Department of Physics and Astronomy at UC Riverside

Bahram Mobasher
Professor of Physics and Observational Astronomy
Ph.D. 1988, University of Durham, Durham, U.K.

Observational Astronomy

E-mail: mobasher@ucr.edu
Phone: (951) 827-7190
Fax: (951) 827-4529

My main research interest is focused on the study of formation and evolution of galaxies using multi-waveband galaxy surveys obtained with 8-10 meter ground-based telescopes (Gemini, VLT, Subaru, Keck) and space facilities (HST, Spitzer, GALEX, Chandra). Using the observational data and stellar synthesis models, I study properties of galaxies as a function of star formation rate, morphology, environment, color, luminosity and redshift. Over the past two years I have been involved in the following studies: developing the Balmer Break technique for identifying very high redshift massive galaxies, using the combined HST and Spitzer data; discovery of an extremely massive and evolved galaxy at z ~ 7 (this has been the subject of extensive press release); study of the density-morphology relation over the largest dynamic range in density, and its evolution with redshift; planning and execution of the new Hubble Ultra Deep Field (HUDF), doubling the area and increasing the depth of the original HUDF; finding candidates for galaxies at z ~ 8 through searches for J-band dropouts in the new HUDF; study of the properties (SFR, mass, extinction) of high redshift galaxies selected through narrow-band Lyman emission (LAE) at z = 5.7 and comparison with Lyman Break Galaxies at the same redshift; developing a photometric redshift code to calculate redshift, spectral types and stellar masses of galaxies. I have been seriously involved in the Great Observatories Origins Deep Survey (GOODS) and The Cosmic Evolution Survey (COSMOS) projects. I have a.lso been leading parts of the new HST treasury project to survey the core and outskirt of the Coma cluster. In the following, I summarize some of my on-going research activities and future plans:

Balmer Break Technique for finding Very High-z Massive Galaxies
I developed a technique for selecting massive high redshift galaxies using the combined near-IR and mid-IR (Spitzer) observations. At z ~ 7, the Balmer Break shifts between K (2.2 μm) and Spitzer (3.6 μm) bands. Also, at this redshift, Spitzer light samples rest-frame near-IR wavelength which is proportional to the stellar mass in galaxies. Furthermore, the strength of Balmer discontinuity correlates with the age of stellar population. Based on these properties, I demonstrate that galaxies with red m_2.2 - m_3.6 and blue m_1.6 -m_2.2 colors and no detection in optical bands, qualify as candidates for mature massive galaxies at z > 5 (Wiklind, Mobasher et al 2006). Using these criteria, we found a total of 18 such candidates over the 160 arcmin² area of the GOODS-S field. Presence of such population of high redshift massive galaxies has significant implications for galaxy formation scenarios (i.e., monolithic collapse vs. Cold Dark Matter) and will require serious revision of current theories.

Discovery of a Very Massive and Evolved Galaxy at z ~ 7
Using the Balmer Break technique (above), I discovered a very massive (6 x 10¹¹ M_sun) and evolved (800 Myrs) galaxy at z ~ 7 in the HUDF (Mobasher, et. al., 2005, ApJ 635, 832). This is an object which is not detected in the J-band (1.2 µm) and the extremely deep images (HUDF-ACS) at shorter wavelengths but is well-detected in H-band (1.6 µm) and Spitzer (3-8 µm). Discovery of this object is widely reported in the press recently. This has enormous implications towards galaxy formation scenarios and for reionisation of the Universe, as was explored by us (Panagia, Fall, Mobasher, et al., 2005, ApJL ApJ 633, L1).

Search for Very High Redshift Galaxies

Near-Infrared Search for Galaxies at z ~ 7
In collaboration with Masami Ouchi (Hubble Fellow at STScI), I have designed a project to perform deep Suprime-Cam imaging (with Subaru 8m class Telescope) of the GOODS-N field at 1 using our custom-built ZR band. Although the spectral coverage of the ZR is bluer than J-band, this reaches 26.3 mag (AB) over 900 arcmin2 area, which is about 100 times more efficient than J-band imaging with the latest near-IR camera (MOIRCS on Subaru). We to 60 z'-drop-out candidates at z > 7, using the z' -ZR colors. This allows us to derive rest-frame UV luminosity function of galaxies at Z 7, which reveals key information on the cosmic star formation history and reionisation of the Universe, estimate stellar mass using Spitzer GOODS-N data and extending the study of mass assembly rate of galaxies to the highest redshifts.

Wide-Field Searches for Lyα Emitters at z = 5.7
I used Suprime-Cam on Subaru (8m) and carried out narrow-band (NB816 filter) surveys of both the GOODS and COSMOS fields, targetting galaxies at z ~ 5.7 (in collaboration with Yoshi Taniguchi (Tohoku University, Japan)). Further 14 nights of observing time on Subaru is awarded (in 2006) to perform narrow- and intermediate-band surveys of the COSMOS field. This will be the first, wide-field search for Lyα emitters (LAEs) at z = 5.7, providing a large, complete, and very homogeneous sample of LAEs. Using the combined HST-ACS/Subaru-NB816/Spitzer surveys of the GOODS and COSMOS, I am investigating the following issues:

I selected the narrow-band filter (NB816) to be as close as possible to that of the HST/ACS i-band. This allows us, for the first time, to investigate detailed rest-frame ACS morphologies of LAEs at z = 5.7. I have compared this with rest-frame morphology of LBGs at z ~ 5-6. Fitting Sersic profiles (r^-1/n), I found differences between LAE and Lyman Break Galaxy (LBG) morphologies at z > 5.

Many of the LAE candidates at z ~ 5.7 in the GOODS fields are detected with Spitzer (IRAC 3.6-8.0 µm). Using their observed SEDs, I estimate their stellar mass and SFRs and compare these with LBGs at the same redshift. I also compare the stellar mass vs. morphology relation for LAEs and LBGs, to investigate if these galaxies have had different formation histories.

The LAE survey in the COSMOS field is the largest and the most homogeneous survey of these galaxies. I use this to establish the LAE luminosity function z ~ 5.7 (corresponding to rest-frame UV) and the star formation rate contributed by these galaxies at this redshift. I will also use this sample to explore the contribution of LAEs to reionisation.

I have planned the up-coming Narrow (NB)- and Intermediate (IA) band Subaru observations of COSMOS to use the following filters: NB704 and NB711 (to identify LAEs in the range 4.7 < z < 4.9 to compare with our current LAEs at z ~ 5.7); IA827 (to find LBGs at z = 5.7 to compare with LAEs at the same redshift); IA427 (to select LBGs at 1.5 < z < 3.5) and IA574 (to constrain the escape fraction of the ionizing radiation).

The New Hubble Ultra-Deep Field (HUDF-2)
I have played a substantial role in developing, planning and executing the HUDF-2, increasing the size and depth of the original HUDF in both optical (ACS) and near-IR (NICMOS) wavelengths.

The HUDF-2 ACS and NICMOS data are now taken and I am currently performing searches for i -(z ~ 5) and J-band (z ~ 7) dropout candidates. Using simulation of galaxies to establish the completeness in the combined HUDF surveys, I estimate limits on the space density of these candidates and their rest-frame properties.

Astrophysics of Star Formation Processes
My aim here is to measure the Star Formation Rate (SFR) in galaxies using independent diagnostics, understand the physics of the star formation (SF) process and constrain the parameters which affect it, as summarised below:

Near-Infrared Paschen α Line as a Dust-free Measure, of the SFR
I have performed near-Infrared spectroscopy on a sample of spectroscopically confirmed star-forming galaxies (in the SA57 field) and measured their Paschen α line strengths. The strength of Pa α line, like Hα, is expected to be correlated with SFR in galaxies. Moreover, SFR estimated from the near-IR light will not be affected by dust obscuration. I will study the correlation between Pa α and other SF diagnostics (Hα, VV, V-band and 1.4 GHz radio) and use these to calibrate Pa α as a new star formation indicator and to constrain obscuration of SFR with dust.

Star Formation Rate at z ~ 2 from Near-infrared Spectroscopy and 24µm Observation
One of the most fundamental problems currently facing observational astronomy is accurate measurement of the SFR at z ~ 2. Whether the SFR continues the increasing trend seen at z < 2, starts to decrease or levels-off, has important implications towards evolution of galaxies and their metal enrichment history. There are two reasons for present uncertainties in the estimated SFRs at z ~ 2; lack of reliable SFR diagnostics at this redshift and limited understanding of dust extinction and its behavior with redshift. Here, I introduce two independent star formation indicators at z ~ 2 and test these on an existing sample of galaxies. At z ~ 2, Hα line (which is a reliable SF indicator) shifts to near-Infrared wavelengths and therefore, near-Infrared spectroscopy would allow a dust-free estimate of the SFR. Moreover, the PAH feature (8.0 µm which is considered as a new SF indicator, shifts to 24 µm at z ~2, and can be measured by the observed Spitzer-MIPS flux.

I have performed a pilot study of this project by performing near-Infrared spectroscopy of a sample of confirmed star-forming galaxies in the range 1.8 < z < 2.2, using CISCO on Subaru. This has confirmed feasibility of this study. I am now planning to carry out near-infrared spectroscopy of a sample of galaxies in GOODS fields, covering the range 1.8 < z < 2.2 and selected through their MIPS (24 µm flux. The aim is to measure their dust-free SFR from Hα line strengths and rest-frame PAH features. This provides two completely independent and dust-free estimates for the SFR at z ~ 2. I plan to compare these with the SFR at z ~ 2, estimated from rest-frame UV 2800 Å flux (which is affected by dust) to constrain the role of dust extinction in SFRs at this redshift.

Comparative Study of Star Formation Rates from Different Diagnostics
We now have different star formation diagnostics in both GOODS and COSMOS surveys, including: GALEX Near- and far-UV, rest-frame UV 2800 Å, rest-frame U-band, Hα, radio (1.4 GHz) and PAH features (8.0 µm). I have started a comprehensive study of the SFRs using these diagnostics. My aim here is to address the following questions: When applying different star formation indicators on the same galaxy, why do we get different SFRs?; How does the scatter in SFRs between independent indicators correlate with physical properties of galaxies (i.e. color, morphology, mass)?; Is there a relation between SFRs with morphology, dust extinction and stellar mass of galaxies?; How could one constrain the dust extinction in measuring SFRs?; How does local environment affect the SFRs? These are all fundamental questions which will be addressed in this study.

Star-formation rate as a function of morphology and environment
Using the ACS and ground-based COSMOS data, we measure the morphology and star formation rate (SFR), from radio/U-band/GALEX-UV luminosities and address the following questions: How does the star formation density vs. redshift relation behave for different types of galaxies (i.e., interacting vs. isolated) in the range 0 < z < 1?; What fraction of the SFR in this redshift range is due to galaxy mergers/interaction?; What are the main parameters which govern star formation activity in galaxies (environment, morphology, mass etc.)?; Why do galaxies with the same Hubble types have different SFRs?; Is there a relation between stellar mass and morphology in galaxies?; and How this evolves with redshift?

Probe of Dark Energy and its Evolution
I was a core member of the team exploring dark energy and its evolution by searching for high redshift Type Ia SNe in the GOODS field. We found six of the highest redshift SNe currently known, all at z > 1.25. For the first time, we measured the equation of state for dark energy and its evolution with redshift (Riess, et al., 2004 ApJ. 607, 665). I am a member of the PANS team (Probe of Acceleration Now with SNe-PANS) which is currently extending this study by searching for more SNe at 1 < z < 1.5, using a total of 250 HST orbits. The observations for this project are now completed and we are in the process of analysing the data. Preliminary analysis of these data confirms the need for dark energy in the form of cosmological constant. Recently, we are awarded more time on HST to continue the study of dark energy through searches for SNe Type Ia.

I am a full member of the SNAP collaboration. My responsibility in the SNAP team is to perform simulations of multi-waveband catalogs and photometric redshifts, based on simulated Spectral Energy Distributions (SEDs) of galaxies to optimise SNAP filter shapes and wavelength coverage. This allows to decide on the required photometric data and their associated errors.

Photometric Redshifts and Stellar of Galaxies
I have developed a code, measuring photometric redshift, spectral type and stellar mass of galaxies by comparing observed Spectral Energy Distribution (SED) of galaxies with model SEDs and estimating the best-fit values for the parameters using luminosity function priors. This is extensively used for scientific analysis of GOODS, COSMOS and HUDF data. I compared photometric redshifts for a sample of galaxies with available spectroscopic redshifts and found rms = 0.031. I have extended the code to included Spitzer (IRAC) photometry and estimate photometric redshifts for Active Galaxy Nuclei (Type II AGNs)-(rms = 0.04), Lyman Break Galaxies (rms = 0.05), Lyman Alpha Emitters (rms = 0.06).

Formation of Dwarf Galaxies
I have selected a sample of dwarf galaxies with M < 10⁹ using the stellar masses estimated by fitting the observed Spectral Energy Distributions (SEDs) of galaxies in COSMOS. I then constructed the rest-frame UV 2500 A luminosity function and luminosity density of this mass-selected sample of dwarfs out to z ~ 1, using the available multiwaveband data. My aim here is to investigate if the UV radiation from young stars has delayed formation of dwarf galaxies, as can be revealed from the presence of a cut-off in the UV luminosity function of mass-selected dwarfs.

HST Treasury Survey of the Coma Cluster Galaxies
I am the core member (Deputy PI) of the HST Coma treasury proposal which was recently approved (168 HST orbits). This is to use the unique spatial resolution of HST and ACS to construct a Treasury imaging survey of the core and infall region of the richest local cluster, Coma. My main science objectives are:

To study of the structure of the dwarf galaxies, including scaling laws, nuclear structure and morphology, to compare with hierarchical and evolutionary models of their formation.

Study of stellar populations from colors and color gradients, and how the internal chemical evolution of galaxies is affected by interaction with the cluster gaseous and galaxy environment.

To determine effect of the cluster environment upon morphological features, disks, bulges and bars, by comparing these structure in the Coma sample with field galaxy samples.

Identification of samples of dwarf galaxy for further study with the new generation of multiobject and spatially resolved spectrographs on 8-10 metre telescopes such as Gemini, Keck and Subaru.

This is the first such survey of a nearby rich cluster and will provide key database for many studies and comparison with samples at higher redshift.

Keck Spectroscopy of Very High Redshift Galaxies
I plan to carry out an extensive program of optical and near-infrared spectroscopy of different populations of high redshift galaxy candidates (discussed in sections I and III above) in GOODS, COSMOS and Hubble Ultra Deep Field (HUDF), using DEIMOS and NIRSPEC instruments on the Keck telescopes.

I have formulated a large spectroscopic project to perform optical spectroscopy (with KeckDEIMOS) of the new population of massive and evolved galaxies at 5 < z < 7, found through the Balmer Break technique (Section I). The aim is to establish spectroscopic redshifts for this new population of galaxies. Confirmation of redshifts of these objects is essential in establishing their stellar mass and age. However, since Balmer Break Galaxies are evolved systems with ages 300-800 Myrs, they are expected to have weak emission lines, making an unambigous measurement of their spectroscopic redshifts difficult. This requires long integration on 8-10m class telescopes. Using deep optical, near-infrared and Spitzer images, we have currently identified ~18 Balmer Break candidates at 5 < z < 7 with stellar masses 5 x 10¹⁰ < M_O < 6 x 10¹¹ and ages 300-800 Myrs. Confirmation of the predicted redshifts for even a few of these objects will support our selection technique and the estimated parameters for the population of Balmer Break Galaxies. Presence of a large population of very massive and old galaxies at z > 5, has enormous implications for scenarios of formation of galaxies (i.e., monolithic collapse vs. Cold Dark Matter) and the dark matter halos hosting them.

I have been awarded four nights on the Keck to use DEIMOS to perform optical spectroscopy of a sample of 168 Narrow-band selected Lyman α emitters (LAEs) at z ~ 5.7, found in the COSMOS field (in collaboration with P. Capak and N. Z. Scoville, Caltech). This is the largest sample of LAEs in a single field currently available. These observations will allow studies of the stellar mass, star formation activity and clustering of galaxies at z > 5 and will be used to constrain the mass assembly rate of galaxies out to z ~ 6. Once redshifts are known, near-infrared spectroscopy (with NIRSPEC) of these galaxies allows measurement of metallicity and star formation rates for these objects, extending the mass-metallicity and mass-SFR relations to the highest possible redshifts.

I plan to perform near-infrared spectroscopy of galaxies at z > 7, selected through our nearinfrared (1 µm) search with Subaru Telescope (section IILl). We expect over 80 such candidates from our search. Spectroscopic confirmation of the redshifts of these galaxies will provide significant number of available galaxies at a redshift close to reionisation epoch. This allows study of rest-frame UV luminosity function of galaxies at z ~ 7, constrain the evolution of number density of galaxies to this redshift and their star formation properties.

Selected Publications

Mobasher, B.; Jogee, S.; Dahlen, T.; De Mallo, D.; Lucas, R.A.; Conselice, C.; Grogin, N.; Livio, M., Structure and Evolution of Starburst and Normal Galaxies, Ap. J. 2004, 600, 143.

Mobasher, B., et al., Photometric Redshifts to Galaxies in the Southern GOODS Field, Ap. J. 2004, 600, 167.

Riess, A.G., Strolger, L., Tonry, J., Casertano, S., Ferguson, H.C., Mobasher, B., Challis, P., Fillipenko, A.V., Saurabh, J., et al., Type Ia Supernovae Discoveries at z > 1 from the Hubble Space Telescope: Evidence for Past Deceleration and Constraints on Dark-Energy Evolution, Ap. J. 2004, 607, 665. (Ranked as the second most cited paper in astrophysics in 2005).

Dahlen, T., Strolger, L., Riess, A.G., Mobasher, R, Chary, R., et al., High Redshift Supernovae Rates, Ap. J. 2004, 613, 189.

Dahlen, T., Mobasher, B., Dickinson, M., Ferguson, H., and Giavalisco, M., Evolution of Rest-Frame Optical and Near-Infrared Galaxy Luminosity Functions and Their Type-Dependence, Ap. J. 2005, 631, 126.

Mobasher, B., et al., Evidence for a Massive Post-Starburst Galaxy at z ~ 6.5, Ap. J 2005, 635, 832. (This was ranked among the top 100 most outstanding science results in 2005, selected by the Science News and Discover magazines).

Panagia, N., Fall, M., Mobasher, B., Dickinson, M., Ferguson, R., Giavalisco, M., and Wiklind, T., Direct Evidence for an Early Reionization of the Universe?, Ap. J. Lett., 2005, 633, 1.

Ajiki, M., Mobasher, B., Taniguchi, Y., Shioya, Y. et al., Narrow-band Survey of the GOODS Fields: Search for Lyman α Emitters at z = 5.7, Ap.J. 2006, 638, 596.

Capak, P., Abraham, R., Ellis, R.S., Mobasher, B., Scoville, N.Z., Sheth, K., The Effects of Environment on Morphological Evolution at 0 < z < 1.2 in the COSMOS Survey, Ap. J. Suppl. 2007, 172, 284.