The Nature of Faint MIPS24 Sources - Cornell Universityastrosun2.astro.cornell.edu/academics/courses/a671... · 2006. 4. 17. · Lagache et al., 2003. A671 Nature of Faint MIPS24

The Nature of Faint MIPS24 Sources

Astronomy 671Jason Marshall

A671 Nature of Faint MIPS24 Sources 2

Infrared/Sub-mm GalaxiesDeep IR/Sub-mm surveys have revealed a population of IR-luminous galaxies at intermediate and high redshift

ISO @ 15 µmSCUBA @ 850 µm

Number counts from these surveys indicate an order of magnitude excess of these sources than would be obtained if the IR galaxy luminosity function were constant with redshiftRapid increase in comoving#density from z=0 to z~1 Lagache et al., MNRAS, 338, 2003

A671 Nature of Faint MIPS24 Sources 3

IR Galaxy SEDs and 24/850 µm Counts

MIPS 24 µm counts at 1<z<3 integrate over PAH features and small (hot) dust grain continuum

SCUBA 850 µm counts integrate over the far-IR thermal dust continuum

z=3z=0

A671 Nature of Faint MIPS24 Sources 4

Is the z=1 Turnover Real?

ISO number counts indicated a rising luminosity function from z=0 to z=1, followed by a turnoverIt is unclear if this turnover is real, or if it is biased by the 7.7µm PAH feature which is centered in the bandpass at z=1

A671 Nature of Faint MIPS24 Sources 5

Spitzer IRAC/MIPS Observations of ELAIS-N1

Part of GOODS programIntegration of 4620 s “per detector pixel” (?)165 arcmin2 total area, a portion of which…

Gray-scale is the MIPS24 image (stretch in µJy)Dots represent IRAC counterparts—more on that later

SExtractor (Source Extractor) was used to identify “sources”

Chary et al., ApJS, 154, 2004

A671 Nature of Faint MIPS24 Sources 6

24 µm Galaxy Counts & Completeness Correction

Monte Carlo simulations were used to determine “completeness”

Artificial sources (of varying flux levels) were added to the MIPS imageSExtractor was re-run to determine the rate of detection as a function of source brightnessThis is used to determine the correction factor for each flux bin

Completeness-corrected counts and models are plotted in the top panel

Chary et al., 2004

A671 Nature of Faint MIPS24 Sources 7

Quick Primer on IR Number Count Models

The model of Lagache et al., 2003 in brief…Flux at wavelength λ0 of a source at redshift z with luminosity L:

Number of sources per solid angle and redshift interval:

The differential counts at flux S is then:

The goal is to build models of N0 (number of sources per unit volume and luminosity interval as a function of redshift) that reproduce the observed galaxy counts

k is the k-correction

dV/dz is the differential volume element fixed by the cosmology

A671 Nature of Faint MIPS24 Sources 8

Quick Primer on IR Number Count Models

Use starburst and normal galaxy template SEDs to calculate k-corrections and come up with functional forms that re-produce the observations (figure on the right):

Lagache et al., 2003

A671 Nature of Faint MIPS24 Sources 9

Comparison with Simulated Images

Center image is the MIPS24 observed dataLeft and right images are simulations based on the number count models

Left one has too many bright sourcesRight one has too many faint sources

Chary et al., 2004

A671 Nature of Faint MIPS24 Sources 10

Comparison of the MIPS24 Counts with Models

All models are too high at bright flux densities (S > 500 µJy)

This isn’t really resolved yet (apparently), but is a real effect…perhaps the ISO data on which the models are based was noisy and thus erroneous

Some models seem to work better than others at lower flux densities

Chary et al., 2004

A671 Nature of Faint MIPS24 Sources 11

Predicted Redshift Distribution of MIPS24 Sources

The models are used to predict the redshift distribution of galaxies

Bimodal distribution is a result of the 8 and 12 micron PAH features redshiftinginto and out of the 24 micron bandLIRGs are 60%ULIRGs are 10%The rest are normal

Chary et al., 2004

A671 Nature of Faint MIPS24 Sources 12

NIR Counterparts to MIPS24 Sources

“All” MIPS24 sources have IRAC counterpartsNature of the sources…

Chary et al., 2004