Search

This paper presents an ensemble data assimilation method using the pseudo ensembles generated by denoising diffusion probabilistic model. Since the model is trained against noisy and sparse observation data, this model can produce divergent ensembles close to observations. Thanks to the variance in generated ensembles, our proposed method displays better performance than the well-established ensemble data assimilation method when the simulation model is imperfect.

Recent developments in machine learning have enabled accurate predictions of the dynamics of slow structural relaxation in glass-forming systems. However, existing machine-learning models for these tasks are mostly designed such that they learn a single dynamic quantity and relate it to the structural features of glassy liquids. In this study, we propose a graph neural network model, ``BOnd TArgeting Network (BOTAN)'', that learns relative motion between neighboring pairs of particles, in addition to the self-motion of particles. By relating the structural features to these two different dynamical variables, the model autonomously acquires the ability to discern how different dynamical processes, strain fluctuations and particle rearrangements, affect the self-motion of particles undergoing slow relaxation, and thus can predict with high precision how slow structural relaxation develops in space and time., Comment: 11 pages, 10 figures, a typo fixed in the title; accepted in J. Chem. Phys

We developed a high-speed image reduction pipeline using Graphics Processing Units (GPUs) as hardware accelerators. Astronomers desire detecting EM counterpart of gravitational-wave sources as soon as possible for sharing positional information to organize systematic follow-up observations. Therefore, high-speed image processing is important. We developed a new image reduction pipeline for our robotic telescope system, which uses a GPU via a Python package CuPy to achieve high-speed image processing. As a result, the processing speed was increased by more than a factor of forty to that of the current pipeline, while maintaining the same functions., Comment: 24 pages, 18 figures, 6 tables, accepted for publication in PASJ, code available at this http URL https://github.com/MNiwano/Eclaire

GRB 041006 was detected by HETE-2 at 12:18:08 UT on 06 October 2004. This GRB displays a soft X-ray emission, a precursor before the onset of the main event, and also a soft X-ray tail after the end of the main peak. The light curves in four different energy bands display different features; At higher energy bands several peaks are seen in the light curve, while at lower energy bands a single broader bump dominates. It is expected that these different features are the result of a mixture of several components each of which has different energetics and variability. To reveal the nature of each component, we analysed the time resolved spectra and they are successfully resolved into several components. We also found that these components can be classified into two distinct classes; One is a component which has an exponential decay of $E_{p}$ with a characteristic timescale shorter than $\sim$ 30 sec, and its spectrum is well represented by a broken power law function, which is frequently observed in many prompt GRB emissions, so it should have an internal-shock origin. Another is a component whose $E_{p}$ is almost unchanged with characteristic timescale longer than $\sim$ 60 sec, and shows a very soft emission and slower variability. The spectrum of the soft component is characterized by either a broken power law or a black body spectrum. This component might originate from a relatively wider and lower velocity jet or a photosphere of the fireball. By assuming that the soft component is a thermal emission, the radiation radius is initially $4.4 \times 10^{6}$ km, which is a typical radius of a blue supergiant, and its expansion velocity is $2.4 \times 10^{5}$ km/s in the source frame., Comment: 19 pages, 10 figures, accepted for publication in PASJ, replaced with the accepted version (minor correction)

We observed an X-ray afterglow of GRB 060904A with the Swift and Suzaku satellites. We found rapid spectral softening during both the prompt tail phase and the decline phase of an X-ray flare in the BAT and XRT data. The observed spectra were fit by power-law photon indices which rapidly changed from $\Gamma = 1.51^{+0.04}_{-0.03}$ to $\Gamma = 5.30^{+0.69}_{-0.59}$ within a few hundred seconds in the prompt tail. This is one of the steepest X-ray spectra ever observed, making it quite difficult to explain by simple electron acceleration and synchrotron radiation. Then, we applied an alternative spectral fitting using a broken power-law with exponential cutoff (BPEC) model. It is valid to consider the situation that the cutoff energy is equivalent to the synchrotron frequency of the maximum energy electrons in their energy distribution. Since the spectral cutoff appears in the soft X-ray band, we conclude the electron acceleration has been inefficient in the internal shocks of GRB 060904A. These cutoff spectra suddenly disappeared at the transition time from the prompt tail phase to the shallow decay one. After that, typical afterglow spectra with the photon indices of 2.0 are continuously and preciously monitored by both XRT and Suzaku/XIS up to 1 day since the burst trigger time. We could successfully trace the temporal history of two characteristic break energies (peak energy and cutoff energy) and they show the time dependence of $\propto t^{-3} \sim t^{-4}$ while the following afterglow spectra are quite stable. This fact indicates that the emitting material of prompt tail is due to completely different dynamics from the shallow decay component. Therefore we conclude the emission sites of two distinct phenomena obviously differ from each other., Comment: 19 pages, 9 figures, accepted for publication in PASJ (Suzaku 2nd Special Issue)

A long X-ray flash was detected and localized by the instruments aboard the High Energy Transient Explorer II (HETE-2) at 00:03:30 UT on 2004 September 16. The position was reported to the GRB Coordinates Network (GCN) approximately 2 hours after the burst. This burst consists of two peaks separated by 200 s, with durations of 110 s and 60 s. We have analyzed the energy spectra of the 1st and 2nd peaks observed with the Wide Field X-Ray Monitor (WXM) and the French Gamma Telescope (FREGATE). We discuss the origin of the 2nd peak in terms of flux variabilities and timescales. We find that it is most likely part of the prompt emission, and is explained by the long-acting engine model. This feature is similar to some bright X-ray flares detected in the early afterglow phase of bursts observed by the Swift satellite., Comment: 9 pages, 4 figures, Accepted for publication in PASJ