The Narrow Line Region Current Models and Future

The Narrow Line Region Current Models and Future Questions Brent Groves Max Planck Institute for Astrophysics Invited Talk The Central Engine in AGN

Model Aims • Fit Emission Lines • Flux • Equivalent Width/ wrt AGN continuum • Emission Line Ratios • Line width • Fit Continuum • IR • Provide descriptions/predictions of NLR ISM state and ionizing continuum Brent Groves -NLR Models

Ratio vs. Ratio • Emission Line Ratio diagrams • Visual tool for determining relationships & NLR parameters • Useful for diagnosing • exciting mechanisms • AGN NLR occupy • specific regions in • each diagram • Baldwin et al 1981 • Vellieux & Osterbrock • Many others. . . Brent Groves -NLR Models Kewley et al 2006 : astro-ph/0605681

Exciting Mechanisms • Photoionization • Excitation by UV & X-ray photons originating from the central source • Shock Ionization • Collisional Excitation from either a jet or winds arising from the central source Brent Groves -NLR Models

Hot on Photons • Controlling parameters • Gas Abundances, Z • Gas density, n. H • Ionizing spectrum, Lν • Column depth of model NH • Incident flux/ radius from central source/ ionization parameter Brent Groves -NLR Models

Hot on Photons • Simple power-law ionizing spectrum, range of n. H, and U • Good: • Can reproduce strong Ratios like [OIII]λ 5007/Hβ & [NII] λ 6583/Hα • Agrees with Lbol ∝L[OIII] • Bad: • Cannot reproduce both high- and low- ionization line strengths simultaneously • simple models ruled out (Stasińska 1984) Brent Groves -NLR Models

Shocking Results • Controlling Parameters • Gas Abundances, Z • Pre-shock gas density, n. H • Shock Velocity, Vs • Magnetic Parameter, B/n 1/2 • Bad: • Does not produce strong ratios • Good: • In some specific objects however. . . Brent Groves -NLR Models

Fitting Failures? Allen, Dopita, & Tsvetanov (1998) Brent Groves -NLR Models

Shockingly Fast? • Fast shocks (Vs > 150 km s-1) produce ionizing photons (Dopita & Sutherland 1995, 96) • Post shock gas cools producing photons which diffuse upstream & downstream • Upstream ionizes pre-shock gas • Combination of collisional & photoionization • Good: Correct Ratios • Bad: Velocity signatures Brent Groves -NLR Models

Fitting Failures? Allen, Dopita, & Tsvetanov (1998) Brent Groves -NLR Models

Multi-Component • Uses the combination of two or more photoionization models to reproduce both high and low ionization lines • Morganti et al. (1981) - Cen. A • Kraemer & Crenshaw (2000) • Komossa & Schulz (1997) Brent Groves -NLR Models

Multi-Component • Uses the combination of two or more photoionization models to reproduce both high and low ionization lines • Morganti et al. (1981) - Cen. A • Kraemer & Crenshaw (2000) • Komossa & Schulz (1997) • Murayama & Taniguchi (1998); • Nagao et al (2001) - HINER • Baskin & Laor (2005) - OIII • strength Murayama & Taniguchi (1998) Brent Groves -NLR Models

Multi-Component • Uses the combination of two or more photoionization models to reproduce both high and low ionization lines • Problem: Too many free variables • More Physically Constrained combinations • Combination of Matter bounded and Ionization bounded clouds (Binette, Wilson & Storchi-Bergmann) • Local Optimally-emitting Clouds (Ferguson et al. ) Brent Groves -NLR Models

A Matter of Bounding AM/I models use a varying contribution of Matter bounded and Ionization bounded clouds • Uses a set ionization parameter • IB clouds see absorbed ionizing spectrum • IB clouds also different pressure to MB clouds • Spectrum controlled by ratio of components, AM/I Brent Groves -NLR Models

A Matter of Bounding AM/I models use a varying contribution of Matter bounded and Ionization bounded clouds Binette, Wilson & Storchi-Bergmann (1996) Allen, Dopita, & Tsvetanov (1998) Brent Groves -NLR Models

LOC, stock & barrel. . . • Spectra dominated by selection effects we see the strongest emitting clouds for each line • Each line emits strongest near critical density • recreates linewidth-ne critical relation Ferguson et al (1997) Brent Groves -NLR Models

LOC, stock & barrel. . . Spectra dominated by selection effects - we see the strongest emitting clouds for each line • Determine emission line flux over range of parameters (U, n) • The total line flux is integral of cloud distribution function ψ(r, n) Ferguson et al (1997) Brent Groves -NLR Models

Dusty, Prad Dominated Clouds • At high U 0 (U 0≥ 10 -2 ), dust dominates the opacity • Hence, dust dominates radiation pressure • In an isobaric system Pgas gradient ≈ Prad gradient • Radiation pressure on dust dominates NLR cloud structure! Brent Groves -NLR Models

A Bit of Self Control • For U≥ 10 -2, density near IF dependent on incident U 0 • U ∝U 0, but U ∝ 1/n. H Ionisation Front Brent Groves -NLR Models

A Bit of Self Control • For U≥ 10 -2, density near IF dependent on incident U 0 • U ∝U 0, but U ∝ 1/n. H • Local U becomes independent of U 0!! • Self regulatory mechanism for constant emission lines!! Brent Groves -NLR Models

Dusty Work • Dust also: • Hardens the radiation field (cf AM/I models) • Competes with Hydrogen for ionizing photons Brent Groves -NLR Models

Dusty Work • Dust also: Bock et al. (2000) • Hardens the radiation field (cf AM/I models) • Competes with Hydrogen for ionizing photons • Increased Temperature through Photoelectric heating (ie ROIII increased) • Depletes Refractory Elements (Good and Bad) • Corresponding IR emission Brent Groves -NLR Models

Collapsing Curves • Models cluster in the region of observations Models fail with high ionization coronal lines Iron lines like Fe. VII overdepleted Groves, Dopita & Sutherland (2004) Brent Groves -NLR Models

Big Picture? Cecil et al. (2002) Brent Groves -NLR Models Dopita et al. (2002)

Big Picture • Individual NLR consist of many clouds • range of densities & pressure • possible abundance & dust variations • range of absorbing columns & variation of incident spectra • Best - Multi component clouds? Brent Groves -NLR Models

Big Picture • Individual NLR consist of many clouds • range of densities & pressure • possible abundance & dust variations • range of absorbing columns & variation of incident spectra • Best - Multi component clouds? • AGN surveys (ie SDSS) • Wider range of parameters • possible variation in incident spectra • Simpler one or tow component cloud • determine average properties and relationships • Contribution of Shocks. . . (Contini, Viegas & Prieto (2002, 04) Brent Groves -NLR Models

Sweet Separation Groves, Heckman & Kauffmann (2006) Kewley, Groves et al. (2006) Brent Groves -NLR Models

Sweet Separation Groves et al 2004 NLR 2 Zo model Dopita et al 2006 SB Mcl/Po=1 model Brent Groves -NLR Models

Heavy Metal Groves, Heckman & Kauffmann (2004) Brent Groves -NLR Models

Seeing the Unseen The Obscured EUV Continuum Martins, Viegas & Grünwald (1999) NGC 4151: Alexander et al (1999) NGC 1068: Alexander et al (2000) Brent Groves -NLR Models

When does it End? Brent Groves -NLR Models

A Model Ending • With the current NLR models we have a good physical Understanding of NLR and its appearance & emission lines • The models can be used to determine properties of the NLR and AGN (but Caution needed!) • Models are still limited. . . • but heading towards a full 3 D, dynamical picture of the NLR. . . • One step at a time! Brent Groves -NLR Models