Thomas Cornish
Gather.town id
CSF14
Poster Title
Exploring the environments of SMGs: a wide-field narrowband study
Institution
Lancaster University
Abstract (short summary)
Submillimetre galaxies (SMGs) are thought to play a significant role in the evolution of the cosmic star formation rate density, contributing as much as ∼20–30% at z ∼ 2–3. It has been hypothesised that these extreme star-forming systems are the progenitors of local early-type cluster galaxies. If true, this would imply that SMGs should reside in galaxy cluster progenitors at high redshift. Whilst there are well-known examples of SMGs residing in protoclusters, these systems were selected for follow-up because of their high galaxy or SMG density. To explore the environments of SMGs in an unbiased way we have undertaken a narrowband VLT/HAWK-I study of H-alpha and [OIII] emitters around three ALMA-identified and spectroscopically-confirmed SMGs at z ∼ 2.3 and z ∼ 3.3, which were selected with no prior knowledge of their environments. On average, these SMGs reside in environments which are ∼2–4x overdense compared to the field. Our results suggest that SMGs do tend to reside in protocluster-like environments, supporting the claim that they likely evolve into the passive early-type galaxies observed in local clusters.
Plain text (extended) Summary
Submillimeter galaxies (SMGs) are massive galaxies that are extremely bright at sub-mm wavelengths. They are typically at z ~ 2–4 and have star formation rates of ~1000 M☉/yr.
SMG redshifts, masses, and star formation histories are consistent with them being the elusive progenitors of local massive early-type cluster galaxies. SMGs are thus expected to reside in protocluster environments, but this hasn’t yet been properly tested.
Whilst previous studies have found individual SMGs in protoclusters, these were selected for follow-up due to being in overdensities and thus such studies can’t tell us about the overall SMG population (e.g. Casey et al., 2015; Oteo et al., 2018). Statistical clustering measurements are promising, but are challenging to interpret due to heavy reliance on photometric redshifts and/or single-dish sub-mm surveys (e.g. Hickox et al., 2012; Wilkinson et al., 2017).
Instead, we use HAWK-I/VLT wide-field narrowband data to investigate the environments of three SMGs, selected based only on redshifts and with no prior knowledge of their environments. Two of these SMGs are located at z ~ 2.3, while the third is at z ~ 3.3.
Each pointing was observed through a broadband filter and a narrowband filter to identify H𝛼 and [OIII] emission at z ~ 2.3 and z ~ 3.3 via their colour excess. Figure 1 shows an example
of a galaxy we identify as an SMG companion using this method: it appears bright in the narrowband image but dim in the broadband image, such that it can still be easily seen in the residual image (obtained by subtracting the broadband image from the narrowband image).
Photometric redshifts were then used to distinguish H𝛼 and [OIII] emitters from low-redshift interlopers. Figure 2 shows the redshifts of our candidate line emitters, plotted as a function of Σ, which describes the significance of a source’s narrowband excess based on the photometric errors (see e.g. Bunker et al., 1995). 39 and 17 H𝛼 emitters were identified near the two z ~ 2.3 SMGs, while 7 [OIII] emitters were identified near the z ~ 3.3 SMG.
Figure 3 shows the positions of our candidate H𝛼 and [OIII] emitters on the sky, along with the target SMGs. For reference, Figure 4 (adapted from Muldrew et al., 2015) depicts how the HAWK-I field of view compares to the scale of a typical simulated protocluster at z = 2, showing that our HAWK-I data would only cover a small - but still significant - fraction of a protocluster.
Figure 5 shows luminosity functions for our H𝛼 and [OIII] candidates, and compares them to analogous luminosity functions obtained from blank-field studies (Sobral et al., 2013; Khostovan et al., 2015). The SMG fields appear significantly (>~5⨉) denser than the blank fields, consistent with the hypothesis that SMGs are an early stage in the evolution of local early-type galaxies. Completeness corrections will likely enhance the effect.
In conclusion, we have conducted a wide-field narrowband study in search of star-forming galaxies around three SMGs, with no prior knowledge of their environments. Our results imply overdensities of star-forming galaxies around the SMGs. This suggests that SMGs do typically reside in protoclusters, and are thus consistent with being the progenitors of local massive early-type cluster galaxies.
SMG redshifts, masses, and star formation histories are consistent with them being the elusive progenitors of local massive early-type cluster galaxies. SMGs are thus expected to reside in protocluster environments, but this hasn’t yet been properly tested.
Whilst previous studies have found individual SMGs in protoclusters, these were selected for follow-up due to being in overdensities and thus such studies can’t tell us about the overall SMG population (e.g. Casey et al., 2015; Oteo et al., 2018). Statistical clustering measurements are promising, but are challenging to interpret due to heavy reliance on photometric redshifts and/or single-dish sub-mm surveys (e.g. Hickox et al., 2012; Wilkinson et al., 2017).
Instead, we use HAWK-I/VLT wide-field narrowband data to investigate the environments of three SMGs, selected based only on redshifts and with no prior knowledge of their environments. Two of these SMGs are located at z ~ 2.3, while the third is at z ~ 3.3.
Each pointing was observed through a broadband filter and a narrowband filter to identify H𝛼 and [OIII] emission at z ~ 2.3 and z ~ 3.3 via their colour excess. Figure 1 shows an example
of a galaxy we identify as an SMG companion using this method: it appears bright in the narrowband image but dim in the broadband image, such that it can still be easily seen in the residual image (obtained by subtracting the broadband image from the narrowband image).
Photometric redshifts were then used to distinguish H𝛼 and [OIII] emitters from low-redshift interlopers. Figure 2 shows the redshifts of our candidate line emitters, plotted as a function of Σ, which describes the significance of a source’s narrowband excess based on the photometric errors (see e.g. Bunker et al., 1995). 39 and 17 H𝛼 emitters were identified near the two z ~ 2.3 SMGs, while 7 [OIII] emitters were identified near the z ~ 3.3 SMG.
Figure 3 shows the positions of our candidate H𝛼 and [OIII] emitters on the sky, along with the target SMGs. For reference, Figure 4 (adapted from Muldrew et al., 2015) depicts how the HAWK-I field of view compares to the scale of a typical simulated protocluster at z = 2, showing that our HAWK-I data would only cover a small - but still significant - fraction of a protocluster.
Figure 5 shows luminosity functions for our H𝛼 and [OIII] candidates, and compares them to analogous luminosity functions obtained from blank-field studies (Sobral et al., 2013; Khostovan et al., 2015). The SMG fields appear significantly (>~5⨉) denser than the blank fields, consistent with the hypothesis that SMGs are an early stage in the evolution of local early-type galaxies. Completeness corrections will likely enhance the effect.
In conclusion, we have conducted a wide-field narrowband study in search of star-forming galaxies around three SMGs, with no prior knowledge of their environments. Our results imply overdensities of star-forming galaxies around the SMGs. This suggests that SMGs do typically reside in protoclusters, and are thus consistent with being the progenitors of local massive early-type cluster galaxies.
URL
Email: t.cornish@lancaster.ac.uk; Twitter: @ThomasCornish5
Poster file