Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
N. ?lvarez Crespo, V. Smol?i?, A. Finoguenov, L. Barrufet, M. Aravena
AAstronomy & Astrophysics manuscript no. main © ESO 2021January 11, 2021
Environments of a sample of AzTEC submillimetre galaxies in theCOSMOS field
N. Álvarez Crespo , , V. Smol˘ci´c , A. Finoguenov , L. Barrufet and M. Aravena Department of Physics, University of Zagreb, Bijeni˘cka cesta 32, 10000 Zagreb, Croatia European Space Agency (ESA), European Space Astronomy Centre (ESAC), Camino Bajo del Castillo s / n, 28692 Villanueva dela Cañada, Madrid, Spaine-mail: [email protected] Department of Physics, University of Helsinki, Gustaf Hällströmin katu 2, FI-00014 Helsinki, Finland Geneva Observatory, University of Geneva, Ch. des Mail-lettes 51, 1290 Versoix, Switzerland Núcleo de Astronomía de la Facultad de Ingeniería y Ciencias, Universidad Diego Portales, Av. Ejército Libertador 441, Santiago,ChileJanuary 11, 2021
ABSTRACT
Aims.
Submillimetre galaxies (SMGs) are bright sources at submillimetre wavelengths ( F µ m > z > Methods.
We present an environmental study of a sample of 116 SMGs in 96 ALMA observation fields, which were initially dis-covered with the AzTEC camera on ASTE and identified with high-resolution 1.25 mm ALMA imaging within the COSMOS surveyfield, having either spectroscopic or unambiguous photometric redshift. We analysed their environments making use of the latestrelease of the COSMOS photometric catalogue, COSMOS2015, a catalogue that contains precise photometric redshifts for more thanhalf a million objects over the 2deg COSMOS field. We searched for dense galaxy environments computing the so-called overdensityparameter as a function of distance within a radius of 5’ from the SMG. We validated this approach spectroscopically for those SMGsfor which spectroscopic redshift is available. As an additional test, we searched for extended X-ray emission as a proxy for the hotintracluster medium, performing an X-ray stacking analysis in the 0.5 – 2 keV band with a 32" aperture and our SMG position usingall available
XMM–Newton and
Chandra
X-ray observations of the COSMOS field.
Results.
We find that 27% (31 out of 116) of the SMGs in our sample are located in a galactic dense environment; this fractionthat is similar to previous studies. The spectroscopic redshift is known for 15 of these 31 sources, thus this photometric approach istested using spectroscopy. We are able to confirm that 7 out of 15 SMGs lie in high-density peaks. However, the search for associatedextended X-ray emission via an X-ray stacking analysis leads to a detection that is not statistically significant.
Key words. galaxies: clusters: general – galaxies: evolution – galaxies: formation – large-scale structure – submillimetre: galaxies
1. Introduction
Submillimetre galaxies (SMGs) are amongst the most lumi-nous dusty galaxies in the Universe (Wilkinson et al. 2017),which emit most of their energy at submillimetre (sub-mm)wavelengths, from 200 µ m to about 1mm (Geach et al. 2017).They were initially discovered in extragalactic submillimetresurveys using the Submillimetre Common-User Bolometer Ar-ray (SCUBA; Holland et al. 1999) (Barger et al. 1998; Hugheset al. 1998). These galaxies are highly star forming, reachingstar formation rates (SFRs) up to thousands of M (cid:12) yr − (see e.g.Casey et al. 2014; Rowan-Robinson et al. 2018). The bulk ofthis population has a redshift distribution that ranges between z ∼ z ∼ M (cid:12) and sizes of around 1 Mpc (see e.g. Bahcall 1996). Their pro-genitors are proto-clusters, which are structures found at red-shifts 2 < z < ff orts have beenmade to search, identify, and characterise them (see e.g. Ouchiet al. 2005; Galametz et al. 2010; Toshikawa et al. 2012). The de-tection of proto-clusters is important to understand hierarchicalstructure formation and stellar mass growth in galaxies at earlytimes.Most likely, SMGs are the progenitors of massive ellipti-cal galaxies observed in the local universe (see e.g. Lilly et al.1999; Fu et al. 2013; Toft et al. 2014). Since early-type galaxiesare predominantly found in clusters, it is important to addressthe question of whether SMGs are preferably found in regionswith enhanced galaxy density. Studying the clustering properties Article number, page 1 of 12 a r X i v : . [ a s t r o - ph . GA ] J a n & A proofs: manuscript no. main of SMGs can also provide constraints on their nature in a cos-mological context. Some models depict SMGs as a long-livedepisode of star formation in the most massive galaxies, driven bythe early fast collapse of the dark matter halo (Xia et al. 2012),yielding strong clustering for these sources. On the other hand,other models in which SMGs are short-lived bursts in less mas-sive galaxies, predict weaker clustering (Almeida et al. 2011).So far, most of the studies on the environments of SMGs havebeen done in individual sources or small samples (Ivison et al.2000; Scott et al. 2002; Aravena et al. 2010; Oteo et al. 2018; Hillet al. 2020), which has not really allowed us to test the abundanceof structures around these galaxies and has forbidden clusteringmeasurements.The Cosmic Evolution Survey (COSMOS; Scoville et al.2007) includes multiwavelength imaging and spectroscopy fromX-ray to radio wavelengths over an area of 1.4 × ffi cient depth to provide a comprehensive view of galaxy for-mation and large-scale structure. In this work we aim to improvethe clustering measurements of SMGs by extending the currentstatistical sample. We evaluate overdensities of a sample of 116SMGs in the COSMOS field that were initially discovered withthe AzTEC camera on the Atacama Submillimeter TelescopeExperiment (ASTE; Aretxaga et al. 2011), and subsequentlyidentified with high-resolution 1.25 mm ALMA imaging in 96di ff erent fields (Brisbin et al. 2017).This paper is organised as follows. In Sect. 2 we introduceour SMG sample and the COSMOS catalogues used for the en-vironmental study. In Sect. 3 we describe the methodology usedto measure the overdensities and their significance and false de-tection rate. In Sect. 4 we present the results of our analysis.Then in Sect. 5 we verify the overdensities using spectroscopicredshift. Later in Sect. 6 we discuss our results comparing themto what previously found in the literature, and finally in Sect. 7we present our conclusions.Unless otherwise stated, we assume a flat Λ CDM cosmol-ogy with a Hubble constant H =
73 km s − Mpc − , total matterdensity Ω m = Ω Λ =
2. Data
The sources studied in this work are the so-called "strict" sub-sample from Brisbin et al. (2017). These 116 SMGs have beendetected using high spatial resolution ( ∼ / ASTEsources with 1.1 mm flux densities ≥ z = ± / C1a, C2a, C3a, C5, C6a, C6b and C17), and the restwere taken from the COSMOS spectroscopic redshift catalogue(available internally for members of the COSMOS collabora-tion). For those sources with photometric redshift, they weremeasured by Brisbin et al. (2017) cross correlating the ALMApositions with the latest release of the COSMOS photometriccatalogue, COSMOS2015. The complete list of SMGs is given in Table A.1 of Bris-bin et al. (2017). Throughout this paper we use the nomencla-ture given by these authors; for alternative names, see the sec-ond column of their Table A.1. Of these 96 di ff erent ALMA ob-servations, 19 are multiple component observation fields, mean-ing there are several resolved SMGs inside each field. The 39SMGs belonging to multi-component systems are described byan alphabetical tag in descendant order of brightness for each ob-serving field. Some of these sources are physically related, whileothers belong to the same ALMA observation only as a result ofchance alignment. This is explored in more detail in Sect. 6.1. To study the environments of SMGs, we made use of the lat-est version of the COSMOS photometric catalogue (COS-MOS2015 hereafter; Laigle et al. 2016). This catalogue in-cludes photometric measurements from the ultraviolet (UV) tothe infrared (IR) wavelengths, including 6 broad optical bands( B, V, g, r, i, z ++ ), 12 medium bands, and 2 narrow bands, aswell as Y, J H and K S data from the UltraVISTA Data Release 2new HyperSuprime-Cam Subaru Y band and new SPLASH 3.6and 4.5 µ m Spitzer / Infrared Array Camera (IRAC) data (Sanderset al. 2007; Capak et al. 2007; McCracken et al. 2012; Ilbert et al.2013).Furthermore, we used a catalogue of spectroscopic redshiftsin the COSMOS field available internally for members of theCOSMOS collaboration. It is composed of 36,274 spectroscopicredshifts, both available only internally to the COSMOS collab-oration and from the following surveys:1. The zCOSMOS-bright survey contributed 8,608 galaxies at0.1 ≤ z ≤ ≤ z ≤ of the COSMOS field (Lilly et al. inprep.).3. The 6,617 galaxies with high- quality spectra from theDEIMOS 10K Spectroscopic Survey Catalog of the COS-MOS Field, which is a survey that samples a broad redshiftdistribution in up to z = < z < of the COSMOS field at 2 < z < < z < / FORS-2 instrument (Comparat et al. 2015).7. The spectroscopic survey of galaxies in the COSMOS fieldusing the Fiber Multi-object Spectrograph (FMOS), a near-IR instrument on the Subaru Telescope at 1.34 ≤ z ≤ ≤ z ≤ For more information we refer to the COSMOS webpage http://cosmos.astro.caltech.edu .Article number, page 2 of 12. Álvarez Crespo et al.: Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
9. A sample of 26 galaxies at 0.82 ≤ z ≤ ≤ z ≤ ≤ z ≤ ≤ z ≤ ffi ciently used to search for overdensities andespecially at high redshifts ( z > ∆ z / (1 + z spec ), where ∆ z is the di ff erence betweenthe spectroscopic and photometric redshifts. They find the stan-dard deviation of the distribution to be σ ∆ z / (1 + z spec ) = z phot ≤ σ ∆ z / (1 + z spec ) = z phot >
3. Methodology
To search for overdensities in the SMG fields, we use the galax-ies in the COSMOS2015 photometric catalogue that lie withinthe redshift range z phot = z S MG ± σ ∆ z / (1 + z S MG ) (1 + z S MG ) fromeach SMG, being σ ∆ z / (1 + z spec ) = z phot ≤ σ ∆ z / (1 + z spec ) = z phot > r = δ g ( r ) = Σ r ( r ) − Σ bg Σ bg = Σ r ( r ) Σ bg − , (1)where Σ r is the local galaxy surface density calculated as Σ r = N r / A r , where N r is the number of galaxies within the given ra-dius and redshift bin in a search window area of A r = π × r .Correspondingly, Σ bg is the background galaxy surface densityand is defined as Σ bg = N bg / A bg . The quantity N bg is the numberof galaxies satisfying the photometric redshift interval within theentire area A bg of the COSMOS field, to take masked areas dueto saturated stars and / or corrupted data into account. Conform-ing to this definition, δ g ( r ) > δ g ( r ) < The probability of observing ≥ N r objects when the expectednumber is n r = Σ bg × A r is analytically defined by the Pois-son distribution p ( ≥ N r , n r ) = − Σ N r i = ( e − n r n ir / i !). We calculatethe significance of the overdensity parameter by computing the Poisson probability for each di ff erent radius and the value δ g ( r )is considered as robust when p ≤ z S MG byrandomly shifting their positions over the inner 1 deg of theCOSMOS field. For each one of these mock catalogues that cor-respond to a certain SMG at z S MG , we generate other 1,000 mockcatalogues by randomly distributing in the sky the number ofgalaxies at z phot = z S MG ± σ ∆ z / (1 + z S MG ) (1 + z S MG ) in the COS-MOS2015 photometric catalogue, leading to 10,000 mock cat-alogues for each SMG. The false detection probability is givenby the fraction of events in which N r are found within a radius r over these 10,000 mock catalogues.
4. Analysis and results
The results from our overdensity analysis are represented in thefigures in the Appendix. For each SMG we show the overden-sity parameter δ g as function of the projected radius r measuredfrom the central SMG, from 0.1’ up to 5’. For each point we in-dicate the number of sources N r within the given radius, includ-ing the SMG. When the value of δ g has a Poisson probability p ( ≥ N r , n r ) ≤ ≥ N r objects defined analytically by a Poisson distribu-tion within a radius r is p ≤ P FD (calculated numerically using mockcatalogues at random SMG positions) of finding a number ofsources ≥ N r within a radius r is ≤
5% (see Sect. 3.2).To evaluate the significance of δ g , we start by the smallestconsidered radius 0’.5. If p ( ≥ N r , n r ) and P FD >
5% for thatradius, we continue to evaluate δ g at the following values ofthe radius up to r = ∼ ff erence at a 95% con-fidence level. The physical distances corresponding to the radiusof the overdensities as reported in the second column of Table 1vary from 232 pkpc at the z min = z mean = z max = X-ray extended emission is an evidence of hot intergalacticmedium in clusters of galaxies. Hence, we search for extendedX-ray emission around the SMGs positions. The entire COS-MOS region has been mapped through 54 overlapping
XMM-Newton pointings and additional
Chandra observations (Elviset al. 2009; Civano et al. 2016) and previous e ff orts have beenmade to search for X-ray emitting galaxy clusters in this field(Hasinger et al. 2007; Finoguenov et al. 2007; George et al.2011).In this work we first cross matched the positions of ourSMGs with the revised catalogue of extended X-ray sourcesin the COSMOS field, which contains 247 X-ray groups with M c = × - 3 × M (cid:12) at a redshift range 0.08 ≤ z < Article number, page 3 of 12 & A proofs: manuscript no. main
Table 1.
SMGs found in an overdensity.SMG r N r Poisson probability False detectionname [’] p ( ≥ N r , n r ) probability P FD AzTEC / C5 ∗ / C6a ∗ / C6b ∗ / C9a 0.5 4 0.115 0.006AzTEC / C9b ∗ / C9c ∗ / C17 ∗ / C25 ∗ / C28a ∗ / C28b 0.5 4 0.112 0.005AzTEC / C33a 0.5 3 0.112 0.032AzTEC / C34a 0.5 4 0.130 0.014AzTEC / C43b 5.0 148 0.034 0.063AzTEC / C45 ∗ / C48b 0.5 4 0.142 0.021AzTEC / C50 0.5 4 0.091 0.002AzTEC / C51b 0.5 4 0.154 0.033AzTEC / C52 ∗ / C55b 0.5 5 0.120 0.006AzTEC / C59 ∗ / C60b 0.5 2 0.029 0.008AzTEC / C61 ∗ / C65 ∗ / C71b ∗ / C79 0.5 3 0.117 0.039AzTEC / C99 0.5 3 0.115 0.029AzTEC / C100a 2.5 48 0.040 0.093AzTEC / C101a 5.0 160 0.049 0.095AzTEC / C117 0.5 5 0.142 0.023AzTEC / C118 ∗ / C122a 1.0 12 0.023 0.067Column description: (1): SMG name according to Brisbin et al. (2017)nomenclature; (2): radius of δ g statistically significant; (3): number ofsources within r ; (4) Poisson probability p ( ≥ N r , n r ); (5) false detectionprobability P FD . ∗ SMGs with spectroscopic redshift.
Chandra and
XMM-Newton mosaic image in the 0.5 - 2 keVband (Gozaliasl et al. 2019). This catalogue has a total clusterflux depth of 3 × − erg cm − s − , covers an area of 2.1 deg , Fig. 1.
In grey (dashed) the redshift distribution of the SMGs found in asignificant overdensity and in orange (solid) those that are not found ina high-density environment. There is no significant di ff erence betweentheir redshift distribution at a 95% confidence level according to the K-Stest. and the precision of the centres for extended sources goes downto 5". None of the positions of our SMGs match any of the X-raygroups reported in the catalogue.However, since only 11 of 116 SMGs from our sample are atthe same redshift as the aforementioned X-ray group catalogue,we then extracted the stacked X-ray flux outside of the alreadydetected groups in the same Chandra and
XMM-Newton mosaicimage in the 0.5 - 2 keV band. If the SMG is at the centre ofa cluster, it should be located near to the centre of a virialiseddark matter halo. The size of this halo at the typical redshift ofSMGs is a few hundreds of pkpcs, so we used a 32" aperture sizecentred in the SMG, following Finoguenov et al. (2009, 2015).Prior to this we subtracted background and point-like sourcessuch X-ray jets and cluster cores. We detected an average flux of1.7 ± × − erg s − cm − , which is a marginal detection sowe ruled out the possibility of significant X-ray extended emis-sion for our sample. Using the mean redshift of the sample of2.3, the halo mas (M200) that corresponds to the marginal X-rayflux detection is (2 . ± . × M (cid:12) . We note that this mass isconsistent with the typical SMG mass inferred in the clusteringstudies.
5. Spectroscopic verification of overdensitycandidates
As a secondary test, we now evaluate the overdensity parame-ter δ g for those 15 SMGs having spectroscopic redshift foundin an overdensity (see Table 1 in Sect. 4 ) using spectroscopicredshift catalogues, to confirm what we previously found usingphotometry. For each SMG with a spectroscopic redshift in Ta-ble 1, we calculated the overdensity parameter using the sourcesfrom the spectroscopic catalogue described in Sect. 2, within aradius of r = z spec = z S MG ± σ ∆ z / (1 + z S MG ) (1 + z S MG ), being σ ∆ z / (1 + z spec ) = z ≤ σ ∆ z / (1 + z spec ) = z > Σ bg is calculated by randomly locating 1,000SMGs in the non-uniform footprint of the COSMOS field cov-ered by the spectroscopic catalogue, as reported in Sect. 2, with arandom redshift in the interval 0 < z < . + σ ∆ z / (1 + z S MG ) (1 + . z ≤ z > . − σ ∆ z / (1 + z S MG ) (1 + . < z < δ g along with its as-sociated uncertainty is calculated using a truncated Gaussian tothe distribution of the number of galaxies generated in the 1,000simulations at z ≤ z > / z ≤ δ g ≥ σ , confirming the photometric identified overdensi-ties for AzTEC / C5, AzTEC / C6a, AzTEC / C6b, AzTEC / C17,AzTEC / C52, AzTEC / C59, and AzTEC / C118. We note that thecompilation of the spectroscopic catalogue is strongly redshiftdependent, where ∼
99% of the observed sources lie at z ≤ ff erent catalogues, thespectral coverage is not completely uniform. These e ff ects limitthe conclusions that can be extracted from this analysis and canonly be used as a confirmation for those cases showing positiveresults. Article number, page 4 of 12. Álvarez Crespo et al.: Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
6. Discussion
The sample in this paper contains 116 di ff erent SMGs, observedin 96 di ff erent ALMA fields, meaning 19 out of 96 observa-tions contain more than one SMG. These are the so-called multi-component systems and 39 of the 116 SMGs are one of them.In the following we unravel which SMGs belonging to multi-component systems are gravitationally bounded, and those forwhich this is simply due to chance alignment.The SMGs AzTEC / C6a and AzTEC / C6b are separated by ∆ z = ff erentiates betweenchance and physical associations of ALMA sources. However,both SMGs reside in a high density environment, which pointstowards a physical association.Similarly, although each SMG of the triplet in the fieldAzTEC / C9 lies within 13” of its closest neighbour, a bit higherthan the threshold, all of these are found within an overdensityso they are very likely physically associated. The redshifts forboth AzTEC / C28a and AzTEC / C28b are compatible with beingidentical as is their separation; they are physically associated andare found in an overdensity.Although redshifts in AzTEC / C43a and AzTEC / C43b areconsistent with being identical, we find only one SMG in anoverdensity, that is AzTEC / C43b. This could happen if SMG isnot located at the centre of the overdensity, so only one of thecomponents would be found using our method. The same occursfor the field AzTEC / C48, while AzTEC / C48a and AzTEC / C48bare physically related, only AzTEC / C48b is found in a dense en-vironment.Although photometric redshifts for AzTEC / C55a andAzTEC / C55b are compatible with being identical ( z AzTEC / C a = + . − . and z AzTEC / C b = + . − . ), considering their phys-ical separation (17.2", 41 kpc at the measured redshift) itis very unlikely that they are physically associated and onlyAzTEC / C55b is located in an overdensity. AzTEC / C34a andAzTEC / C34b are not physically related since their redshiftsdi ff er strongly ( z AzTEC / C a = + . − . and z AzTEC / C b = + . − . ), their association is only due to chance alignmentand we only find AzTEC / C34a in an overdensity. AzTEC / C60aand AzTEC / C60b are not physically associated ( z AzTEC / C a = + . − . and z AzTEC / C b = + . − . ) and only AzTEC / C60b isfound in an overdensity. The same situation is found for the fieldAzTEC / C100; only AzTEC / C100a is found in a dense environ-ment since AzTEC / C100a and AzTEC / C100b ( z AzTEC / C a = + . − . and z AzTEC / C b = + . − . ) results from a chancealignment. Early attempts to measure the environment surrounding SMGsinvolved their projected two-dimensional distribution of the skyusing projected two-dimensional angular correlation function(ACF). The results for these early e ff orts are ambiguous, mostlybecause of the lack of redshifts to trace the three-dimensionalstructure (see Scott et al. 2002; Borys et al. 2003; Almaini et al.2003; Borys et al. 2004; Blain et al. 2004; Weiß et al. 2009;Hickox et al. 2012). Williams et al. (2011) analysed severalSMGs in the COSMOS field, assuming various redshift distri-butions to estimate their de-projected ACF. These authors couldonly set upper limits to the correlation length. Nonetheless, ACFstudies have the important limitation that they are only able to measure the average clustering properties of a population, miss-ing the individual di ff erences between its components.We do not find a significant redshift di ff erence between thoseSMGs lying in density peaks and those located in environmentsthat are indistinguishable from field galaxies. The correlation be-tween redshift and clustering in SMGs is yet unclear. For in-stance, studying the clustering of galaxies selected in the IRACbands, Farrah et al. (2006) did not find any strong redshift evolu-tion in their sample. Wilkinson et al. (2017) however find that onaverage, SMGs at z > B zK galaxies in the COSBO field, the in-ner 20’ ×
20’ region of the COSMOS field (Bertoldi et al. 2007).Two of the sources in their sample, AzTEC / C6a and AzTEC / C7,also appear in ours. We find AzTEC / C6a in a overdensity too,while for AzTEC / C7 the overdensity is not significant in bothstudies. Additionally, we find a similar percentage of SMGs inoverdensities, since they find 30% of their sample in overdensi-ties and we find ∼
27% and similar physical sizes, although oursample spans a higher redshift interval. AzTEC / C6 belongs to aproto-cluster, it has been previously found within an X-ray emit-ting region with 17 spectroscopically confirmed member galax-ies (Casey et al. 2015; Wang et al. 2016; Cucciati et al. 2018).Smolˇci´c et al. (2017) performed an analysis using a simi-lar method over a smaller sample of SMGs in the COSMOSfield, and ten of our sources overlap (it is important to note theyuse a di ff erent nomenclature; the list of corresponding namesis given in Table A.1 of Brisbin et al. (2017). Nonetheless adirect comparison can be performed only for the "small over-densities" part of their study, where the physical distances anal-ysed are similar to those of this study. Considering only theirkpc-scale study, the fraction of SMGs found in a significantoverdensity (i.e. ∼ / C3a, AzTEC / C14, AzTEC / C22a, and AzTEC / C22b(AzTEC2, AzTEC9, AzTEC11S, and AzTEC11N, respectivelyin Smolˇci´c et al. (2017)) are not found in an overdensityin both studies. However, for AzTEC / C2a and AzTEC / C18(AzTEC8, AzTEC12, respectively in their nomenclature), nei-ther of us find an overdensity at small scales, but they findhigh galactic density when looking at Mpc scales. The SMGsAzTEC / C5, AzTEC / C6a, and AzTEC / C17 (AzTEC1, Cosbo 3,and J1000 + ff erence we encounter is with the source AzTEC / C42(AzTEC5), which they find in a significantly dense environmentwhile we do not. This could be because the redshift value theyuse is z phot = . + . . (Smolˇci´c et al. 2012), while a more up-to-date value is available for our study z phot = . + . . (Smolˇci´cet al. 2017).
7. Summary and conclusions
We explore the clustering properties of a sample of 116 SMGsdescribed as the "strict" sample in Brisbin et al. (2017), drawnfrom a S / N-limited sample initially discovered with the AzTECcamera on ASTE, and identified with high-resolution 1.25 mmALMA imaging. Their redshifts lie within the interval 0.829 < z < δ g in an interval ∆ z andin steps of r = σ ∆ z / (1 + z spec ) = z phot ≤ Article number, page 5 of 12 & A proofs: manuscript no. main and σ ∆ z / (1 + z spec ) = z phot > ∼ / ASTE sources selected within the COSMOS sur-vey field are located in high-density environments, suggest-ing that a fraction of the SMGs are linked to formation ofstructures.(b) For those SMGs found lying in overdensities with spectro-scopic redshifts (15 out of 31), the photometric approach istested using spectroscopically verified overdensities, whichare able to confirm 7 of these SMGs high-density peaks.However, because of the lack of completeness of the spec-troscopic catalogue used for this analysis, this approach canonly be used as a lower limit.(c) We search for extended X-ray emission around SMGs viamatching the positions of our SMGs to those of the revisedcatalogue of extended X-ray sources in the COSMOS field,which contains combined
XMM-Newton and
Chandra datain the 0.5 - 2 keV band, with negative results. Moreover, weperform an X-ray stacking analysis in the 0.5 - 2 keV bandusing a 32" aperture size, but the average flux found is 1.7 ± × − erg s − cm − ; this contribution is smaller than1 σ so not statistically significant.This is consistent with previous results that show a similarfraction of SMGs in high-density environments at these angularscales (Aravena et al. 2010; Smolˇci´c et al. 2017). About one-third of our sources are related to the formation of structuresat high redshift since they are located in regions with enhancedgalaxy density, a fraction that is similar to previous studies (Ar-avena et al. 2010; Smolˇci´c et al. 2017). We do not appreciate acorrelation between redshift and clustering strength.The reason why we do not find a higher fraction of SMGs as-sociated with strong galaxy overdensities could be due to biasesthat a ff ect our data. It is a known e ff ect that using photomet-ric redshifts leads to a weaker value of δ g (Chiang et al. 2013),therefore our results might be biased towards the most promi-nent overdensities and we could be missing the not-so-dense en-vironments. To overcome all the caveats present in this work andanalyse the environments of SMGs in a robust way, further theo-retical and observational e ff orts are needed, such as a dedicatedspectroscopic campaign of the galaxies in the area surroundingthe SMGs such as that performed for AzTEC / C6a.
Acknowledgements.
We thank the anonymous referee for her / his insightful com-ments that led to improvements in the paper. This research was funded by the Eu-ropean Union’s Seventh Framework programme under grant agreement 337595(ERC Starting Grant, "CoSMass"). NAC and LB are supported by EuropeanSpace Agency (ESA) Research Fellowships. We thank B. C. Lemaux and G.Zamorani for helpful discussion and comments that led to improvements in thispaper. We gratefully acknowledge the contributions of the entire COSMOS col-laboration consisting of more than 100 scientists. More information on the COS-MOS survey is available at . Wethank the VUDS team for making the data in the COSMOS field available priorto public release. References
Almaini, O., Scott, S. E., Dunlop, J. S., et al. 2003, MNRAS, 338, 303Almeida, C., Baugh, C. M., & Lacey, C. G. 2011, MNRAS, 417, 2057Aravena, M., Bertoldi, F., Carilli, C., et al. 2010, ApJ, 708, L36Aretxaga, I., Wilson, G. W., Aguilar, E., et al. 2011, MNRAS, 415, 3831Bahcall, N. A. 1996, ArXiv Astrophysics e-prints [ astro-ph/9611148 ]Barger, A. J., Cowie, L. L., Sanders, D. B., et al. 1998, Nature, 394, 248Bertoldi, F., Carilli, C., Aravena, M., et al. 2007, ApJS, 172, 132Blain, A. W., Chapman, S. C., Smail, I., & Ivison, R. 2004, ApJ, 611, 725 Blain, A. W., Smail, I., Ivison, R. J., Kneib, J.-P., & Frayer, D. T. 2002,Phys. Rep., 369, 111Borys, C., Chapman, S., Halpern, M., & Scott, D. 2003, MNRAS, 344, 385Borys, C., Scott, D., Chapman, S., et al. 2004, MNRAS, 355, 485Brisbin, D., Miettinen, O., Aravena, M., et al. 2017, A&A, 608, A15Capak, P., Aussel, H., Ajiki, M., et al. 2007, ApJS, 172, 99Casey, C. M., Cooray, A., Capak, P., et al. 2015, ApJ, 808, L33Casey, C. M., Narayanan, D., & Cooray, A. 2014, Phys. Rep., 541, 45Chiang, Y.-K., Overzier, R., & Gebhardt, K. 2013, ApJ, 779, 127Civano, F., Marchesi, S., Comastri, A., et al. 2016, ApJ, 819, 62Comparat, J., Richard, J., Kneib, J.-P., et al. 2015, A&A, 575, A40Cucciati, O., Lemaux, B. C., Zamorani, G., et al. 2018, A&A, 619, A49Elvis, M., Civano, F., Vignali, C., et al. 2009, ApJS, 184, 158Farrah, D., Lonsdale, C. J., Borys, C., et al. 2006, ApJ, 641, L17Fasano, G. & Franceschini, A. 1987, MNRAS, 225, 155Finoguenov, A., Connelly, J. L., Parker, L. C., et al. 2009, ApJ, 704, 564Finoguenov, A., Guzzo, L., Hasinger, G., et al. 2007, ApJS, 172, 182Finoguenov, A., Tanaka, M., Cooper, M., et al. 2015, A&A, 576, A130Fu, H., Cooray, A., Feruglio, C., et al. 2013, Nature, 498, 338Galametz, A., Vernet, J., De Breuck, C., et al. 2010, A&A, 522, A58Geach, J. E., Dunlop, J. S., Halpern, M., et al. 2017, MNRAS, 465, 1789George, M. R., Leauthaud, A., Bundy, K., et al. 2011, ApJ, 742, 125Gozaliasl, G., Finoguenov, A., Tanaka, M., et al. 2019, MNRAS, 483, 3545Hasinger, G., Capak, P., Salvato, M., et al. 2018, ApJ, 858, 77Hasinger, G., Cappelluti, N., Brunner, H., et al. 2007, ApJS, 172, 29Hayward, C. C., Behroozi, P. S., Somerville, R. S., et al. 2013, MNRAS, 434,2572Hickox, R. C., Wardlow, J. L., Smail, I., et al. 2012, MNRAS, 421, 284Hill, R., Chapman, S., Scott, D., et al. 2020, MNRASHolland, W. S., Robson, E. I., Gear, W. K., et al. 1999, MNRAS, 303, 659Hughes, D. H., Serjeant, S., Dunlop, J., et al. 1998, Nature, 394, 241Ilbert, O., McCracken, H. J., Le Fèvre, O., et al. 2013, A&A, 556, A55Ivison, R. J., Dunlop, J. S., Smail, I., et al. 2000, ApJ, 542, 27Kartaltepe, J. S., Sanders, D. B., Le Floc’h, E., et al. 2010, ApJ, 721, 98Kriek, M., Shapley, A. E., Reddy, N. A., et al. 2015, ApJS, 218, 15Krogager, J.-K., Zirm, A. W., Toft, S., Man, A., & Brammer, G. 2014, ApJ, 797,17Laigle, C., McCracken, H. J., Ilbert, O., et al. 2016, ApJS, 224, 24Larson, D., Dunkley, J., Hinshaw, G., et al. 2011, ApJS, 192, 16Le Fèvre, O., Tasca, L. A. M., Cassata, P., et al. 2015, A&A, 576, A79Lilly, S. J., Eales, S. A., Gear, W. K. P., et al. 1999, ApJ, 518, 641Lilly, S. J., Le Brun, V., Maier, C., et al. 2009, ApJS, 184, 218Lilly, S. J., Le Fèvre, O., Renzini, A., et al. 2007, ApJS, 172, 70McCracken, H. J., Milvang-Jensen, B., Dunlop, J., et al. 2012, A&A, 544, A156Oke, J. B. 1974, ApJS, 27, 21Onodera, M., Renzini, A., Carollo, M., et al. 2012, ApJ, 755, 26Oteo, I., Ivison, R. J., Dunne, L., et al. 2018, ApJ, 856, 72Ouchi, M., Shimasaku, K., Akiyama, M., et al. 2005, ApJ, 620, L1Overzier, R. A., Guo, Q., Kau ff mann, G., et al. 2009, MNRAS, 394, 577Peacock, J. A. 1983, MNRAS, 202, 615Roseboom, I. G., Bunker, A., Sumiyoshi, M., et al. 2012, MNRAS, 426, 1782Rowan-Robinson, M., Wang, L., Farrah, D., et al. 2018, A&A, 619, A169Sanders, D. B., Salvato, M., Aussel, H., et al. 2007, ApJS, 172, 86Scott, S. E., Fox, M. J., Dunlop, J. S., et al. 2002, MNRAS, 331, 817Scoville, N., Aussel, H., Brusa, M., et al. 2007, ApJS, 172, 1Silverman, J. D., Kashino, D., Sanders, D., et al. 2015, ApJS, 220, 12Smolˇci´c, V., Miettinen, O., Tomiˇci´c, N., et al. 2017, A&A, 597, A4Smolˇci´c, V., Aravena, M., Navarrete, F., et al. 2012, A&A, 548, A4Spergel, D. N., Bean, R., Doré, O., et al. 2007, ApJS, 170, 377Springel, V., White, S. D. M., Jenkins, A., et al. 2005, Nature, 435, 629Stach, S. M., Dudzeviˇci¯ut˙e, U., Smail, I., et al. 2019, MNRAS, 487, 4648Strandet, M. L., Weiss, A., Vieira, J. D., et al. 2016, ApJ, 822, 80Tasca, L. A. M., Le Fèvre, O., Ribeiro, B., et al. 2017, A&A, 600, A110Toft, S., Smolˇci´c, V., Magnelli, B., et al. 2014, ApJ, 782, 68Toshikawa, J., Kashikawa, N., Ota, K., et al. 2012, ApJ, 750, 137Wang, T., Elbaz, D., Daddi, E., et al. 2016, ApJ, 828, 56Weiß, A., Kovács, A., Coppin, K., et al. 2009, ApJ, 707, 1201Wilkinson, A., Almaini, O., Chen, C.-C., et al. 2017, MNRAS, 464, 1380Williams, C. C., Giavalisco, M., Porciani, C., et al. 2011, ApJ, 733, 92Xia, J.-Q., Negrello, M., Lapi, A., et al. 2012, MNRAS, 422, 1324Zabl, J., Nørgaard-Nielsen, H. U., Fynbo, J. P. U., et al. 2015, MNRAS, 451,2050 Article number, page 6 of 12. Álvarez Crespo et al.: Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
Appendix A: Figures
Fig. .1.
Overdensity parameter δ g vs. radius r measured from the central SMG. Open dots represent when the only source measured within thatradius is the target SMG, and black dots represent when at least one more source than the central SMG is observed. Error bars represent Poissonianerrors. The dot is enclosed in a square when Poisson probability is p ≤ r is indicated next to each point. A horizontal line represents δ g =
0. Article number, page 7 of 12 & A proofs: manuscript no. main
Fig. .1. continued.Article number, page 8 of 12. Álvarez Crespo et al.: Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
Fig. .1. continued. Article number, page 9 of 12 & A proofs: manuscript no. main
Fig. .1. continued.Article number, page 10 of 12. Álvarez Crespo et al.: Environments of a sample of AzTEC submillimetre galaxies in the COSMOS field
Fig. .1. continued. Article number, page 11 of 12 & A proofs: manuscript no. main