THE EMISSION LINE RATIOS: A TOOL FOR INVESTIGATING THE PHYSICS OF THE LINE EMITTING REGIONS IN AGN Dragana Ilić, Jelena Kovačević, Luka Popović 1. Department of Astronomy, Faculty of Mathematics, Belgrade 2. Astronomical Observatory Belgrade

Outline  AGN properties  emission line regions  the Broad Line Region (BLR)  physical properties  the Boltzmann plot method and the photoionization models  method to estimate the BLR physical properties June 02, th BSAC, Chepelare, Bulgaria 2

Active Galactic Nuclei (AGN)  AGN phenomena – ubiquitous!  AGN properties:  compact size  great luminosities L ~ erg/s  broad continuum (at all wavelengths)  intensive broad & narrow emission lines  variability; polarized radiation  strong radio-sources  many different types June 02, th BSAC, Chepelare, Bulgaria 3

AGN – The unification model  AGN consists of:  massive black hole (10 7 – M ๏ )  accretion disc  emission line regions the Broad Line Region – BLR the Narrow Line Region – NLR  torus  jets (outflowing material) June 02, th BSAC, Chepelare, Bulgaria 4

AGN – the schematic view June 02, th BSAC, Chepelare, Bulgaria 5

Emission Line Spectra June 02, th BSAC, Chepelare, Bulgaria 6 Mrk 817 (Ilic et al. 2006) HH HβHβ HγHγ Broad Emission Lines (BELs) - only permitted transitions FWHM ~ km/s

Hβ line region June 02, th BSAC, Chepelare, Bulgaria 7 Optical Fe II ( Å) emission Narrow Emission Lines (NELs) - permitted & forbidden FWHM ~ km/s

What EL can tell us? June 02, th BSAC, Chepelare, Bulgaria 8  Kinematics of the region  velocities (line widths)  size (reverberation – time delays)  geometry (line shapes)  Physical conditions of the region  temperature  density  ionization state - line ratios - presence/absence of some lines

Broad Line Region (BLR)  FWHM ~ up to 10,000 km/s  dimensions ~ lt-days up to lt-month  complex line shapes (large widths; double- peak lines; asymmetry) => complex and stratified region (at least 2 subregions)  geometry not known: more than one proposed model (eg. Biconical ejection, two-component model, the rotational accretion disk model...) June 02, th BSAC, Chepelare, Bulgaria 9

 photoionization (main heating source?)  plasma more likely closer to stellar atmospheres than photoionized nebulae (Osterbrock 1989)  There are many problems!! such as: 1. temperature and density estimates 2. problems of the Fe II emission The BLR physics June 02, th BSAC, Chepelare, Bulgaria 10 ionizing photons BLR Only estimates, not directly determined from BEL ratios! There is practically no direct information on the temperature in the BLR! (Osterbrock & Ferland 200 6) There is practically no direct information on the temperature in the BLR! (Osterbrock & Ferland 200 6) BLR temperature T e ~ 10 4 K BLR density N e ~ cm -3

Properties of the plasma in... ...stellar atmospheres T e ~ K N e ~ cm -3 ...BLR T e ~ 10 4 K N e ~ cm -3 ...emission nebulae T e ~ 10 4 K N e ~ cm -3 June 02, th BSAC, Chepelare, Bulgaria 11

 photoionization (main heating source?)  plasma more likely closer to stellar atmospheres than photoionized nebulae (Osterbrock 1989)  There are many problems!! such as: 1. temperature and density estimates 2. problems of the Fe II emission The BLR physics June 02, th BSAC, Chepelare, Bulgaria 12 ionizing photons BLR Optical Fe II ( Å) lines Many problems related to the Fe II multiplet, such as: - Photoionization models can not explain the total (UV and optical Fe II) emission - Geometrical place of Fe II emission region in AGN structure ? - etc…

The Boltzmann Plot (BP)  optically thin plasma, small changes in T & n (Griem 1997, Popović 2003, 2006) I lu ≈ (hc/λ) g u A ul l (N 0 /z) exp(-E u /kT)  for lines from one spectral series (e.g. Balmer lines) if the population of upper states (n>3) follow Boltzman distribution => excitation T F n = (F ul λ)/(g u A ul ) => log(F n ) = B − A·E u June 02, th BSAC, Chepelare, Bulgaria 13 F ul ≈ (hc/ λ ) g u A ul l (N 0 /z) exp(-E u /kT) A = log 10 e/(k B T)

BP method in the BLR of AGN  the BP method is applicable for approx 50% of cases (La Mura et al. 2007)  why BP works only for some AGN type 1?  If BP works in case of a BLR,  the BLR is in PLTE, so how this can be explained with the photoionization model? June 02, th BSAC, Chepelare, Bulgaria 14 Using the classical photoionization model of the BLR  under some conditions the populating of upper levels in the Balmer lines is driven by temperature  the BP method can be applied for the BLR diagnostics

CLOUDY simulations (Ferland 1998)  set a minimal number of input parameters of standard BLR and vary hydrogen density (n H = 10 8 – cm -3 ), ionizing flux (F H = ) and column density (N H = – cm -2 )  analyze some emission line ratios generated by the code: the Balmer lines and HeII4686/ HeI5876 ratio  apply the BP method on Balmer lines and calculate the BP temperature T BP  consider the average temperature T av of the region June 02, th BSAC, Chepelare, Bulgaria 15

The error of the BP applied on the CLOUDY’s Balmer line series June 02, th BSAC, Chepelare, Bulgaria 16 N H = cm -2 N H = cm -2 The parameter space where f<10% is well defined and constrained (in a similar range for different N H )

 the phyisical parameters of the simulations for which the BP method could be applied (f < 10%) follow some relations even when N H changes  the average temperature and the helium lines ratio R= HeII4686/ HeI5876 : T av = A+B · R  the hydrogen density and the helium line ratio R: log n H = D/(C + R) [n H is in 10 7 cm −3 ] June 02, th BSAC, Chepelare, Bulgaria 17 Ilić et al. 2010, sent to MNRAS

 SDSS imaged 8,400 square degrees of the sky in five optical bandpasses (230 million objects)  obtained spectra of 930,000 galaxies, 120,000 quasars, and 225,000 stars June 02, th BSAC, Chepelare, Bulgaria 18 SDSS Galaxy Map

SDSS sample  90 objects, selected to have:  z < 0.4  entire Balmer line series covered, and clearly recognized, at least up to Hδ  broad component detectable for each Balmer line  measure Balmer line fluxes  applied BP method  measure HeII 4686 and HeI 5876 June 02, th BSAC, Chepelare, Bulgaria 19

Results for the SDSS sample  T av = K  n H = cm -3  correlations between the BLR physical parameters (average temperature and hydrogen density) and the FWHM and FWZI of the BELs June 02, th BSAC, Chepelare, Bulgaria 20 Ilić et al. 2010, sent to MNRAS

 302 AGN spectra, SDSS (DR7)  Fe II line emission analysed in details  study origin of Fe II and other properties Kovačević et al. 2010, ApJS, accepted The AGN sample and Fe II template (see talk of Jelena K. in the next session) 21

Conclusions  for a limited space of physical parameter (n H and Φ H ) the photoionization can affect the Balmer lines in such way that the BP method is working  in that case, the BP excitation temperature can be very useful for diagnostics of the BLR Te  for the same space of physical parameters, the ratio of the helium lines (He II λ4686/He I λ5876) can indicate the BLR hydrogen density June 02, th BSAC, Chepelare, Bulgaria 22

Conclusions  the average temperature and hydrogen density in the BLR: T av = K n H = cm -3  we found correlations between the BLR physical properties and the geometry (FWHM and FWZI) June 02, th BSAC, Chepelare, Bulgaria 23

Thank you for your attention! June 02, th BSAC, Chepelare, Bulgaria

 Paolo Padovani (European Southern Observatory)  Evanthia Hatziminaoglou (European Southern Observatory)  Milan S. Dimitrijević (Astronomical Observatory Belgrade)  Petr Skoda (Astronomical Observatory Ondrejov, Czech Republic )  Igor Chilingarian (Observatoire Paris-Site de Meudon, France)  Milcho Tsvetkov (Sofia Sky Archive Data Center, BAS)  Katja Tsvetkova (Sofia Sky Archive Data Center, BAS)  Žarko Mijajlović (Faculty of Mathematics, Univ. Belgrade) AI-VO, Belgrade, June 29 – July 1

AstroMundus  2-years Erasmus Mundus Masters Course (120 ECTS) in Astrophysics offered by: University of Innsbruck, Padova, Rome, Goettingen and Belgrade  students carry out their master studies in at least 2 and up to 4 countries  Joint Master Degree issued by all partner universities June 02, 2010

Lab vs. observations  Broad lines are really broad! June 02, th BSAC, Chepelare, Bulgaria 27

 Over 8 years of operations, it obtained deep, multi- color images covering more than a 1/4 of the sky and created 3-dimensional maps containing more than 930,000 galaxies and 120,000 quasars  2.5-meter telescope in New Mexico  the 120-megapixel camera, 1.5 square degrees of sky  pair of spectrographs fed by optical fibers  A custom-designed set of software pipelines kept pace with the enormous data flow from the telescope June 02, th BSAC, Chepelare, Bulgaria 28

Origin of Fe II emission  correlations between widths (and shifts) of Fe II and ILR component of Hα and Hβ  ILR origin of Fe II Kovačević et al. 2010, ApJS, accepted June 02, th BSAC, Chepelare, Bulgaria 29

EW FeII – EW [OIII] anticorrelation  could there be luminosity dependance? June 02, th BSAC, Chepelare, Bulgaria 30 Kovačević et al. 2009, ApJS, accepted

EW FeII – EW [OIII] anticorrelation  Baldwin effect (L~EW -k ) - could be the cause… June 02, th BSAC, Chepelare, Bulgaria 31 total 302 FWHM Hb<3000km/s FWHM Hb>3000km/s