Learning Transient Universe in Near-Ultraviolet By Wide-angle Cameras

TL;DR

This paper proposes a space-based near-ultraviolet sky patrol mission using multiple wide-angle cameras to detect and study transient astronomical events like supernova shock breakouts and stellar flares.

Contribution

It introduces a novel mission concept with eight small wide-field NUV cameras for real-time detection of transient events across a large sky area.

Findings

Simulations show effective detection of galactic and extragalactic transients.

The proposed system can monitor 3000 square degrees simultaneously.

Real-time onboard software enables prompt transient detection.

Abstract

We perform a detailed analysis and simulations on the transient detection capability in the near-ultraviolet (NUV) band by focusing on some major local transient events. These events include the tidal disruption event due to a supermassive blackhole, the shock breakout of a core-collapse supernova and the flare of a late-type star. Our simulations show that a set of small wide-angle NUV cameras can allow us to detect and study numerous galactic and extra-galactic transient events. Based on the analysis and simulations, here we propose a space-based NUV sky patrol mission by updating the proposal that was originally submitted to the Chinese Space Station mission in 2011. The mission proposed here is composed of a set of eight small wide-field NUV cameras each with a diameter of 20cm. The total sky area simultaneously covered by the NUV cameras is as large as 3000$\mathrm{deg^2}$. The survey cadence ranges from 30 to 300s. The transient events are required to be detected by a dedicated on-board software in real time.

Tables

Table 1. Table 1: UV brightness estimated from different soft X-ray fluxes.

$f_x$	${\mathrm{mag}}_{\mathrm{AB}}$	$\Gamma$	Transients
$\mathrm{erg}{s}^{-1}{\mathrm{cm}}^{-2}$		mag
(1)	(2)	(3)	(4)
${10}^{-7}$	14.3	1.0	LGRB
${10}^{-8}$	17.7	1.0	LLGRB
	20.0	0.5	SGRB
${10}^{-9}$	19.3	1.0	LLGRB
	12.6	2.0	SBO
${10}^{-10}$	15.1	2.0	SBO, TDE
	9.8		stellar flare
${10}^{-11}$	17.6	2.0	SBO,TDE
	12.3		stellar flare

Table 2. Table 2: The parameters used for detection rate estimations.

Item	Value
(1)	(2)
$d$	20cm
$\eta$	0.3
RN	$4e^{-}$
CCD pixels	4k$\times$4k
$n_{\mathrm{pix}}$	9 pixels
${\lambda }_0$	2000Å
$\Delta \lambda$	500Å
Exposure time	TDEs: 3000s, SBOs: 300s and flares: 30s
$S/N$ threshold	7

Equations

$$f_{\nu}=EN(E)\frac{dE}{d\nu}=\frac{\lambda f_{\mathrm{X}}}{c}\bigg{[}\frac{2-\Gamma}{(E_{2}/E)^{2-\Gamma}-(E_{1}/E)^{2-\Gamma}}\bigg{]}$$

$$f_{\nu}=EN(E)\frac{dE}{d\nu}=\frac{\lambda f_{\mathrm{X}}}{c}\ln^{-1}\bigg{(}\frac{E_{2}}{E_{1}}\bigg{)}$$

$$\mathrm{ma}{\mathrm{g}}_{\mathrm{AB}}=-2.5\log f_{\nu }-48.6+A_{\lambda }$$

$$\sigma T_{d}^{4}=\frac{3GM_{\mathrm{BH}}\dot{M}f}{8\pi R^{3}}\bigg{\{}\frac{1}{2}+\bigg{[}\frac{1}{4}+\frac{3}{2}f\bigg{(}\frac{\dot{M}}{\eta\dot{M}_{\mathrm{Edd}}}\bigg{)}^{2}\bigg{(}\frac{R_{\mathrm{S}}}{R}\bigg{)}^{2}\bigg{]}^{1/2}\bigg{\}}^{-1}$$

$$f_{\mathrm{out}}=\frac{2}{\pi}\arctan\bigg{[}\frac{1}{7.5}\bigg{(}\frac{\dot{M}_{\mathrm{fb}}}{\dot{M}_{\mathrm{Edd}}}-1\bigg{)}\bigg{]}$$

$${\dot{M}}_{\mathrm{fb},\mathrm{p}}=1.9\times 10^{26}{M}_{\mathrm{BH},6}^{-1/2}{m}_{\ast }^2{\beta }^3{x}_{\ast }^{-3/2}\mathrm{g}{\mathrm{s}}^{-1}$$

$$r_{\mathrm{ph}}\approx 1.4\times 10^{15}\bigg{(}\frac{f_{\mathrm{out}}}{f_{\upsilon}}\bigg{)}\beta^{5/2}M_{\mathrm{BH,6}}^{-5/6}x_{*}^{-1}m_{*}^{11/6}\ \mathrm{cm}$$

$$T_{\mathrm{ph}}\approx 1.6\times 10^{4}\bigg{(}\frac{f_{\mathrm{out}}}{f_{\upsilon}}\bigg{)}^{-1/12}\beta^{-23/24}M_{\mathrm{BH,6}}^{-41/73}x_{*}^{1/3}m_{*}^{-53/72}\ \mathrm{K}$$

$$L_{\nu }=4{\pi }^2{\int }_{R_{\mathrm{in}}}^{R_{\mathrm{out}}}R{B}_{\nu }(T_d)dR+4{\pi }^2{r}_{\mathrm{ph}}^2{B}_{\nu }(T_{\mathrm{ph}})$$

$$\frac{S}{N}\simeq \frac{\eta N_{\ast }t}{\sqrt{\eta N_{\ast }t+n_{\mathrm{pix}}(\eta N_{b}t+R{N}^2)}}$$

$$N_{*}=\frac{1}{16h}\bigg{(}\frac{d}{d_{L}}\bigg{)}^{2}\int_{\nu_{1}}^{\nu_{2}}L_{\nu}d\ln\nu$$

$$N_b=\frac{1}{4}\pi d^2{\int }_{\Delta {\Omega }_p}{\int }_{{\nu }_1}^{{\nu }_2}{R}_{\nu }d\nu d\Omega$$

$${\Gamma }_{\mathrm{TDE}}={\int }_{\Delta \Omega }{\int }_0^{z_{\max }}{\int }_{M_{\mathrm{BH}}}{\int }_{R_p}^{R_T}\frac{dn}{d\ln M_{\mathrm{BH}}}\frac{d\gamma }{d\ln R_p}\frac{dV}{dzd\Omega }d\ln R_{p}d\ln M_{\mathrm{BH}}dzd\Omega$$

$${\Gamma }_{\mathrm{SBO}}={\int }_{\Delta \Omega }{\int }_0^{z_{\max }}{f}_{\mathrm{SNII}}{R}_{\mathrm{SN}}\frac{dV}{dzd\Omega }dzd\Omega$$

$$\psi (z)=0.015\frac{(1+z{)}^{2.7}}{1+[(1+z)/2.9{]}^{5.6}}{\mathrm{M}}_{\odot }\mathrm{y}{\mathrm{r}}^{-1}\mathrm{Mp}{\mathrm{c}}^{-3}$$

$${\Gamma }_{\mathrm{flares}}={\int }_{\Delta \Omega }{\int }_0^{r_{E_1}}{\int }_{m_1}^{m_2}4\pi r^2{\rho }_{\star }{\int }_{E_1}^{E_2}\frac{dn}{dm}\frac{dN}{dE}dEdrd\Omega dm$$

Full text

methods: observational — ultraviolet: general — surveys — stars: flare — supernovae: general — galaxies: nuclei

††journal: PASP

1 Introduction

Time-domain astronomy, which mainly explores the Universe by searching for and studying transients in different wavelength bands, is a hot topic of modern astronomy. Transient phenomena have been widely explored in high energy bands (i.e., X-ray and $\gamma$-ray) in the past decades, thanks to the great success achieved by the Neil Gehrels *Swift *Observatory (Gehrels et al. 2004) and the Fermi Gamma-ray Space Telescope (e.g., Atwood et al. 2009; Meegan et al. 2009). This success enables us to make great progress in the understanding of Gamma-ray bursts (GRBs, see a review in Woosley & Bloom 2006). The capability of exploring transients in X-ray and $\gamma$-ray will be continued by forthcoming missions, including SVOM (Wei et al. 2016) and the Einstein Probe (EP) (Yuan et al. 2015).

In addition to the high energy bands, transient phenomena have been comprehensively probed by a lot of optical time-domain surveys in recent years, because of the great progress made in both detector and computation technology in the past decades. Some ongoing optical survey programs include: the Pan-STARRS1 survey (PS1; Chambers et al. 2016), the Asteroids Terrestrial-impact Last Alert System (ATLAS; Tonry et al. 2018), the Palomar Transient Factory (PTF; Law et al. 2009) and the successor Zwicky Transient Facility (ZTF; Kulkarni 2018), the All-Sky Automated Survey for Supernovae (ASAS-SN; Shappee et al. 2014), Pi of the sky (Burd et al. 2005), RAPTOR (Vestrand et al. 2002), and the Ground Wide-Angle Cameras system (GWAC, Wei et al. 2016). These optical transient surveys discovered an abundance of supernovae (SNe) in diverse types and the related phenomena (e.g., Gal-Yam et al. 2014; Arcavi et al. 2017; Whitesides et al. 2017), a number of tidal disruption events (TDEs) induced by supermassive blackholes (SMBHs, e.g., Holoien et al. 2018 and reference therein), “Changing-look” active galactic nuclei (CL-AGNs, e.g., Gezari et al. 2017; Wang et al. 2018), and some mysterious fast-evolving luminous transients without a convincing explanation (e.g., Rest et al. 2018 and references therein). The forthcoming Large Synoptic Survey Telescope (LSST) project (LSST Science Collaboration et al. 2017) will be a milestone for optical transient survey. The transient sky has been monitored by LOFAR (e.g., Gunst & Bentum 2007; de Vos et al. 2007) and the SKA in radio.

In addition to electromagnetic radiation, a non-electromagnetic message transmitted by a gravitational-wave (e.g., the Advanced LIGO, Harry 2010; Aasi et al. 2015; LIGO Scientific Collaboration et al., 2015, the Advanced Virgo, Acernese et al. 2009; Accadia et al. 2012; Acernese et al. 2015; KAGRA, Somiya 2012; Aso et al. 2013) and by neutrino events (e.g., the IceCube experiment, Aartsen et al. 2015) has been used to explore transient phenomena. Operation of Advanced LIGO opened an era of multi-message astronomy by successfully detecting the gravitational-wave signals emitted from the coalescence of a binary black holes (BBHs, e.g., GW 150914, Abbott et al. 2016) and the coalescence of a binary neutron stars (BNS, GW 170817. e.g., Abbott et al. 2017). The coalescence of binary neutron stars was confirmed by the detection of electromagnetic emission from the associated kilonova (e.g., Shappee et al. 2017; Drout et al. 2017; Evans et al. 2017).

Time-domain astronomy is, however, rarely developed in ultraviolet (UV) bands, even though the study of transients in UV can potentially address some major scientific questions (e.g., Sagiv et al. 2014; Brosch et al. 2014). In fact, observations carried out by Swift/UVOT and *GALEX *(Martin et al. 2005) returned exciting results on transients (e.g., Soderberg et al. 2008; Schawinski et al. 2008), although both instruments are not designed to search for transients based on a large field-of-view (FoV). The main scientific goal of *GALEX *is to study star formation and galaxy evolution in UV bands, which results in very limited time-domain observations.

In the development of the Chinese Space Station mission, we proposed a wide-angle time domain survey in near-UV (NUV) as early as in 2011. Subsequent studies, however, indicated that the Space Station is not an ideal platform for this survey. The depth of the survey is limited because the on-board cameras can only work in drift scan mode.

In this paper, we perform a detailed analysis on the detection capability of an NUV wide-angle time domain survey. The analysis focuses on some important scientific objects. One can see from our analysis that an abundance of results can be obtained by a set of small NUV cameras. Based on this analysis, we then propose a small satellite mission dedicated to the NUV wide-angle time domain survey.

The paper is organized as follows. Section 2 describes the major scientific motivations for the proposed NUV transient survey by supplementing the CL-AGN phenomenon. The analysis and simulation of detection capability are presented in Section 3. Section 4 describes the concept of the proposed dedicated NUV transient survey mission. A $\Lambda$CDM cosmology with parameters $H_{0}=70\mathrm{km\ s^{-1}\ Mpc^{-1}}$, $\Omega_{m}=0.3$, and $\Omega_{\Lambda}=0.7$ is adopted throughout the paper

2 Scientific Objectives

For transient phenomena that can be preferably explored in the UV band, we refer readers to the excellent review given by Sagiv et al. (2014, and references therein). The phenomena described in that review include: the supernova shock breakout (SBO) resulting from the explosive death of a massive star; the TDE of a main-sequence star (or a white dwarf) due to the tidal force of a supermassive (or intermediate massive) blackhole; the variability of AGNs; and the flare of stars.

Besides the topics described in the review, here we pay more attention to the CL-AGN phenomenon from a scientific perspective, because the phenomenon is a hot and challenging topic in modern astronomy. According to their observed optical spectra, AGNs can be classified into Type-1 and Type-2 AGNs. The spectra of Type-1 AGNs have both broad ($\mathrm{FWHM>1000\ km\ s^{-1}}$) and narrow ($\mathrm{FWHM\sim 10^{2}\ km\ s^{-1}}$) Balmer emission lines. On the contrary, only narrow Balmer emission lines can be identified in Type-2 AGNs. The two types can be unified by the widely accepted unified model based on the orientation effect caused by the dust torus (see Antonucci 1993 for a review). This successful model has, however, been recently challenged by the discovery of so-called CL-AGNs that show a change in their spectral types on a time scale of several years.

Although both “turn-on” and “turn-off” type transitions have been revealed in past years, there are only 40 identified CL-AGNs at present (e.g., Shapovalova et al. 2010; Shappee et al. 2014; LaMassa et al. 2015; McElroy et al. 2016; Runnoe et al. 2016; Gezari et al. 2017; Yang et al. 2018; Ruan et al. 2016; MacLeod et al. 2016; Wang et al. 2018). The discovery of CL-AGNs is actually a hard and expensive task, which requires repeated spectroscopy to identify a change in the spectral type of the Balmer line profiles. At the same time, an optical transient survey is not a good way to select CL-AGN candidates because of the serious contamination caused by AGN’s normal variation in optical bands.

The origin of CL-AGNs is still under debate, even though there is accumulating evidence supporting the fact that it is likely due to a variation in SMBH accretion rate that results from either a viscous radial inflow or disk instability (e.g., Yang et al. 2018; Wang et al. 2018; Gezari et al. 2017). In addition, other explanations include: (1) a variation in the obscuration if the torus has a patchy configuration (e.g., Elitzur 2012); (2) an accelerating outflow launched from the central SMBH (e.g., Shapovalova et al. 2010); and (3) a TDE (e.g., Merloni et al. 2015; Blanchard et al. 2017).

Besides the prominent line profile change, CL-AGNs are typically accompanied by a significant variation in their blue featureless continuum, because Type-1 and Type-2 AGNs differ significantly in their UV continuum. This significant difference in continuum means an NUV transient survey is the best way to find both “turn-on” and “turn-off” CL-AGNs in the local Universe.

3 Detection Rate Predictions

For a wide-angle time domain survey in NUV, the detection rates of some important scientific objects are predicted in this section.

3.1 UV Brightness Estimated from X-ray Flux

At the beginning, we estimate the NUV brightness of some transients from their soft X-ray flux. A powerlaw photon spectrum of $N(E)=N_{0}E^{-\Gamma}$ is adopted for GRBs, SBOs, and TDEs. The corresponding X-ray flux $f_{\mathrm{X}}$ within the energy range from $E_{1}$ to $E_{2}$ can be written as an integral of $f_{\mathrm{X}}=\int_{E_{1}}^{E_{2}}N(E)EdE=N_{0}\int_{E_{1}}^{E_{2}}E^{1-\Gamma}dE$, which allows us to obtain the specific flux $f_{\nu}$ at a given wavelength $\lambda$ as

[TABLE]

when $\Gamma\neq 2$, and

[TABLE]

when $\Gamma=2$. Based on the traditionally used Band function (Band et al. 1993), the values of $\Gamma$ are adopted to be 1.0 and 0.5 for Long GRBs and short GRBs, according to the BASTE and Fermi observations (e.g., Abdo et al. 2009; Zhang et al. 2011, Nova et al. 2011). A soft spectrum with a $\Gamma=2.0$ is used for SBOs and TDEs. In fact, the XMM-Newton spectrum of TDE RXJ1242 C1119 has an index of $\Gamma=2.5$, although the initial spectrum taken by ROSAT has an index of $\Gamma\simeq 5$ (e.g., Komossa 2017; Komossa et al. 2004; Halpern et al. 2004). In fact, similar spectral hardening has been observed in a few TDEs (e.g., Komossa & Bade 1999; Nikolajuk & Walter 2013). Soderberg et al. (2008) revealed a X-ray spectrum with a photon index of $\Gamma=2.3\pm 0.3$ in SBO event of SN 2008D.

We estimate the UV brightness of a stellar flare from its soft X-ray flux according to the Neupert effect which suggests a correlation between soft X-ray luminosity and UV energy release (e.g., Gudel et al. 2002; Hawley et al. 1995). A average ratio of X-ray to NUV specific luminosity of 20 is adopted in our estimation.

The corresponding magnitude in the AB system is then determined from the definition (Fukugita et al. 1996)

[TABLE]

where $A_{\lambda}$ denotes the extinction at wavelength $\lambda$. Table 1 lists the calculated magnitudes at different X-ray flux levels, in which $\lambda=2800$Å, $A_{\lambda}(2800\AA)=0.2$mag, $E_{1}=0.5$keV and $E_{2}=4.0$keV are adopted. The $A_{\lambda}(2800\AA)$ is estimated from the $V$-band extinction through the extinction curve of LMC provided in Gordon et al. (2003), by assuming $A_{V}=0.1$mag.