Publications

The concentration of nitrous oxide (N₂O), an ozone-depleting greenhouse gas, is rapidly increasing in the atmosphere. Most atmospheric N₂O originates in terrestrial ecosystems, of which the majority can be attributed to microbial cycling of nitrogen in agricultural soils. Here, we demonstrate how the abundance of nitrogen cycling genes vary across intensively managed agricultural fields and adjacent restored wetlands in the Sacramento-San Joaquin Delta in California, USA. We found that the abundances of nirS and nirK genes were highest at the intensively managed organic-rich cornfield and significantly outnumber any other gene abundances, suggesting very high N₂O production potential. The quantity of nitrogen transforming genes, particularly those responsible for denitrification, nitrification and DNRA, were highest in the agricultural sites, whereas nitrogen fixation and ANAMMOX was strongly associated with the wetland sites. Although the abundance of nosZ genes was also high at the agricultural sites, the ratio of nosZ genes to nir genes was significantly higher in wetland sites indicating that these sites could act as a sink of N₂O. These findings suggest that wetland restoration could be a promising natural climate solution not only for carbon sequestration but also for reduced N₂O emissions.

Currently, the most successful learning models in computer vision are based on learning successive representations followed by a decision layer. This is usually actualized through feedforward multilayer neural networks, e.g. ConvNets, where each layer forms one of such successive representations. However, an alternative that can achieve the same goal is a feedback based approach in which the representation is formed in an iterative manner based on a feedback received from previous iteration's output. We establish that a feedback based approach has several fundamental advantages over feedforward: it enables making early predictions at the query time, its output naturally conforms to a hierarchical structure in the label space (e.g. a taxonomy), and it provides a new basis for Curriculum Learning. We observe that feedback networks develop a considerably different representation compared to feedforward counterparts, in line with the aforementioned advantages. We put forth a general feedback based learning architecture with the endpoint results on par or better than existing feedforward networks with the addition of the above advantages. We also investigate several mechanisms in feedback architectures (e.g. skip connections in time) and design choices (e.g. feedback length). We hope this study offers new perspectives in quest for more natural and practical learning models.

The optical afterglow of long-duration GRB 071003 is among the brightest yet to be detected from any GRB, with A ≈ 12 mag in KAIT observations starting 42 s after the GRB trigger, including filtered detections during prompt emission. However, our high-S/N afterglow spectrum displays only extremely weak absorption lines at what we argue is the host redshift of z = 1.60435, in contrast to the three other, much stronger Mg IT absorption systems observed at lower redshifts. Together with Keck adaptive optics observations, which fail to reveal a host galaxy coincident with the burst position, our observations suggest a halo progenitor and offer a cautionary tale about the use of Mg n for GRB redshift determination. We present early- through late-time observations spanning the electro-magnetic spectrum, constrain the connection between the prompt emission and early variations in the light curve (we observe no correlation), and discuss possible origins for an unusual, marked rebrightening that occurs a few hours after the burst: likely either a late-time refreshed shock or a wide-angle secondary jet. Analysis of the late-time afterglow is most consistent with a wind environment, suggesting a massive star progenitor. Together with GRB 070125, this may indicate that a small but significant portion of star formation in the early universe occurred far outside what we consider a normal galactic disk. © 2008. The American Astronomical Society. All rights reserved.

Copper deficiency is implicated in a variety of genetic, neurological, cardiovascular, and metabolic diseases. Current approaches for addressing copper deficiency rely on generic copper supplementation, which can potentially lead to detrimental off-target metal accumulation in unwanted tissues and subsequently trigger oxidative stress and damage cascades. Here we present a new modular platform for delivering metal ions in a tissue-specific manner and demonstrate liver-targeted copper supplementation as a proof of concept of this strategy. Specifically, we designed and synthesized an N-acetylgalactosamine-functionalized ionophore, Gal-Cu(gtsm), to serve as a copper-carrying "Trojan Horse" that targets liver-localized asialoglycoprotein receptors (ASGPRs) and releases copper only after being taken up by cells, where the reducing intracellular environment triggers copper release from the ionophore. We utilized a combination of bioluminescence imaging and inductively coupled plasma mass spectrometry assays to establish ASGPR-dependent copper accumulation with this reagent in both liver cell culture and mouse models with minimal toxicity. The modular nature of our synthetic approach presages that this platform can be expanded to deliver a broader range of metals to specific cells, tissues, and organs in a more directed manner to treat metal deficiency in disease.

Abstract We report our photometric and spectroscopic follow-up observations of the superoutbursts of three dwarf novae (DNe; GWAC 180415A, GWAC 181017A, and GWAC 181211A) identified independently by the Ground Wide-angle Cameras system, one of the ground-based instruments of the China–France SVOM mission. Based on a combination of our photometry and that taken from the American Association of Variable Star Observers, our period analysis of the superhumps enables us to determine the mass ratios to be 0.0967–0.1163, 0.1879–0.1883, and 0.0981–0.1173 for GWAC 180415A, GWAC 181017A, and GWAC 181211A, respectively. GWAC 180415A can be firmly identified as a WZ sge-type DNe due to its long duration (∼2 weeks) multiple rebrightenings with amplitudes of 3–4 mag, the early superhump associated with a double-wave modulation and the low-mass ratio. The inferred low-mass ratio and location in the <?CDATA $\varepsilon \mbox{--}{P}_{\mathrm{orb}}$?> diagram suggest that GWAC 181211A is a WZ sge-type DNe candidate. The measured Balmer decrements suggest the Balmer line emission is produced from an optical thick region in GWAC 180415A and GWAC 181017A, and from an optically thin region in GWAC 181211A.

A strongly interacting plasma of linearly dispersing electron and hole excitations in two spatial dimensions (2D), also known as a Dirac fluid, can be captured by relativistic hydrodynamics and shares many universal features with other quantum critical systems. Here, we propose a one-dimensional (1D) model to capture key aspects of the 2D Dirac fluid while including lattice effects and being amenable to nonperturbative computation. When interactions are added to the Dirac-like 1D dispersion without opening a gap, we show that this kind of irrelevant interaction is able to preserve Fermi-liquid-like quasiparticle features while relaxing a zero-momentum charge current via collisions between particle-hole excitations, leading to resistivity that is linear in temperature via a mechanism previously discussed for large-diameter metallic carbon nanotubes. We further provide a microscopic lattice model and obtain numerical results via density-matrix renormalization group simulations, which support the above physical picture. The limits on such fast relaxation at strong coupling are of considerable interest because of the ubiquity of bad metals in experiments.

We report observations of 240 Cepheid variables obtained with the Near Infrared Camera (NICMOS) through the F160W filter on the Hubble Space Telescope (HST). The Cepheids are distributed across six recent hosts of Type Ia supernovae (SNe Ia) and the "maser galaxy" NGC 4258, allowing us to directly calibrate the peak luminosities of the SNe Ia from the precise, geometric distance measurements provided by the masers. New features of our measurement include the use of the same instrument for all Cepheid measurements across the distance ladder and homogeneity of the Cepheid periods and metallicities thus necessitating only a differential measurement of Cepheid fluxes and reducing the largest systematic uncertainties in the determination of the fiducial SN Ia luminosity. The NICMOS measurements reduce differential extinction in the host galaxies by a factor of 5 over past optical data. Combined with an expanded of 240 SNe Ia at z<0.1 which define their magnitude-redshift relation, we find H_0=74.2 +/-3.6, a 4.8% uncertainty including both statistical and systematic errors. We show that the factor of 2.2 improvement in the precision of H_0 is a significant aid to the determination of the equation-of-state of dark energy, w = P/(rho c^2). Combined with the WMAP 5-year measurement of Omega_M h^2, we find w= -1.12 +/- 0.12 independent of high-redshift SNe Ia or baryon acoustic oscillations (BAO). This result is also consistent with analyses based on the combination of high-z SNe Ia and BAO. The constraints on w(z) now with high-z SNe Ia and BAO are consistent with a cosmological constant and improved by a factor of 3 from the refinement in H_0 alone. We show future improvements in H_0 are likely and will further contribute to multi-technique studies of dark energy.