Radiation Backgrounds Observations across the Electromagnetic Spectrum Günther Hasinger, MPE Garching & TUM The History of Nuclear Black Holes in Galaxies Harvard University, May 18, 2006
WMAP mm Sky Backgrounds ROSAT 0.1-2 kev DIRBE FIR RXTE 2-10 kev visual EGRET γ
Background Energy Distribution Zodiacal Light NIR Excess (COBE/DIRBE & IRTS/NIRS) Intensity [nw m -2 sr -1 ] Cirrus Frequency [Hz] G.H. (2000)
NIR EBL Excess Pop III Stars or Zodi COBE/DIRBE & IRTS/NIRS Zodiacal light x 0.23 Excess Galaxies Dwek et al., 2005 Pop III stars: z min -z max =7-15 or 9-30 Also: mini-qsos pose problems with soft X-ray background!
Background Energy Distribution Zodiacal Light NIR Excess (COBE/DIRBE & IRTS/NIRS) Intensity [nw m -2 sr -1 ] Cirrus H.E.S.S. upper limits Frequency [Hz]
Optical EBL by TeV absorption H.E.S.S. data on 2 blazars at z=0.17-0.19 Aharonian et al., 2006, Nature 440, 1018 It is very hard to measure a cosmic background!
Background Energy Distribution Intensity [nw m -2 sr -1 ] Spitzer / HST resolved sources H.E.S.S. upper limits Frequency [Hz]
Dole et al., 2006
Background Energy Distribution Intensity [nw m -2 sr -1 ] H.E.S.S. upper limits ROSAT Galactic/WHIM Spitzer / HST QSO SED (Elvis et al.) Spitzer AGN Treister et al.06 ROSAT/XMM/ Chandra (Worsley et al.) Silva et al. 04 Sy2 SED (Sturm et al. 06) RXTE (Revnivtsev et al.) Frequency [Hz] Galactic Gamma bkg New EGRET bkg (Strong et al., 04) EGRET resolved
Mid-IR AGN resolved by Spitzer AGN make ~3-10% of the MIR background. Integrating over all wavebands, their contribution to the total light could be ~10-15%...... Considering, that M BH ~10-3 M Bulge and energy production efficiency ε BH ~ 100ε *, estimate L BH ~ 0.1 L * Hasinger 1999, Fabian & Iwasawa 1999 Groth Strip: Barmby et al., 2006 GOODS: Treister et al., 2006
X-ray Background Spectrum Worsley et al., 2004 Courtesy Brusa, Comastri, Gilli It is hard to measure a cosmic background!
XMM Slew 4yr ½yr erosita 4yr XMM N. Brandt & G.H. 2005, ARA&A 43, 827
Lockman Hole Red (more soft X-rays) Blue (only hard X-rays) 1 Msec ROSAT XMM-Newton
X-ray Background average rest-frame spectra type-1 AGN EW~600eV type-2 AGN EW~400eV Streblyanskaya et al., 2004 Large equivalent width can be explained by <Z> ~ 3 x solar metallicity. BH spin measurement within reach. Good news for XEUS!
The final goal: XEUS (20xx?)
Black Hole Spin Non-rotating Black Hole (Schwarzschild-metric): R min = 6 GM/c 2 = 3 R S Schwarzschild Maximally rotating Black Hole (Kerr-metric): R min = GM/c 2 = R S /2 Kerr Schwarzschild (ε <0.05) Kerr (ε <0.34)
Throughput Matters! z=1 L X =10 44 erg/s F X =3.6 x 10 14 cgs 1 Msec Con-X TES 1 Msec XEUS WFI
Cosmos Survey Subaru XMM-Newton Suprimcam PI: Y. HST PI: Taniguchi G. ACS Hasinger PI: N. Scoville 2deg 2
Chandra Deep Field South Chandra 1 Msec (Giacconi et al.) Chandra 4x250 ksec (PI: N. Brandt) XMM-Newton 400 ksec (PI: J. Bergeron)
1.216 AGN - Ty 1, unobs 2.28p Ultra Deep AGN Field - opt. faint 1.82p AGN - opt. faint 0.773 AGN - XBONG 0.665 AGN - XBONG 1.65p AGN - opt. faint CDF-S ACS UDF 1.69p AGN - opt. faint 0.456 Starburst 3.193 AGN - Ty 1, unobs 4.29p AGN - opt. faint 1.309 AGN - XBONG 0.438 Starburst 0.414 Starburst 1.53p AGN - opt. faint 3.064 AGN - Ty 2 Courtesy: S. Beckwith
3 AGN zoo 3 B V i z Mainieri 2003, PhD thesis
3 AGN zoo 3 B V i z
Chandra Deep Field IDs Extreme X-ray/optical objects EXOs (e.g. Koekemoer) CDFN: Triangles CDFS: Squares dot: unidentified 0<z<0.5 0.5<z<1 1<z<2 2<z<6 Brandt & G.H. 2005
L X vs. redshift CDFN: Triangles CDFS: Squares I=15-20 I=20-22 I=22-23 I>23 Spectroscopic Desert Brandt & G.H. 2005
CDFS Optical IDs ombining all forces: ORS Spectro-z Szokoly et al., 2004) ombo-17 multiband Wolf et al., 2004) Number of Sources ORS/ISAAC photo-z Zheng et al., 2004 ainieri et al., 2004) Redshift > 95% have spectro- or photo-z thanks to VLT, GOODS, GEMS, ACF UDS etc. (only possible in CDFS!!!) Photo-z fill in spectroscopic desert, but still peak at z~0.7
Type1/Type2 at the same Redshift LX=44.6 HR=-0.54 Sy1 LX=43.8 HR=-0.46 Sy1 LX=43.6 HR=-0.46 Sy1 LX=43.1 HR=0.06 Sy2 LX=42.9 HR=1.00 Sy2 LX=42.7 HR=1.00 Sy2 Szokoly et al., 2004
Optically identified hard samples type-1: optical BLAGN, or galaxy with L X >42, HR<-0.2 type-2: optical NLAGN, or galaxy with L X >42, HR>-0.2 729 AGN, optical/nir completeness ~80-90%
Type 2 fraction f(l X ) Local Seyferts Fraction of type-2 s decreases with luminosity. Powerful AGN can clean out their environment! Barger et al. significantly incomplete for Sy1!
Type 2 fraction f(z) No significant evolution in absorption fraction detected
XLF 0.5-2 kev type-1 AGN Luminositydependent density evolution (LDDE) confirmed G.H., Miyaji & Schmidt, 2005, A&A 441, 417
Space/Luminosity density evolution G.H., Miyaji & Schmidt, 2005
Densities in soft and hard band Brandt & G.H., 2005 G.H., Miyaji & Schmidt, 2005 Ueda et al., 2003, based on ~1000 AGN-1 based on ~250 AGN Very similar behaviour in hard and soft band. Soft samples go deeper and are more complete.
Direct comparison of hard vs. Soft XLF m2 m1 Gilli, Comastri & G. H., 2006
Background Synthesis Model Ingredients: AGN X-ray luminosity function AGN cosmological evolution canonical AGN spectrum AGN absorption distribution variation of type1/type2 ratio? absorbed unabsorbed Many different models can fit the XRB spectrum Need other constraints Prediction: hard sources in Chandra/XMM surveys; type-2 QSO! Uncertainty in background spectrum important Comastri et al., 1995; Gilli et al. 2000-2006
Most recent Population Synthesis Model Gilli, Comastri & G.H., 2006
Most recent Population Synthesis Model Gilli, Comastri & G.H., 2006
Most recent Population Synthesis Model total type-1 C-thin type-2 C-thin type-2 C-thick Gilli, Comastri & G.H., 2006
Constraining the N H distribution still significant uncertainties and limited predictive power Need: identified samples of 10-30 kev sources & better determination of 20-40 kev background See Dwelly et al. 2005
SDSS QLF (Richards et al., 2005) vs. Soft XLF (Hasinger et al., 2005) Miyaji 2006, priv. comm.
Hao et al., astro-ph/0501042 SDSS Seyfert emission line LF (z=0) 2dF/SDSS T. Miyaji, priv. comm. Comparison
Extending Soltan s Argument Local BH Mass Function SDSS optical QSO LF Hard X-ray AGN LF Soft X-ray AGN LF Efficiency ε=0.1 Eddington Ratio λ=1.0 Marconi et al., 2006
Local BH mass vs. accreted BH mass function Accreted Black Hole mass function derived from X-ray background can be compared with the mass function of dormant relic black holes in local galaxies (Soltan 1982). These two estimates can be reconciled, if an energy conversion efficiency of ε=0.1 is assumed. Such high efficiency requires a Kerr-BH! Average Eddington Ratio is λ~0.3 λ: Eddington Ratio Schwarzschild Kerr ε: Conversion efficiency Marconi et al., 2006, MNRAS
E-CDFS XMM/COSMOS erosita XEUS Wall et al., 2005 High Luminosity QSO QSO at very high z?? Radio QSO (Wall et al., 2005) Soft X-ray ROSAT/Chandra/ XMM (G.H., Miyaji & Schmidt 2005) Chandra/ROSAT(Silverman et al. 2004) Optical QSOs (Schmidt et al., 1995, Fan et al. (2001,2004) Very large solid angle & deep surveys required to study z>5 QSOs
The new Spectrum-X-Gamma Mission Lobster ART-XC erosita Mission is approved on the Russian side. Launch possible on Soyus/Fregat from Kourou or Baikonur in 2010/11. Equatorial LEO orbit guarantees excellent background, other orbits still under discussion. DLR has earmarked funding profile for erosita Mission will be proposed for ESA & PPARC co-funding in 2006.
Instrument is in Phase-B now! Dark Universe Observatory XMM pn ROSITA (on-axis)
XMM Slew 4yr ½yr erosita Expect: 100.000 clusters (DE!) 3 Million AGN [225, 60, 18, 5] @ z > [ 6, 7, 8, 9] Lots of other stuff!
Thank you Intensity [nw m -2 sr -1 ] very much! Frequency [Hz]