Search

The 21 cm transition from neutral Hydrogen promises to be the best observational probe of the Epoch of Reionisation (EoR). This has led to the construction of low-frequency radio interferometric arrays, such as the Hydrogen Epoch of Reionization Array (HERA), aimed at systematically mapping this emission for the first time. Precision calibration, however, is a requirement in 21 cm radio observations. Due to the spatial compactness of HERA, the array is prone to the effects of mutual coupling, which inevitably lead to non-smooth calibration errors that contaminate the data. When unsmooth gains are used in calibration, intrinsically spectrally-smooth foreground emission begins to contaminate the data in a way that can prohibit a clean detection of the cosmological EoR signal. In this paper, we show that the effects of mutual coupling on calibration quality can be reduced by applying custom time-domain filters to the data prior to calibration. We find that more robust calibration solutions are derived when filtering in this way, which reduces the observed foreground power leakage. Specifically, we find a reduction of foreground power leakage by 2 orders of magnitude at k=0.5.

Interferometric experiments designed to detect the highly redshifted 21-cm signal from neutral hydrogen are producing increasingly stringent constraints on the 21-cm power spectrum, but some k-modes remain systematics-dominated. Mutual coupling is a major systematic that must be overcome in order to detect the 21-cm signal, and simulations that reproduce effects seen in the data can guide strategies for mitigating mutual coupling. In this paper, we analyse 12 nights of data from the Hydrogen Epoch of Reionization Array and compare the data against simulations that include a computationally efficient and physically motivated semi-analytic treatment of mutual coupling. We find that simulated coupling features qualitatively agree with coupling features in the data; however, coupling features in the data are brighter than the simulated features, indicating the presence of additional coupling mechanisms not captured by our model. We explore the use of fringe-rate filters as mutual coupling mitigation tools and use our simulations to investigate the effects of mutual coupling on a simulated cosmological 21-cm power spectrum in a "worst case" scenario where the foregrounds are particularly bright. We find that mutual coupling contaminates a large portion of the "EoR Window", and the contamination is several orders-of-magnitude larger than our simulated cosmic signal across a wide range of cosmological Fourier modes. While our fiducial fringe-rate filtering strategy reduces mutual coupling by roughly a factor of 100 in power, a non-negligible amount of coupling cannot be excised with fringe-rate filters, so more sophisticated mitigation strategies are required., Comment: 19 pages, 12 figures, submitted to MNRAS

Radio interferometers targeting the 21cm brightness temperature fluctuations at high redshift are subject to systematic effects that operate over a range of different timescales. These can be isolated by designing appropriate Fourier filters that operate in fringe-rate (FR) space, the Fourier pair of local sidereal time (LST). Applications of FR filtering include separating effects that are correlated with the rotating sky vs. those relative to the ground, down-weighting emission in the primary beam sidelobes, and suppressing noise. FR filtering causes the noise contributions to the visibility data to become correlated in time however, making interpretation of subsequent averaging and error estimation steps more subtle. In this paper, we describe fringe rate filters that are implemented using discrete prolate spheroidal sequences, and designed for two different purposes -- beam sidelobe/horizon suppression (the `mainlobe' filter), and ground-locked systematics removal (the `notch' filter). We apply these to simulated data, and study how their properties affect visibilities and power spectra generated from the simulations. Included is an introduction to fringe-rate filtering and a demonstration of fringe-rate filters applied to simple situations to aid understanding., Comment: 21 pages, 18 figures, submitted to Monthly Notices of the Royal Astronomical Society

This paper presents the design and deployment of the Hydrogen Epoch of Reionization Array (HERA) phase II system. HERA is designed as a staged experiment targeting 21 cm emission measurements of the Epoch of Reionization. First results from the phase I array are published as of early 2022, and deployment of the phase II system is nearing completion. We describe the design of the phase II system and discuss progress on commissioning and future upgrades. As HERA is a designated Square Kilometer Array (SKA) pathfinder instrument, we also show a number of "case studies" that investigate systematics seen while commissioning the phase II system, which may be of use in the design and operation of future arrays. Common pathologies are likely to manifest in similar ways across instruments, and many of these sources of contamination can be mitigated once the source is identified.

Detection of the faint 21 cm line emission from the Cosmic Dawn and Epoch of Reionisation will require not only exquisite control over instrumental calibration and systematics to achieve the necessary dynamic range of observations but also validation of analysis techniques to demonstrate their statistical properties and signal loss characteristics. A key ingredient in achieving this is the ability to perform high-fidelity simulations of the kinds of data that are produced by the large, many-element, radio interferometric arrays that have been purpose-built for these studies. The large scale of these arrays presents a computational challenge, as one must simulate a detailed sky and instrumental model across many hundreds of frequency channels, thousands of time samples, and tens of thousands of baselines for arrays with hundreds of antennas. In this paper, we present a fast matrix-based method for simulating radio interferometric measurements (visibilities) at the necessary scale. We achieve this through judicious use of primary beam interpolation, fast approximations for coordinate transforms, and a vectorised outer product to expand per-antenna quantities to per-baseline visibilities, coupled with standard parallelisation techniques. We validate the results of this method, implemented in the publicly-available matvis code, against a high-precision reference simulator, and explore its computational scaling on a variety of problems., Comment: 25 pages, 20 figures, submitted to RAS Techniques and Instruments, matvis is publicly available at https://github.com/HERA-Team/matvis

Observations with radio arrays that target the 21-cm signal originating from the early Universe suffer from a variety of systematic effects. An important class of these are reflections and spurious couplings between antennas. We apply a Hamiltonian Monte Carlo sampler to the modelling and mitigation of these systematics in simulated Hydrogen Epoch of Reionisation Array (HERA) data. This method allows us to form statistical uncertainty estimates for both our models and the recovered visibilities, which is an important ingredient in establishing robust upper limits on the Epoch of Reionisation (EoR) power spectrum. In cases where the noise is large compared to the EoR signal, this approach can constrain the systematics well enough to mitigate them down to the noise level for both systematics studied. Where the noise is smaller than the EoR, our modelling can mitigate the majority of the reflections with there being only a minor level of residual systematics, while cross-coupling sees essentially complete mitigation. Our approach performs similarly to existing filtering/fitting techniques used in the HERA pipeline, but with the added benefit of rigorously propagating uncertainties. In all cases it does not significantly attenuate the underlying signal., Comment: 19 pages, 14 figures, submitted to MNRAS

The key to detecting neutral hydrogen during the epoch of reionization (EoR) is to separate the cosmological signal from the dominating foreground radiation. We developed direct optimal mapping (DOM) to map interferometric visibilities; it contains only linear operations, with full knowledge of point spread functions from visibilities to images. Here, we demonstrate a fast Fourier transform-based image power spectrum and its window functions computed from the DOM images. We use noiseless simulation, based on the Hydrogen Epoch of Reionization Array Phase I configuration, to study the image power spectrum properties. The window functions show $<10^{-11}$ of the integrated power leaks from the foreground-dominated region into the EoR window; the 2D and 1D power spectra also verify the separation between the foregrounds and the EoR., Comment: Published in ApJ

Observations of 21cm line from neutral hydrogen promise to be an exciting new probe of astrophysics and cosmology during the Cosmic Dawn and through the Epoch of Reionization (EoR) to when dark energy accelerates the expansion of the Universe. At each of these epochs, separating bright foregrounds from the cosmological signal is a primary challenge that requires exquisite calibration. In this paper, we present a new calibration method called nucal that extends redundant-baseline calibration, allowing spectral variation in antenna responses to be solved for by using correlations between visibilities measuring the same angular Fourier modes at different frequencies. By modeling the chromaticity of the beam-weighted sky with a tunable set of discrete prolate spheroidal sequences (DPSS), we develop a calibration loop that optimizes for spectrally smooth calibrated visibilities. Crucially, this technique does not require explicit models of the sky or the primary beam. With simulations that incorporate realistic source and beam chromaticity, we show that this method solves for unsmooth bandpass features, exposes narrowband interference systematics, and suppresses smooth-spectrum foregrounds below the level of 21cm reionization models, even within much of the so-called "wedge" region where current foreground mitigation techniques struggle. We show that this foreground subtraction can be performed with minimal cosmological signal loss for certain well-sampled angular Fourier modes, making spectral-redundant calibration a promising technique for current and next-generation 21cm intensity mapping experiments., Comment: 22 pages, 10 figures, Accept for Publication, MNRAS

Radio interferometers aiming to measure the power spectrum of the redshifted 21 cm line during the Epoch of Reionisation (EoR) need to achieve an unprecedented dynamic range to separate the weak signal from overwhelming foreground emissions. Calibration inaccuracies can compromise the sensitivity of these measurements to the effect that a detection of the EoR is precluded. An alternative to standard analysis techniques makes use of the closure phase, which allows one to bypass antenna-based direction-independent calibration. Similarly to standard approaches, we use a delay spectrum technique to search for the EoR signal. Using 94 nights of data observed with Phase I of the Hydrogen Epoch of Reionization Array (HERA), we place approximate constraints on the 21 cm power spectrum at $z=7.7$. We find at 95% confidence that the 21 cm EoR brightness temperature is $\le$(372)$^2$ "pseudo" mK$^2$ at 1.14 "pseudo" $h$ Mpc$^{-1}$, where the "pseudo" emphasises that these limits are to be interpreted as approximations to the actual distance scales and brightness temperatures. Using a fiducial EoR model, we demonstrate the feasibility of detecting the EoR with the full array. Compared to standard methods, the closure phase processing is relatively simple, thereby providing an important independent check on results derived using visibility intensities, or related., Comment: 16 pages, 14 figures, accepted for publication by MNRAS

Radio Frequency Interference (RFI) is one of the systematic challenges preventing 21cm interferometric instruments from detecting the Epoch of Reionization. To mitigate the effects of RFI on data analysis pipelines, numerous inpaint techniques have been developed to restore RFI corrupted data. We examine the qualitative and quantitative errors introduced into the visibilities and power spectrum due to inpainting. We perform our analysis on simulated data as well as real data from the Hydrogen Epoch of Reionization Array (HERA) Phase 1 upper limits. We also introduce a convolutional neural network that capable of inpainting RFI corrupted data in interferometric instruments. We train our network on simulated data and show that our network is capable at inpainting real data without requiring to be retrained. We find that techniques that incorporate high wavenumbers in delay space in their modeling are best suited for inpainting over narrowband RFI. We also show that with our fiducial parameters Discrete Prolate Spheroidal Sequences (DPSS) and CLEAN provide the best performance for intermittent ``narrowband'' RFI while Gaussian Progress Regression (GPR) and Least Squares Spectral Analysis (LSSA) provide the best performance for larger RFI gaps. However we caution that these qualitative conclusions are sensitive to the chosen hyperparameters of each inpainting technique. We find these results to be consistent in both simulated and real visibilities. We show that all inpainting techniques reliably reproduce foreground dominated modes in the power spectrum. Since the inpainting techniques should not be capable of reproducing noise realizations, we find that the largest errors occur in the noise dominated delay modes. We show that in the future, as the noise level of the data comes down, CLEAN and DPSS are most capable of reproducing the fine frequency structure in the visibilities of HERA data., Comment: 21 pages, 13 figures

We report the most sensitive upper limits to date on the 21 cm epoch of reionization power spectrum using 94 nights of observing with Phase I of the Hydrogen Epoch of Reionization Array (HERA). Using similar analysis techniques as in previously reported limits (HERA Collaboration 2022a), we find at 95% confidence that $\Delta^2(k = 0.34$ $h$ Mpc$^{-1}$) $\leq 457$ mK$^2$ at $z = 7.9$ and that $\Delta^2 (k = 0.36$ $h$ Mpc$^{-1}) \leq 3,496$ mK$^2$ at $z = 10.4$, an improvement by a factor of 2.1 and 2.6 respectively. These limits are mostly consistent with thermal noise over a wide range of $k$ after our data quality cuts, despite performing a relatively conservative analysis designed to minimize signal loss. Our results are validated with both statistical tests on the data and end-to-end pipeline simulations. We also report updated constraints on the astrophysics of reionization and the cosmic dawn. Using multiple independent modeling and inference techniques previously employed by HERA Collaboration (2022b), we find that the intergalactic medium must have been heated above the adiabatic cooling limit at least as early as $z = 10.4$, ruling out a broad set of so-called "cold reionization" scenarios. If this heating is due to high-mass X-ray binaries during the cosmic dawn, as is generally believed, our result's 99% credible interval excludes the local relationship between soft X-ray luminosity and star formation and thus requires heating driven by evolved low-metallicity stars., Comment: 57 pages, 37 figures. Updated to match the accepted ApJ version. Corresponding author: Joshua S. Dillon

Combining the visibilities measured by an interferometer to form a cosmological power spectrum is a complicated process. In a delay-based analysis, the mapping between instrumental and cosmological space is not a one-to-one relation. Instead, neighbouring modes contribute to the power measured at one point, with their respective contributions encoded in the window functions. To better understand the power measured by an interferometer, we assess the impact of instrument characteristics and analysis choices on these window functions. Focusing on the Hydrogen Epoch of Reionization Array (HERA) as a case study, we find that long-baseline observations correspond to enhanced low-k tails of the window functions, which facilitate foreground leakage, whilst an informed choice of bandwidth and frequency taper can reduce said tails. With simple test cases and realistic simulations, we show that, apart from tracing mode mixing, the window functions help accurately reconstruct the power spectrum estimator of simulated visibilities. The window functions depend strongly on the beam chromaticity, and less on its spatial structure - a Gaussian approximation, ignoring side lobes, is sufficient. Finally, we investigate the potential of asymmetric window functions, down-weighting the contribution of low-k power to avoid foreground leakage. The window functions presented here correspond to the latest HERA upper limits for the full Phase I data. They allow an accurate reconstruction of the power spectrum measured by the instrument and will be used in future analyses to confront theoretical models and data directly in cylindrical space., Comment: 18 pages, 19 figures, accepted for publication in MNRAS

Motivated by the desire for wide-field images with well-defined statistical properties for 21cm cosmology, we implement an optimal mapping pipeline that computes a maximum likelihood estimator for the sky using the interferometric measurement equation. We demonstrate this direct optimal mapping with data from the Hydrogen Epoch of Reionization (HERA) Phase I observations. After validating the pipeline with simulated data, we develop a maximum likelihood figure-of-merit for comparing four sky models at 166MHz with a bandwidth of 100kHz. The HERA data agree with the GLEAM catalogs to <10%. After subtracting the GLEAM point sources, the HERA data discriminate between the different continuum sky models, providing most support for the model of Byrne et al. 2021. We report the computation cost for mapping the HERA Phase I data and project the computation for the HERA 320-antenna data; both are feasible with a modern server. The algorithm is broadly applicable to other interferometers and is valid for wide-field and non-coplanar arrays., Comment: 16 pages, 10 figures, 2 tables, published on ApJ

We introduce CALAMITY, a precision bandpass calibration method for radio interferometry. CALAMITY can solve for direction independent gains with arbitrary frequency structure to the high precision required for 21 cm cosmology with minimal knowledge of foregrounds or antenna beams and does not require any degree of redundancy (repeated identical measurements of the same baseline). We have achieved this through two key innovations. Firstly, we model the foregrounds on each baseline independently using a flexible and highly efficient set of basis functions that have minimal overlap with 21 cm modes and enforce spectral smoothness in the calibrated foregrounds. Secondly, we use an off-the-shelf GPU accelerated API (TENSORFLOW) to solve for per-baseline foregrounds simultaneously with per-frequency antenna gains in a single optimization loop. GPU acceleration is critical for our technique to be able to solve for the large numbers of foreground and gain parameters simultaneously across all frequencies for an interferometer with $\gtrsim 10$ antennas in a reasonable amount of time. In this paper, we give an overview of our technique and using realistic simulations and demonstrate its performance in solving for and removing pathological gain structures to the level necessary to measure fluctuations in the 21 cm emission field from Hydrogen gas during the Cosmic Dawn and Reionization. If you want to start using CALAMITY now, you can find a tutorial notebook at https://github.com/aewallwi/calamity/blob/main/examples/Calamity_Tutorial.ipynb ., Comment: 19 pages, 13 figures, comments welcome. Replacement incorporates minor corrections in response to community feedback. Submitted to ApJ

We derive a general formalism for interferometric visibilities, which considers first-order antenna-antenna coupling and assumes steady-state, incident radiation. We simulate such coupling features for non-polarized skies on a compact, redundantly-spaced array and present a phenomenological analysis of the coupling features. Contrary to previous studies, we find mutual coupling features manifest themselves at nonzero fringe rates. We compare power spectrum results for both coupled and non-coupled (noiseless, simulated) data and find coupling effects to be highly dependent on LST, baseline length, and baseline orientation. For all LSTs, lengths, and orientations, coupling features appear at delays which are outside the foreground 'wedge', which has been studied extensively and contains non-coupled astrophysical foreground features. Further, we find that first-order coupling effects threaten our ability to average data from baselines with identical length and orientation. Two filtering strategies are proposed which may mitigate such coupling systematics. The semi-analytic coupling model herein presented may be used to study mutual coupling systematics as a function of LST, baseline length, and baseline orientation. Such a model is not only helpful to the field of 21cm cosmology, but any study involving interferometric measurements, where coupling effects at the level of at least 1 part in 10^4 could corrupt the scientific result. Our model may be used to mitigate coupling systematics in existing radio interferometers and to design future arrays where the configuration of array elements inherently mitigates coupling effects at desired LSTs and angular resolutions.

The Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX) is a radio interferometer array currently in development, with an initial 256-element array to be deployed at the South African Radio Astronomy Observatory (SARAO) Square Kilometer Array (SKA) site in South Africa. Each of the 6m, $f/0.23$ dishes will be instrumented with dual-polarisation feeds operating over a frequency range of 400-800 MHz. Through intensity mapping of the 21 cm emission line of neutral hydrogen, HIRAX will provide a cosmological survey of the distribution of large-scale structure over the redshift range of $0.775 < z < 2.55$ over $\sim$15,000 square degrees of the southern sky. The statistical power of such a survey is sufficient to produce $\sim$7 percent constraints on the dark energy equation of state parameter when combined with measurements from the Planck satellite. Additionally, HIRAX will provide a highly competitive platform for radio transient and HI absorber science while enabling a multitude of cross-correlation studies. In this paper, we describe the science goals of the experiment, overview of the design and status of the sub-components of the telescope system, and describe the expected performance of the initial 256-element array as well as the planned future expansion to the final, 1024-element array., Comment: 24 pages, 6 figures. Updated to match published version

We present a framework for identifying and flagging malfunctioning antennas in large radio interferometers. We outline two distinct categories of metrics designed to detect outliers along known failure modes of large arrays: cross-correlation metrics, based on all antenna pairs, and auto-correlation metrics, based solely on individual antennas. We define and motivate the statistical framework for all metrics used, and present tailored visualizations that aid us in clearly identifying new and existing systematics. We implement these techniques using data from 105 antennas in the Hydrogen Epoch of Reionization Array (HERA) as a case study. Finally, we provide a detailed algorithm for implementing these metrics as flagging tools on real data sets., Comment: 31 pages, 17 figures

Recently, the Hydrogen Epoch of Reionization Array (HERA) collaboration has produced the experiment's first upper limits on the power spectrum of 21-cm fluctuations at z~8 and 10. Here, we use several independent theoretical models to infer constraints on the intergalactic medium (IGM) and galaxies during the epoch of reionization (EoR) from these limits. We find that the IGM must have been heated above the adiabatic cooling threshold by z~8, independent of uncertainties about the IGM ionization state and the nature of the radio background. Combining HERA limits with galaxy and EoR observations constrains the spin temperature of the z~8 neutral IGM to 27 K < T_S < 630 K (2.3 K < T_S < 640 K) at 68% (95%) confidence. They therefore also place a lower bound on X-ray heating, a previously unconstrained aspects of early galaxies. For example, if the CMB dominates the z~8 radio background, the new HERA limits imply that the first galaxies produced X-rays more efficiently than local ones (with soft band X-ray luminosities per star formation rate constrained to L_X/SFR = { 10^40.2, 10^41.9 } erg/s/(M_sun/yr) at 68% confidence), consistent with expectations of X-ray binaries in low-metallicity environments. The z~10 limits require even earlier heating if dark-matter interactions (e.g., through millicharges) cool down the hydrogen gas. Using a model in which an extra radio background is produced by galaxies, we rule out (at 95% confidence) the combination of high radio and low X-ray luminosities of L_{r,\nu}/SFR > 3.9 x 10^24 W/Hz/(M_sun/yr) and L_X/SFR<10^40 erg/s/(M_sun/yr). The new HERA upper limits neither support nor disfavor a cosmological interpretation of the recent EDGES detection. The analysis framework described here provides a foundation for the interpretation of future HERA results., Comment: 40 pages, 19 figures, accepted to ApJ

We report upper-limits on the Epoch of Reionization (EoR) 21 cm power spectrum at redshifts 7.9 and 10.4 with 18 nights of data ($\sim36$ hours of integration) from Phase I of the Hydrogen Epoch of Reionization Array (HERA). The Phase I data show evidence for systematics that can be largely suppressed with systematic models down to a dynamic range of $\sim10^9$ with respect to the peak foreground power. This yields a 95% confidence upper limit on the 21 cm power spectrum of $\Delta^2_{21} \le (30.76)^2\ {\rm mK}^2$ at $k=0.192\ h\ {\rm Mpc}^{-1}$ at $z=7.9$, and also $\Delta^2_{21} \le (95.74)^2\ {\rm mK}^2$ at $k=0.256\ h\ {\rm Mpc}^{-1}$ at $z=10.4$. At $z=7.9$, these limits are the most sensitive to-date by over an order of magnitude. While we find evidence for residual systematics at low line-of-sight Fourier $k_\parallel$ modes, at high $k_\parallel$ modes we find our data to be largely consistent with thermal noise, an indicator that the system could benefit from deeper integrations. The observed systematics could be due to radio frequency interference, cable sub-reflections, or residual instrumental cross-coupling, and warrant further study. This analysis emphasizes algorithms that have minimal inherent signal loss, although we do perform a careful accounting in a companion paper of the small forms of loss or bias associated with the pipeline. Overall, these results are a promising first step in the development of a tuned, instrument-specific analysis pipeline for HERA, particularly as Phase II construction is completed en route to reaching the full sensitivity of the experiment., Comment: Accepted to ApJ. https://reionization.org/science/public-data-release-1/

Precision calibration poses challenges to experiments probing the redshifted 21-cm signal of neutral hydrogen from the Cosmic Dawn and Epoch of Reionization (z~30-6). In both interferometric and global signal experiments, systematic calibration is the leading source of error. Though many aspects of calibration have been studied, the overlap between the two types of instruments has received less attention. We investigate the sky based calibration of total power measurements with a HERA dish and an EDGES style antenna to understand the role of auto-correlations in the calibration of an interferometer and the role of sky in calibrating a total power instrument. Using simulations we study various scenarios such as time variable gain, incomplete sky calibration model, and primary beam model. We find that temporal gain drifts, sky model incompleteness, and beam inaccuracies cause biases in the receiver gain amplitude and the receiver temperature estimates. In some cases, these biases mix spectral structure between beam and sky resulting in spectrally variable gain errors. Applying the calibration method to the HERA and EDGES data, we find good agreement with calibration via the more standard methods. Although instrumental gains are consistent with beam and sky errors similar in scale to those simulated, the receiver temperatures show significant deviations from expected values. While we show that it is possible to partially mitigate biases due to model inaccuracies by incorporating a time-dependent gain model in calibration, the resulting errors on calibration products are larger and more correlated. Completely addressing these biases will require more accurate sky and primary beam models., Comment: 16 pages, 13 figures, 1 table; accepted for publication in MNRAS main journal

We describe the validation of the HERA Phase I software pipeline by a series of modular tests, building up to an end-to-end simulation. The philosophy of this approach is to validate the software and algorithms used in the Phase I upper limit analysis on wholly synthetic data satisfying the assumptions of that analysis, not addressing whether the actual data meet these assumptions. We discuss the organization of this validation approach, the specific modular tests performed, and the construction of the end-to-end simulations. We explicitly discuss the limitations in scope of the current simulation effort. With mock visibility data generated from a known analytic power spectrum and a wide range of realistic instrumental effects and foregrounds, we demonstrate that the current pipeline produces power spectrum estimates that are consistent with known analytic inputs to within thermal noise levels (at the 2 sigma level) for k > 0.2 h/Mpc for both bands and fields considered. Our input spectrum is intentionally amplified to enable a strong `detection' at k ~0.2 h/Mpc -- at the level of ~25 sigma -- with foregrounds dominating on larger scales, and thermal noise dominating at smaller scales. Our pipeline is able to detect this amplified input signal after suppressing foregrounds with a dynamic range (foreground to noise ratio) of > 10^7. Our validation test suite uncovered several sources of scale-independent signal loss throughout the pipeline, whose amplitude is well-characterized and accounted for in the final estimates. We conclude with a discussion of the steps required for the next round of data analysis., Comment: 32 pages, 20 figures. Submitted to the Astrophysical Journal

As current- and next-generation astronomical instruments come online, they will generate an unprecedented deluge of data. Analyzing these data in real time presents unique conceptual and computational challenges, and their long-term storage and archiving is scientifically essential for generating reliable, reproducible results. We present here the real-time processing (RTP) system for the Hydrogen Epoch of Reionization Array (HERA), a radio interferometer endeavoring to provide the first detection of the highly redshifted 21 cm signal from Cosmic Dawn and the Epoch of Reionization by an interferometer. The RTP system consists of analysis routines run on raw data shortly after they are acquired, such as calibration and detection of radio-frequency interference (RFI) events. RTP works closely with the Librarian, the HERA data storage and transfer manager which automatically ingests data and transfers copies to other clusters for post-processing analysis. Both the RTP system and the Librarian are public and open source software, which allows for them to be modified for use in other scientific collaborations. When fully constructed, HERA is projected to generate over 50 terabytes (TB) of data each night, and the RTP system enables the successful scientific analysis of these data., Comment: 15 pages, 3 figures, published in Astronomy and Computing

Precise measurements of the 21 cm power spectrum are crucial for understanding the physical processes of hydrogen reionization. Currently, this probe is being pursued by low-frequency radio interferometer arrays. As these experiments come closer to making a first detection of the signal, error estimation will play an increasingly important role in setting robust measurements. Using the delay power spectrum approach, we have produced a critical examination of different ways that one can estimate error bars on the power spectrum. We do this through a synthesis of analytic work, simulations of toy models, and tests on small amounts of real data. We find that, although computed independently, the different error bar methodologies are in good agreement with each other in the noise-dominated regime of the power spectrum. For our preferred methodology, the predicted probability distribution function is consistent with the empirical noise power distributions from both simulated and real data. This diagnosis is mainly in support of the forthcoming HERA upper limit, and also is expected to be more generally applicable., Comment: 35 Pages, 9 Figures, 4 Tables. Replaced with accepted ApJ version; some clarifying text added in response to referee comments with no changes to results

This paper describes the design, implementation, and verification of a test-bed for determining the noise temperature of radio antennas operating between 400-800MHz. The requirements for this test-bed were driven by the HIRAX experiment, which uses antennas with embedded amplification, making system noise characterization difficult in the laboratory. The test-bed consists of two large cylindrical cavities, each containing radio-frequency (RF) absorber held at different temperatures (300K and 77K), allowing a measurement of system noise temperature through the well-known 'Y-factor' method. The apparatus has been constructed at Yale, and over the course of the past year has undergone detailed verification measurements. To date, three preliminary noise temperature measurement sets have been conducted using the system, putting us on track to make the first noise temperature measurements of the HIRAX feed and perform the first analysis of feed repeatability., Comment: 19 pages, 12 figures

The Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX) is a planned interferometric radio telescope array that will ultimately consist of 1024 close packed 6 m dishes that will be deployed at the SKA South Africa site. HIRAX will survey the majority of the southern sky to measure baryon acoustic oscillations (BAO) using the 21 cm hyperfine transition of neutral hydrogen. It will operate between 400-800 MHz with 391 kHz resolution, corresponding to a redshift range of $0.8 < z < 2.5$ and a minimum $\Delta z/z$ of ~0.003. One of the primary science goals of HIRAX is to constrain the dark energy equation of state by measuring the BAO scale as a function of redshift over a cosmologically significant range. Achieving this goal places stringent requirements on the mechanical and optical design of the HIRAX instrument which are described in this paper. This includes the simulations used to optimize the instrument, including the dish focal ratio, receiver support mechanism, and instrument cabling. As a result of these simulations, the dish focal ratio has been reduced to 0.23 to reduce inter-dish crosstalk, the feed support mechanism has been redesigned as a wide (35 cm diam.) central column, and the feed design has been modified to allow the cabling for the receiver to pass directly along the symmetry axis of the feed and dish in order to eliminate beam asymmetries and reduce sidelobe amplitudes. The beams from these full-instrument simulations are also used in an astrophysical m-mode analysis pipeline which is used to evaluate cosmological constraints and determine potential systematic contamination due to physical non-redundancies of the array elements. This end-to-end simulation pipeline was used to inform the dish manufacturing and assembly specifications which will guide the production and construction of the first-stage HIRAX 256-element array., Comment: 20 pages, 11 figures

The central challenge in 21~cm cosmology is isolating the cosmological signal from bright foregrounds. Many separation techniques rely on the accurate knowledge of the sky and the instrumental response, including the antenna primary beam. For drift-scan telescopes such as the Hydrogen Epoch of Reionization Array \citep[HERA, ][]{DeBoer2017} that do not move, primary beam characterization is particularly challenging because standard beam-calibration routines do not apply \citep{Cornwell2005} and current techniques require accurate source catalogs at the telescope resolution. We present an extension of the method from \citet{Pober2012} where they use beam symmetries to create a network of overlapping source tracks that break the degeneracy between source flux density and beam response and allow their simultaneous estimation. We fit the beam response of our instrument using early HERA observations and find that our results agree well with electromagnetic simulations down to a -20~dB level in power relative to peak gain for sources with high signal-to-noise ratio. In addition, we construct a source catalog with 90 sources down to a flux density of 1.4~Jy at 151~MHz., Comment: 22 pages, 22 figures, accepted for publication in ApJ

Characterizing the epoch of reionization (EoR) at $z\gtrsim 6$ via the redshifted 21 cm line of neutral Hydrogen (HI) is critical to modern astrophysics and cosmology, and thus a key science goal of many current and planned low-frequency radio telescopes. The primary challenge to detecting this signal is the overwhelmingly bright foreground emission at these frequencies, placing stringent requirements on the knowledge of the instruments and inaccuracies in analyses. Results from these experiments have largely been limited not by thermal sensitivity but by systematics, particularly caused by the inability to calibrate the instrument to high accuracy. The interferometric bispectrum phase is immune to antenna-based calibration and errors therein, and presents an independent alternative to detect the EoR HI fluctuations while largely avoiding calibration systematics. Here, we provide a demonstration of this technique on a subset of data from the Hydrogen Epoch of Reionization Array (HERA) to place approximate constraints on the brightness temperature of the intergalactic medium (IGM). From this limited data, at $z=7.7$ we infer "$1\sigma$" upper limits on the IGM brightness temperature to be $\le 316$ "pseudo" mK at $\kappa_\parallel=0.33$ "pseudo" $h$ Mpc$^{-1}$ (data-limited) and $\le 1000$ "pseudo" mK at $\kappa_\parallel=0.875$ "pseudo" $h$ Mpc$^{-1}$ (noise-limited). The "pseudo" units denote only an approximate and not an exact correspondence to the actual distance scales and brightness temperatures. By propagating models in parallel to the data analysis, we confirm that the dynamic range required to separate the cosmic HI signal from the foregrounds is similar to that in standard approaches, and the power spectrum of the bispectrum phase is still data-limited (at $\gtrsim 10^6$ dynamic range) indicating scope for further improvement in sensitivity as the array build-out continues., Comment: 22 pages, 12 figures (including sub-figures). Published in PhRvD. Abstract may be slightly abridged compared to the actual manuscript due to length limitations on arXiv

We introduce DAYENU, a linear, spectral filter for HI intensity mapping that achieves the desirable foreground mitigation and error minimization properties of inverse co-variance weighting with minimal modeling of the underlying data. Beyond 21 cm power-spectrum estimation, our filter is suitable for any analysis where high dynamic-range removal of spectrally smooth foregrounds in irregularly (or regularly) sampled data is required, something required by many other intensity mapping techniques. Our filtering matrix is diagonalized by Discrete Prolate Spheroidal Sequences which are an optimal basis to model band-limited foregrounds in 21 cm intensity mapping experiments in the sense that they maximally concentrate power within a finite region of Fourier space. We show that DAYENU enables the access of large-scale line-of-sight modes that are inaccessible to tapered DFT estimators. Since these modes have the largest SNRs, DAYENU significantly increases the sensitivity of 21 cm analyses over tapered Fourier transforms. Slight modifications allow us to use DAYENU as a linear replacement for iterative delay CLEANing (DAYENUREST). We refer readers to the Code section at the end of this paper for links to examples and code., Comment: Accepted to MNRAS, 19 pages, 17 figures. Replacement matches accepted version

The key challenge in the observation of the redshifted 21-cm signal from cosmic reionization is its separation from the much brighter foreground emission. Such separation relies on the different spectral properties of the two components, although, in real life, the foreground intrinsic spectrum is often corrupted by the instrumental response, inducing systematic effects that can further jeopardize the measurement of the 21-cm signal. In this paper, we use Gaussian Process Regression to model both foreground emission and instrumental systematics in $\sim 2$ hours of data from the Hydrogen Epoch of Reionization Array. We find that a simple co-variance model with three components matches the data well, giving a residual power spectrum with white noise properties. These consist of an "intrinsic" and instrumentally corrupted component with a coherence-scale of 20 MHz and 2.4 MHz respectively (dominating the line of sight power spectrum over scales $k_{\parallel} \le 0.2$ h cMpc$^{-1}$) and a baseline dependent periodic signal with a period of $\sim 1$ MHz (dominating over $k_{\parallel} \sim 0.4 - 0.8$h cMpc$^{-1}$) which should be distinguishable from the 21-cm EoR signal whose typical coherence-scales is $\sim 0.8$ MHz., Comment: 15 pages, 15 figures, 1 table, Accepted to MNRAS

In 21 cm cosmology, precision calibration is key to the separation of the neutral hydrogen signal from very bright but spectrally-smooth astrophysical foregrounds. The Hydrogen Epoch of Reionization Array (HERA), an interferometer specialized for 21 cm cosmology and now under construction in South Africa, was designed to be largely calibrated using the self-consistency of repeated measurements of the same interferometric modes. This technique, known as "redundant-baseline calibration" resolves most of the internal degrees of freedom in the calibration problem. It assumes, however, on antenna elements with identical primary beams placed precisely on a redundant grid. In this work, we review the detailed implementation of the algorithms enabling redundant-baseline calibration and report results with HERA data. We quantify the effects of real-world non-redundancy and how they compare to the idealized scenario in which redundant measurements differ only in their noise realizations. Finally, we study how non-redundancy can produce spurious temporal structure in our calibration solutions--both in data and in simulations--and present strategies for mitigating that structure., Comment: 24 Pages, 19 Figures. Updated to match the accepted MNRAS version

We discuss absolute calibration strategies for Phase I of the Hydrogen Epoch of Reionization Array (HERA), which aims to measure the cosmological 21 cm signal from the Epoch of Reionization (EoR). HERA is a drift-scan array with a 10 degree wide field of view, meaning bright, well-characterized point source transits are scarce. This, combined with HERA's redundant sampling of the uv plane and the modest angular resolution of the Phase I instrument, make traditional sky-based and self-calibration techniques difficult to implement with high dynamic range. Nonetheless, in this work we demonstrate calibration for HERA using point source catalogues and electromagnetic simulations of its primary beam. We show that unmodeled diffuse flux and instrumental contaminants can corrupt the gain solutions, and present a gain smoothing approach for mitigating their impact on the 21 cm power spectrum. We also demonstrate a hybrid sky and redundant calibration scheme and compare it to pure sky-based calibration, showing only a marginal improvement to the gain solutions at intermediate delay scales. Our work suggests that the HERA Phase I system can be well-calibrated for a foreground-avoidance power spectrum estimator by applying direction-independent gains with a small set of degrees of freedom across the frequency and time axes., Comment: Accepted to ApJ

We present a study of internal reflection and cross coupling systematics in Phase I of the Hydrogen Epoch of Reionization Array (HERA). In a companion paper, we outlined the mathematical formalism for such systematics and presented algorithms for modeling and removing them from the data. In this work, we apply these techniques to data from HERA's first observing season as a method demonstration. The data show evidence for systematics that, without removal, would hinder a detection of the 21 cm power spectrum for the targeted EoR line-of-sight modes in the range 0.2 < k_parallel < 0.5\ h^-1 Mpc. After systematic removal, we find we can recover these modes in the power spectrum down to the integrated noise-floor of a nightly observation, achieving a dynamic range in the EoR window of 10^-6 in power (mK^2 units) with respect to the bright galactic foreground signal. In the absence of other systematics and assuming the systematic suppression demonstrated here continues to lower noise levels, our results suggest that fully-integrated HERA Phase I may have the capacity to set competitive upper limits on the 21 cm power spectrum. For future observing seasons, HERA will have upgraded analog and digital hardware to better control these systematics in the field., Comment: Submitted to ApJ

The detection of the Epoch of Reionization (EoR) delay power spectrum using a "foreground avoidance method" highly depends on the instrument chromaticity. The systematic effects induced by the radio-telescope spread the foreground signal in the delay domain, which contaminates the EoR window theoretically observable. Applied to the Hydrogen Epoch of Reionization Array (HERA), this paper combines detailed electromagnetic and electrical simulations in order to model the chromatic effects of the instrument, and quantify its frequency and time responses. In particular, the effects of the analogue receiver, transmission cables, and mutual coupling are included. These simulations are able to accurately predict the intensity of the reflections occurring in the 150-m cable which links the antenna to the back-end. They also show that electromagnetic waves can propagate from one dish to another one through large sections of the array due to mutual coupling. The simulated system time response is attenuated by a factor $10^{4}$ after a characteristic delay which depends on the size of the array and on the antenna position. Ultimately, the system response is attenuated by a factor $10^{5}$ after 1400 ns because of the reflections in the cable, which corresponds to characterizable ${k_\parallel}$-modes above 0.7 $h\;\rm{Mpc}^{-1}$ at 150 MHz. Thus, this new study shows that the detection of the EoR signal with HERA Phase I will be more challenging than expected. On the other hand, it improves our understanding of the telescope, which is essential to mitigate the instrument chromaticity., Comment: 13 pages, 17 figures - Submitted to MNRAS - 2nd revision

Development of the hardware, data analysis, and simulation techniques for large compact radio arrays dedicated to mapping the 21 cm line of neutral hydrogen gas has proven to be more difficult than imagined twenty years ago when such telescopes were first proposed. Despite tremendous technical and methodological advances, there are several outstanding questions on how to optimally calibrate and analyze such data. On the positive side, it has become clear that the outstanding issues are purely technical in nature and can be solved with sufficient development activity. Such activity will enable science across redshifts, from early galaxy evolution in the pre-reionization era to dark energy evolution at low redshift., Comment: 10 pages + references, 2 figures, 1 table; APC white paper submitted to the Astro2020 decadal survey. arXiv admin note: text overlap with arXiv:1810.09572

In this white paper, we lay out a US roadmap for high-redshift 21 cm cosmology (30 < z < 6) in the 2020s. Beginning with the currently-funded HERA and MWA Phase II projects and advancing through the decade with a coordinated program of small-scale instrumentation, software, and analysis projects targeting technology development, this roadmap incorporates our current best understanding of the systematics confronting 21 cm cosmology into a plan for overcoming them, enabling next-generation, mid-scale 21 cm arrays to be proposed late in the decade. Submitted for consideration by the Astro2020 Decadal Survey Program Panel for Radio, Millimeter, and Submillimeter Observations from the Ground as a Medium-Sized Project., Comment: 10 pages (plus a cover page and references), 6 figures. Submitted as a APC White Paper for Astro2020

The Cosmic Dawn and Reionization epochs remain a fundamental but challenging frontier of astrophysics and cosmology. We advocate a large-scale, multi-tracer approach to develop a comprehensive understanding of the physics that led to the formation and evolution of the first stars and galaxies. We highlight the line intensity mapping technique to trace the multi-phase reionization topology on large scales, and measure reionization history in detail. Besides 21cm, we advocate for Lya tomography mapping during the epoch of Wouthuysen-Field coupling as an additional probe of the cosmic dawn era., Comment: Astro2020 Science White Paper; 5 pages, 1 figure

We use a semi-analytic model to explore the potential impact of a brief and violent period of radio-loud accretion onto black-holes (The Radio Scream) during the Cosmic Dawn on the HI hyperfine 21 cm signal. We find that radio emission from super-massive black hole seeds can impact the global 21 cm signal at the level of tens to hundreds of percent provided that they were as radio loud as $z\approx1$ black holes and obscured by gas with column depths of $N_\text{H}\gtrsim 10^{23}$ cm$^{-2}$. We determine plausible sets of parameters that reproduce some of the striking features of the EDGES absorption feature including its depth, timing, and side steepness while producing radio/X-ray backgrounds and source counts that are consistent with published limits. Scenarios yielding a dramatic 21 cm signature also predict large populations of $\sim \mu$Jy point sources that will be detectable in future deep surveys from the Square Kilometer Array (SKA). Thus, 21 cm measurements, complemented by deep point source surveys, have the potential to constrain optimistic scenarios where super-massive black-hole progenitors were radio-loud., Comment: 21 pages. 15 figures. Accepted to MNRAS. Replacement matches accepted version

Radio Frequency Interference (RFI) is an ever-present limiting factor among radio telescopes even in the most remote observing locations. When looking to retain the maximum amount of sensitivity and reduce contamination for Epoch of Reionization studies, the identification and removal of RFI is especially important. In addition to improved RFI identification, we must also take into account computational efficiency of the RFI-Identification algorithm as radio interferometer arrays such as the Hydrogen Epoch of Reionization Array grow larger in number of receivers. To address this, we present a Deep Fully Convolutional Neural Network (DFCN) that is comprehensive in its use of interferometric data, where both amplitude and phase information are used jointly for identifying RFI. We train the network using simulated HERA visibilities containing mock RFI, yielding a known "ground truth" dataset for evaluating the accuracy of various RFI algorithms. Evaluation of the DFCN model is performed on observations from the 67 dish build-out, HERA-67, and achieves a data throughput of 1.6$\times 10^{5}$ HERA time-ordered 1024 channeled visibilities per hour per GPU. We determine that relative to an amplitude only network including visibility phase adds important adjacent time-frequency context which increases discrimination between RFI and Non-RFI. The inclusion of phase when predicting achieves a Recall of 0.81, Precision of 0.58, and $F_{2}$ score of 0.75 as applied to our HERA-67 observations., Comment: 11 pages, 7 figures

The separation of cosmological signal from astrophysical foregrounds is a fundamental challenge for any effort to probe the evolution of neutral hydrogen during the Cosmic Dawn and epoch of reionization (EoR) using the 21 cm hyperfine transition. Foreground separation is made possible by their intrinsic spectral smoothness, making them distinguishable from spectrally complex cosmological signal even though they are ~5 orders of magnitude brighter. Precisely calibrated radio interferometers are essential to maintaining the smoothness and thus separability of the foregrounds. One powerful calibration strategy is to use redundant measurements between pairs of antennas with the same physical separation in order to solve for each antenna's spectral response without reference to a sky model. This strategy is being employed by the Hydrogen Epoch of Reionization Array (HERA), a large radio telescope in South Africa that is now observing while being built out to 350 14-m dishes. However, the deviations from perfect redundancy inherent in any real radio telescope complicate the calibration problem. Using simulations of HERA, we show how calibration with antenna-to-antenna variations in dish construction and placement generally lead to spectral structure in otherwise smooth foregrounds that significantly reduces the number of cosmological modes available to a 21 cm measurement. However, we also show that this effect can be largely eliminated by a modified redundant-baseline calibration strategy that relies predominantly on short baselines., Comment: 14 pages, 12 figures. Updated to match the accepted MNRAS version

Interferometric arrays seeking to measure the 21 cm signal from the Epoch of Reionization must contend with overwhelmingly bright emission from foreground sources. Accurate recovery of the 21 cm signal will require precise calibration of the array, and several new avenues for calibration have been pursued in recent years, including methods using redundancy in the antenna configuration. The newly upgraded Phase II of Murchison Widefield Array (MWA) is the first interferometer that has large numbers of redundant baselines while retaining good instantaneous UV-coverage. This array therefore provides a unique opportunity to compare redundant calibration with sky-model based algorithms. In this paper, we present the first results from comparing both calibration approaches with MWA Phase II observations. For redundant calibration, we use the package OMNICAL, and produce sky-based calibration solutions with the analysis package Fast Holographic Deconvolution (FHD). There are three principal results. (1) We report the success of OMNICAL on observations of ORBComm satellites, showing substantial agreement between redundant visibility measurements after calibration. (2) We directly compare OMNICAL calibration solutions with those from FHD, and demonstrate these two different calibration schemes give extremely similar results. (3) We explore improved calibration by combining OMNICAL and FHD. We evaluate these combined methods using power spectrum techniques developed for EoR analysis and find evidence for marginal improvements mitigating artifacts in the power spectrum. These results are likely limited by signal-to-noise in the six hours of data used, but suggest future directions for combining these two calibration schemes., Comment: 20 pages, 11 figures. Accepted to ApJ

New massively redundant low frequency arrays allow for a novel investigation of closure relations in interferometry. We employ commissioning data from the Hydrogen Epoch of Reionization Array to investigate closure quantities in this densely packed grid array of 14m antennas operating at 100 MHz to 200 MHz. We investigate techniques that utilize closure phase spectra for redundant triads to estimate departures from redundancy for redundant baseline visibilities. We find a median absolute deviation from redundancy in closure phase across the observed frequency range of about 4.5deg. This value translates into a non-redundancy per visibility phase of about 2.6deg, using prototype electronics. The median absolute deviations from redundancy decrease with longer baselines. We show that closure phase spectra can be used to identify ill-behaved antennas in the array, independent of calibration. We investigate the temporal behavior of closure spectra. The Allan variance increases after a one minute stride time, due to passage of the sky through the primary beam of the transit telescope. However, the closure spectra repeat to well within the noise per measurement at corresponding local sidereal times (LST) from day to day. In future papers in this series we will develop the technique of using closure phase spectra in the search for the HI 21cm signal from cosmic reionization., Comment: 32 pages. 11 figures. Accepted to Radio Science

We estimate the 21 cm Radio Background from accretion onto the first intermediate-mass Black Holes between $z\approx 30$ and $z\approx 16$. Combining potentially optimistic, but plausible, scenarios for black hole formation and growth with empirical correlations between luminosity and radio-emission observed in low-redshift active galactic nuclei, we find that a model of black holes forming in molecular cooling halos is able to produce a 21 cm background that exceeds the Cosmic Microwave Background (CMB) at $z \approx 17$ though models involving larger halo masses are not entirely excluded. Such a background could explain the surprisingly large amplitude of the 21 cm absorption feature recently reported by the EDGES collaboration. Such black holes would also produce significant X-ray emission and contribute to the $0.5-2$ keV soft X-ray background at the level of $\approx 10^{-13}-10^{-12}$ erg sec$^{-1}$ cm$^{-2}$ deg$^{-2}$, consistent with existing constraints. In order to avoid heating the IGM over the EDGES trough, these black holes would need to be obscured by Hydrogen column depths of $ N_\text{H} \sim 5 \times 10^{23} \text{cm}^{-2}$. Such black holes would avoid violating contraints on the CMB optical depth from Planck if their UV photon escape fractions were below $f_{\text{esc}} \lesssim 0.1$, which would be a natural result of $N_\text{H} \sim 5 \times 10^{23} \text{cm}^{-2}$ imposed by an unheated IGM., Comment: 10 pages, 7 figures, accepted to ApJ, replacement to match submitted version

Foreground power dominates the measurements of interferometers that seek a statistical detection of highly-redshifted HI emission from the Epoch of Reionization (EoR). The chromaticity of the instrument creates a boundary in the Fourier transform of frequency (proportional to $k_\parallel$) between spectrally smooth emission, characteristic of the strong synchrotron foreground (the "wedge"), and the spectrally structured emission from HI in the EoR (the "EoR window"). Faraday rotation can inject spectral structure into otherwise smooth polarized foreground emission, which through instrument effects or miscalibration could possibly pollute the EoR window. Using data from the HERA 19-element commissioning array, we investigate the polarization response of this new instrument in the power spectrum domain. We perform a simple image-based calibration based on the unpolarized diffuse emission of the Global Sky Model, and show that it achieves qualitative redundancy between the nominally-redundant baselines of the array and reasonable amplitude accuracy. We construct power spectra of all fully polarized coherencies in all pseudo-Stokes parameters. We compare to simulations based on an unpolarized diffuse sky model and detailed electromagnetic simulations of the dish and feed, confirming that in Stokes I, the calibration does not add significant spectral structure beyond the expected level. Further, this calibration is stable over the 8 days of observations considered. Excess power is seen in the power spectra of the linear polarization Stokes parameters which is not easily attributable to leakage via the primary beam, and results from some combination of residual calibration errors and actual polarized emission. Stokes V is found to be highly discrepant from the expectation of zero power, strongly pointing to the need for more accurate polarized calibration., Comment: 21 pages, 12 figures, submitted to ApJ