Conference (website) at Stockholm University

(Day 1) Wednesday, 13 August 2014

Mellema – Introduction

• Photon Underproduction Crisis (Kollmeier+ 2014)
• Counting photons: the SF efficiency and UV escape fraction are degenerate, and functions of halo mass and time.
• Topology of ionized regions from 21-cm can indicate which sources are responsible for reionization.

Gnedin – Cosmic Reionization on Computers (CROC)

• Star formation and feedback (subgrid models) are the biggest problems in reionization and galaxy formation simulations.
• Note: In a star-forming cloud, 90% of the ionizing photons will be produced before the first SN occurs.
• Range of “escape fractions”: from the star forming core, molecular cloud, neutral ISM, outflows, and halo gas
• (Becker+ 2014) Gunn-Peterson trough optical depths at several redshift ranges in z = 5-6. The CROC simulations match these observational constraints.
• Gas fractions: match results from Okamoto+ (2008) after reionization.
• Molecular fractions: All halos that are photo-evaporated (< 3 x 10^8 Msun) contain no molecular gas.

Mostardi Rehagen – A Critical Test of LyC Leakage in Star-forming galaxies at z ~ 3

• (Mostardi+ 2013) Narrow band searches for LyC emission.  Out of 48 LBGs, 4 had LyC detected.  7 out of 90 LAEs have LyC detected.
• LyC escape fraction (measured from fluxes): 1-2% (LBGs) and 5-15% (LAEs)
• Comoving LyC emissivity = 9-15 x 10^24 erg/s/Hz/Mpc^3
• HST follow-ups are underway, and they confirm that there is no containment from background galaxies. The colors indicate that they have young stellar populations in a dusty environment.

Zackrisson – Probing LyC leakage from galaxies in the reionization epoch

• (Zackrisson+ 2013) Utilize the fact that LyC absorption will be converted into nebular emission.
• Above the Balmer break, the nebular emission lines increase with decreasing escape fraction.
• Dependency on f_esc in the UV slope vs. the EW of the H-beta line. See Shimizu+ (2014) for application of this model to SPH simulations.
• Dust has a large effect on this relationship and may need to be calibrated.

Vanzella – Investigating the ionizing escaping radiation at high redshift (z>3)

• Out of 100 LBGs, there is one LyC detection. f_esc > 50%, no Ly-alpha line.
• However, there are a few other candidates with possible LyC leakage.
• Ongoing spectroscopic search for z = 6-7 galaxies. Expect 100 z ~ 6 galaxies and 150 z ~7 galaxies.
• Detected an extremely blue galaxy at z = 6, beta = -3.08 ± 0.26, M_UV = -18.9.
• The blue galaxies have stronger Ly-alpha lines from an analysis of stacked spectra.
• (Vanzella+ 2014) Frontier Fields: Detected z = 6.39 LAE with a magnification factor of 17.

Diaz – The environment of CIV absorption systems in the post-reionization Universe

• (Becker+ 2014) QSO embedded in a 100 cMpc HII region, but the line of sight reaches a region with a large scatter in optical depth at z = 5.7.
• (Diaz+ 2014) Line of sight distribution of LAEs and found 3 within 1 cMpc of the CIV absorption.
• (Steidel+ 2010) Statistical study of impact parameters of CIV systems within 100 kpc.
• (Diaz+ in prep) How can CIV and CV exist at such large distances from the QSO?

Bergvall – Lyman continuum escape and the local starburst population

• Tadpole galaxies may have higher escape fractions because the IGM wind can photo-evaporate or “push” the gas away from the stars in the galaxy.
• Searched for LyC emission at z ~ 0.3 and found 4 galaxies at >3% leakage. They have extremely blue colors.
• The typical stellar ages in these galaxies are 50-100 Myr. The galaxies with escaping photons favor smaller mass (10^9 – 10^10 Msun) galaxies.
• The low detection of local LyC leakers may be an observational bias that favors stellar ages that are too low to allow outflows to develop.

Fernandez – Modeling Infrared Observations of High Redshift Galaxies to Determine the Escape Fraction

• Using the Near-IR background to constrain the sources of reionization, and applying an analytical model of the NIR to N-body simulations (Fernandez+ 2010).
• Fractional anisotropy: dI/I = sqrt( l (l+1) C_l / (2 pi I^2) ) is useful to measure the properties of the reionizers (Fernandez+ 2012).
• Halos with large f_esc values are dimmer than ones with low f_esc because of nebular emission lines. Dust affects the fractional anisotropy at wavelengths 10-100 µm.
• (Fernandez+ 2013) Also using cross-correlations between 21-cm and NIRB observations to constrain the SFR. The strongest anti-correlation occurs at 50% ionized.

Leitherer – Pushing the Cosmic Origins Spectrograph to the Lyman Limit

• Observed Tol 0440-381, Tol 1247-232, and Mrk 54 with FUSE and then followed up with COS.
• Tol 1247 has some very high ionization lines, indicating a large amount of massive star formation.
• The galaxies have escape fractions (including dust) are 1.5%, 1.7%, and 0.5% for the galaxies, respectively. However, the standard COS pipeline over-predicts the continuum, which results in an increase in f_esc by a factor of ~2.
• New stellar models that include better Wolf-Rayet (WR) models and rotation produce more LyC emission. Metallicity has little effect on this emission, though.

Borthakur – A Local Clue to the Reionization of the Universe

• (Veilleux 2005) Review on outflows, which are important in creating bubbles and channels in which LyC can escape.
• Searching for Lyman-alpha analogs to LBGs in the local Universe. They have high sSFRs (~10 Gyr^-1) and compact with ~10^9 Msun of stars within 200 pc. They have stellar masses between 10^10.5 to 10^11 Msun and SFRs around 50 Msun/yr.
• They have an SED that is consistent with an instantaneous burst with an age of 3 Myr with WR features. In a BPT plot, they lie to the left of the star forming sequence in the [SII]/H-alpha measure.

Fremling – The Escape of Ionizing Radiation from Local Starburst Galaxies

• Using a resolved extinction map of Tol 1247-232 to calculate an intrinsic H-alpha emission map that then allows for the calculation of the ionization photon production rate. They also investigate the impact of dust absorption.
• Measure 5% and 50% of ionizing photons are absorbed in Tol 1247-232 and Haro 11, corresponding to an escape fraction of 1% and 9%, respectively.

Lee – A New Look at Diffuse Ionized Gas in Dwarf Galaxies

• Hints of a deficiency in dwarfs in previous studies when comparing H-alpha and SFR, where the former is commonly used as a proxy for SFR. Could there be H-alpha emission below the detection limit? Use deeper imaging to resolve this issue.
• (Lee, J.C. 2009, 2011) Showed there is a steeper slope (or a break) in the H-alpha/SFR correlation below ~0.01 Msun/yr.
• What’s causing this break? Non-standard IMF? IMF is not fully populated? Bursty SF? Metallicity effects in stellar models? Leakage of LyC? Undetected diffuse ionized gas?
• Using a 6m telescope instead of a 2.3m one, they resolve the diffuse ionized gas into individual HII regions and a diffuse H-alpha emission outside of the star forming regions.

Keating – The role of environment in the near-zones of z ~ 7 QSOs

• Using zoom-in simulations with RT post-processing (RADAMESH) to compute the ionization state near QSOs.
• Simulate three different regions (normal, intermediate, and overdense). They also vary the neutral fraction in simulation, where they calculate the propagation of the I-front. The HII regions do not depend on the environment, but it is highly dependent on the neutral fraction.
• Trying to match the near-zone reaction in the Mortlock z ~ 7 QSO, where the spectra and near zone size agrees better with a more neutral environment.

(Day 2) Thursday, 14 August 2014

Finlator – Metal Absorbers in Inhomogeneous Reionization Models

• What do metal absorbers teach us about galaxies, reionization, and the UVB? (e.g. Kulharni+ 2013; Oh 2002)
• Need to account for the local UV and metallicity field around galaxies to calculate accurate absorption systems from simulations.
• Assumes an escape fraction decreasing with halo mass and increasing with redshift in the RadHydro simulations.
• (Finlator+ 2013) Marginal agreement with the observed OI abundance, comparing with Becker+ (2011). The CII/CIV ratio is sensitive to the UVB (slope and normalization).
• Lower CII and higher CIV when taking into account spatial variations in the UVB instead of a uniform one.
• CII is weakly underproduced when compared to observations of Becker+ (2011) and D’Odorico+ (2013). Could be solved by 5-10% of the outflows being cold and dense.

Siana – Deep HST Searches for Lyman Continuum from Galaxies at z~2–3

• Dwarf galaxies (M_V > -18) emit the majority of the UV emissivity at z > 2. What is the escape fraction? Currently unobservable. “Bursty” star formation complicates the interpretation of observables.
• Focusing on 3 LBGs at z ~ 3.1 with HST in LyC and with Keck in Ly-alpha. Finding LyC emission in galaxies that were offset from the LBGs. Follow-up with Keck NIR Spectroscopy, and found that they were contaminated by foreground galaxies! 🙁 Leads to the conclusion that massive galaxies do not dominate the UVB.
• (Alavi+ 2014) Galaxy LF at z ~ 2, steep faint-end slope and no sign of turnover at M_1500 = -13. Around 60% of the UV emissivity is produced by dwarf galaxies.
• No dwarf galaxies (out of 12) behind Abell 1689 with detected LyC emission.

Faisst – Constraints on re-ionization at z~8 using Lyman-alpha emitters

• Probing reionization by searching for the clustering of LAEs at z ~ 8.
• No luck with detecting the brightest LAE candidates at z > 7.5
• Suggests a neutral fraction of >50% at z > 7, in tension with CMB measurements? However there are many unknowns in the conversion from X_Lya to x_HI.

Oey – Constraints on Lyman Continuum Optical Depth from Ly-alpha, C II, and C II*

• Green Pea Galaxies (Gardamone+ 2009): High sSFRs (100 Gyr^-1), strong emission in [OIII] and suppressed [OII] emission.
• Detecting low-ionization states CII and CII* in these objects, which can be correlated to Ly-alpha emission (Shapley+ 2002, Prochaska+ 2011).
• (Jaskat & Oey 2014) Ly-alpha, CII, SiII, CII*, SiII* probe the neutral gas optical depth and geometry.

Micheva – New analysis of LyC leaking z~3 galaxies in the SSA22 field

• Large LBGs are more often LyC leakers.
• There are 36 LyC candidates at z ~ 3 out of ~200 galaxies.

Iwata – Constraints on LyC escape fraction from direct observations of z~3 galaxies

• After removing foregrounds, 4-5% of LBGs and LAEs show strong LyC emission.
• Some LyC sources have very large f_LyC / f_LyA intrinsic ratios.
• LyC often comes from spatially often sub-clumps that may be infalling into larger galaxies.
• Escape fraction could be bimodal with fractions at zero and unity. Line of sight down the barrel of an ionized column?
• There are 9 genuine LyC detections in the SSA22 field, suggesting an escape fraction of ~9% if the distribution is bimodal. Should detect ~50 LyC objects with new observations with Hyper Suprime Cam, and there is a 60% probability of a non-detection at z ~ 3.

Yajima – The Escape of Ionizing Photons from Lyman-alpha Emitters at the Epoch of Reionization

• Studying the escape fraction in the 60 most massive galaxies in a zoom-in calculation with a (5 Mpc/h)^3 high-resolution volume.
• Dust-to-gas ratio (0.008 * Z/Zsun) from Draine+ (2007), size distribution from Ferrara & ??? (2001)
• Finds some slight decrease in f_esc with respect to halo mass (median of 20% at 10^9 Msun and 15% at 10^11 Msun). No strong correlation between EW of Ly-alpha and f_esc. L_Lya is proportional to (dN_ion/dt) * (1 – f_esc). No redshift dependence. Finds an increase in f_esc above 10^12 Msun.
• At the peak of cosmic SFR, the average HI fraction in halos reaches a global minimum; however, the dust masses increase at the same redshift, limiting f_esc.
• Cold gas accretion causes efficient Ly-alpha cooling and an inflow feature in the Ly-alpha line profiles.
• (Yajima+ 2014) Calculated the escape fractions of Ly-alpha, UV and Lyman continuum, agreeing with observations.
• When f_esc in LyC is compared to f_{esc,LyA}, there is a weak correlation and large scatter.

Cooke – Not the usual suspects: Uncovering key contributors to cosmic reionization

• Note: LBGs are preferentially selected galaxies with low LyC emission, which are the wrong objects to constrain the escape fraction.
• (Cooke+ 2014) Galaxy color-color evolutionary tracks are very sensitive to LyC flux, where galaxies with high (10-20%) escape fraction fall out of the traditional LBG selection region.
• For z ~ 3 LBGs, they find an intrinsic escape fraction of 8.6 ± (1.0 – 0.6)% in 796 spectra.
• In the VVDS Ultra Deep Survey, they fall in the range 5-50% of escape fraction, using the composite method of Cooke+ (2014). Looking at the spectra, their method is pretty accurate to within f_esc = ±3-5%.
• Following up with MOSFIRE and ZFOURGE, they have a 100% (!) success in finding LyC galaxies.

Kim – Variation in the escape fraction of ionizing photons from galaxies and the redshifted 21-cm power spectrum during reionization

• Using semi-analytic models of galaxy formation (GALFORM) to source a semi-numeric calculation of reionization.
• Predictions for 21-cm power spectra are very sensitive to the galaxy properties (which is a good thing!)
• However, there is little difference in the power spectra on the sensitivity of f_esc on halo mass.

Park – The cross-power spectrum between 21cm emission and galaxies in hierarchical galaxy formation models

• (Park+ 2014) The cross-correlation between 21cm emission and galaxies can be used to estimate the sizes of HII regions, galaxy properties, etc.

Rutkowski – Lyman-Continuum Leakage in Dwarf Star-Forming Galaxies at z~1.2

• (in prep) “Conservative” selection sample of ~230 galaxies, and stacked them to get an upper limit of f_esc, which they find to be 4.8% (1-sigma). Consistent with previous work (Cowie+ 2009).
• In the high H-alpha EW (>200A) sample, they also have a non-detection, which places an upper limit of 22%. These are identified in HST IR that are low-mass (M_star ~ 10^8 Msun) galaxies.

Fleming – A Study of Lyman Alpha Emission from Low-redshift Galaxies and its Potential use as a Tracer of Lyman Continuum Escape

• SubLyme: Predicted detection of ~125 galaxies in 18 months at various magnitudes, masses, and redshifts. Targeting a redshift range of 0.2-0.5. f_esc upper limits of 3-5%.

Kaurov – Comparison of numerical and analytical methods for studying cosmic Reionization

• Using the results of star formation in a cosmological simulation in analytic models of reionization (excursion set – Furlanetto+ 2004; PDF – Miralda-Escudé+ 2000).
• Can modify the original Furlanetto description to include small-scale structure and then again to include outside-in reionization from the PDF method.  Interesting new approach of analyzing numerical simulations of reionization.

Hariharan – Enabling Radiative Transfer on AMR grids in CRASH

• Overview of the implementation and applying it to the density field of the Santa Barbara cluster comparison project.