Classification of stellar spectra with SVM based on within-class scatter and between-class scatter

Support Vector Machine (SVM) is a popular data mining technique, and it has been widely applied in astronomical tasks, especially in stellar spectra classification. Since SVM doesn’t take the data distribution into consideration, and therefore, its classification efficiencies can’t be greatly improved. Meanwhile, SVM ignores the internal information of the training dataset, such as the within-class structure and between-class structure. In view of this, we propose a new classification algorithm-SVM based on Within-Class Scatter and Between-Class Scatter (WBS-SVM) in this paper. WBS-SVM tries to find an optimal hyperplane to separate two classes. The difference is that it incorporates minimum within-class scatter and maximum between-class scatter in Linear Discriminant Analysis (LDA) into SVM. These two scatters represent the distributions of the training dataset, and the optimization of WBS-SVM ensures the samples in the same class are as close as possible and the samples in different classes are as far as possible. Experiments on the K-, F-, G-type stellar spectra from Sloan Digital Sky Survey (SDSS), Data Release 8 show that our proposed WBS-SVM can greatly improve the classification accuracies.     

Chromospheric activity in $\epsilon$ Eridani: results from theoretical wave studies

This work discusses theoretical models of chromospheric heating for \(\epsilon\) Eridani by shock waves. Self-consistent, nonlinear and time-dependent ab-initio numerical computations for the excitation of the atmosphere (i.e., arrays of flux tubes) are pursued based on waves generated in stellar convective zones. Based on previous studies the magnetic filling factor is estimated according to the stellar rotational period, although general models are described as well. The Ca II H+K fluxes are computed assuming partial redistribution (PRD). Time-dependent ionization notably affects the resulting Ca II fluxes, as expected. The emergent Ca II H+K fluxes are based on two-component models, consisting of a dominant magnetic component (as given by longitudinal tube waves) and a subordinate acoustic component. The Ca II fluxes as obtained are smaller by about a factor of 2 than those given by observations. Possible reasons for this discrepancy include (1) inherent limitations of our theoretical approach as it is based on 1-D rather than 3-D modelling and/or (2) the existence of additional heating processes in \(\epsilon\) Eridani (a young star) not included here.     

The astrometric signal of microlensing events caused by free floating planets

Astrometric observations of microlensing events can be used to obtain important information about lenses. During these events, the shift of the position of the multiple image centroid with respect to the source star location can be measured. This effect, which is expected to occur on scales from micro-arcseconds to milli-arcseconds, depends on the lens-source-observer system physical parameters. Here, we consider the astrometric and photometric observations by space and ground-based telescopes of microlensing events towards the Galactic bulge caused by free floating planets (FFPs). We show that the efficiency of astrometric signal on photometrically detected microlensing events tends to increase for higher FFP masses in our Galaxy. In addition, we estimate that during five years of the Gaia observations, about a dozen of microlensing events caused by FFPs are expected to be detectable.     

Nonlinear model to study filamentation instability of circularly polarized dispersive Alfvén waves in solar wind plasmas

This paper examines the small-scale solar wind turbulence driven in view of the Alfvén waves subjected to ponderomotive nonlinearity. Filamentation instability is known to take place for the case of dispersive Alfvén wave (DAW) propagating parallel to the ambient magnetic field. The ponderomotive force associated with DAW is responsible for wave localization and these webs of filaments become more intense and irregular as one proceeds along the spatial domain. The ponderomotive force associated with pump changes with pump parameters giving rise to different evolution patterns. This paper studies in detail the nonlinear evolution of filamentation instability introduced by dispersive Alfven waves (DAWs) which becomes dispersive on account of the finite frequency of DAW i.e., pump frequency is comparable to the ion cyclotron frequency. We have explicitly obtained the perturbation dynamics and then examined the impact of pump magnitude on the driven magnetic turbulence using numerical simulation. The results show steepening at small scales with increasing pump amplitude. The compressibility associated with acoustic fluctuations may explain the variation in spectral scaling of solar wind turbulence as observed by Alexandrova et al. (Astrophys. J. 674:1157, 2008).     

Temperature evolution in the presence of anisotropic stresses

In a recent paper by Das et al. (Astrophys. Space Sci. 361:99, 2016) it was shown that the sign of the anisotropy parameter plays a pivotal role in determining the time of formation of the horizon of a collapsing radiating star. Spurred on by this observation we investigate the impact of the (an)isotropy featured in the (Das et al. in Astrophys. Space Sci. 361:99, 2016) collapsing model on the temperature profiles of the evolving system. We show that the temperature within the stellar interior is increased as the anisotropy in the pressure grows. Relaxational effects due to heat dissipation within the core further enhances the temperature at each interior point of the stellar distribution.     

Radio variability of the blazar S5 0716+714: a ∼6.1 year quasi-periodicity

We report a ～6.1 yr quasi-periodicity for the blazar S5 0716+714 using the radio light curves at 4.8, 8 and 14.5 GHz observed by the University of Michigan Radio Astronomical Observatory (UMRAO) from 1981 to 2012, by means of the Jurkevich, discrete correlation function (DCF) and power spectral analysis techniques. There are a general correlation among light curves at different frequencies and a time lag of \(170\pm 10\) days between 4.8 and 14.5 GHz light curves can be confirmed. We also estimate the orbit parameters assuming a binary black hole system, and the magnetic field strength under the jet comoving frame.     

Retrieval of atmospheric mass densities in lower thermosphere below 200 km from precise orbit of re-entry object CZ-3B R/B by analytical and numerical methods

Taking the re-entry object CZ-3B R/B (COSPAR identifier 2012-018D, NORAD catalog number 38253) as an example, retrieval of atmospheric mass densities in lower thermosphere below 200 km from its rebuilt precise orbit is studied in this paper. Two methodologies, i.e. analytical and numerical methods, are adopted in the retrieval. Basic principles of these two methodologies are briefly introduced. Based on the short-arc sparse observational data accumulated in the high accuracy re-entry prediction, orbit determinations of re-entry object CZ-3B R/B are performed sectionally, and then its precise orbit is rebuilt. According to the orbit theory, the variation of orbital semi-major axis of re-entry object CZ-3B R/B induced by atmospheric drag perturbation only is derived from the rebuilt precise orbit. In the derivation of secular change of the orbital semi-major axis of re-entry object CZ-3B R/B induced by atmospheric drag perturbation only, the time-span is set as one minute tentatively. And then retrieval results of atmospheric mass densities in lower thermosphere below 200 km by analytical and numerical methods are presented, as well as their bias deviations from the calculated results of the NRLMSISE-00 empirical model of the atmosphere. Setting bias deviation bands, the corresponding ‘confidence coefficients’ of the retrieved atmospheric mass densities with respect to the model values are given. Average bias deviations of the retrieved atmospheric mass densities by analytical and numerical methods from the model values are also calculated respectively. On the whole, the retrieved atmospheric mass densities by numerical method approach to the model values more closely; the differences between the retrieved results and the model values are relatively smaller at the peaks of atmospheric mass densities than the other places.     

A model of the late universe with viscous Zel’ldovich fluid and decaying vacuum

Many have speculated about the presence of a stiff fluid in very early stage of the universe. Such a stiff fluid was first introduced by Zel’dovich. Recently the late acceleration of the universe was studied by taking bulk viscous stiff fluid as the dominant cosmic component, but the age predicted by such a model is less than the observed value. We consider a flat universe with viscous stiff fluid and decaying vacuum energy as the cosmic components and found that the model predicts a reasonable background evolution of the universe with de Sitter epoch as end phase of expansion. More over, the model also predicts a reasonable value for the age of the present universe. We also performed a dynamical system analysis of the model and found that the end de Sitter phase predicted by the model is stable.     

Dark energy cosmological models with cosmic string

In this paper we have studied the anisotropic Kantowski-Sachs, locally rotationally symmetric (LRS) Bianchi type-I and LRS Bianchi type-III geometries filled with dark energy and one dimensional cosmic string in the Saez-Ballester theory of gravitation. To get physically valid solution we take hybrid expansion law of the average scale factor which is a product of power and exponential type of functions that results in time dependent deceleration parameter (\(q\)). The equation of state parameter of dark energy (\(\omega _{\mathit{de}}\)) has been discussed and we have observed that for the three models it crosses the phantom divide line (\(\omega _{\mathit{de}} = -1\)) and shows quintom like behavior. The density of dark energy (\(\rho _{\mathit{de}}\)) is an increasing function of redshift and remains positive throughout the evolution of the universe for the three models. Moreover in Kantowski-Sachs and LRS Bianchi type-I geometries the dark energy density dominates the string tension density (\(\lambda \)) and proper density (\(\rho \)) throughout the evolution of the universe. The physical and geometrical aspects of the statefinder parameters (\(r,s\)), squared speed of sound (\(v_{s}^{2} \)) and \(\omega _{\mathit{de}}\)–\(\omega ^{\prime }_{\mathit{de}}\) plane are also discussed.     

Solar Microflares Observed by SphinX and RHESSI

In 2009, the Russian Complex Orbital Observations Near-Earth of Activity of the Sun (CORONAS-Photon) spacecraft was launched, carrying the Polish Solar PHotometer In X-rays (SphinX). The SphinX was most sensitive in the spectral range 1.2?–?15 keV, thus an excellent opportunity appeared for comparison with the low-energy end of Ramaty High Energy Solar Spectroscopic Imager (RHESSI) spectra. Common spectral measurements with these instruments cover the range where most of the flare energy is accumulated. We have chosen four consecutive small solar events observed on 4 July 2009 at 13:43 UT, 13:48 UT, 13:52 UT, and 13:55 UT (RHESSI flare peak times) and used them to compare the data and results from the two instruments. Moreover, we included Geostationary Operational Environmental Satellite (GOES) records in our analysis. In practice, the range of comparison performed for SphinX and RHESSI is limited roughly to 3?–?6 keV. RHESSI fluxes measured with a use of one, four, and nine detectors in the 3?–?4 keV energy band agree with SphinX measurements. However, we observed that SphinX spectral irradiances are three times higher than those of RHESSI in the 4?–?6 keV energy band. This effect contributes to the difference in obtained emission measures, but the derived temperatures of plasma components are similar. RHESSI spectra were fitted using a model with two thermal components. We have found that the RHESSI hot component is in agreement with GOES, and the RHESSI hotter component fits the SphinX flaring component well. Moreover, we calculated the so-called thermodynamic measure and the total thermal energy content in the four microflares that we studied. The results obtained show that SphinX is a very sensitive complementary observatory for RHESSI and GOES.