The Origin and Structure of Elliptical Galaxies Structure formation by hierarchical merging. Spiral galaxies: merging of small, gas-rich substructures. Elliptical Galaxies: merging of equal-mass substructures (Toomre & Toomre 1972). Spiral galaxies Elliptical Galaxies (Moore 2000)

The Origin and Structure of Elliptical Galaxies Collisionless merger Gas-rich merger ( Hernquist 92, 93, Naab & Burkert 93) ( Bekki 99; Mihos & Hernquist 96; Springel et al. 05) ( Naab & Burkert )

Testing the Merger Hypothesis The merger hypothesis has been investigated numerically (Gerhard 81;Barnes 92; Hernquist 92,93; Mihos & Hernquist 93,96) Remnants spheroidal and follow a de Vaucouleurs law slowly rotating and anisotropic v (Binney 82) Anisotropy: v

Violent Relaxation of Collisionless Particle Systems CDM simulation DM density profile NFW (Hetznecker & Burkert)

Isophotal Shapes More insight requires a more detailed investigation of departures from pure elliptical isophotes Boxy and disky ellipticals: boxy: disky: (Bender 1987)

Disky and Boxy Ellipticals Boxy and disky ellipticals form two distinct groups with different kinematical properties (Bender et al. 88,89). Violent relaxation cannot explain these observations. Can the merger scenario explain this dichotomy? boxy disky

Initial Conditions High-resolution merger simulations with mass ratios 1:1, 2:1, 3:1, and 4:1 (Naab & Burkert 99 - 06). Spirals consist of disk, bulge and dark halo. face on Edge on

Anisotropy of Merger Remnants versus v Equal mass mergers are anisotropic and boxy 3:1 and 4:1 remnants are fast, isotropic rotators and disky 2:1 remnants show intermediate properties 1:1 2:1 3:1 4:1 Anisotropy and isophotal shape depends on the mass ratio.

The Frequency of Boxy and Disky Ellipticals The ratio of boxy-to-disky ellipticals should not depend on luminosity Khochfar & Burkert 03

The Frequency of Boxy and Disky Ellipticals The ratio of boxy-to-disky ellipticals should not depend on luminosity observations theory Khochfar & Burkert 03

Spheroidal and Mixed Mergers The merger fraction between galaxies of different morphologies depends strongly on redshift and mass (Khochfar & Burkert 03). E-E E-E 1 2 3 4 -22 -20 -18 z Massive boxy ellipticals might result from E-E mergers.

Spheroidal Merger Simulations E-E merger simulation 1 S-S 3:1 E-E -1 -2 (Naab, Khochfar & Burkert 05) E-E merger remnants are boxy and very anisotropic even for large mass ratios of 3:1 or 4:1.

More than 20% gas accretion Assumption 1: E-E mergers boxy Assumption 2: More than 20% gas accretion disky BH feedback Assumption 3: Star formation suppressed for (Naab, Khochfar & Burkert 05)

Energetic Feedback by Central Black Hole (Springel et al. 05)

Kinematics of Merger Remnants The main difference between disky, unequal mass merger remnants and boxy, equal-mass mergers is not their anisotropy but their rotation. (Burkert & Naab 05)

Kinematics of Merger Remnants (Burkert & Naab 05)

SAURON Integral Field Kinematics (De Zeeuw et al. 92) M 31 Intrinsic properties from axisymmetric three-integral Schwarzschild models.

SAURON Integral Field Kinematics (Cappellari et al. astro-ph/0509470) boxy disky 0.6 0.4 0.2 Disky ellipticals are anisotropic with high ellipticities. Boxy ellipticals are isotropic and round.

boxy disky 0.6 0.4 0.2 (Burkert & Naab 06)

What‘s about early-type mergers? prolate oblate

Summary The major merger scenario can explain the origin of fast rotating, disk ellipticals equal-mass merger slowly rotating, boxy ellipticals unequal-mass merger All ellipticals show a large spread in anisotropy. Ellipticals with assembled from early-type mergers. Central BH feedback suppresses star formation for Major mergers are not consistent with the structure of ellipticals as predicted from Schwarzschild models.

The End

The Merging History of the Universe Extended Press-Schechter theory (Kauffmann & White 1993; Khochfar & Burkert 2000, 2001, 2002) Frequency of minor or major mergers. Redshift dependence Final mass dependence N z Disky : Boxy = 5 : 3, independent of z and M

The Skewness of Elliptical Galaxies Bender, Saglia & Gerhard (94) studied the line-of-sight velocity distribution of elliptical galaxies. They detected small deviations from a Gaussian distribution that can be characterized by Gauss-Hermite-functions. > 0: shift to negative v < 0: shift to positive v

Rotation of Merger Remnants v versus 1:1 2:1 3:1 4:1 Equal mass mergers rotate slowly. 3:1 and 4:1 remnants rotate fast. 2:1 remnants show intermediate rotation Rotation depends on the mass ratio.

The Skewness of Elliptical Galaxies Bender, Saglia & Gerhard (94) studied the line-of-sight velocity distribution of elliptical galaxies. > 0: shift to negative v < 0: shift to positive v Ellipticals show a distinct correlation between their local and . 3 h v (Pickney et al. 03)

The Skewness of Elliptical Galaxies Bender, Saglia & Gerhard (94) studied the line-of-sight velocity distribution of elliptical galaxies. > 0: shift to negative v < 0: shift to positive v Ellipticals show a distinct correlation between their local and . 3 h v (Bender et al. 94)

The Skewness of Collisionless Merger Remnants 2 dimensional analysis of the LOSVD on a grid superimposed on the merger remnant. Collisionless merger major axis v > 0 v < 0 v

The Role of Gas in Major Mergers

Gas-rich Merger Simulations Formation of a gaseous disk.

Evidence for Gaseous Mergers 2 dimensional analysis of the LOSVD on a grid superimposed on the merger remnant. Gaseous merger v > 0 v < 0 major axis v > 0 v < 0 v

The Early-Type Angular Momentum Problem Faint ellipticals rotate faster than the fastest simulated remnants. v Gas cannot solve this problem (Naab & Burkert 04) Angular momentum problem for early-type galaxies. (Cretton, Naab, Burkert, Rix 2000)

Conclusions Elliptical galaxies formed by major mergers. Kinematics and isophotal shape of merger remnants is determined by their mass ratio: 1:1-2:1 anisotropic + boxy 3:1-4:1 fast rotating + disky. Low luminosity ellipticals experienced substantial gas inflow after the last major merger disky 2-component systems with a disk embedded in a spheroidal component Their LOS velocity distribution also indicates that the progenitors were gas-rich spiral galaxies. Very luminous, boxy ellipticals might have formed from gas-poor, early-type mergers.

Statistics of Boxy and Disky Ellipticals The CDM-model predicts an overabundance of disky ellipticals which should be independent of mass. Disky : Boxy = 5 : 3 log I This ratio agrees with observations of intermediate- and low mass ellipticals. Properties of ellipticals with Boxy More massive than disk galaxies High metallicities and enhanced (Mehlert et al. 00,03) log Formation redshift:

Projection Effects Projection effects lead to a large scatter but do not change the result. v v v v

The central stellar disk in NGC 3623 (de Zeeuw et al. 02) I (Naab & Burkert) 3:1 merger

Central Properties of Merger Remnants Sauron data M32 with gas 3:1 merger simulation I

Luminous ellipticals did not form through Additional Problems Luminous ellipticals: are more massive than disk galaxies formed at high redshifts have high metallicities (Mehlert et al. 03, Thomas et al. 03) Luminous ellipticals did not form through major mergers

The End

Problem 2: Anisotropy 1:1 2:1 3:1 4:1 (Naab & Burkert 2003)

Problem 1: Isophotal Shapes 20% bulge 10% bulge no bulge with gas (Naab, Burkert 2003)

Projection effects Projection effects lead to a large scatter but do not change the result. v v v v Violent relaxation alone cannot explain this dichotomy.

Spiral 1:1 Merger E-E Merger Gas-rich 3:1 merger

Elliptical Galaxies The de Vaucouleurs law: Cluster galaxies Collaborators: Th. Naab, S. Khochfar, R. Jesseit, M. Wetzstein The de Vaucouleurs law: Cluster galaxies Spheroidal stellar systems. No star formation.

Elliptical Galaxies The Sersic Law Cluster galaxies Collaborators: Th. Naab, S. Khochfar, Roland Jesseit The Sersic Law Cluster galaxies Spheroidal stellar systems. No star formation. n -12 -24 1 10 Graham et al. 02 Trujillo & Burkert 04

Kinematical Properties Massive elliptical galaxies are anisotropic, triaxial systems (e.g. Binney 82; de Zeeuw 97; Cretton et al. 00) Ellipticity distribution Anisotropy v (Binney & Tremaine 87) (Binney 82) v Anisotropy:

The Cosmological Angular Momentum Problem Major mergers are required to produce extended galactic disks. (D‘Onghia & Burkert, astro-ph/0402504) bulgeless disks (van den Bosch et al. 01) minor mergers The merging history of Es cannot be studied in a cosmological context.

The Formation of Elliptical Galaxies

Collisionless Merger Simulations Equal-mass merger Analysis of the remnant based on the same methods as used by observers.