UV slopes of Starforming Galaxies in Strong Lensing fields at the Epoch of Reionization with JWST

The gist: The study measures UV slopes for statistically significant samples of galaxies deep into the Epoch of Reionization, up to redshift z > 7.5, using strong lensing fields around massive galaxy clusters with JWST data.

Context and Motivation

UV slopes (β) are a powerful diagnostics for galaxies at the epoch of reionization, tracing star formation, ISM ionization, and the escape fraction fesc of ionizing photons Studies at low and intermediate z find suggest a gradual reddening of the UV slope with time and steeper slopes for fainter galaxies With the James Webb Space Telescope (JWST), it has been possible to measure the UV slope for statistically significant samples of galaxies deep into the EoR, up to redshift z ≳ 10 and for increasingly faint galaxies Recent studies of β with JWST at z > 7 reveal a flattening of the trend of the UV slope with respect to redshift and UV magnitude The low-brightness regime is of particular interest for reionization, as most of the recent models of this process posit that numerous faint galaxies are the prime drivers.

Methods and Data Sources

The researchers utilized NIRCam and NIRSpec observations from several surveys across 7 strong lensing fields to find galaxies down to MUV ∼ −16 and for redshifts between 7.5 < z ≲ 12.5 They measure their β with a forward-modelling procedure The sample selection involved obtaining a photometric estimate of redshift using the EAzY SEDfitting code and applying criteria such as S/N 7.5 A total of 378 objects were included in their final high-redshift selection. Spectroscopic redshift (zspec) was used in place of zEAzY,ML when identifying z > 7.5 galaxies, and the criterion zEAzY,16% > 6.0 was ignored altogether For the spectroscopic sample, they first adjusted each of the 45 spectra to each object’s photometry by obtaining the Spectrum-to-Flux Ratio and fitting a second degree Chebyshev polynomial. The spectroscopic UV-continuum slope index βspec was then estimated using their photometry-corrected spectrum by fitting a power-law through the Calzetti UV-continuum windows.

Key Findings on UV Slopes

The analysis of the whole sample yielded an average value of β = −2.3 ± 0.4, with values ranging from-4.0 to-0.5 They found no significant trend between β and redshift, with dβ/dz = 0.02+0.02−0.03 over the range 7.5 7 They also found a weak trend between β and MUV, dβ/dMUV = −0.06+0.03−0.02 across their whole sample, suggesting a very weak trend of UV slope with galaxy brightness. This weak relation is in agreement with what is found by Saxena et al. (2024), where for a subsample at z > 8 they find a similar trend.

LyC Escape Fraction Estimation

The escape fraction of LyC photons (fesc) was estimated for a subsample of 14 galaxies observed with NIRSpec and for which they have emission line measurements They found an average escape fraction of 0.26 ± 0.22 across the subsample, a moderately high value with respect to what predicted to obtain neutral IGM at z ∼ 6 This result is in line with previous values found in literature, but the wide scatter puts it in agreement with values found by Mascia et al. (2024).

Discussion and Conclusions

The absence of a trend of β with redshift suggests a flattening of the overall slope at higher redshifts, which may indicate that a wider portion of galaxies at high-z have their UV emission dominated by the nebular continuum with an intrinsic β ∼ −2.3 The weak trend between β and MUV suggests a lack of dust or an absence of correlation between dust and SFR for galaxies in the interval 7.5 < z < 12.5, with more massive galaxies not having had yet enough time to build up a dust reservoir with respect to less massive ones The average fesc is high for what is necessary to achieve an ionized IGM by z ∼ 6, but the extrapolation model from a low-redshift sample may be overestimating its value.

Summary and Conclusions

The study uses photometric and spectroscopic data from JWST NIRCam and NIRSpec instrument gathered by 6 surveys across 7 strong lensing fields to analyze the UV slopes (β) in a sample of 378 galaxies at 7.5 < z < 12.5 They find no significant trend between β and redshift, with dβ/dz = 0.00+0.02−0.03 over the range 7.5 7 found in recent works They also find a weak trend between β and MUV dβ/dMUV = −0.06 ± 0.03−0.02 over the whole UV magnitude range of −22 < MUV < −16, also consistent with recent results at highredshift (Saxena et al. 2024, Austin et al. 2024) They find an average escape fraction of 0.26±0.22 across their sample, slightly higher than the averages found in Mascia et al. (2024) for objects at = 8 but with a much greater dispersion. This is a relatively high value of fesc with respect to what predicted to obtain neutral IGM at z ∼ 6.

How it works

1. The researchers use NIRCam photometry to calculate the UV slopes of high-redshift z > 7.5 galaxies in the CLU and NCF fields They use a direct power-law fit (fλ ∝ λ β) of the available medium and wide-band photometry For each object, they estimate the UV slope with a Markov-chain Monte Carlo (MCMC) fit that convolves the model UV slope Fλ = A · λ β with each filter’s response function.

2. They compare photometric (βphot) and spectroscopic (βspec) values for these objects, finding an average value ∆β = βphot − βspec = −0.2 ± 0.4 This shows that on average the values for the UV slope calculated from photometry are consistent with spectroscopy.

3. They extrapolate LyC escape fractions using a model from Mascia et al. (2024) that uses the UV slope, Hβ equivalent width and galaxy sizes to predict its LyC escape fraction This extrapolation yields an average = 0.26 ± 0.22 between redshifts of ∼ 8 and ∼ 11.

Key Takeaways

**)& We find no significant trend between β and redshift, with dβ/dz = 0.00+0.02−0.03 over the range 7.5 7 found in recent works The flattening of this relation may indicate that a wider portion of galaxies at high-z have their UV emission dominated by the nebular continuum with an intrinsic β ∼ −2.3. The weak trend between β and MUV suggests a lack of dust, or at least to an absence of correlation between dust and SFR for galaxies in the interval 7.5 < z < 12.5, with more massive galaxies not having had yet enough time to build up a dust reservoir with respect to less massive ones. The average fesc is high for what is necessary to achieve an ionized IGM by z ∼ 6, but the extrapolation model from a low-redshift sample may be overestimating its value. The wide scatter in the subsample and the extrapolation obtained from low-z galaxies lead us to take this result with a grain of salt. The average fesc is 0.26±0.22 across our sample, slightly higher than the averages found in Mascia et al. (2024) for objects at = 8 but with a much greater dispersion. The confidence intervals were obtained with a MC method by resampling the value of each parameter and coefficient 1000 times within its distribution. The detailed results are reported in Table 2.

1. Improved galaxy property prediction for reionization models: An AI system can now predict UV slope beta using NIRCam and NIRSpec photometry from strong lensing fields to constrain galaxy properties such as star formation history and dust content, directly testing models of the Epoch of Reionization by measuring UV slopes (beta) for galaxies at the epoch of reionization.

2. Enhanced LyC escape fraction estimation: The system can estimate the LyC escape fraction using a model calibrated on low-redshift data, leveraging the relation log10fesc = A + B · EW(Hβ) + C · re + D · β, allowing for more robust predictions of ionizing photon leakage in high-redshift galaxies.

3. Robust photometric redshift selection: The AI can implement a rigorous selection pipeline using zEAzY SEDfitting code criteria, such as S/N 7.5, to select high-redshift candidates while filtering out interlopers based on photometric constraints.

4. Spectral slope consistency verification: The system can compare photometric and spectroscopic measurements of UV slopes, identifying discrepancies as they occur in specific regions like the central clump versus aperture photometry, ensuring that the derived βphot is consistent with βspec.

UV slopes (β) are a powerful diagnostics for galaxies at the Epoch of Reionization, tracing star formation, ISM ionization, and the escape fraction f esc of ionizing photons. Studies at low and intermediate z find a gradual β reddening with time and steeper slopes for fainter galaxies, however recent JWST studies reveal a flattening of this trend at z>7. We want measure β for galaxies at z>7.5 using the strong lensing around massive galaxy clusters to observe high-redshift and faint galaxies. The low-brightness regime is of particular interest for reionization, as most of the recent models of this process posit that numerous faint galaxies are the prime drivers of reionization. We use NIRCam and NIRSpec data from CANUCS, Technicolor, JUMPS, Silver Bullet, UNCOVER and MEGASCIENCE across 7 strong lensing fields. We find galaxies down to M UV about-16 and 7.5<z<12.5. We measure with a forward-modelling procedure and estimate f esc for a subsample with emission line data using a relation, calibrated from a low-z sample, with UV slope, galaxy size and H β equivalent width. We find 378 galaxies (45 with spectrum), yielding average values β=-2.3 plus or minus0.4, z=8.5 plus or minus1.0, and M UV=-18 plus or minus1. We find no significant β evolution across our redshift range, suggesting a flattening of the β-z trend above z about7.5. We find a weak negative trend between β and M UV. For 14 galaxies we estimate an average f esc=0.26 plus or minus0.22. The flat trend of β at z>7.5 suggests similar properties between 300 and 600 Myr after the Big Bang. The weak trend between β and M UV suggests an analogous composition for low- and high-mass galaxies' ISM, likely due to a lack of time for dust buildup. While average f esc is higher than necessary to ionize the IGM by z 6, the model extrapolated at low-z may overestimate its value.