Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.astro.spbu.ru/staff/resh/Lectures/lec8.pdf Äàòà èçìåíåíèÿ: Mon Jan 26 21:20:07 2009 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 23:35:11 2012 Êîäèðîâêà:

Galaxies at high redshifts

Expected proper ties of high-z galaxies

Why search for high-z galaxies? · Studying distant galaxies can provide impor tant constraints on galaxy formation theory. · One can obtain impor tant insight into physical processes at early epoch (e.g. star formation). · Cosmological tests.

To find a distant galaxy, one must have some idea of what one is searching for. · Surface density 105-6 galaxies/ o or 1 galaxy per (5 - 10 )2 (assuming constant space density ­ no mergers). Primeval galaxies should be very numerous. · Redshift of galaxy formation Two simple arguments: ­ Overlap argument The present-day ratio
distance between g alaxies diameter of a g alaxy

101 - 102 .


Expected proper ties of high-z galaxies
Distance between galaxies in expanding Universe changes as 1 1+z . Hence at z 10 - 100 galaxies must overlap (assuming constant physical diameters of galaxies). ­ Density argument According to spherical collapse model (e.g. Gunn & Gott 1972), self-gravitating, virialized dark halos form when their mean density exceeds 200 times the background at formation: > 200 cr = 200 0r (1 + z )3 . c MW: r = 8 kpc, V = 200 km/s = 10- Therefore, z (M W ) 15
23

Expected proper ties of high-z galaxies
· Angular sizes If the observable protogalaxy phase were the end product of the monolitic collapse of a massive cloud of gas, then the bulk of its star formation might occur in a small region kiloparsecs in size (1 kpc 0. 15 at z = 5). On the other hand, some models of galaxy formation predict that young galaxies will be lumpy in structure and extended in size, primarily due to the fact that in these models galaxy formation is a hierachical process. Such an object would have a size of perhaps 5 - 20 ( 10 - 100 kpc). One can resolve galaxy-size object at ANY redshift (10 kpc vs. z )

g/cm3 .

z 10 - 20 ­ epoch of galaxies formation ( 200 - 500 mln. years)

Expected proper ties of high-z galaxies

Expected proper ties of high-z galaxies
The highest redshift objects (circles ­ galaxies, crosses ­ QSOs)

· Luminosities Par tridge & Peebles (1967): "galaxies should go through a phase of high luminosity in early stages of their evolution. The estimated luminosity for a galaxy resembling our own is 3 â 1046 ergs/sec, roughly 700 times higher than the present luminosity." (MB -27m and M 24m - 25m ) · Spectrum Par tridge & Peebles (1967): up to (5 - 10)% of the bolometric luminosity may be emit° ted in Ly (W(Ly) 10 - 100A), rendering the line potentially detectable out to the highest redshifts.


Why is it so difficult to find high-z galaxies?

Why is it so difficult to find high-z galaxies?

· K -correction · Cosmological dimming (Tolmen effect) How does surface brightness depend on distance in a Euclidean static universe?.. In the expanding Universe Io z=1 z=5 µ 3.m 0, µ 8.m 0.
bs

=

Itrue . (1 + z )4

The expansion of the Universe provides astronomers with the benefit that recession velocities can be translated into radial distances. It also presents the challenge that sources observed at different redshifts are sampled, by any par ticular instrument, at different rest-frame frequencies. The transformations between observed and rest-frame broad-band photometric measurements involve terms known as "K -corrections". K (z ) = -2.5 lg [ L(1 ) (1 + z )], L(0 )

L(1 ) ­ intrinsic luminosity of the source, L(0 ) ­ observed luminosity (monochromatic correction in mag.).

Why is it so difficult to find high-z galaxies?

Why is it so difficult to find high-z galaxies?
Morphological K -correction:

To determine the appropriate K -corrections, the spectral energy distribution of the galaxy has to be convolved with the transmission function of the filter in the reast-frame and at redshift of the galaxy. This is a simple calculation once the spectrum of the object is known.


Identification of very distant galaxies
Spectral features of high-z objects

Identification of very distant galaxies
Spectral features of high-z objects

· Strong Ly emission Population synthesis models predict that a young, dust free, star-forming galaxy should show strong Ly emission with ° intrinsic W(Ly) 100 A. The flux in the Ly line is expected to be disrectly propor tional to the SFR and 3-6% of the bolometric luminosity are emitted in Ly. ° · An intrinsically flat spectrum for > 912 A Population synthesis models prediction: flat spectral energy distribution between the Lyman-limit and the Balmer-limit ° (3646A). This is due to UV-radiation from hot, shor t-lived (< 108 yr) massive stars.

· A pronounced drop of the SED at the Lyman limit, e.g. a complete ° absence of flux at wavelength below the Lyman limit ( < 912 A) ­ The feature forms in the stellar atmosphere of massive stars as a result of the hydrogen ionization edge and is quite pronounced, with a discontinuity of an order of magnitude. ­ The break in the stellar continuum is made more pronounced by the photoelectric absorption of the interstellar HI gas, which is abundunt in young galaxies, and also by intervening HI gas. ­ The UV spectrum of sources at high z is also subject to additional opacity owing to line blanketing by the intervening Ly forest that ° ° dims the continuum between 912 A and 1216 A by an amount that depends on z .

Identification of very distant galaxies
Search techniques

Identification of very distant galaxies
Search techniques

· Search for objects with prominent emission lines with narrow-band filters. This requires the detection of emission line objects at a wavelength at which the redshifted lines (Ly, H etc.) are expected. One uses narrow band imaging with a spectral resolution of a few handred to several thousand km/s. Subsequent spectroscopic observations are necessary to establish weather the detected emission line is in fact the high redshift line one was searching for.

· Search for objects with unusual broad-band colors.

Ly emitters (LAEs)

Lyman-break galaxies (LBGs)


Identification of very distant galaxies
Search techniques

Identification of very distant galaxies
Search techniques

Where to look for distant galaxies? ­ Fields in which high-z objects are already detected (clustering of galaxies) ­ Search in the surrounding of galaxy clusters (gravitational lensing) ­ Extremely deep fields: HDF-N, HDF-S, HUDF, SDF etc.

­ Serendipitous discoveries

Identification of very distant galaxies
Search techniques

Identification of very distant galaxies
Search techniques

Examples of spectra
All the methods does work! Current status: >50 LAEs at z > 5 (z = 6.96 ­ most distant) 5000 LBGs at 1500 LBGs at >500 LBGs at z >100 LBGs at z z4 z5 6 > 5 with spectroscopically-confirmed redshifts


Identification of very distant galaxies
Search techniques

LBGs

Vanzella et al. (2007): composite spectra of B , V , and i drop galaxies (the success rate of the redshift identifications to be 70%).

First modern searches for galaxies at high redshift using the Lyman-break technique ­ Guhathakur ta et al. (1990), Steidel & Hamilton (1992, 1993), etc. Verma et al. (2007) presented the proper ties of 21 LBGs at z 5 (spectra + 10-band photometry in the range 0.45-8 µm). Main results: typical SFR 40 M /yr, stellar mass 2 · 109 M , age < 100 Myr, size r1/2 1 kpc. On average, LBGs at z 5 are 10 times less massive and are significantly younger than LBGs at z 3. Progenitors of early-type galaxies or bulges?

LBGs

LBGs

Luminosity function of LBGs
Bouwens et al. (2007): 4671, 1416, and 627 B , V , and i dropouts (z 4, z 5, and z 6) in several deep fields of the HST. The LF parameters for the rest-frame LFs at different z .

Spatial distribution
Ouchi et al. (2004): 2600 LBGs with z=3.5-5.2 in the SDF.


LBGs

LBGs

Surface brightness profiles
Hathi et al. (2008) used the stacked HUDF images to analyze the average surface brightness profiles of z 4 - 6 galaxies (30 objects at z 4, 30 at z 5, and 30 at z 6). 0.2cm From these stacked images, they are able to study averaged radial structure at much higher signal-to-noise ratio than possible for an individual faint object.

Composite images for z 4, 5, and 6 (from left to right) objects. Each stamp is 1. 53 on a side.

LBGs

LBGs

Main conclusions: ­ The shape of the average surface brightness profiles shows that even the faintest z 4 - 6 objects are resolved. ­ The average surface brightness profiles display breaks at a radius that progresses towards lower redshift from r = 0. 27 (1.6 kpc) at z 6 to r = 0. 35 (2.5 kpc) at z 4. The radius where surface brightness profiles star t do deviate significantly from an r1/n profile might serve as a "virial clock" that traces the time since the onset of the last major merger, accretion event or global starburst in these objects. The limits to dynamical age estimates for the galaxies from their profile shapes ( 100 Myr) are comparable with the SED ages obtained from the broadband colors.


LAEs
The systematic search for Ly-emitting high-z galaxies for a long time been a business without the success expected from early predictions (e.g. Par tridge & Peebles 1967). First results ­ end of 90th (e.g. Hu et al. 1998 with the 10-m Keck telescope). Kodaira et al. (2003):

LAEs

Typical characteristics of LAEs
­ Compact ( 1 kpc) ­ SFR2­50 M /yr ­ Masses 109 - 1010 M ­ Luminosity function Ouchi et al. (2008) presented LFs of Ly emitters at z 3, 4, and 6 in a 1 deg2 sky of the Subaru/XMM­Newton Deep Survey Field (858 photometrically-selected candidates + 84 confirmed LAEs). They derived the LFs of Ly and ° UV-continuum ( 1500 A).

(a) z = 6.541, FWHM=4.4 kpc, SFR> 9 M /yr (b) z = 6.578, FWHM<4.4 kpc, SFR> 5 M /yr

LAEs

LAEs

Spatial distribution
Shimasaku et al. (2003): 43 LAEs at z = 4.86 in the SDF Shimasaku et al. (2006): 89 LAEs at z = 5.7 in the SDF

­ The apparent Ly LF shows no significant evolution beween z = 3 and 6. ­ The UV LF of LAEs increases from z = 3 to 6, indicating that galaxies with Ly emission are more common at earlier epochs. ­ The ratio in number density of LAEs to LBGs increases from z = 3 to 6: galaxies with Ly emission are more common at high z .


LAEs

LAEs
The survey has yielded six promising (> 5 ) candidate LAEs which lie between z = 8.7 and z = 10.2. All but one of the candidates remain undetected in deep HST optical images and lower redshift line interpretations can be excluded, with reasonable assumptions, through the nondetection of secondary emission in fur ther spectroscopy. At least two of the candidates are likely to be a z 9: 8.99 and 9.32.

LAEs at z 9?
Stark et al. (2007) presented new observational constraints on the abundance of faint high-z LAEs secured from a deep Keck nearinfrared spectroscopic survey which utilizes the strong magnification provided by lensing galaxy clusters at intermediate redshift (z = 0.2 - 0.5) (9 clusters). Stark et al. have under taken a systematic search for line emission in the J -band (1.143-1.375 µm) within carefully-selected regions which offer very high magnifications (10­50 times) for background sources with redshifts z 10.

LAEs

Summary

Main conclusions: ­ Assuming two or more of the LAE candidates are real, then the cumulative abundance of low luminosity galaxies (defined as those with L > 1041.5 erg/s) is at least 0.3 Mpc-3 . Such a large abundance of low luminosity LAEs suppor ts the contention of a steep faint end slope for the star-forming luminosity function at z 10. ­ The first glimpse at the z 10 Universe suggests that low luminosity star-forming galaxies contribute a significant propor tion of the UV photons necessary for cosmic reionization.

z 5 galaxies:
­ compact (1­5 kpc) ­ asymmetric ­ high rest-frame surface brightness and luminosity (µ0 (B ) 18m / , L L ) ­ young ( 100 Myr) ­ low mass ( 109 - 1010 M ) ­ high SFR ( 101 - 102 M /yr) ­ spatial density spatial density of bright galaxies at z = 0 ­ evidence of LSS


Summary

Summary
Some impor tant questions (Schaerer 2007):
How do different high-z populations such as LAE and LBG fit together?

These characteristics are very strongly biased by the selection procedure itself, and it is therefore unclear to what extent they reflect actual proper ties of all objects located at z 5. The observed objects can be "building blocks" that later merge and accrete the surrounding matter to form the galaxies we now know in our vicinity. On the other hand, some of these objects can represent bulges of massive spirals under formation or elliptical galaxies.



Are there other currently unknown populations? What are the evolutionary links between these populations and galaxies at lower redshift? What is the metallicity of the high-z galaxies? Where is Pop III? What is the star formation history of the universe during the first Gyr after Big Bang? Are there dusty galaxies at z 6? How, where, when, and how much dust is produced at high redshift? Which are the sources of reionisation? And, are these currently detectable galaxies or very faint low mass objects? What is the histor y of cosmic reionisation?

Summary

Summary

Useful literature