|
|
The Home Page of David Rafferty
|
|
| Home |
|
Research |
|
Publications |
|
Résumé |
|
Links |
|
|
|
|
|
|
|
|
|
|
|
Papers
- D. A. Rafferty, B. R. McNamara, & P. E. J. Nulsen, "The Regulation of Cooling and Star Formation in Luminous Galaxies by AGN Feedback and the Cooling-Time/Entropy Threshold for the Onset of Star Formation," ApJ, 687, 899 (2008).
AbstractUsing broadband optical imaging and Chandra X-ray data for a sample of 46 cluster central dominant galaxies (CDGs), we investigate the connection between star formation, the intracluster medium (ICM), and the central active galactic nucleus (AGN). We report the discovery of a remarkably sharp threshold for the onset of star formation that occurs when the central cooling time of the hot atmosphere falls below ~ 5x108 yr, or equivalently when the central entropy falls below ~ 30 keV cm2. In addition to this criterion, star formation in cooling flows also appears to require that the X-ray and galaxy centroids lie within ~ 20 kpc of each other, and that the jet (cavity) power is smaller than the X-ray cooling luminosity. These three critieria, together with the high ratio of cooling time to AGN outburst (cavity) age across our sample, directly link the presence of star formation and AGN activity in CDGs to cooling instabilities in the intracluster plasma. Our results provide compelling evidence that AGN feedback into the hot ICM is largely reponsible for regulating cooling and star formation in the cores of clusters, leading to the significant growth of supermassive black holes in CDGs at late times.
- G. M. Voit, K. W. Cavagnolo, M. Donahue, D. A. Rafferty, B. R. McNamara, & P. E. J. Nulsen, "Conduction and the Star Formation Threshold in Brightest Cluster Galaxies," ApJ, 681, L5 (2008).
AbstractCurrent models of galaxy evolution suggest that feedback from active galactic nuclei is needed to explain the high-luminosity cutoff in the galaxy luminosity function. Exactly how an AGN outflow couples with the ambient medium and suppresses star formation remains poorly understood. However, we have recently uncovered an important clue to how that coupling might work. Observations of H-alpha emission and blue light from the universe's most luminous galaxies, which occupy the centers of galaxy clusters, show that star formation happens only if the minimum specific entropy of the intracluster gas is <~30 keV cm2. Here we suggest that this threshold for star formation is set by the physics of electron thermal conduction, implying that conduction is critical for channeling AGN energy input toward incipient star-forming regions and limiting the progress of star formation.
- S. Diehl, H. Li, C. L. Fryer, & D. A. Rafferty, "Constraining the Nature of X-ray Cavities in Clusters and Galaxies," ApJ, 687, 173 (2008).
AbstractWe present results from an extensive survey of 64 cavities in the X-ray halos of clusters, groups and normal elliptical galaxies. We show that the evolution of the size of the cavities as they rise in the X-ray atmosphere is inconsistent with the standard model of adiabatic expansion of purely hydrodynamic models. We also note that the majority of the observed bubbles should have already been shredded apart by Rayleigh-Taylor instabilities if they were of purely hydrodynamic nature. Instead we find that the data agrees much better with a model where the cavities are magnetically dominated and inflated by a current-dominated magneto-hydrodynamic jet model, recently developed by Li et al. (2006) and Nakamura et al. (2006). We conduct complex Monte-Carlo simulations of the cavity detection process including incompleteness effects to reproduce the cavity sample's characteristics. We find that the current-dominated model agrees within 1 sigma, whereas the other models can be excluded at >5 sigma confidence. However, this assessment is dependent on our correct understanding of the detectability of cavities in X-ray atmospheres, and will await confirmation when automated cavity detection tools become available in the future. Our results have considerable impact on the energy budget associated with active galactic nucleus feedback.
- D. A. Rafferty, B. R. McNamara, P. E. J. Nulsen, & M. W. Wise, "The Feedback-Regulated Growth of Black Holes and Bulges through Gas Accretion and Starbursts in Cluster Central Dominant Galaxies," ApJ, 652, 216 (2006).
AbstractWe present an analysis of the growth of black holes through accretion and bulges through star formation in 33 galaxies at the centers of cooling flows. Most of these systems show evidence of cavities in the intracluster medium (ICM) inflated by radio jets emanating from their active galactic nuclei (AGN).
We present a new and extensive analysis of X-ray cavities in these systems. We find that AGN are energetically able to balance radiative losses (cooling) from the ICM in more than half of our sample. Using a subsample of 17 systems, we examine the relationship between cooling and star formation. We find that the star
formation rates are approaching or are comparable to X-ray and far UV limits on the rates of gas condensation onto the central galaxy. The remaining radiative losses could be offset by AGN feedback. The vast gulf between radiative losses and the sink of cooling material, which has been the primary objection to cooling flows,
has narrowed and, in some cases, is no longer a serious issue. Using the cavity (jet) powers, we place strong lower limits on the rate of growth of supermassive black holes in central galaxies, and we find that they are growing at an average rate of ~ 0.1 Msun yr-1, with some systems growing as quickly
as ~ 1 Msun yr-1. We find a trend between bulge growth (star formation) and black hole growth that is approximately in accordance with the slope of the local (Magorrian) relation between black hole and bulge mass. However, the large scatter in the trend suggests that bulges and black holes do not always grow in lock step. With the exception of the rapidly accreting supercavity systems (e.g, MS 0735.6+7421), the black holes are accreting well below their Eddington rates. Most systems could be powered by Bondi accretion from the hot ICM, provided the central gas density increases into the Bondi radius as rho ~ r-1. However, if the slope of the gas density profile flattens into a core, as observed in M87, Bondi accretion is unlikely to be driving the most powerful outbursts.
- B. R. McNamara, D. A. Rafferty, L. Birzan, J. Steiner, M. W. Wise, P. E. J. Nulsen, C. L. Carilli, R. Ryan, & M. Sharma, "The Starburst in the Abell 1835 Cluster Central Galaxy: A Case Study of Galaxy Formation Regulated by an Outburst from a Supermassive Black Hole," ApJ, 648, 164 (2006).
AbstractWe present an optical, X-ray, and radio analysis of the starburst in the Abell 1835 cluster's central cD galaxy. The dense gas surrounding the galaxy is radiating X-rays with a luminosity of ~1 × 1045 erg/s consistent with a cooling rate of ~1000-2000 solar masses per year.
However, new Chandra and XMM-Newton observations find less than 200 solar masses per year of gas cooling below ~2 keV, a level that is consistent with the cD's current star formation rate of 100-180 solar masses per year. One or more heating agents (feedback) must then be replenishing the remaining
radiative losses. The heat fluxes from supernova explosions and thermal conduction alone are unable to do so. However, a pair of X-ray cavities from an AGN outburst has deposited ~1.7 × 1060 erg into the surrounding gas over the past 40 Myr. The corresponding jet power 1.4 × 1045
erg/sec is enough to offset most of the radiative losses from the cooling gas. The jet power exceeds the radio synchrotron power by ~4000 times, making this one of the most radiatively inefficient radio sources known. The large jet power implies that the cD's supermassive black hole accreted at a mean rate of ~0.3 solar masses per year over the last 40 Myr or so, which is a small fraction of the Eddington accretion rate for a 10E9 solar mass black hole. The ratio of the bulge growth rate through star formation and the black hole growth rate through accretion is consistent with the slope of the (Magorrian) relationship between bulge and central black hole mass in nearby quiescent galaxies. The consistency between net cooling, heating (feedback), and the cooling sink (star formation) in this system resolves the primary objection to traditional cooling flow models.
- A. J. Levan, N. R. Tanvir, A. S. Fruchter, E. Rol, J. P. U. Fynbo, J. Hjorth, G. Williams, E. Bergeron, D. Bersier, T. Gray, P. Jakobsson, K. Nilsson, E. Olszewski, R. S. Priddey, D. A. Rafferty, & J. Rhoads, "The Faint Afterglow and Host Galaxy of the Short-Hard GRB 060121" ApJ, 648, L9 (2006).
AbstractWe present optical and X-ray observations of the afterglow and host galaxy of the short-hard GRB 060121. The faint R-band afterglow is seen to decline as t-0.66 while the X-ray falls as t-1.18, indicating the presence of the cooling break between the two frequencies.
However, the R-band afterglow is very faint compared to the predicted extrapolation of the X-ray afterglow to the optical regime (specifically, betaOX ~ 0.2), while the K-band is consistent with this extrapolation (betaKX ~ 0.6), demonstrating suppression of the optical flux. Late time
HST observations place stringent limits on the afterglow R-band flux implying a break in the R-band lightcurve. They also show that the burst occurred at the edge of a faint red galaxy which most likely lies at a significantly higher redshift than the previous optically identified short-duration bursts. Several neighboring galaxies also have very red colors that are similarly suggestive of higher redshift. We consider possible explanations for the faintness and color of the burst. Our preferred model is that the burst occurred at moderately high redshift and was significantly obscured; however, it is also possible that the burst lies at z > 4.5 in which case the faintness of the R-band afterglow could be attributed to the Lyman-break. We discuss the implications that either scenario would have for the nature of the progenitors of short bursts.
- P. E. J. Nulsen, D. C. Hambrick, B. R. McNamara, D. A. Rafferty, L Birzan, M. W. Wise, & L. P. David, "The Powerful Outburst in Hercules A," ApJ, 625, L9 (2005).
AbstractThe radio source Hercules A resides at the center of a cooling flow cluster of galaxies at redshift z=0.154. A Chandra X-ray image reveals a shock front in the intracluster medium (ICM) surrounding the radio source, about 160 kpc from the active galactic nucleus (AGN) that hosts it. The shock has a Mach number of 1.65, making it the strongest of the cluster-scale shocks driven by an AGN outburst found so far. The age of the outburst is ~=5.9 × 107 yr, its energy ~3 × 1061 ergs, and its mean power ~1.6 × 1046 ergs s-1. As for the other large AGN outbursts in cooling flow clusters, this outburst overwhelms radiative losses from the ICM of the Hercules A Cluster by a factor of ~100. It adds to the case that AGN outbursts are a significant source of
preheating for the ICM. Unless the mechanical efficiency of the AGN in Hercules A exceeds 10%, the central black hole must have grown by more than 1.7 × 108 Msun to power this one outburst.
- B. R. McNamara, P. E. J. Nulsen, M. W. Wise, D. A. Rafferty, C. Carilli, C. L. Sarazin, & E. L. Blanton, "The Heating of Gas in a Galaxy Cluster by X-ray Cavities and Large-Scale Shock Fronts," Nature, 433, 45 (2005).
AbstractMost of the baryons in galaxy clusters reside between the galaxies in a hot, tenuous gas. The densest gas in their centres should cool and accrete onto giant central galaxies at rates of 10-1,000 solar masses per year. No viable repository for this gas, such as clouds or new stars, has been found.
New X-ray observations, however, have revealed far less cooling below X-ray temperatures than expected, altering the previously accepted picture of cooling flows. As a result, most of the gas must be heated to and maintained at temperatures above ~2keV (ref. 3). The most promising heating mechanism is powerful radio jets emanating from supermassive black holes in the central galaxies of clusters. Here we report the discovery of giant cavities and shock fronts in a distant (z = 0.22) cluster caused by an interaction between a radio source and the hot gas surrounding it. The energy involved is ~6 × 1061 erg, the most powerful radio outburst known. This is enough energy to quench a cooling flow for several Gyr, and to provide ~1/3keV per particle of heat to the surrounding cluster.
- L. Birzan, D. A. Rafferty, B. R. McNamara, M. W. Wise, & P. E. J. Nulsen, "The Systematic Properties of X-ray Cavities," ApJ, 607, 800 (2004).
AbstractWe present an analysis of 16 galaxy clusters, one group and one galaxy drawn from the Chandra X-ray Data Archive. These systems possess prominent X-ray surface brightness depressions associated with cavities or bubbles that were created by interactions
between powerful radio sources and the surrounding hot gas. The minimum energy associated with the cavities ranges from pV~1055 ergs in galaxies, groups, and poor clusters to pV~1060 ergs in rich clusters. We evaluate the hypothesis that cooling in the hot gas can be
quenched by energy injected into the surrounding gas by the rising bubbles. Nearly half of the systems in this study may have instantaneous mechanical luminosities large enough to balance cooling, at least for a short period of time, if the cavities are filled with a relativistic gas. We find a trend
or upper envelope in the distribution of central X-ray luminosity versus instantaneous mechanical luminosity with the sense that the most powerful cavities are found in the most X-ray-luminous systems. Such a trend would be expected if many of these systems produce bubbles at a rate that scales in proportion to the cooling rate of the surrounding gas. Finally, we use the X-ray cavities to measure the mechanical power of radio sources over six decades of radio luminosity, independently of the radio properties themselves. We find that the ratio of the instantaneous mechanical (kinetic) luminosity to the 1.4~GHz synchrotron luminosity ranges from a few to roughly a thousand. This wide range implies that the 1.4 GHz synchrotron luminosity is an unreliable gauge of the mechanical power of radio sources.
Posters and Proceedings
- D. A. Rafferty, B. R. McNamara, & P. E. J. Nulsen, "Chandra X-ray Observations of Cavity Systems," presented at "The Cool, the Colder, and the Cold" workshop in Leiden, The Netherlands, September, 2008.
- D. A. Rafferty, B. R. McNamara, & P. E. J. Nulsen, "The Regulation of Star Formation by AGN Feedback," presented at the "Radio Galaxies in the Chandra Era" meeting in Cambridge, MA, July, 2008.
Abstract We use broadband optical imaging and Chandra X-ray data for a sample of 46 cluster central galaxies to investigate the connection between star formation, the intracluster medium, and the central active galactic nucleus (AGN). By comparing the optical and X-ray-derived properties on similar spatial scales, we find that the central galaxy is likely to experience significant star formation when: 1) the X-ray and galaxy centroids are within 20 kpc of each other, 2) the central cooling time of the hot atmosphere is much less than 8x108 yr, and 3) the ratio of cavity power to X-ray cooling luminosity is approximately less than unity. These conditions are consistent with the idea that cooling and star formation at the centers of cooling flows are regulated by AGN feedback.
- D. A. Rafferty, B. R. McNamara, P. E. J. Nulsen, & M. W. Wise, "The Growth of Black Holes and Bulges at the Cores of Cooling Flows," presented at the "Heating vs. Cooling in Galaxies and Clusters of Galaxies" conference in Garching, Germany, August, 2006.
AbstractCentral cluster galaxies (cDs) in cooling flows are growing rapidly through gas accretion and star formation proceeding at cosmologically interesting rates. At the same time, AGN outbursts fueled by accretion onto supermassive black holes are generating X-ray cavity systems and driving outflows that exceed those in powerful quasars. We show that the resulting bulge and black hole growth follows a trend that is roughly consistent with the slope of the local (Magorrian) relation between bulge and black hole mass for nearby quiescent ellipticals. However, a large scatter suggests that cD bulges and black holes do not always grow in lock-step. New measurements made with XMM, Chandra, and Fuse of the condensation rates in cooling flows are now approaching or are comparable to the star formation rates, alleviating the need for an invisible sink of cold matter. We show that the remaining radiation losses can be offset by AGN outbursts in more than half of the systems in our sample, indicating that the level of cooling and star formation is regulated by AGN feedback.
- D. A. Rafferty, B. R. McNamara, L. Birzan, P. E. J. Nulsen, & M. W. Wise, "Feedback and Star Formation in Cluster Cores," presented at the September, 2004, meeting of the High Energy Astrophysics Division of the AAS.
AbstractUsing an analysis of 16 galaxy clusters, one group and one galaxy drawn from the Chandra X-ray Data Archive, we evaluate the hypothesis that cooling of the intracluster medium (ICM) can be quenched by energy dissipated by rising bubbles created by radio sources. We find that the instantaneous mechanical luminosities in the bubbles required to offset cooling range between 1-20 pV per bubble. Nearly half of the systems in this study may have instantaneous mechanical luminosities large enough to balance cooling. For the remaining systems in our sample, the energy dissipated by the bubbles alone cannot balance cooling unless additional shock heating is occurring. The residual cooling of gas in the systems should lead to star formation in the central galaxy whose rate scales in proportion to the cooling rate. To investigate this scenario, we derive central system properties from the X-ray data and compare these to the central colors.
- D. A. Rafferty, B. R. McNamara, L. Birzan, P. E. J. Nulsen, & M. W. Wise, "Feedback and Star Formation in Cluster Cores," presented at the "Riddle of Cooling Flows" conference in Charlottesville, VA, May, 2003.
|
|
|
|
|
|
|
This page last updated January 25, 2009.
|
|
|