We present multi-wavelength high-spatial resolution (~0.1'', 70 pc)observations of UGC 4211 at z=0.03474, a late-stage major galaxy merger at theclosest nuclear separation yet found in near-IR imaging (0.32'', ~230 pcprojected separation). Using Hubble Space Telescope/STIS, VLT/MUSE+AO,Keck/OSIRIS+AO spectroscopy, and ALMA observations, we show that the spatialdistribution, optical and NIR emission lines, and millimeter continuum emissionare all consistent with both nuclei being powered by accreting supermassiveblack holes (SMBHs). Our data, combined with common black hole massprescriptions, suggests that both SMBHs have similar masses, log MBH~8.1(south) and log MBH~8.3 (north), respectively. The projected separation of 230pc (~6X the black hole sphere of influence) represents the closest-separationdual AGN studied to date with multi-wavelength resolved spectroscopy and showsthe potential of nuclear (<50 pc) continuum observations with ALMA to discoverhidden growing SMBH pairs. While the exact occurrence rate of close-separationdual AGN is not yet known, it may be surprisingly high, given that UGC 4211 wasfound within a small, volume-limited sample of nearby hard X-ray detected AGN.Observations of dual SMBH binaries in the sub-kpc regime at the final stages ofdynamical friction provide important constraints for future gravitational waveobservatories.

Accurate active galactic nucleus (AGN) identifications in large galaxysamples are crucial to assess the role of AGN and AGN feedback in thecoevolution of galaxies and their central supermassive black holes. Emissionline flux ratio diagnostics are the most common technique for identifying AGNin optical spectra. New large samples of integral field unit observations allowThe exploration of the role of aperture size used for the classification. Inthis paper, we present galaxy classifications for all 10,010 galaxies observedwithin the Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) survey.We use Baldwin-Philips-Terlevich line flux ratio diagnostics combined with anH$\alpha$ equivalent threshold in 60 apertures of varying size for theclassification and provide the corresponding catalogs. MaNGA-selected AGNprimarily lie below the main sequence of star-forming galaxies, reside withmassive galaxies with stellar masses of $\sim 10^{11}$~M$_{\odot}$ and a medianH$\alpha$-derived star formation rate of $\sim 1.44$M$_{\odot}$~yr$^{-1}$. Wefind that the number of `fake' AGN increases significantly beyond selectionapertures of $>$~1.0~R$_{eff}$ due to increased contamination from diffuseionized gas (DIG). A comparison with previous works shows that the treatment ofthe underlying stellar continuum and flux measurements can significantly impactgalaxy classification. Our work provides the community with AGN catalogs andgalaxy classifications for the full MaNGA.

We present JADES JWST/NIRSpec spectroscopy of GN-z11, the most luminouscandidate $z>10$ Lyman break galaxy in the GOODS-North field with$M_{UV}=-21.5$. We derive a redshift of $z=10.603$ (lower than previousdeterminations) based on multiple emission lines in our low and mediumresolution spectra over $0.8-5.3\,\mu$m. We significantly detect the continuumand measure a blue rest-UV spectral slope of $\beta=-2.4$. Remarkably, we seespatially-extended Lyman-$\alpha$ in emission (despite the highly-neutral IGMexpected at this early epoch), offset 555 km/s redward of the systemicredshift. From our measurements of collisionally-excited lines of both low- andhigh-ionization (including [O II] $\lambda3727$, [Ne III] $\lambda 3869$ and CIII] $\lambda1909$) we infer a high ionization parameter ($\log U\sim -2$). Wedetect the rarely-seen N IV] $\lambda1486$ and N III]$\lambda1748$ lines inboth our low and medium resolution spectra, with other high ionization linesseen in low resolution spectrum such as He II (blended with O III]) and C IV(with a possible P-Cygni profile). Based on the observed rest-UV line ratios,we cannot conclusively rule out photoionization from AGN. The high C III]/He IIratios, however, suggest a likely star-formation explanation. If the observedemission lines are powered by star formation, then the strong N III]$\lambda1748$ observed may imply an unusually high $N/O$ abundance. Balmeremission lines (H$\gamma$, H$\delta$) are also detected, and if powered by starformation rather than an AGN we infer a star formation rate of $\sim 20-30M_{\odot}\,\rm yr^{-1}$ (depending on the IMF) and low dust attenuation. OurNIRSpec spectroscopy confirms that GN-z11 is a remarkable galaxy with extremeproperties seen 430 Myr after the Big Bang.

Recent observations of GN-z11 with JWST/NIRSpec revealed numerous oxygen,carbon, nitrogen, and helium emission lines at $z=10.6$. Using the measuredline fluxes, we derive abundance ratios of individual elements within theinterstellar medium (ISM) of this super-luminous galaxy. Driven by theunusually-bright NIII] $\lambda$1750 and NIV] $\lambda$1486 emission lines (andby comparison faint OIII] $\lambda\lambda$1660, 1666 lines), our fiducial modelprefers log(N/O)>-0.25, greater than four times solar and in stark contrast tolower-redshift star-forming galaxies. The derived log(C/O)>-0.78, ($\approx$30% solar) is also elevated with respect to galaxies of similar metallicity(12+log(O/H)$\approx7.82$), although less at odds with lower-redshiftmeasurements. Given the long timescale typically expected to enrich nitrogenwith stellar winds, traditional scenarios require a very fine-tuned formationhistory to reproduce such an elevated N/O. We find no compelling evidence thatnitrogen enhancement in GN-z11 can be explained by enrichment from metal-freePopulation III stars. Interestingly, yields from runaway stellar collisions ina dense stellar cluster or a tidal disruption event provide promising solutionsto give rise to these unusual emission lines at $z=10.6$, and explain theresemblance between GN-z11 and a nitrogen-loud quasar. These recentobservations showcase the new frontier opened by JWST to constrain galacticenrichment and stellar evolution within 440 Myr of the Big Bang.

Difference-in-differences is without a doubt the most widely used method forevaluating the causal effect of a hypothetical intervention in the possiblepresence of confounding bias due to hidden factors. The approach is typicallyused when both pre- and post-exposure outcome measurements are available, andone can reasonably assume that the additive association of the unobservedconfounder with the outcome is equal in the two exposure arms, and constantover time; a so-called parallel trends assumption. The parallel trendsassumption may not be credible in many practical settings, including if theoutcome is binary, a count, or polytomous, and more generally, when theunmeasured confounder exhibits non-additive effects on the distribution of theoutcome, even if such effects are constant over time. We introduce analternative approach that replaces the parallel trends assumption with an oddsratio equi-confounding assumption, which states that confounding bias for thecausal effect of interest, encoded by an association between treatment and thepotential outcome under no-treatment can be identified with a well-specifiedgeneralized linear model relating the pre-exposure outcome and the exposure. Asthe proposed method identifies any causal effect that is conceivably identifiedin the absence of confounding bias, including nonlinear effects such asquantile treatment effects, the approach is aptly called UniversalDifference-in-differences (UDiD). Both fully parametric and more robustsemiparametric UDiD estimators are described and illustrated in a real-worldapplication concerning the causal effects of maternal Zika virus infection onbirth rate in Brazil.