اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدThe Feasibility and Flexibility of Selecting Quasars by Variability Using Ensemble Machine Learning Algorithms
58
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
In this work we train three decision-tree based ensemble machine learning algorithms (Random Forest Classifier, Adaptive Boosting and Gradient Boosting Decision Tree respectively) to study quasar selection in the variable source catalog in SDSS Stripe 82. We build training and test samples (both containing 1:1 of quasars and stars) using the spectroscopic confirmed sources in SDSS DR14 (including 8330 quasars and 3966 stars). We find that, trained with variation parameters alone, all three models can select quasars with similarly and remarkably high precision and completeness ($sim$ 98.5% and 97.5%), even better than trained with SDSS colors alone ($sim$ 97.2% and 96.5%), consistent with previous studies. Through applying the trained models on the variable sources without spectroscopic identifications, we estimate the spectroscopically confirmed quasar sample in Stripe 82 variable source catalog is $sim$ 93% complete (95% for $m_i<19.0$). Using the Random Forest Classifier we derive the relative importance of the observational features utilized for classifications. We further show that even using one- or two-year time domain observations, variability-based quasar selection could still be highly efficient.
قيم البحث
اقرأ أيضاً
We present the second Multi-Epoch X-ray Serendipitous AGN Sample (MEXSAS2), extracted from the 6th release of the XMM Serendipitous Source Catalogue (XMMSSC-DR6), cross-matched with Sloan Digital Sky Survey quasar catalogues DR7Q and DR12Q. Our sampl
e also includes the available measurements for masses, bolometric luminosities, and Eddington ratios. Analyses of the ensemble structure function and spectral variability are presented, together with their dependences on such parameters. We confirm a decrease of the structure function with the X-ray luminosity, and find a weak dependence on the black hole mass. We introduce a new spectral variability estimator, taking errors on both fluxes and spectral indices into account. We confirm an ensemble softer when brighter trend, with no dependence of such estimator on black hole mass, Eddington ratio, redshift, X-ray and bolometric luminosity.
We study the extreme ultraviolet (EUV) variability (rest frame wavelengths 500 - 920 $AA$) of high luminosity quasars using HST (low to intermediate redshift sample) and SDSS (high redshift sample) archives. The combined HST and SDSS data indicates a
much more pronounced variability when the sampling time between observations in the quasar rest frame is $> 2times 10^{7}$ sec compared to $< 1.5times 10^{7}$ sec. Based on an excess variance analysis, for time intervals $< 2times 10^{7}$ sec in the quasar rest frame, $10%$ of the quasars (4/40) show evidence of EUV variability. Similarly, for time intervals $>2times 10^{7}$ sec in the quasar rest frame, $55%$ of the quasars (21/38) show evidence of EUV variability. The propensity for variability does not show any statistically significant change between $2.5times 10^{7}$ sec and $3.16times 10^{7}$ sec (1 yr). The temporal behavior is one of a threshold time interval for significant variability as opposed to a gradual increase on these time scales. A threshold time scale can indicate a characteristic spatial dimension of the EUV region. We explore this concept in the context of the slim disk models of accretion. We find that for rapidly spinning black holes, the radial infall time to the plunge region of the optically thin surface layer of the slim disk that is responsible for the preponderance of the EUV flux emission (primarily within 0 - 7 black hole radii from the inner edge of the disk) is consistent with the empirically determined variability time scale.
We constrain the average episodic quasar lifetime (as in steady-state accretion) using two statistics of quasars that are recently turned off (i.e., dimmed by a large factor): 1) the fraction of turned-off quasars in a statistical sample photometrica
lly observed over an extended period (e.g., $Delta t=20$ yrs); 2) the fraction of massive galaxies that show orphan broad MgII emission, argued to be short-lived echoes of recently turned-off quasars. The two statistics constrain the average episodic quasar lifetime to be hundreds to thousands of years. Much longer (or shorter) episodic lifetimes are strongly disfavored by these observations. This average episodic lifetime is broadly consistent with the infall timescale (viscous time) in the standard accretion disk model for quasars, suggesting that quasar episodes are governed by accretion disk physics rather than by the gas supply on much larger scales. Compared with the cumulative quasar lifetime of $sim 10^6-10^8,$yrs constrained from quasar clustering and massive black hole demographics, our results suggest that there are $sim 10^3-10^5$ episodes of quasar accretion during the assembly history of the supermassive black hole. Such short episodes should be clustered over intervals of $sim 10^4,$yrs to account for the sizes of ionized narrow-line regions in quasars. Our statistical argument also dictates that there will always be a small fraction of extreme variability quasars caught in state transitions over multi-year observing windows, despite the much longer episodic lifetime. These transitions could occur in a rather abrupt fashion during non-steady accretion.
We present the ensemble variability analysis results of quasars using the Dark Energy Camera Legacy Survey (DECaLS) and the Sloan Digital Sky Survey (SDSS) quasar catalogs. Our dataset includes 119,305 quasars with redshifts up to 4.89. Combining the
two datasets provides a 15-year baseline and permits analysis of the long timescale variability. Adopting a power-law form for the variability structure function, $V=A(t/1yr)^{gamma}$, we use the multi-dimensional parametric fitting to explore the relationships between the quasar variability amplitude and a wide variety of quasar properties, including redshift (positive), bolometric luminosity (negative), rest-frame wavelength (negative), and black hole mass (uncertain). We also find that $gamma$ can be also expressed as a function of redshift (negative), bolometric luminosity (positive), rest-frame wavelength (positive), and black hole mass (positive). Tests of the fitting significance with the bootstrap method show that, even with such a large quasar sample, some correlations are marginally significant. The typical value of $gamma$ for the entire dataset is $gtrsim 0.25$, consistent with the results in previous studies on both the quasar ensemble variability and the structure function. A significantly negative correlation between the variability amplitude and the Eddington ratio is found, which may be explained as an effect of accretion disk instability.
The variability of quasars across multiple wavelengths is a useful probe of physical conditions in active galactic nuclei. In particular, variable accretion rates, instabilities, and reverberation effects in the accretion disk of a supermassive black
hole (SMBH) are expected to produce correlated flux variations in UV and optical bands. Recent work has further argued that binary quasars should exhibit strongly correlated UV and optical periodicities. Strong UV-optical correlations have indeed been established in small samples of up to approximately 30 quasars with well-sampled light curves, and have extended the bluer-when-brighter trend previously found within the optical bands. Here we further test the nature of quasar variability by examining the observed-frame UV-optical correlations in a large sample of 1,315 bright quasars with overlapping UV and optical light curves for the Galaxy Evolution Explorer (GALEX) and the Catalina Real-time Transient Survey (CRTS), respectively. We find that strong correlations exist in this much larger sample, but we rule out, at approximately 95% confidence, the simple hypothesis that the intrinsic UV and optical variations of all quasars are fully correlated. Our results therefore imply the existence of physical mechanism(s) that can generate uncorrelated optical and UV flux variations.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
