James Don Neill

Modern astronomical data reduction pipelines must handle increasing data volumes, complex instrument configurations, and the need for reproducibility across diverse computing environments. While most existing data reduction pipelines (DRPs) work effectively within their intended scope, they often rely on home-grown orchestration that makes monitoring, debugging, and extending them increasingly costly as instruments and data rates evolve. As part of developing the LRIS-2 data reduction pipeline for the W. M. Keck Observatory, we introduce a prototype architecture that adapts mature data engineering practices, in particular workflow orchestration with Prefect, to produce a more transparent, modular, and resilient reduction system. The pipeline is represented as a directed acyclic graph (DAG) of observable, state-aware tasks, enabling selective re-execution of individual steps, real-time monitoring through a live dashboard, and parallel execution of independent steps. We demonstrate the architecture on a working prototype that reduces both real LRIS-1 calibration data and synthetic LRIS-2 frames, sourcing scientific primitives from established open-source libraries for slit tracing and calibration combination under a single orchestration layer. Measured results include a 55× reduction in wall-clock time on an unchanged re-run due to caching, correct subgraph-only recomputation on a parameter change, a 2.6× parallel speedup using process-based distributed execution, and isolated, non-fatal failure handling for corrupted input frames. We report these results candidly, including where the architecture’s dual-layer (Python/C) primitive model remains future work, and discuss the tradeoffs of adopting data-engineering orchestration patterns in an astronomical context.

For photometry of galaxies in the infrared that extend to very low surface brightnesses, it is necessary to observe with telescopes in space. The Wide-field Infrared Survey Explorer (WISE) and NEOWISE provide sensitive all-sky access to mid-infrared flux from galaxies at 3.4 μ m (W1) and 4.6 μ m (W2) bands. This study is complemented by the availability of accurate tip of the red giant branch distances for a large fraction of the targets. In this work, photometry is assembled on a complete volume-limited sample of galaxies between 1 and 10 Mpc brighter than absolute magnitude −13 in the W1 band. Stellar masses are inferred from W1 fluxes and measured distances. While most of the galaxies are dwarfs, most of the stellar mass is concentrated in a small number of giants.

Accurate chemical compositions of star-forming regions provide a critical diagnostic tool for characterizing the star formation history and gas flows that regulate galaxy formation. However, the abundance discrepancy factor (ADF) between measurements derived from the “direct” optical electron temperature ( T _e ) method and those from recombination lines (RLs) introduces a ∼0.2 dex systematic uncertainty in the oxygen abundance. The degree of uncertainty for other elements is unknown. We conduct a comprehensive analysis of O ⁺⁺ and N ⁺ ion abundances using optical and far-infrared (far-IR) spectra of a star-forming region within the nearby dwarf galaxy Haro 3, which exhibits a typical ADF. Assuming homogeneous conditions, the far-IR emission indicates an oxygen abundance higher than that derived using the T _e method and consistent with the RL value, as expected from temperature fluctuations, whereas the far-IR nitrogen abundance is too large to be explained by temperature fluctuations. A two-phase analytical model reveals that differential dust obscuration associated with temperature inhomogeneity is likely required to explain all the emission-line ratios, and that the total oxygen metallicity of two phases is consistent with the RL metallicity. Our findings underscore the critical importance of resolving the cause of abundance discrepancies and understanding the biases between different metallicity methods. This work presents a promising methodology, and we identify further approaches to address the dominant sources of uncertainty.

To reproduce observed galaxy properties, cosmological simulations require that massive galaxies experience feedback from active galactic nuclei, which regulates star formation within those galaxies. However, the energetics and timescales of these feedback processes are poorly constrained. We combined optical, infrared, submillimeter, and radio observations of the active galaxy VV 340a, which is hosting a low-power jet launched from a supermassive black hole at its center. We found that the jet undergoes precession, with a period of (8.2 ± 5.5) × 10 ⁵ years, and drives an outflow of gas at a rate of 19.4 ± 7.9 solar masses per year. The jet shocks the gas, producing highly ionized plasma that extends several kiloparsecs from the nucleus. The outflow ejects sufficient gas from the galaxy to influence its star-formation rate.

We present the discovery and characterisation of two Ly α nebulae (LANs), RO-1001-Sat and RO-0959-Sat, as satellite structures of two giant LANs at z = 2.920 and 3.092. These two satellite LANs are found neighbouring two out of four known giant LANs at z ∼ 3 in our Multi-Unit Spectroscopic Explorer (MUSE) follow-up observations, reinforcing the idea that Ly α emission can be used as a tracer of massive dark matter halos at high z . This high occurrence of massive satellite halos is in agreement with simulations. With sizes of ≃80 × 160 and 80 × 100 pkpc ² , the two nebulae are both ∼300 pkpc from the main LANs. With Ly α emission only shifted by ≃100 − 300 km s ⁻¹ between each of the two pairs, these two satellite structures are likely connected to their main halos by the large-scale structure. RO-1001-Sat and RO-0959-Sat are estimated to have log( M _h / M _⊙ )≃13.2 ± 0.3 and 12.8 ± 0.3, putting them potentially close to the regime of cold-mode accretion according to several models, which suggests that cold streams should be able to penetrate the virial radii to reach the halo centres without being shock-heated. The central brightest galaxies in the two halos are morphologically distinct despite having a similar stellar mass of ∼10 ¹¹ M _⊙ , one being an elliptical quiescent galaxy in RO-1001-Sat and the other being a dusty star-forming spiral in RO-0959-Sat. Intriguingly, the quiescent galaxy aligns well with the peak of the LAN as well as the potential well of the host halo, making it the first clear-cut case in which the cold gas ought to be accreting onto the galaxy but with no observable star formation, either due to morphological quenching or, more likely, radio-mode feedback from an active galactic nucleus, as supported by excess yet weak radio emission. Finally, we show a tentative detection of a Ly α filament connecting RO-1001 and RO-1001-Sat. This work shows how panoramic MUSE (and in the future, BlueMUSE) observations of massive halo seeds can be used to efficiently search for additional halos, unveiling their large-scale structure and enabling the study of Ly α -selected galaxy groups.

We present the first detailed, spatially resolved study of the circumgalactic medium (CGM) surrounding the low-mass, low-redshift star-forming galaxy J0910b (SDSS J091458.61+110845.1; M _⋆ ∼ 10 ^9.2 M _⊙ , z = 0.096), focusing on its gas dynamics, ionization mechanisms, and kinematics. J0910b is the lowest-mass galaxy in our low-redshift CGM survey using the Keck Cosmic Web Imager (KCWI). Using KCWI integral field spectroscopy and CLOUDY modeling, we map the ionization structure and velocity field of the CGM through H α , [O III ] λ 5007, and [O II ] λλ 3727, 3729 emission. We detect a predominantly cool ( T ∼ 10 ⁴ K) ionized halo extending to ∼35 kpc (∼0.35 R _vir ), with a complex structure including a counterrotating component and two filamentary inflows. The counterrotating CGM contains ∼7.4 × 10 ⁹ M _⊙ of ionized gas—nearly 5 times the stellar mass—and exhibits a positive radial metallicity gradient, rising from 0.2 Z _⊙ at 9 kpc to 0.65 Z _⊙ at 16 kpc. Velocity dispersion maps reveal a shell-like structure near the interstellar medium-CGM interface, suggesting feedback-driven outflows that redistributed enriched gas. In contrast, the filaments show low metallicities (∼0.07 Z _⊙ ), high specific angular momentum ( j ∼ 3.7 × 10 ³ kpc km s ⁻¹ ), and an accretion rate of 0.05 M _⊙ yr ⁻¹ , consistent with an intergalactic origin. Together, these components reveal a dynamically rich CGM. The counterrotating gas dominates the halo’s mass, accounts for ∼33% of the cosmic baryon budget, and must be continuously replenished ( t _cool ∼ 0.3–2 Myr). These results suggest a long-lived, regulated CGM that governs star formation through angular momentum exchange and gas recycling.

We present high-resolution Keck Cosmic Web Imager and MUSE integral field unit spectroscopy of VV 114, a local IR-luminous merger undergoing a vigorous starburst and showing evidence of galactic-scale feedback. The high-resolution data allow for spectral deblending of the optical emission lines and reveal a broad emission line component (σ _broad ∼ 100–300 km s⁻¹) with line ratios and kinematics consistent with a mixture of ionization by stars and radiative shocks. The shock fraction (percentage of ionization due to shocks) in the high-velocity gas is anticorrelated with the projected surface number density of resolved star clusters, and we find that the radial density profiles around clusters are fit well by models of adiabatically expanding cluster winds driven by massive stellar winds and supernovae (SNe). The total kinetic power estimated from the cluster wind models matches the wind + SN mechanical energy deposition rate estimated from the soft-band X-ray luminosity, indicating that at least 70% of the shock luminosity in the galaxy is driven by the star clusters. Hubble Space Telescope narrowband near-IR imaging reveals embedded shocks in the dust-buried IR nucleus of VV 114E. Most of the shocked gas is blueshifted with respect to the quiescent medium, and there is a close spatial correspondence between the shock map and the Chandra soft-band X-ray image, implying the presence of a galactic superwind. The energy budget of the superwind is in close agreement with the total kinetic power of the cluster winds, confirming the superwind is driven by the starburst.

We present high-resolution Keck Cosmic Web Imager (KCWI) and MUSE IFU spectroscopy of VV 114, a local infrared-luminous merger undergoing a vigorous starburst and showing evidence of galactic-scale feedback. The high-resolution data allow for spectral deblending of the optical emission lines and reveal a broad emission line component ($σ_{\rm{broad}} \sim$~100--300 km s$^{-1}$) with line ratios and kinematics consistent with a mixture of ionization by stars and radiative shocks. The shock fraction (percent ionization due to shocks) in the high velocity gas is anticorrelated with projected surface number density of resolved star clusters, and we find radial density profiles around clusters are well fit by models of adiabatically expanding cluster winds driven by massive stellar winds and supernovae (SNe). The total kinetic power estimated from the cluster wind models matches the wind+SNe mechanical energy deposition rate estimated from the soft band X-ray luminosity, indicating that at least 70\% of the shock luminosity in the galaxy is driven by the star clusters. \emph{Hubble Space Telescope} narrow band near-infrared imaging reveals embedded shocks in the dust-buried infrared nucleus of VV 114E. Most of the shocked gas is blueshifted with respect to the quiescent medium, and there is a close spatial correspondence between the shock map and the \emph{Chandra} soft band X-ray image, implying the presence of a galactic superwind. The energy budget of the superwind is in close agreement with the total kinetic power of the cluster winds, confirming the superwind is driven by the starburst.

We present the discovery of 14 new (and the recovery of four known) low accretion rate magnetic white dwarfs in post-common envelope binaries that emit strong cyclotron emission using the Zwicky Transient Facility (ZTF) light curves, doubling the known sample size. In addition, we discovered a candidate magnetic period bouncer and recovered three known ones. We confirmed the presence of cyclotron emission using low-resolution spectra in 19 objects. Using the ZTF light curves, follow-up spectra, and the spectral energy distribution, we measured the orbital period, magnetic field strength, and white dwarf temperature of each system. Although the phase-folded light curves have diverse shapes and show a much larger variability amplitude, we show that their intrinsic properties (e.g. period distribution, magnetic field strength) are similar to those of previously known systems. The diversity in light curve shapes can be explained by differences in the optical depth of the accretion spot and geometric differences, the inclination angle, and the magnetic spot latitude. The evolutionary states of the longer period binaries are somewhat uncertain but vary; we found systems consistent with being pre-polars, detached polars, or low-state polars. In addition, we discovered two new low-state polars that likely have brown dwarf companions and could be magnetic period bouncers.

Supernovae (SNe) come in various flavors and are classified into different types based on emission and absorption lines in their spectra. SN candidates are now abundant with the advent of large systematic sky surveys like the Zwicky Transient Facility (ZTF), however, the identification bottleneck lies in their spectroscopic confirmation and classification. Fully robotic telescopes with dedicated spectrographs optimized for SN follow-up have eased the burden of data acquisition. However, the task of classifying the spectra still largely rests with the astronomers. Automating this classification step reduces human effort and can make the SN type available sooner to the public. For this purpose, we have developed a deep-learning based program for classifying core-collapse supernovae (CCSNe) with ultra-low resolution spectra from the SED-machine spectrograph on the Palomar 60 inch telescope. The program consists of hierarchical classification task layers, with each layer composed of multiple binary classifiers running in parallel to produce a reliable classification. The binary classifiers utilize recurrent neural networks and convolutional neural networks architecture and are designed to take multiple inputs to supplement spectra with g- and r-band photometry from ZTF. On non-host-contaminated and good quality SEDM spectra (“gold” test set) of CCSNe, CCSNscore is ∼94% accurate in distinguishing between hydrogen-rich (Type II) and hydrogen-poor (Type Ibc) CCSNe. With light curve input, CCSNscore classifies ∼83% of the gold set with high confidence (score ≥0.8 and score-error < 0.05), with ∼98% accuracy. Based on SNIascore’s and CCSNscore’s real-time performance on bright transients (m _pk ≤ 18.5) and our reporting criteria, we expect ∼0.5% (∼4) true SNe Ia to be misclassified as SNe Ibc and ∼6% (∼17) of true CCSNe to be misclassified between Type II and Type Ibc annually on the Transient Name Server.