河原 創

We present an auto-differentiable spectral modeling of exoplanets and brown dwarfs. This model enables a fully Bayesian inference of the high--dispersion data to fit the ab initio line-by-line spectral computation to the observed spectrum by combining it with the Hamiltonian Monte Carlo in recent probabilistic programming languages. An open source code, exojax, developed in this study, was written in Python using the GPU/TPU compatible package for automatic differentiation and accelerated linear algebra, JAX (Bradbury et al. 2018). We validated the model by comparing it with existing opacity calculators and a radiative transfer code and found reasonable agreements of the output. As a demonstration, we analyzed the high-dispersion spectrum of a nearby brown dwarf, Luhman 16 A and found that a model including water, carbon monoxide, and $\mathrm{H_2/He}$ collision induced absorption was well fitted to the observed spectrum ($R=10^5$ and 2.28-2.30 $\mu$m). As a result, we found that $T_0=1295_{-32}^{+35}$ K at 1 bar and C/O $=0.62 \pm 0.03$, which is slightly higher than the solar value. This work demonstrates the potential of full Bayesian analysis of brown dwarfs and exoplanets as observed by high-dispersion spectrographs and also directly-imaged exoplanets as observed by high-dispersion coronagraphy.

<jats:title>Abstract</jats:title> <jats:p>We report the first detection of a hydroxyl radical (OH) emission signature in the planetary atmosphere outside the solar system, in this case, in the dayside of WASP-33b. We analyze high-resolution near-infrared emission spectra of WASP-33b taken using the InfraRed Doppler spectrograph on the 8.2 m Subaru telescope. The telluric and stellar lines are removed using a detrending algorithm, <jats:sc>SysRem</jats:sc>. The residuals are then cross-correlated with OH and H<jats:sub>2</jats:sub>O planetary spectrum templates produced using several different line lists. We check and confirm the accuracy of OH line lists by cross-correlating with the spectrum of GJ 436. As a result, we detect the emission signature of OH at <jats:italic>K</jats:italic> <jats:sub>p</jats:sub> of <jats:inline-formula> <jats:tex-math> <?CDATA ${230.9}_{-7.4}^{+6.9}$?> </jats:tex-math> <jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="apjlabec71ieqn1.gif" xlink:type="simple" /> </jats:inline-formula> km s<jats:sup>−1</jats:sup> and <jats:italic>v</jats:italic> <jats:sub>sys</jats:sub> of −0.3<jats:inline-formula> <jats:tex-math> <?CDATA ${}_{-5.6}^{+5.3}$?> </jats:tex-math> <jats:inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="apjlabec71ieqn2.gif" xlink:type="simple" /> </jats:inline-formula> km s<jats:sup>−1</jats:sup> with a signal-to-noise ratio (S/N) of 5.4 and a significance of 5.5<jats:italic>σ</jats:italic>. Additionally, we marginally detect H<jats:sub>2</jats:sub>O emission in the <jats:italic>H</jats:italic>-band with an S/N of 4.0 and a significance of 5.2<jats:italic>σ</jats:italic> using the POKAZATEL line list. However, no significant signal is detected using the HITEMP 2010, which might be due to differences in line positions and strengths, as well as the incompleteness of the line lists. Nonetheless, this marginal detection is consistent with the prediction that H<jats:sub>2</jats:sub>O is mostly thermally dissociated in the upper atmosphere of the ultra-hot Jupiters. Therefore, along with CO, OH is expected to be one of the most abundant O-bearing molecules in the dayside atmosphere of ultra-hot Jupiters and should be considered when studying their atmospheres.</jats:p>

We report the discovery of three edge-on binaries with white dwarf (WD) companions that gravitationally magnify (instead of eclipsing) the light of their stellar primaries, as revealed by a systematic search for pulses with long periods in the Kepler photometry. We jointly model the self-lensing light curves and radial-velocity orbits to derive the WD masses, all of which are close to 0.6 solar masses. The orbital periods are long, ranging from 419 to 728 days, and the eccentricities are low, all less than 0.2. These characteristics are reminiscent of the orbits found for many blue stragglers in open clusters and the field, for which stable mass transfer due to Roche-lobe overflow from an evolving primary (now a WD) has been proposed as the formation mechanism. Because the actual masses for our three WD companions have been accurately determined, these self-lensing systems would provide excellent tests for models of interacting binaries.

We report high-resolution spectroscopic detection of TiO molecular signature in the day-side spectra of WASP-33b, the second hottest known hot Jupiter. We used High-Dispersion Spectrograph (HDS; R similar to 165,000) in the wavelength range of 0.62-0.88 m with the Subaru telescope to obtain the day-side spectra of WASP-33b. We suppress and correct the systematic effects of the instrument, the telluric and stellar lines by using SYSREM algorithm after the selection of good orders based on Barnard star and other M-type stars. We detect a 4.8 sigma signal at an orbital velocity of K-p= +237.5(-5.0)(+13.0) km s(-1) and systemic velocity V-sys= -1.5(10.5)(+4.0) km s(-1), which agree with the derived values from the previous analysis of primary transit. Our detection with the temperature inversion model implies the existence of stratosphere in its atmosphere, however, we were unable to constrain the volume-mixing ratio of the detected TiO. We also measure the stellar radial velocity and use it to obtain a more stringent constraint on the orbital velocity, Kp=239.0(1.0)(+2.0) km s(-1). Our results demonstrate that high-dispersion spectroscopy is a powerful tool to characterize the atmosphere of an exoplanet, even in the optical wavelength range, and show a promising potential in using and developing similar techniques with high-dispersion spectrograph on current 10m-class and future extremely large telescopes.

We consider the time-frequency analysis of a scattered light curve of a directly imaged exoplanet. We show that the geometric effect due to planetary obliquity and orbital inclination induce the frequency modulation of the apparent diurnal periodicity. We construct a model of the frequency modulation and compare it with the instantaneous frequency extracted from the pseudo-Wigner distribution of simulated light curves of a cloudless Earth. The model provides good agreement with the simulated modulation factor, even for the light curve with Gaussian noise comparable to the signal. Notably, the shape of the instantaneous frequency is sensitive to the difference between the prograde, retrograde, and pole-on spin rotations. While our technique requires the albedo map to be static, it does not need to solve the albedo map of the planet. The time-frequency analysis is complementary to other methods which utilize the amplitude modulation. This paper demonstrates the importance of the frequency domain of the photometric variability for the characterization of directly imaged exoplanets in future research.

We visually inspected the light curves of 7557 Kepler Objects of Interest (KOIs) to search for single. transit events (STEs) that were possibly due to long-period giant planets. We identified 28 STEs in 24 KOIs, among which 14 events are newly reported in this paper. We estimate the radius and orbital period of the objects causing STEs by fitting the STE light curves simultaneously with the transits of the other planets in the system or with. prior information on the host star density. As a result, we found that STEs in seven of those systems are consistent with Neptune- to Jupiter-sized objects of orbital periods ranging from a few to similar to 20 years. We also estimate that. greater than or similar to 20% of the compact multi-transiting systems host cool giant planets with periods greater than or similar to 3 years on the basis of their occurrence in the KOIs with multiple candidates, assuming the small mutual inclination between inner and outer planetary orbits.

We consider the effect of planetary spin on the planetary radial velocity (PRV) in dayside spectra of exoplanets. To understand the spin effect qualitatively, we derive an analytic formula of the intensity-weighted radial velocity from the planetary surface on the following assumptions: (1) constant and solid rotation without precession, (2) stable and uniform distribution of molecules/atoms, (3) emission models from the dayside hemisphere, and (4) a circular orbit. On these assumptions, we find that the curve of the PRV is distorted by the planetary spin and this anomaly is characterized by the spin radial velocity at the equator and a projected angle on a celestial plane between the spin axis and the axis of orbital motion lambda(p) in a manner analogous to the Rossiter-McLaughlin effect. The latter can constrain the planetary obliquity. Creating mock PRV data with 3 km s(-1) accuracy, we demonstrate how lambda(p) and the spin radial velocity at the equator are estimated. We find that the stringent constraint of eccentricity is crucial to detect the spin effect. Though our formula is still qualitative, we conclude that the PRV in the dayside spectra will be a powerful means for constraining the planetary spin.

The oxygen absorption line imprinted in the scattered light from Earth-like planets has been considered the most promising metabolic biomarker for exolife. We examine the feasibility of the detection of the 1.27 mu m oxygen band from habitable exoplanets, in particular, around late-type stars observed with a future instrument on a 30 m class ground-based telescope. We analyzed the night airglow around 1.27 mu m with the IRCS/echelle spectrometer on Subaru and found that the strong telluric emission from atmospheric oxygen molecules declines by an order of magnitude by midnight. By compiling nearby star catalogs combined with the sky background model, we estimate the detectability of the oxygen absorption band from an Earth twin, if it exists, around nearby stars. We find that the most dominant source of photon noise for the oxygen 1.27 mu m band detection comes from the night airglow if the contribution of the stellar point-spread function (PSF) halo is suppressed enough to detect the planet. We conclude that the future detectors, for which the detection contrast is limited by photon noise, can detect the oxygen 1.27 mu m absorption band of Earth twins for similar to 50 candidates of the late-type star. This paper demonstrates the importance of deploying a small inner working angle as an efficient coronagraph and extreme adaptive optics on extremely large telescopes, and clearly shows that doing so will enable the study of potentially habitable planets.

We perform a systematic X-ray analysis of six giant radio relics of four<br /> clusters with Suzaku satellite by compiling new analysis of CIZA 2242.8-5301,<br /> Zwcl 2341.1-0000, and Abell 3667 and previous results of A3667 and A3376.<br /> Especially we first observed the narrow (~50 kpc) relic of CIZA 2242.8-5301 by<br /> Suzaku satellite, which enable us to reduce the projection effect. From the<br /> spectroscopic temperature profiles across the relic, we find that temperature<br /> profiles exhibits significant jumps across the relics for CIZA 2242.8-5301,<br /> A3376, A3667NW and A3667SE. We evaluated the Mach number from X...

Aiming at obtaining detailed information on the surface environment of Earth analogs, Kawahara & Fujii proposed an inversion technique of annual scattered light curves named spin-orbit tomography (SOT), which enables us to sketch a two-dimensional albedo map from annual variation of the disk-integrated scattered light, and demonstrated the method with a planet in a face-on orbit. We extend it to be applicable to general geometric configurations, including low-obliquity planets like the Earth in inclined orbits. We simulate light curves of the Earth in an inclined orbit in three photometric bands (0.4-0.5 mu m, 0.6-0.7 mu m, and 0.8-0.9 mu m) and show that the distribution of clouds, snow, and continents is retrieved with the aid of the SOT. We also demonstrate the SOT by applying it to an upright Earth, a tidally locked Earth, and Earth analogs with ancient continental configurations. The inversion is model independent in the sense that we do not assume specific albedo models when mapping the surface, and hence applicable in principle to any kind of inhomogeneity. This method can potentially serve as a unique tool to investigate the exohabitats/exoclimes of Earth analogs.

We develop an inversion technique of annual scattered light curves to sketch a two-dimensional albedo map of exoplanets in face-on orbits. As a test bed for future observations of extrasolar terrestrial planets, we apply this mapping technique to simulated light curves of a mock Earth-twin at a distance of 10 pc in a face-on circular orbit. A primary feature in recovered albedo maps traces the annual mean distribution of clouds. To extract information of other surface types, we attempt to reduce the cloud signal by taking the difference of two bands. We find that the inversion of reflectivity difference between 0.8-0.9 and 0.4-0.5 mu m bands roughly recovers the continental distribution, except for high latitude regions persistently covered with clouds and snow. The inversion of the reflectivity difference across the red edge (0.8-0.9 and 0.6-0.7 mu m) emphasizes the vegetation features near the equator. The planetary obliquity and equinox can be estimated simultaneously with the mapping under the presence of clouds. We conclude that the photometric variability of the scattered light will be a powerful means for exploring the habitat of a second Earth.

Scattered lights from terrestrial exoplanets provide valuable information about their planetary surface. Applying the surface reconstruction method proposed by Fujii et al. to both diurnal and annual variations of scattered light, we develop a reconstruction method of land distribution with both longitudinal and latitudinal resolutions. We find that one can recover a global map of an idealized Earth-like planet on the following assumptions: (1) cloudlessness, (2) a face-on circular orbit, (3) known surface types and their reflectance spectra, (4) lack of atmospheric absorption, (5) known rotation rate, (6) a static map, and (7) the absence of a moon. Using the dependence of light curves on planetary obliquity, we also show that the obliquity can be measured by adopting the chi(2) minimization or the extended information criterion. We demonstrate the feasibility of our methodology by applying it to a multi-band photometry of a cloudless model Earth with future space missions such as the occulting ozone observatory (O3). We conclude that future space missions can estimate both the surface distribution and the obliquity at least for cloudless Earth-like planets within 5 pc.