Search for: All records

Abstract Generative Adversarial Networks trained on samples of simulated or actual events have been proposed as a way of generating large simulated datasets at a reduced computational cost. In this work, a novel approach to perform the simulation of photodetector signals from the time projection chamber of the EXO-200 experiment is demonstrated. The method is based on a Wasserstein Generative Adversarial Network — a deep learning technique allowing for implicit non-parametric estimation of the population distribution for a given set of objects. Our network is trained on real calibration data using raw scintillation waveforms as input. We find that it is able to produce high-quality simulated waveforms an order of magnitude faster than the traditional simulation approach and, importantly, generalize from the training sample and discern salient high-level features of the data. In particular, the network correctly deduces position dependency of scintillation light response in the detector and correctly recognizes dead photodetector channels. The network output is then integrated into the EXO-200 analysis framework to show that the standard EXO-200 reconstruction routine processes the simulated waveforms to produce energy distributions comparable to that of real waveforms. Finally, the remaining discrepancies and potential ways to improve the approach further are highlighted. 

Interactions between plants and herbivores are central in most ecosystems, but their strength is highly variable. The amount of variability within a system is thought to influence most aspects of plant-herbivore biology, from ecological stability to plant defense evolution. Our understanding of what influences variability, however, is limited by sparse data. We collected standardized surveys of herbivory for 503 plant species at 790 sites across 116° of latitude. With these data, we show that within-population variability in herbivory increases with latitude, decreases with plant size, and is phylogenetically structured. Differences in the magnitude of variability are thus central to how plant-herbivore biology varies across macroscale gradients. We argue that increased focus on interaction variability will advance understanding of patterns of life on Earth.

 

ABSTRACT We report the discovery of the closest known black hole candidate as a binary companion to V723 Mon. V723 Mon is a nearby ($d\sim 460\, \rm pc$), bright (V ≃ 8.3 mag), evolved (Teff, giant ≃ 4440 K, and Lgiant ≃ 173 L⊙) red giant in a high mass function, f(M) = 1.72 ± 0.01 M⊙, nearly circular binary (P = 59.9 d, e ≃ 0). V723 Mon is a known variable star, previously classified as an eclipsing binary, but its All-Sky Automated Survey, Kilodegree Extremely Little Telescope, and Transiting Exoplanet Survey Satellite light curves are those of a nearly edge-on ellipsoidal variable. Detailed models of the light curves constrained by the period, radial velocities, and stellar temperature give an inclination of $87.0^{\circ ^{+1.7^\circ }}_{-1.4^\circ }$, a mass ratio of q ≃ 0.33 ± 0.02, a companion mass of Mcomp = 3.04 ± 0.06 M⊙, a stellar radius of Rgiant = 24.9 ± 0.7 R⊙, and a giant mass of Mgiant = 1.00 ± 0.07 M⊙. We identify a likely non-stellar, diffuse veiling component with contributions in the B and V band of ${\sim }63{{\ \rm per\ cent}}$ and ${\sim }24{{\ \rm per\ cent}}$, respectively. The SED and the absence of continuum eclipses imply that the companion mass must be dominated by a compact object. We do observe eclipses of the Balmer lines when the dark companion passes behind the giant, but their velocity spreads are low compared to observed accretion discs. The X-ray luminosity of the system is $L_{\rm X}\simeq 7.6\times 10^{29}~\rm ergs~s^{-1}$, corresponding to L/Ledd ∼ 10−9. The simplest explanation for the massive companion is a single compact object, most likely a black hole in the ‘mass gap’.