Slide 1

The Cosmic Near-Infrared Background: Remnant light form early stars Journal Club talk 3.12.2010 A. B. Fry Ferdinand & Komatsu 2006 (F&K06) also Ferdinand & Komatsu et al. 2010 (F&K10)

Slide 2

The Cosmic Near-Infrared Background: Remnant light form early stars The cosmic infrared background radiation is the diffuse light from faint galaxies that remains after Milky Way and zodiacal light is subtracted. The near-infrared (1-3 um) background light (NIRB) from redshifted stars at z~10 contributes to this backgorund. The NIRB offers invaluable information regarding the physics of cosmic reionization that is difficult to probe by other means. Fernandez & Komatsu explore the uncertainty in the background intensity from these stars due to metallicity, the mass spectrum, and other parameters. The authors show that contrary to previous results that the stellar component of the NIRB could come from stars with some metals (Z=1/50 solar). A follow up paper examines intensity fluctuations of the mean NIRB which could further constrain high-z galaxy populations.

Slide 3

Slide 4

The cosmic infrared background radiation (CIBR) is the diffuse light from faint galaxies that remains after Milky Way and zodiacal light is subtracted.

Slide 5

http://www.astro.ucla.edu/~wright/CIB R/ The black data points between 1 and 300 microns on this graph come from the DIRBE experiment on the COBE satellite. The red data points are from Wright, E.L. 2001 (ApJ, 553, 538) which use a different zodiacal light model than the one used by Hauser et al. (1998, ApJ, 508, 25). The blue lower limit symbols are based on integrating galaxy counts, while the purple upper limit symbols are based on limits on photon-photon collisions from gamma-ray astronomy. The black data points at wavelengths shorter than 1 microns come from Dube, Wickes & Wilkinson (1979, ApJ, 232, 333), Toller (1983, ApJL, 266, 79), and Hurwitz, Martin & Bowyer (1991, ApJ, 372, 167). The curve is the Lambda CDM model with the Salpeter IMF from Primack et al., multiplied by a factor of 1.84, and with modifications for wavelengths longer than 300 microns to fit the FIRAS distortion limits, and for wavelengths shorter than 0.8 microns to fit the optical and UV data.

Slide 6

Lets focus on the near infrared (1-3 m) background radiation (NIRB) where redshifted ultraviolet light from early stars at z~10 contributes.

Slide 7

The observed NIRB seems too large to be accounted for by the integrated light from galaxies, it could come from early stars

Slide 8

For example, suppose that most reionization occurred at z=9. Then ultraviolet photons (~1000 ) produced at this redshift during reionizaiton will then be redshifted to the near infrared regime (~1m).

Slide 9

The NIRB offers invaluable information regarding the physics of cosmic reionization that is difficult to probe by other means.

Slide 10

In F&K06 they discuss the simplified physics of the NIRB, explore different metallicities and initial mass spectra of the first stars, and provide a relation to the NIBR and star formation rate. They predict the average intensity at 1-2 m (units of nW m -2 sr -1, just like previous plots) which is a function of the mass spectrum of early stars, the star formation rate, metallicity. Etc.

Slide 11

The mean background intensity is computed in terms of the volume emissivity, p( ,z ), which is a function of the mass spectrum of early stars, the star formation rate, metallicity. Etc. The background intensity* (Peacock 1999, p. 91): * Redshift affects the flux density in several ways Photon energies and arrival rates are redshfited reducing the flux density by a factor of (1+z) 2 The bandwidth d is reduced by a factor of 1+z so the energy flux per unit bandwidth does down by one power of 1+z The observed photons at frequency 0 were emitted at a frequency 0 (1+z) si the flux density is the luminosity at this frequency divided by the total area divided by 1+z

Slide 12

p( ,z ) is the volume emissivity in units of energy per unit time per unit frequency and unit commoving volume: The sum over takes into account the various radiative process contributions to the emissivity

Slide 13

p( ,z ) is the volume emissivity in units of energy per unit time per unit frequency and unit commoving volume: p * is the continuum emission form the stars themselves p line is the emission from recombination lines p cont is free-free and free-bound continuum emissions p 2 is the two-photon emission

Slide 14

p( ,z ) is the volume emissivity in units of energy per unit time per unit frequency and unit commoving volume: The dimensionless quantity represents a ratio of the mass-weighted average total radiative energy to the stellar rest-mass energy in a unit frequency interval

Slide 15

p( ,z ) is the volume emissivity in units of energy per unit time per unit frequency and unit commoving volume: d * /dt is the mean star formation rate at the redshift of interest in units of M yr -1 Mpc -3. It is very uncertain! L is a time averaged luminosity for the radiative process (m) is the stellar main-sequence lifetime

Slide 16

p( ,z ) is the volume emissivity in units of energy per unit time per unit frequency and unit commoving volume: Finally, f(m) is the mass spectrum. They use three different versions

Slide 17

Salpeter Larson: Top-heavy:

Slide 18

From 6 to 8 M , the O/Ne/Mg core of the star collapses, or the star ejects its outer envelope, leaving a white dwarf or neutron star. From 8 to 25 M , the iron core collapses, the star explodes as a supernova, and a neutron star is left as a remnant. A significant amount of metals are ejected. From 25 to 40 M , there is a weak supernova and a black hole is created by fallback. The amount of metals that are ejected into the IGM decreases sharply, leaving most of the metals locked in the black hole. From 40 to 100 M , the star directly collapses into a black hole. The only metals produced are from mass loss during the stars life. From 100 to 140 M , a pulsational pair instability supernova results. This ejects the outer envelope of the star, and then the core collapses into a black hole. Metals in the outer envelope pollute the IGM. From 140 to 260 M , a pair instability supernova results, which completely disrupts the star and leaves no remnant. All the metals are ejected into the IGM. Above 260 M , the star collapses directly into a black hole, and there is no enrichment of the IGM. The Fate of Massive Stars

Slide 19

Slide 20

The previous equations 1-3 allow us to predict the average intensity (in units of nW m -1 sr -1 ) at 1-2 m:

Slide 21

Equations 1-3 allow us to predict the average intensity (in units of nW m -1 sr -1 ) at 1-2 m:

Slide 22

3 to 1 m corresponds to.414 to 1.24 eV The spectrum of the NIRB

Slide 23

The dependence on the initial mass spectrum f(m) is such that heaver mass spectra tend to give higher background intensities. For metal- rich stars Ly emission dominates. For metal-poor stars there is a significant contribution form the stars themselves. The predicted sensitivity is not sensitive to stellar metallicity The spectrum of the NIRB

Slide 24

Metallicity changes the hardness of the stellar spectrum, it affects the ratio of energy in the Ly and two-photon emission to stellar emission energy: the harder the spectrum is the more ionizing photons are emitted and thus the more the Ly and two photon-emission energies. The spectrum of the NIRB

Slide 25

F&K06 find that a population of metal-poor stars do not overproduce metals that we observe in the universe today, except for the Larson mass function upper 1 value for the star formation rate.

Slide 26

The uncertainty in measurements of the NIRB are massive. They vary from 2-50 nW m -2 sr -1 at 1- 2 m including the upper and lower 1 bounds. The mean SFR d * /dt is constrained to.3-12 M yr -1 Mpc -3.

Slide 27

Measuring the absolute value of the NIRB is very difficult due to systematic uncertainty in zodiacal light subtraction, however, fluctuations could be measured without this absolute calibration!

Slide 28

F&K 2010 et al. present calculations of the power spectrum and metallicity/initial-mass- spectrum dependence of the NIRB fluctuations, as well as dependence on the star formation efficiency and the escape fraction of ionizing photons. Fluctuations

Slide 29

Slide 30

F&K 2010 et al. present calculations of the power spectrum and metallicity/initial-mass- spectrum dependence of the NIRB fluctuations, as well as dependence on the star formation efficiency and the escape fraction of ionizing photons. Fluctuations

Slide 31

Luminosity-density power spectrum of halos with Pop II stars with an initial mass spectrum, f esc = 0.19, and f* = 0.5, assuming a rectangular bandpass from 1 2 m.

Slide 32

The luminosity-density power spectra of halos are approximately power-laws over the entire range of wave numbers that the simulation covers. The clustering of halos is highly non-linearly biased relative to the underlying matter distribution. The growth of the power spectrum is partly driven by the growth of linear matter fluctuations as well as that of halo bias. Fluctuations

Slide 33

The NIRB intensity contribution from early stars is essentially determined by the mass-weighted mean nuclear burning energy of the stars and the cosmic star formation rate. The intensity is not sensitive to stellar metallicity. Variations in the NIRB can tell us details about the first stars, reionization, and the high redshift host galaxies/IGM. The amplitude of the (observable) angle power spectrum of the mean NIRB is determined by f * while the angular power spectrum of the IGM can probe the ionization history. Conclusions

Slide 34

Image credit: Robert Hurt, SSC, JPL, CalTech, NASA The first stars may have lighted up the cosmos within 200 to 400 million years after the Big Bang, and then clustered together into what later became galaxiesSSCJPLCalTechNASA

Slide 35

*A note about the escape fraction for Ly which varies widely in the literature: the fraction of ionizing photons escaping the nebula does not alter the Ly luminosity very much because all of the ionizing photons will eventually be converted to Ly photons that in turn will escape freely via the cosmological redshift therefore these predictions should be free from uncertainty in the escape fraction.

Slide 36

The amplitude of Cl is, among other things, a sensitive probe of the nature of high-z galaxies.

Slide 37

More

Slide 38

The authors find if the NIRB has a stellar origin metal-free stars are not the only explanation of the excess NIRB; stars with significant metals (e.g., Z = 1/50 Z ) can produce the same amount of background intensity as metal- free stars. We predict * / ~48nWm 2 sr 1, where * is the mean star formation rate at z = 715 (solar masses per year per cubic megaparsec) for stars more massive than 5 solar masses While the star formation rate at z = 715 inferred from the current data is significantly higher than the local rate at z