基于NPSVM模型的超新星识别方法

巡天观测与高能物理、黑洞天文等领域均有密切的联系.基于星系-超新星二分类问题,研究光谱数据预处理,结合余弦相似度改善PCA(Principal Component Analysis)光谱分解特征提取方法,用SDSS(the Sloan Digital Sky Survey)、WISeREP(the Weizmann Interactive Supernova data REPository)组成的5620条光谱数据集训练支持向量机,可以得到0.498%泛化误差的识别模型和新样本分类概率.使用Neyman-Pearson决策方法建立NPSVM(Neyman-Pearson Support Vector Machine)模型可进一步降低超新星的漏判率.     

Ia超新星的几个问题

首先利用Ａｓｉａｇｏ 超新星星表对Ｉａ 超新星作了统计分析。其次用具有精确测光的Ｉａ 超新星对其均匀性及多样性进行了研究。旋涡星系中Ｉａ 超新星的产生率比椭圆星系的要高。最亮的Ｉａ 超新星只出现于晚型旋涡星系中;而旋涡星系及早型的椭圆星系都是暗Ｉａ 超新星的寄主星系。离星系中心越近Ｉａ 超新星的光度弥散有增加的趋势,但这一趋势对蓝Ｉａ 超新星不明显。利用色指数可将Ｉａ 超新星划分为蓝超新星及红超新星。蓝Ｉａ 超新星构成了相对均匀的Ｉａ 超新星样本,是较好的距离指示器;而红Ｉａ 超新星的存在则表明了Ｉａ 超新星整体多样性的特点。最后,我们还探讨了Ｉａ 超新星中碳点火的非线性问题。     

Ia型超新星“SN2013gt”发现记

不知道从什么时候,我开始喜欢上了在没有色彩的星系图片中搜寻未知的星星。     

宇宙命运与Ia型超新星（上）

超新星是一些质量较大的恒星演化到晚期发生爆炸所产生的天体。它们在短时间内很明亮,一颗超新星的亮度可以相当于整个星系。超新星中有一类被称作Ia型超新星,     

Ia型超新星的爆发机制及其前身星模型

Ia型超新星作为测量遥远星系距离——从而测定宇宙膨胀速率——的“标准烛光”，已经成为宇宙学中具有重要意义的天体。从某些方面来讲，Ia型超新星仍属于神秘天体(或爆发事件)，其前身星及爆发模型还没有得到很好的理解，当前的观测不足以对理论模型作出精确的限制。然而有很好的理由相信大多数Ia型超新星可能是由接近钱德拉塞卡质量极限(≈1.39M⊙)的碳—氧白矮星通过聚变中心的碳和氧所引发的热按爆炸产生的。至于这一爆发通过何种机制完成，例如具体到Ia型超新星爆发时的流体动力学过程，仍存在分歧。最近爆燃阶段的三维数值模拟结果似乎表明，在Ia型超新星爆发晚期引入爆轰机制是没有必要的。另一方面，尽管当前的多数证据表明，C—O白矮星 主序星(或红巨星)的演化模式比C-O白矮星 C-O白矮星的演化模式可能更合理，但双简并白矮星的前身星模型并不能被排除，因为它们能解释一些特殊的Ia型超新星爆发。     

宇宙的命运与Ia型超新星（下）

珀尔马特领导的超新星宇宙学计划（SCP）小组最初对超新星观测研究中的许多困难并不完全了解,随着他们逐渐接近成功,天文学家们开始看到此项研究的前途并出现竞争。澳大利亚国立大学的施密特和美国约翰．霍普金斯大学的里斯等人也实施超新星观测项目。     

星周尘埃壳层的消光值

本文首次提出了从观测得到的具有星周尘埃壳层的恒星的能谱分布求取星周尘埃云的消光,并由此可通过改正星周消光改正后的星际消光法求得恒星距离的方法。     

宇宙的命运与Ia型超新星（中）

宇宙膨胀的发现 1923年,俄罗斯宇宙学家亚历山大·弗里德曼首先用∧=0的爱因斯坦场方程建立了宇宙学膨胀模型（∧为宇宙学常数）,他发表论文阐述了膨胀宇宙的思想,即曲率分别为正、负、零时的三种情况,称为弗里德曼宇宙模型。     

Ia型超新星前身星模型及其相关天体研究

Ia型超新星具有可校准的光度,可当作标准烛光,用来测定宇宙学距离,从而探索宇宙的形状.然而,Ia型超新星的前身星仍不清楚,这将直接影响当前宇宙学结果的可靠性.本文在Ia型超新星前身星模型及其相关天体方面做了系统性的工作,下面是我们取得的一些主要的研究结果:(1)近年来人们观测发现,约有一半的Ia型超新星的延迟时标小于100 Myr(Ia型超新星的延迟时标是指从恒星形成后到发生超新星爆炸的时间间隔).这些超新星是怎么来的?为解决这一难题,我     

发现4100多年前的超新星遗迹——尧帝景星(PKS 1209-52)

《竹书纪年》记载尧``元年丙子帝即位'', ``四十二年景星见于翼''. 《论衡》记载``尧时景星见于轸''. 这可能是出现在翼宿和轸宿之间的超新星. 由已知的干支纪年推算尧帝四十二年是公元前2164年. 把足够宽的范围\lk($180^\circ\pm23^\circ$)作``翼、轸''之间的过渡区域, 在格林(Green)超新星遗迹表中搜索到20颗超新星, 只有高银纬遗迹PKS 1209-52 (G296.5$+$10.0)符合尧帝景星``状如半月''的亮度和年龄限制, 它是尧帝景星(SN-B.C.2164)的唯一候选体. 依据超新星的表面亮度$\varSigma$与线直径D之间的$\varSigma\propto D^{-\beta     

Type Ia Supernovae

Summary. Type Ia Supernovae are in many aspects still enigmatic objects. Their observational and theoretical exploration is in full swing, but we still have plenty to learn about these explosions. Recent years have already witnessed a bonanza of supernova observations. The increased samples from dedicated searches have allowed the statistical investigation of Type Ia Supernovae as a class. The observational data on Type Ia Supernovae are very rich, but the uniform picture of a decade ago has been replaced by several correlations which connect the maximum luminosity with light curve shape, color evolution, spectral appearance, and host galaxy morphology. These correlations hold across almost the complete spectrum of Type Ia Supernovae, with a number of notable exceptions. After 150 days past maximum, however, all observed objects show the same decline rate and spectrum. The observational constraints on explosion models are still rather sparse. Global parameters like synthesized nickel mass, total ejecta mass and explosion energetics are within reach in the next few years. These parameters bypass the complicated calculations of explosion models and radiation transport. The bolometric light curves are a handy tool to investigate the overall appearance of Type Ia Supernovae. The nickel masses derived this way show large variations, which combined with the dynamics from line widths, indicate that the brighter events are also coming from more massive objects. The lack of accurate distances and the uncertainty in the correction for absorption are hampering further progress. Improvements in these areas are vital for the detailed comparison of luminosities and the determination of nickel masses. Coverage at near-infrared wavelengths for a statistical sample of Type Ia Supernovae will at least decrease the dependence on the absorption. Some of the most intriguing features of Type Ia Supernovae are best observed at these wavelengths, like the second peak in the light curve, the depression in the J band, and the unblended [Feii] lines in the ashes. Received 24 January 2000 / Published online 8 May 2000     

The Peak Luminosity of Type Ia Supernovae and its Implications for the Cosmic Expansion Rate

1 IN~DUCTIONInterest in take la supernovae (SNe la) has risen dramatically with their application to thecosmological problems since Kowal (1968). Their powerful capabilities as distance indicatorson the cosmic scale have pushed them into the limelight of cosmology. They now provide themain route to the current eXPansion rate and the deceleration of the adverse. (see Branch1998; Perlmutter et al. 1999; mess et al. 1998).SNe la are reversally accepted as the result of thermonuclear eXPlo…     

Distribution of 56Ni Yields of Type Ia Supernovae and its Implication for Progenitors

The amount of 56Ni produced in Type Ia supernova (SN Ia) explosion is probably the most important physical parameter underlying the observed correlation of SN Ia lumi-nosities with their light curves. Based on an empirical relation between the 56Ni mass and the light curve parameter △m15, we obtained rough estimates of the 56Ni mass for a large sample of nearby SNe Ia with the aim of exploring the diversity in SN Ia. We found that the derived 56Ni masses for different SNe Ia could vary by a factor of ten (e.g., MNi = 0.1 - 1.3 M⊙),which cannot be explained in terms of the standard Chandraseldaar-mass model (with a 56Ni mass production of 0.4 - 0.8 M⊙). Different explosion and/or progenitor models are clearly required for various SNe Ia, in particular, for those extremely nickel-poor and nickel-rich producers. The nickel-rich (with MNi 0.8 M⊙) SNe Ia are very luminous and may have massive progenitors exceeding the Chandrasekhar-mass limit since extra progenitor fuel is required to produce more 56Ni to power the light curve. This is also consistent with the find-ing that the intrinsically bright SNe Ia prefer to occur in stellar environments of young and massive stars. For example, 75% SNe Ia in spirals have △m15 < 1.2 while this ratio is only 18% in E/S0 galaxies. The nickel-poor SNe Ia (with MNi < 0.2 M⊙) may invoke the sub-Chandrasekhar model, as most of them were found in early-type E/S0 galaxies dominated by the older and low-mass stellar populations. This indicates that SNe Ia in spiral and E/S0 galaxies have progenitors of different properties.     

Optical photometry and spectroscopy of the Type Ibn supernova SN 2006jc until the onset of dust formation

We present optical UBVRI photometric and spectroscopic data of the Type Ibn supernova SN 2006jc, until the onset of the dust-forming phase. The optical spectrum shows a blue continuum and is dominated by the presence of moderately narrow (velocity ∼2500 km s⁻¹) He  i emission lines superimposed over a relatively weak supernova spectrum. The helium lines are produced in a pre-existing He-rich circumstellar shell. The observed helium line fluxes indicate the circumstellar shell is dense, with a density of  ∼10⁹–10¹⁰ cm⁻³  . The helium mass in this shell is estimated to be  ≲0.07 M_⊙  . The optical light curves show a clear signature of dust formation, indicated by a sharp decrease in the magnitudes around day 50, accompanied by a reddening of the colours. The evolution of the optical light curves during the early phase and that of the uvoir bolometric light curve at all phases is reasonably similar to normal Ib/c supernovae.