1. Observational Evidence of AGN Feedback Author: A.C Fabian reporter: Jun Xu

2. Outline Introduction The Radiative or Wind Mode The Kinetic Mode Summary Future study

3. ABCs of AGN Feedback Radiation, winds and jets from AGN interact with ICM or ISM May terminate star formation in the galaxy and stifle accretion onto the black hole This powerful object may strongly impact on their surrounding area in two different mode — Radiative or Wind Mode and the Kinetic mode

4. Why is AGN feedback important The binding energy of the galaxy bulge, which is of mass M gal, is E gal ≈M gal σ 2 The mass of the black hole is typically observed to be M BH =1.4×10 -3 M gal And assuming a radiative efficiency for the accretion process of 10% E BH /E gal ≈ 1.4×10 -4 (c/σ ) 2 Most galaxies have σ < 400 Km S -1 So E BH /E gal > 80

5. The Radiative or Wind mode Also called quasar mode When AGN is very luminous, accreting close to the Eddington rate Most concerned with pushing cold gas about Strong radiation Energy: heating Momentum: pushing

6. M BH − σ relation For a quasar at the Eddington limit based on Energy gives based on Momentum balance gives Gultekin et al 2009

7. Radiation pressure on dust The quasar is locally at Eddington limit, it must far below the Eddington limit when the moss of the galaxy is included Dust embedded in the gas is expected L edd is reduced by a factor of σ d / σ T, Which is about 1000 for a typical quasar, dropping to 500 for low Eddington ratio objects Is this just a coincidence or underlying reason why M gal /M BH ∼ 1000 ?

8. Radiation pressure on dust If the repeated action of radiation pressure on dust is responsible for M BH − σ relation, then For a constant mass-to-light ration,this corresponds to the Faber- Jackson(1976) relation Since M gal = 2σ 2 r/G, then

9. AGN winds The main interaction may due to winds, not to radiation pressure. To produce M BH ∝ σ 4, L w needs to be proportional to the Eddington limit, which is plausible if wind is accelerated by radiation pressure The commonest way to observe AGN winds is by line absorption of the quasar continuum by intervening wind material

10. AGN winds The X-ray warm absorbers commonly seen in Seyfert galaxies flowing at ~1000 km s -1 are insufficient Faster winds are required, such as those seen in UV observation of BAL quasars and X-ray of some AGN with velocities of tens of thousands of km s -1 some AGN Tombesi et al (2012) estimated mass outflow rate exceeds 5% of the mass accretion rate,and have L w range from 10 42.6 – 10 44.6

11. Galaxy outflows Evident of AGN feedback is clearly seen in some galactic outflows Most of the lower velocity winds are considered to be powered by stellar processes such as supernova Identifying the effects of AGN feedback in outflows relies on observing higher velocity (e.g.> 500 km s -1 ) High fuelling rate leads to high obscuration,makes observation more difficult Many recent reports of outflows hosting AGN. A spectacular example is the 1300 km s-1 outflow in a redshift 6.4 quasar(Maiolino et al 2012)

12. Downsizing problem The most luminous and massive AGN peak at earlier cosmic times Surprising anti-hierarchical behavior in CDM universe Something is quenching quasar

13. The Kinetic Mode Also called radio mode Which uses the mechanical energy of radio-emitting Weak radiation Typical operates when the galaxy has a hot halo and the accreting black hole has powerful jets

14. Cooling flow problem The massive galaxies at centres of groups and clusters are often surrounded by gas with a radiative cooling time short enough that a cooling flow should be taking place(Fabian 1994) A1835 has a star formation rate in a low redshift BCG ~125 M sun yr -1, Less than uninhibited mass cooling rate which is ~1000 M sun yr -1 XMM-Newton RGS provided crucial information against a simple cooling flow mode in that they failed to show the strong lines expected form FeXVII as the gas cooled below 0.7 keV Maybe some thing heating the gas. The likely heat source is the AGN at the centre of the cool core

15. Gas properties of some sample

16. Cooling flow problem The general consensus now is that massive black hole at the centre of the galaxy is feeding energy back into its surrounding at a rate to balancing the loss of energy through cooling The accretion flow onto the black hole generates powerful jets which inflate bubbles of relativistic plasma A study of the brightest 55 clusters originally show that over 70% of th ose cluster where the cooling time is less than 3Gyr, therefore needing heat. An updating make the bubble fraction is >95% due to pr ojection effect

17. Bubbles The bubbles or cavities, commonly seen in deep Chandra images of cool core clusters are blown and powered by jets from central BH The innermost bubbles are usually fairly spherical and are surrounde d by a thick pressure region fronted by a weak shock Thermal energy within that region corresponds to 3.7 times that of a s urrounding region of similar volume to the bubble(Graham 2008) Bubbles rise buoyantly in the surrounding hot atmosphere, turning into ghost bubbles as they become undetectable in high frequency

18. Chandra X-ray images

19. X-ray image of Perseus cluster Fabian et al. 2006

20. Bubbles power versus cooling power

21. Heating/Cooling balance The lack of high star formation rates suggest that cooling does not exceed heating by ten percent or so. The temperature drops indicates that heating does not generally exce ed cooling by much either. This represents a relatively close balance which needs to continue Over tens to hundreds of bubbling cycles If too much gas starts to cool then the accretion rate should increase making the heating rate go up and vice versa

22. Summary There is clearly enough energy and momentum produced by the AGN to expel the interstellar medium of the host galaxy It appears that the radiative or wind mode wad most active when the AGN was a young quasar. And the galaxy had a large component of cold molecular gas and nucleus was probably highly obscured. The kinetic mode is more easily observed, since it is acting now in nearby massive objects. The surrounding gas is hot, highly ionized. Although the gross energetics are roughly understood, the details are not An attractive possibility is that the radiative mode shaped the overall Galaxy and BH mass at early times, and the kinetic mode has since Maintain the situation where needed.

23. Further study Contributions to understanding AGN feedback can be expected from all wavebands Advanced instruments and telescopes which are planned for next few years will help In particular, the JAXA/NASA/ESA X-ray observatory ASTRO-H, to be launched in 2014, will offer non-dispersive high spectral resolution X-ray spectroscopy on a spatial scale of 1.5 arcmin using a microcalori- meter

24. Thanks