail I. , 2024, MNRAS, 529, 2290 
mail I. et al., 2014, ApJ , 782, 19 
mirnov O. M. , Tasse C., 2015, MNRAS , 449, 2668 
mith G. P. , Treu T., Ellis R. S., Moran S. M., Dressler A., 2005, ApJ , 620,

78 
mith C. M. A. , Gear W. K., Smith M. W. L., Papageorgiou A., Eales S. A.,

2019, MNRAS , 486, 4304 
peagle J. S. , Steinhardt C. L., Capak P. L., Silverman J. D., 2014, ApJS ,

214, 15 
NRAS 532, 2842–2859 (2024) 
tach S. M. , Swinbank A. M., Smail I., M. Hilton, Simpson J. M., Cooke E.
A., 2017, ApJ , 849, 154 

talevski M. , Ricci C., Ueda Y., Lira P., Fritz J., Baes M., 2016, MNRAS ,
458, 2288 

tanford S. A. et al., 2006, ApJ , 646, L13 
tern D. et al., 2005, ApJ , 631, 163 
trazzullo V. et al., 2018, ApJ , 862, 64 
winbank A. M. et al., 2014, MNRAS , 438, 1267 
adaki K.-i. et al., 2012, MNRAS , 423, 2617 
asse C. , 2014, preprint ( arXiv:1410.8706 ) 
asse C. et al., 2018, A&A , 611, A87 
ran K.-V. H. et al., 2010, ApJ , 719, L126 
ang T. et al., 2016, ApJ , 828, 56 
ang T. et al., 2017, A&A , 601, A63 
ebb T. M. A. et al., 2013, AJ , 146, 84 
ang G. et al., 2022, ApJ , 927, 192 
un M. S. , Reddy N. A., Condon J. J., 2001, ApJ , 554, 803 

http://dx.doi.org/10.1093/mnras/stt2376
http://dx.doi.org/10.1093/mnrasl/slu180
http://dx.doi.org/10.1088/0067-0049/186/2/341
http://dx.doi.org/10.1088/2041-8205/714/2/L190
http://dx.doi.org/10.1088/0004-637X/711/1/284
http://dx.doi.org/10.1051/0004-6361/201425017
http://dx.doi.org/10.1111/j.1365-2966.2009.15224.x
http://dx.doi.org/10.3847/1538-4357/ac81c5
http://dx.doi.org/10.1088/0004-637X/782/1/19
http://dx.doi.org/10.1093/mnras/stv418
http://dx.doi.org/10.1086/426930
http://dx.doi.org/10.1093/mnras/stz1090
http://dx.doi.org/10.1088/0067-0049/214/2/15
http://dx.doi.org/10.3847/1538-4357/aa93f6
http://dx.doi.org/10.1093/mnras/stw444
http://dx.doi.org/10.1086/506449
http://dx.doi.org/10.1086/432523
http://dx.doi.org/10.3847/1538-4357/aacd10
http://dx.doi.org/10.1093/mnras/stt2273
http://dx.doi.org/10.1111/j.1365-2966.2012.21063.x
http://arxiv.org/abs/1410.8706
http://dx.doi.org/10.1051/0004-6361/201731474
http://dx.doi.org/10.1088/2041-8205/719/2/L126
http://dx.doi.org/10.3847/0004-637X/828/1/56
http://dx.doi.org/10.1051/0004-6361/201526645
http://dx.doi.org/10.1088/0004-6256/146/4/84
http://dx.doi.org/10.3847/1538-4357/ac4971
http://dx.doi.org/10.1086/323145


MeerKAT observations of XCS J2215 2857 

A ON  PLOTS  

I  some selected MeerKAT sources that are J2215 cluster members. The 
b black curve. The stellar attenuation and unattenuated models are shown 
a dust, and A GN emission models are also shown with the green, red, and 
o  star-forming galaxies is shown as the brown curve. The red dots are the 
C served fluxes, whilst the green triangles represent the upper limit values 
f nfrared bands. We assumed the following upper limits: A 4 sigma upper 
l y – 85 μm), 5 sigma upper limit for the 24 μm Spitzer MIPS (0.070 mJy) 
a e of ≈23 mag (Ch1 and Ch2) and ≈20 mag (Ch3 and Ch4). The 5 sigma 
u  mag. J band image is ≈24.4 whilst that of the K s – band is ≈24.5. 

D
ow

nloaded from
PPENDIX  A :  SPECTRAL  E N E R G Y  DISTRI BU TI

n Figs A1 , A2 we show the spectral energy distribution plots for
est-fitting model for each of the cluster galaxies is shown as the 
s the yellow and the dashed blue curv e, respectiv ely. The nebular , 
range curv es, respectiv ely. The non-radio synchrotron emission for
IGALE -generated model fluxes, the purple open circles show the ob
or the radio sources that were not detected in either the optical or i
imit for the SCUBA-2 sources (i.e. 21.6 mJy – 450 μm and 2.52 mJ
nd for the IRAC channels we assumed a magnitude upper limit valu
pper limit of the HST z 850 band is ≈ 26 mag and the i 775 band ≈25.1
MNRAS 532, 2842–2859 (2024) 

Figure A1. Spectral energy distributions of member galaxies of J2215. See Appendix A for details. 

 https://academ
ic.oup.com

/m
nras/article/532/2/2842/7705620 by W

itw
atersrand H

ealth Sciences Library user on 06 Septem
ber 2024



2858 D. Y. Klutse et al. 

MNRAS 532, 2842–2859 (2024) 

Figure A2. Spectral energy distributions of member galaxies of J2215. See Appendix A for details. 

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

atersrand H
ealth Sciences Library user on 06 Septem

ber 2024



MeerKAT observations of XCS J2215 2859 

A

pcigale.ini’ configuration file. This ‘pcigale.ini’ file was run to generate several 

Values 

10, 16, 27, 46, 77, 129, 215, 359, 599, 1000 Myr 

1000, 1291, 1668, 2154, 2782, 3593, 4641, 5994, 7742, 10 000 Myr 

5.0, 50.0, 500 Myr 

0.0, 0.01, 0.05, 0.1, 0.25 

1 (Chabrier) 

0.02 

0.014 

0.2 

0.8 

m 0.15, 0.45, 0.60, 0.90, 1.20, 1.5, 1.8, 2.1, 2.4, 2.8, 0.50, 0.75 

unger stars, i.e. μ 0.25, 0.50, 0.75, 0.44 

0.47, 1.12, 2.5 

0.10, 0.15, 0.20, 1.0 

2.0, 3.0 

7 

GN axis 30, 70 

0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.8, 0.9 

2.58 

0.8 

d the default v alues set by CIGALE V2022.0 

T

©
P
(

D
ow

nloaded from
 https://academ

ic.oup.com
/m

nras/article/532/2/2842/7705620 by W
itw

ater
PPENDIX  B:  CIGALE M O D E L  PA R A M E T E R S  

Table B1. This table shows the model parameters and values set in the ‘
models to fit our observed SED. 

Module Parameter 

sfhdelayed e-folding time of the burst of star formation 

Stellar age 

Age of the last starburst 

Mass fraction of late starburst population 

Simple stellar population (bc03) Initial mass function (imf) 

Bruzual & Charlot ( 2003 ) Metallicity 

Nebular emission Gas metallicity 

Fraction of escaped photons 

Fraction of absorbed photons 

Dust attenuation V -band attenuation in the interstellar mediu

dustattmodifiedCF00 (modified 

Charlot & Fall 2000 attenuation law) 

Attenuation ratio between older stars and yo

= A 

ISM 
v /A 

BC 
v + A 

ISM 
v 

Dust emissions Mass fraction of PAH 

Minimum radiation field 

Power-law slope 

AGN model Average edge-on optical depth at 9.7 μm 

skirtor2016 Inclination, i.e. viewing angle w.r.t. to the A

AGN fraction 

Galaxy radio synchrotron emission FIR/radio correlation coefficient 

spectral index a 

Note . F or parameters and their corresponding v alues that were not listed here, we use
a For parameters relating to star-forming galaxies. 

his paper has been typeset from a T E 

X/L 

A T E 

X file prepared by the author. 
MNRAS 532, 2842–2859 (2024) 

2024 The Author(s). 
ublished 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 
 https://cr eativecommons.or g/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 

srand H
ealth Sciences Library user on 06 Septem

ber 2024

https://creativecommons.org/licenses/by/4.0/

	1 INTRODUCTION
	2 OBSERVATION AND DATA REDUCTION
	3 ANALYSIS
	4 MORPHOLOGIES AND OPTICAL COLOURS OF RADIO SOURCES IN J2215
	5 CLASSIFICATION OF RADIO SOURCES
	6 AGN ACTIVITY WITHIN J2215
	7 STAR FORMATION ACTIVITY WITHIN J2215
	8 SUMMARY
	ACKNOWLEDGEMENTS
	DATA AVAILABILITY
	REFERENCES
	APPENDIX A: SPECTRAL ENERGY DISTRIBUTION PLOTS
	APPENDIX B: cigale MODEL PARAMETERS