Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.604327
Title: Measuring the evolution of the cosmic star formation rate using narrow-band selection
Author: Drake , Alyssa Bryony
Awarding Body: Liverpool John Moores University
Current Institution: Liverpool John Moores University
Date of Award: 2013
Availability of Full Text:
Access through EThOS:
Abstract:
We have used the narrow-band technique to detect a sample of objects with bright emission in the narrow-band using 6 narrow-band filters on The Subaru Telescope and VISTA. We assess objects' narrow-band excess to ultimately compile a catalogue of 5725 emission-line galaxies in 12 rcdshift slices, spanning 10 Gyr of cosmic time. We determine photometric redshifts for the sample using II-band photometry, and use a spectroscopicalIy confirmed subset to fine tune the resultant redshift distribution. We use the maximum-likelihood technique to determine luminosity functions in each redshift slice and model the selection effects inherent in any narrow-band selection statistically. to negate the retrospective corrections ordinarily required. The deep narrow band data are sensitive to very low star formation rates (SFRs), and allow an accurate evaluation of the faint end slope of the Schechter function, a. We find that Q; is particularly sensitive to the assumed faintest broad band magnitude of a galaxy capable of hosting an emission line, and propose that this limit should be empirically motivated. For this analysis we base our threshold on the limiting observed equivalent widths of emission lines in the local Universe. We compute the characteristic SFR of galaxies in each redshift slice, and the integrated SFR density, PSFR. We find our results to be in good agreement with the literature and parametrise the evolution of the SFR density as PSFR ex: (1 + Z)4.58 confirming a steep decline in star formation activity since z ,...... 1.6. We next determine stellar masses for galaxies in the 4 highest redshift slices of the deep Subaru data at z = 0.63, z = 0.83, z = 1.19 and z = 1.46. We apply our maximum likelihood method to determine luminosity functions in a series of mass bins at each redshift, and determine values of PSFR as a function of mass. We use two different prescriptions for extinction as a function of stellar mass to determine dust extinction at Ha , and use the Cardelli Clayton and Mathis (1989) reddening law to extrapoloate values for each at [0 Ill] and [0 Ill. We calculate metallicities for objects in each mass bin, and correct the PsFR accordingly where necessary. We highlight the importance of these corrections for the shape of the overall P SFR as a function of mass, and its normalisation with redshift. We find that the dominant contribution to the PsFR after correction comes from galaxies in the mass range 1010.5 to 1011 .0 Mo in the highest redshift bin. We find that the normalisation of the PSFR decreases with decreasing redshift, however it is unclear whether the peak of the function moves towards lower masses .. Nevertheless we find that galaxies in the lower mass bins, reaching as low as 101.5 Me , play an important part in the contribution to the overall P SFR at all redshifts studied in this work. Lastly we consider the restframe colours of all objects detected in these red shift slices, and deduce that we are detecting two populations of star forming galaxies, following separate SFR-mass relations. We identify the second (higher mass) locus of objects in the highest redshift slice as red sequence galaxies still undergoing star formation. A handful of such objects are also seen in the lower redshift slices . The red objects found at z = 1.46 exhibit specific star formation rates sufficiently high to have assembled their stellar mass at a constant rate since the beginning of the Universe, however by z = 0.83 the same objects display sSFRs too low to have formed their stellar population at the current rate in the time available. These objects therefore must have been forming stars at a higher rate in the past, and so we infer that we are witnessing the "switch off" of star formation in massive galaxies between redshifts z = 1.46 and z = 0.83. ALYSSA We have used the narrow-band technique to detect a sample of objects with bright emission in the narrow-band using 6 narrow-band filters on The Subaru Telescope and VISTA. We assess objects' narrow-band excess to ultimately compile a catalogue of 5725 emission-line galaxies in 12 rcdshift slices, spanning 10 Gyr of cosmic time. We determine photometric redshifts for the sample using lI-band photometry, and use a spectroscopicalIy confirmed subset to fine tune the resultant redshift distribution. We use the maximum-likelihood technique to determine luminosity functions in each redshift slice and model the selection effects inherent in any narrow-band selection statistically. to negate the retrospective corrections ordinarily required. The deep narrow band data are sensitive to very low star formation rates (SFRs), and allow an accurate evaluation of the faint end slope of the Schechter function, a. We find that Q; is particularly sensitive to the assumed faintest broad band magnitude of a galaxy capable of hosting an emission line, and propose that this limit should be empirically motivated. For this analysis we base our threshold on the limiting observed equivalent widths of emission lines in the local Universe. We compute the characteristic SFR of galaxies in each redshift slice, and the integrated SFR density, PSFR. We find our results to be in good agreement with the literature and parametrise the evolution of the SFR density as PSFR ex: (1 + Z)4.58 confirming a steep decline in star formation activity since z ,...... 1.6. We next determine stellar masses for galaxies in the 4 highest redshift slices of the deep Subaru data at z = 0.63, z = 0.83, z = 1.19 and z = 1.46. We apply our maximum likelihood method to determine luminosity functions in a series of mass bins at each redshift, and determine values of PSFR as a function of mass. We use two different prescriptions for extinction as a function of stellar mass to determine dust extinction at Ha , and use the Cardelli Clayton and Mathis (1989) reddening law to extrapoloate values for each at [0 Ill] and [0 Ill. We calculate metallicities for objects in each mass bin, and correct the PsFR accordingly where necessary. We highlight the importance of these corrections for the shape of the overall P SFR as a function of mass, and its normalisation with redshift. We find that the dominant contribution to the PsFR after correction comes from galaxies in the mass range 1010.5 to 1011 .0 Mo in the highest redshift bin. We find that the normalisation of the PSFR decreases with decreasing redshift, however it is unclear whether the peak of the function moves towards lower masses .. Nevertheless we find that galaxies in the lower mass bins, reaching as low as 101.5 Me , play an important part in the contribution to the overall P SFR at all redshifts studied in this work. Lastly we consider the restframe colours of all objects detected in these red shift slices, and deduce that we are detecting two populations of star forming galaxies, following separate SFR-mass relations. We identify the second (higher mass) locus of objects in the highest redshift slice as red sequence galaxies still undergoing star formation. A handful of such objects are also seen in the lower redshift slices . The red objects found at z = 1.46 exhibit specific star formation rates sufficiently high to have assembled their stellar mass at a constant rate since the beginning of the Universe, however by z = 0.83 the same objects display sSFRs too low to have formed their stellar population at the current rate in the time available. These objects therefore must have been forming stars at a higher rate in the past, and so we infer that we are witnessing the "switch off" of star formation in massive galaxies between redshifts z = 1.46 and z = 0.83.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.604327  DOI: Not available
Share: