The Host Galaxy and Redshift of the Repeating Fast Radio Burst FRB 121102
Shriharsh P. Tendulkar, Cees Bassa, James M. Cordes, Geoffery C. Bower, Casey J. Law, Shamibrata Chatterjee, Elizabeth A. K. Adams, Slavko Bogdanov, Sarah Burke-Spolaor, Bryan J. Butler, Paul Demorest, Jason W. T. Hessels, Victoria M. Kaspi, T. Joseph W. Lazio, Natasha Maddox, Benito Marcote, Maura A. McLaughlin, Zsolt Paragi, Scott M. Ransom, Paul Scholz, Andrew Seymour, Laura G. Spitler, Huib J. van Langevelde, Robert S. Wharton
DDraft version January 6, 2017
Typeset using L A TEX twocolumn style in AASTeX61
THE HOST GALAXY AND REDSHIFT OF THE REPEATING FAST RADIO BURST FRB 121102
S. P. Tendulkar, C. G. Bassa, J. M. Cordes, G. C. Bower, C. J. Law, S. Chatterjee, E. A. K. Adams, S. Bogdanov, S. Burke-Spolaor,
7, 8, 9
B. J. Butler, P. Demorest, J. W. T. Hessels,
2, 10
V. M. Kaspi, T. J. W. Lazio, N. Maddox, B. Marcote, M. A. McLaughlin,
8, 9
Z. Paragi, S. M. Ransom, P. Scholz, A. Seymour, L. G. Spitler, H. J. van Langevelde,
12, 17 and R. S. Wharton Department of Physics and McGill Space Institute, McGill University, 3600 University St., Montreal, QC H3A 2T8, Canada ASTRON, the Netherlands Institute for Radio Astronomy, Postbus 2, NL-7990 AA Dwingeloo, The Netherlands Cornell Center for Astrophysics and Planetary Science and Department of Astronomy, Cornell University, Ithaca, NY 14853, USA Academia Sinica Institute of Astronomy and Astrophysics, 645 N. A’ohoku Place, Hilo, HI 96720, USA Department of Astronomy and Radio Astronomy Lab, University of California, Berkeley, CA 94720, USA Columbia Astrophysics Laboratory, Columbia University, New York, NY 10027, USA National Radio Astronomy Observatory, Socorro, NM 87801, USA Department of Physics and Astronomy, West Virginia University, Morgantown, WV 26506, USA Center for Gravitational Waves and Cosmology, West Virginia University, Chestnut Ridge Research Building, Morgantown, WV 26505 Anton Pannekoek Institute for Astronomy, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA Joint Institute for VLBI ERIC, Postbus 2, 7990 AA Dwingeloo, The Netherlands National Radio Astronomy Observatory, Charlottesville, VA 22903, USA National Research Council of Canada, Herzberg Astronomy and Astrophysics, Dominion Radio Astrophysical Observatory, P.O. Box 248,Penticton, BC V2A 6J9, Canada Arecibo Observatory, HC3 Box 53995, Arecibo, PR 00612, USA Max-Planck-Institut f¨ur Radioastronomie, Auf dem H¨ugel 69, Bonn, D-53121, Germany Leiden Observatory, Leiden University, PO Box 9513, 2300 RA Leiden, The Netherlands (Received January 6, 2017; Revised January 6, 2017; Accepted January 6, 2017)
Submitted to ApJABSTRACTThe precise localization of the repeating fast radio burst (FRB 121102) has provided the first unambiguous association(chance coincidence probability p (cid:46) × − ) of an FRB with an optical and persistent radio counterpart. We reporton optical imaging and spectroscopy of the counterpart and find that it is an extended (0 . (cid:48)(cid:48) − . (cid:48)(cid:48)
8) object displayingprominent Balmer and [O
III ] emission lines. Based on the spectrum and emission line ratios, we classify the counterpartas a low-metallicity, star-forming, m r (cid:48) = 25 . z = 0 . (cid:46) M ∗ ∼ − × M (cid:12) , assuming a mass-to-light ratio between2 to 3 M (cid:12) L − (cid:12) . Based on the H α flux, we estimate the star formation rate of the host to be 0 . M (cid:12) yr − and asubstantial host dispersion measure depth (cid:46)
324 pc cm − . The net dispersion measure contribution of the host galaxyto FRB 121102 is likely to be lower than this value depending on geometrical factors. We show that the persistentradio source at FRB 121102’s location reported by Marcote et al. (2017) is offset from the galaxy’s center of lightby ∼
200 mas and the host galaxy does not show optical signatures for AGN activity. If FRB 121102 is typical ofthe wider FRB population and if future interferometric localizations preferentially find them in dwarf galaxies with
Corresponding author: S. P. Tendulkar, C. G. [email protected]; [email protected] a r X i v : . [ a s t r o - ph . H E ] J a n Tendulkar et al. low metallicities and prominent emission lines, they would share such a preference with long gamma ray bursts andsuperluminous supernovae.
Keywords: stars: neutron – stars: magnetars – galaxies: distances and redshifts – galaxies: dwarf –galaxies: ISM he host of FRB 121102 INTRODUCTIONFast radio bursts (FRBs) are bright ( ∼ Jy) and short( ∼ ms) bursts of radio emission that have dispersionmeasures (DMs) in excess of the line of sight DM con-tribution expected from the electron distribution of ourGalaxy. To date 18 FRBs have been reported — mostof them detected at the Parkes telescope (Lorimer et al.2007; Thornton et al. 2013; Burke-Spolaor & Bannister2014; Keane et al. 2012; Ravi et al. 2015; Petroff et al.2015; Keane et al. 2016; Champion et al. 2016; Raviet al. 2016) and one each at the Arecibo (Spitler et al.2014) and Green Bank telescopes (Masui et al. 2015).A plethora of source models have been proposed toexplain the properties of FRBs (see e.g. Katz 2016, fora brief review). According to the models, the excessDM for FRBs may be intrinsic to the source, placing itwithin the Galaxy; it may arise mostly from the inter-galactic medium, placing a source of FRBs at cosmolog-ical distances ( z ∼ . −
1) or it may arise from the hostgalaxy, placing a source of FRBs at extragalactic, butnot necessarily cosmological, distances ( ∼
100 Mpc).Since the only evidence to claim an extragalactic ori-gin for FRBs has been the anomalously high DM, somemodels also attempted to explain the excess DM as apart of the model, thus allowing FRBs to be Galactic.All FRBs observed to date have been detected with sin-gle dish radio telescopes, for which the localization is oforder arcminutes, insufficient to obtain an unambiguousassociation with any object. To date, no independent in-formation about their redshift, environment, and sourcecould be obtained due to the lack of an accurate localiza-tion of FRBs. Keane et al. (2016) attempted to identifythe host of FRB 150418 on the basis of a fading radiosource in the field that was localized to a z = 0 . ≈ × − — the first unam-biguous identification of multi-wavelength counterpartsto FRBs. Independently, Marcote et al. (2017) used theEuropean VLBI Network (EVN) to localize the burstsand the persistent source and showed that both are co-located within ∼
12 milliarcseconds.Here we report the imaging and spectroscopic follow-up of the optical counterpart to FRB 121102 using the8-m Gemini North telescope. OBSERVATIONS AND DATA ANALYSISThe location of FRB 121102 was observed with theGemini Multi-Object Spectrograph (GMOS) instrumentat the 8-m Gemini North telescope atop Mauna Kea,Hawai’i. Imaging observations were obtained with SDSS r (cid:48) , i (cid:48) and z (cid:48) filters on 2016 October 24, 25, and November2, under photometric and clear conditions with 0 . (cid:48)(cid:48)
58 to0 . (cid:48)(cid:48)
66 seeing. Exposure times of 250 s were used in the r (cid:48) filter and of 300 s in the i (cid:48) and z (cid:48) filters with totalexposures of 1250 s in r (cid:48) , 1000 s in i (cid:48) and 1500 s in z (cid:48) . Thedetectors were read out with 2 × . (cid:48)(cid:48)
146 pix − . The images were corrected fora bias offset, as measured from the overscan regions, flatfielded using sky flats and then registered and co-added.The images were astrometrically calibrated againstthe Gaia
DR1 Catalog (Gaia Collaboration et al. 2016).To limit the effects of distortion, the central 2 . (cid:48) × . (cid:48) r (cid:48) , i (cid:48) ,and z (cid:48) images were matched with 35 – 50 unblendedstars yielding an astrometric calibration with 7 – 9 masroot-mean-square (rms) position residuals in each coor-dinate after iteratively removing ∼ − Gaia frame is thus ∼ − Source Extractor (Bertin & Arnouts1996) software to detect and extract sources in the coad-ded images. The r (cid:48) and i (cid:48) images were photometri-cally calibrated with respect to the IPHAS DR2 cat-alog (Barentsen et al. 2014) using Vega-AB magnitudeconversions stated therein. We measure isophotal in-tegrated magnitudes of m r (cid:48) = 25 . ± . m i (cid:48) = 23 . ± . g (cid:48) , r (cid:48) , i (cid:48) , and z (cid:48) bands and will be reported in a subsequent publica-tion.Spectroscopic observations were obtained with GMOSon 2016 November 9 and 10 with the 400 lines mm − grating (R400) in combination with a 1 (cid:48)(cid:48) slit, coveringthe wavelength range from 4650 to 8900 ˚A. A total of Tendulkar et al.
Figure 1.
The co-added spectrum of the host galaxy of FRB 121102, the reference object, and the sky contribution (scaledby 10% and offset by − µ Jy). The spectra have been resampled to the instrumental resolution. Prominent emission lines arelabelled with their rest frame wavelengths. Black horizontal bars denote the wavelength ranges of the filters used for imaging.Most of the wavelength coverage of the z (cid:48) band is outside the coverage of this plot. nine 1800 s exposures were taken with 2 × . (cid:48)(cid:48)
292 pix − and an instrumentalresolution of 4.66 ˚A, sampled at 1 .
36 ˚A pix − . The con-ditions were clear, with 0 . (cid:48)(cid:48) . (cid:48)(cid:48) . (cid:48)(cid:48) . (cid:48)(cid:48) . ◦
6, containing the coun-terpart to FRB 121102 as well as an m r (cid:48) = 24 . m i (cid:48) = 22 . . (cid:48)(cid:48) m i (cid:48) ≈ .
6, about 11% higher thanderived from photometry. Given that the spectrophoto-metric standard was observed on a different night withworse seeing (1 . (cid:48)(cid:48) RESULTS AND ANALYSISThe final combined and calibrated spectrum is shownin Figure 1. Besides continuum emission, which is he host of FRB 121102 Table 1.
Emission line properties.
Line Obs. Flux Width ( σ ) A λ /A V (erg cm − s − ) (˚A) (mag)H β . × − III ] λ . × − III ] λ . × − I ] λ < . × − II ] λ < . × − α . × − II ] λ < . × − II ] λ . × − II ] λ . × − Note —Observed emission line properties from fittingnormalized Gaussians to the rest-wavelength host galaxyspectrum. Upper limits (3- σ ) on line fluxes assumeGaussian widths of σ = 2 ˚A. The absorption A λ /A V atthe observed line wavelengths is taken from Cardelli et al.(1989). To obtain unabsorbed line fluxes, multiply by10 . A λ /A V ) A V , where A V is the Galactic absorptiontowards FRB 121102. weakly detected in the red part of the spectrum, fourstrong emission lines are clearly visible and are iden-tified as H α , H β , and [O III ] λ III ] λ z = 0 . ± . II ] λλ II ] λλ I ] λ σ width values listed in Table 1. Weestimate rest frame equivalent widths for the strongestemission lines; 392 ±
102 ˚A for [O
III ] λ ±
55 ˚A for H α .The ratios of measured line fluxes for [O III ]/H β against [N II ]/H α and [S II ]/H α — the well-known Bald-win, Phillips & Terlevich (BPT) diagram (Baldwin et al.1981) — are shown in Figure 2. The line ratios of thehost galaxy of FRB 121102 are compared to those fromthe SDSS DR12 galaxy sample (Alam et al. 2015). Thelocations below and to the left of the solid and dashedgrey lines indicate that the emission lines are due to starformation and not due to AGN activity (Kewley et al.2001, 2006; Kauffmann et al. 2003). Note that the BPTdiagram line ratios are insensitive to reddening (fromthe Milky Way as well as the host itself).We use the galfit software (Peng et al. 2002, 2010)to constrain the morphology of the optical counterpart.A S´ersic profile (Σ( r ) = Σ e e − κ [( r/R e ) /n − ), convolved Figure 2.
BPT (Baldwin et al. 1981) diagrams of [N II ]/H α and [S II ]/H α against [O III ]/H β for the SDSS DR12 (Alamet al. 2015) galaxy sample with significant ( > σ ) emissionlines. The black symbol with error bars denotes the lo-cation of the host galaxy of FRB 121102. The solid anddashed lines denote the demarcations between star-formingand AGN dominated galaxies, respectively (Kewley et al.2001, 2006; Kauffmann et al. 2003). The region between thetwo curves corresponds to composite objects with AGN andstar formation. with the point-spread-function, was fitted against thespatial profile of the counterpart. For the i (cid:48) -band image,the best fit has an effective radius of R e = 0 . (cid:48)(cid:48) ± . (cid:48)(cid:48) n = 2 . ± .
5, and an ellipticity of b/a = 0 . ± .
13. The lower signal-to-noise of the coun-terpart in the r (cid:48) and z (cid:48) images did not permit meaningfulresults. Instead, we directly fit the spatial profile in allthree bands with a two-dimensional elliptical Gaussianprofile. In the case of the i (cid:48) -band image, the fit provides Tendulkar et al. a position and effective radius, taken as the Gaussian σ , consistent with the S´ersic profile convolved with thepoint-spread-function. The results of the fits are shownin Figure 3.The position and extent of the host galaxy, as ap-proximated with the two-dimensional elliptical Gaussianprofile, agrees well in the r (cid:48) and i (cid:48) bands (semi-majoraxis σ a = 0 . (cid:48)(cid:48)
44 with ellipticity b/a = 0 . z (cid:48) -band has a slightly offset position and appears larger( σ = 0 . (cid:48)(cid:48)
59 with b/a = 0 . r (cid:48) and i (cid:48) bands are dominatedby the bright emission lines of H α , H β , [O III ] λ III ] λ z (cid:48) -band traces thecontinuum flux of the host galaxy. As such, the mor-phology suggests that the host galaxy has at least oneH II region at a slight offset from the galaxy center.Finally, the bottom right panel of Figure 3 plots theGaussian centroids on the International Celestial Refer-ence Frame (ICRF) through the astrometric calibrationof the r (cid:48) , i (cid:48) , and z (cid:48) images against Gaia . The posi-tional uncertainties in each axis are the quadratic sumof the astrometric tie against
Gaia (of order 2 mas) andthe centroid uncertainty on the image (between 20 and50 mas). The
Gaia frame is tied to the ICRF definedvia radio VLBI to a ∼ ±
68 and 163 ±
32 mas in theline-dominated r (cid:48) and i (cid:48) images, and 286 ±
64 mas in thecontinuum-dominated z (cid:48) image. Though offset from thecentroids, the persistent radio source is located withinthe effective radii of the different bands. DISCUSSION AND CONCLUSIONSThe observations presented here confirm the interpre-tation by Chatterjee et al. (2017) that the extendedoptical counterpart associated with FRB 121102 is thehost galaxy of the FRB. Our measurement of the red-shift z = 0 . z DM < .
32 (Chatterjee et al. 2017) and to-gether with the very low chance superposition probabil-ity, firmly places FRB 121102 at a cosmological distance,ruling out all Galactic models for this source.In the following discussion, we assume the cosmolog-ical parameters from the Planck Collaboration et al.(2016) as implemented in astropy.cosmology (AstropyCollaboration et al. 2013), giving a luminosity distanceof D L = 972 Mpc, and 1 (cid:48)(cid:48) corresponding to projectedproper and comoving distances of 3.31 kpc and 3.94 kpc,respectively. Figure 3.
The top left, top right and bottom left pan-els show respective 7 . (cid:48)(cid:48) × . (cid:48)(cid:48) r (cid:48) , i (cid:48) and z (cid:48) images, centered on the optical counterpart toFRB 121102. Each image has been smoothed by a Gaussianwith a width of 0 . (cid:48)(cid:48)
2, while the plus sign and ellipse denotethe position and extent of a two-dimensional Gaussian fitto the spatial profile of the counterpart. The i (cid:48) -band imagealso shows the narrower S´ersic fit by galfit . The bottomright panel combines the positional and morphological mea-surements from the different bands on an astrometric frameof 1 (cid:48)(cid:48) × (cid:48)(cid:48) in size. The colors are identical to those used inthe other panels. The large ellipses denote the extent of theGaussian and S´ersic fits, while the small ellipses denote the1- σ absolute positional uncertainties. The location of thepersistent counterpart as measured with the EVN at 5 GHzby Marcote et al. (2017) is represented by the black cross.The uncertainty in the EVN location is much smaller thanthe size of the symbol.) We use the Schlegel et al. (1998) estimate of the Galac-tic extinction along this line of sight , E B − V = 0 . R V = 3 .
1, we find A V = 2 .
42, and use the Cardelliet al. (1989) Galactic extinction curve to correct thespectrum with band extinctions of A r (cid:48) = 2 . , A i (cid:48) =1 . , and A z (cid:48) = 1 .
16 mag. We note that the Schlaflyet al. (2010); Schlafly & Finkbeiner (2011) recalibratedextinction model predicts a slightly lower extinction of E B − V = 0 . k -correction to the magnitudes as they arenot needed for the precision discussed here. From the IRSA Dust Extinction Calculator http://irsa.ipac.caltech.edu/applications/DUST/ he host of FRB 121102
Burst Energetics
The redshift measurement allows us to put FRB 121102’senergetics on a firmer footing, confirming the energyscale of 10 erg ( δ Ω / π )( A ν / . ν/ A ν and ∆ ν are the fluence andbandwidth, respectively, at observing frequency ν and δ Ω is the opening angle of the bursts. A more detailedanalysis of energetics of individual bursts detected bythe VLA and their rates will be reported in Law C. J.et al. (in preparation).4.2.
Physical Properties of the Host
The host of FRB 121102 is a small galaxy with a diam-eter of (cid:46) z (cid:48) -band image. The absolute magnitudes, includingthe emission line fluxes and after correcting for theMilky Way’s extinction, are M r (cid:48) = − . M i (cid:48) = − . α luminosity of the host galaxy,corrected for Milky Way extinction, is L H α = 2 . × erg s − . The corresponding star formation rate isSFR(H α ) = 7 . × − M (cid:12) yr − × ( L H α / erg s − ) =0 . M (cid:12) yr − (Kennicutt et al. 1994). This value doesnot completely account for the extinction of H α pho-tons in the host galaxy. The correction suggested byKewley et al. (2002) is SFR( IR ) = 2 . × SFR(H α ) . ≈ . M (cid:12) yr − (in the 8–1000 µ m band). This is consistentwith the 3- σ upper limit of < M (cid:12) yr − estimated fromthe ALMA non-detection of the host at 230 GHz assum-ing a submillimeter spectral index α = 3 (Chatterjeeet al. 2017).The mass-to-light ratio Υ ∗ is dependent on the starformation history and the initial mass function for starformation. As an estimate, we use Υ R ∗ ≈ − M (cid:12) L − (cid:12) based on the dynamics of dwarf galaxies with high starformation rates (Lelli et al. 2014), implying a stellarmass M ∗ ∼ − × M (cid:12) . As dwarf galaxies areusually gas-rich (e.g. Papastergis et al. 2012), we expectthat this estimate is a lower limit to the host baryonicmass. We also note that dwarf galaxies are typicallydark matter dominated (Cˆot´e et al. 2000), and so thetotal dynamical mass is likely to be larger.We use the R (Kewley & Dopita 2002), N O N y ) to estimatethe metallicity where, R = log (([O II ] λ III ] λλ , / H β ) ,N ([N II ] λ / H α ) ,O N ([O III ] λ / [N II ] λ × H α/ H β ) , and y = log ([N II ] λ / [S II ] λλ , .
264 log ([N II ] λ / H α ) . As the [O II ] λ II ] λ R ≥ . ,N ≤ − . ,O N ≥ . ,y ≤ − . , where the limits are calculated from the 3- σ limit on[N II ] λ II ] λ σ metallicity limit of log ([O / H]) + 12 < . < . N < . O N and < . ([O / H]) + 12 (cid:46) . ∼
15% of all galaxiesbrighter than M B < −
16 have metallicity lower than8.7 (Graham & Fruchter 2015). This set of galaxies ac-count for less than 20% of the star formation of the localUniverse.The host properties are similar to those of extremeemission line galaxies (EELGs; Atek et al. 2011), young,low-mass starbursts which have emission lines of rest-frame equivalent widths greater than 200˚A.4.3.
Ionized Gas Properties in the Host
The Balmer lines from the host also allow us to esti-mate the properties its ionized ISM and its contributionto the total DM of FRB 121102.The H α surface density for the galaxy with flux F H α ,semi-major axis a , and semi-minor axis b is S (H α ) = F H α πab , ≈ . × − erg cm − s − arcsecond − , ≈
120 Rayleigh , (1)where we have used the extinction corrected flux F H α =2 . × − erg cm − s − and the semi-major and minor We note that the Pettini & Pagel (2004) calibration has highscatter for O N (cid:38) Tendulkar et al. axes ( a = 0 . (cid:48)(cid:48) b/a = 0 .
68) from the i (cid:48) and r (cid:48) images.In the source frame (denoted below by the subscript,‘s’), the surface density is S (H α ) s = (1 + z ) S (H α ) = 243 Rayleigh . (2)For a temperature T = 10 T K, we express the emis-sion measure (EM = (cid:82) n e d s ) given by Reynolds (1977)in the galaxy’s frameEM(H α ) s = 2 . − T . (cid:20) S (H α ) s Rayleigh (cid:21) , ≈
670 pc cm − T . . (3)We get a smaller value from the extinction-correctedH β flux, EM(H β ) s ≈
530 pc cm − . For the calcu-lations below, we proceed with a combined estimate,EM s ≈
600 pc cm − .This value is fairly large compared to measurementsof the local Galactic disk. The WHAM H α survey, forexample, gives values of tens of pc cm − in the Galacticplane and about 1 pc cm − looking out of the plane(Hill et al. 2008). However, lines of sight to distantpulsars and studies of other galaxies give EM values inthe hundreds (Reynolds 1977; Haffner et al. 2009).The estimate for EM s is sensitive to the inferred solidangle of the galaxy and emitting regions. Ongoing ob-servations with the Hubble Space Telescope will betterresolve the H α emitting structures and improve our con-straint on the EM with respect to the location of theburst.The implied optical depth for free-free absorption atan observation frequency ν (in GHz) is τ ff ≈ . × − [(1 + z ) ν GHz ] − . T − . EM s ≈ . × − ν − . T − . . (4)Free-free absorption for FRB 121102 is therefore negligi-ble even at 100 MHz. This suggests that the radio spec-tra of the bursts and possibly the persistent source areunaffected by absorption and are inherent to the emis-sion process or to propagation effects near the sources,confirming the inference made by (Scholz et al. 2016)based on the widely varying spectral shapes of the burstsalone. 4.3.1. Implied DM from H α -emitting Gas The EM implies a DM value sometimes given byDM = (EM f f L ) / , where f f is the volume filling factorof ionized clouds in a region of total size L (Reynolds1977). As summarized in Appendix B of Cordes et al.(2016), additional fluctuations decrease the DM derivedfrom EM, giving a source-frame value, (cid:100) DM s ≈
387 pc cm − L / (cid:20) f f ζ (1 + (cid:15) ) / (cid:21) / × (cid:18) EM600 pc cm − (cid:19) / , (5)where (cid:15) ≤ ζ ≥ L kpc in kpc. Here we have usedEM s = 600 pc cm − and assumed 100% cloud-to-cloudvariations ( ζ = 2) and fully modulated electron densitiesinside clouds ( (cid:15) = 1).The host contribution to the measured DM is a factor(1 + z ) − smaller than the source frame DM . Also, theline of sight to the FRB source may sample only a frac-tion of (cid:100) DM s depending on if it is embedded in or offsetfrom the H α -emitting gas. For an effective path lengththrough the ionized gas L FRB ≤ L , we then have (cid:100) DM(FRB) = (cid:100) DM s z (cid:18) L FRB L (cid:19) ≈
324 pc cm − (cid:18) L FRB L (cid:19) (cid:20) L kpc f f ζ (1 + (cid:15) ) (cid:21) / . (6)This estimate can be compared with empirical con-straints discussed in Chatterjee et al. (2017) on contri-butions from the host and the intergalactic medium(IGM) to the total DM made by subtracting theNE2001 model’s DM contribution from the MilkyWay (Cordes & Lazio 2002) (DM MW = 188 pc cm − )and the Milky Way halo (DM MW halo = 30 pc cm − )from the total DM = 558 pc cm − . This givesDM IGM + DM host = 340 pc cm − . The Milky Waycontributions have uncertain errors but are likely of or-der 20%. The measured redshift implies a mean IGMcontribution DM IGM ≈
200 pc cm − (Ioka 2003; In-oue 2004) but can vary by about ±
85 pc cm − (Mc-Quinn 2014). This yields a range of possible valuesfor DM host : 55 (cid:46) DM host (cid:46)
225 pc cm − that furtherimplies 0 . (cid:46) ( L FRB /L ) (cid:2) L kpc f f /ζ (1 + (cid:15) ) (cid:3) / (cid:46) . − and 70–200 pc cm − , respectively (Manchester et al.2005). This empirically demonstrates that the free elec-tron content of star-forming dwarf galaxies is of the or-der we estimate. The relatively large DM contributionfrom the host galaxy (as inferred from the H α emission)implies that any contributions from the vicinity of the The factor of (1+ z ) − is a combination of the photon redshift,time dilation and the frequency − dependence of cold plasma dis-persion. he host of FRB 121102 <
100 yr) supernova remnant (e.g.Piro 2016).4.4.
Implications for Source Models
Chatterjee et al. (2017) reported the locations of theradio bursts, the optical and variable radio counterpartsand the absence of millimeter-wave and X-ray emission.Marcote et al. (2017) have shown that the bursts and thepersistent radio source are colocated to within a linearprojected separation of 40 pc, suggesting that the twoemission sources should be physically related, thoughnot necessarily the same source. The radio source prop-erties are consistent with a low luminosity AGN or ayoung ( < ∼ M (cid:12) ).We also note that the radio source is offset from the op-tical center of the galaxy by 170–300 mas, correspond-ing to a transverse linear distance of 0.5–1 kpc, nearlya quarter to half of the radial extent, which is not con-sistent with a central AGN, but such offsets have beenseen before in dwarf galaxies, e.g. Henize 2-10 (Reineset al. 2011).The association of an optical/X-ray AGN with a dwarfgalaxy is also extremely rare. A search of emission-linedwarf galaxies (10 . (cid:46) M ∗ (cid:46) . M (cid:12) ) using BPT linediagnostics identified an AGN rate of ∼ α consistent with anAGN. Similarly, an X-ray survey of z < Relation to Dwarf Galaxies
It is interesting to note that the only FRB host di-rectly identified so far is a low metallicity dwarf galaxyrather than, say, an extremely high-star-formation-rategalaxy such as Arp 220 or a galaxy with a very power-ful AGN or some other extreme characteristics. Dwarfgalaxies are also a small fraction of the stellar massin the Universe (Papastergis et al. 2012). Ravi et al.(2016) also suggested that the extremely low scatter-ing of FRB 150807 compared to its DM may be linkedto its origin from a low-mass ( < M (cid:12) ) galaxy. How-ever, the strong polarization and scattering properties ofFRB 110523 do suggest the presence of turbulent mag-netized plasma around the source (Masui et al. 2015),suggesting that individual FRB environments may bequite diverse.If FRBs are indeed more commonly hosted by dwarfgalaxies in the low redshift Universe, they would sharethis preference with two other classes of high-energytransients — long duration gamma-ray bursts and su-perluminous supernovae, both of which prefer low-mass,low-metallicity, and high star formation rate hosts (e.g.,Fruchter et al. 2006; Perley et al. 2013; Vergani et al.2015; Perley et al. 2016, and other works). Indeed,superluminous supernovae are prefentially hosted byEELGs (Leloudas et al. 2015). If this relation is true, itmay point to a link between FRBs and extremely mas-sive progenitor stars, possibly extending to magnetarsthat have been associated with massive progenitor stars(e.g. Olausen & Kaspi 2014).4.5. Future Optical Follow-Up of FRBs
A link between FRBs and dwarf galaxies will impactfuture multi-wavelength follow-up plans. Without theprecise localization for FRB 121102 (Chatterjee et al.2017), the host galaxy is scarcely distinguishable fromother objects in the deep Gemini images.Due to the trade-off between field of view and local-ization precision, FRB search projects that have a largeFRB detection rate such as CHIME (Kaspi V. M. et al,.2017, in preparation), UTMOST (Caleb et al. 2016),and HIRAX (Newburgh et al. 2016) will localize highsignal to noise detections to only sub-arcmin precision.0
Tendulkar et al.
If FRB hosts are star-forming galaxies with strong emis-sion lines, slitless objective prism spectroscopy could ef-ficiently distinguish these objects from a field of starsand elliptical galaxies, leading to putative host identifi-cations without very precise localization. However, thisstrongly depends on the link between FRBs and theirhost properties and the homogeneity of FRBs — whichwill first have to be confirmed with more interferometriclocalizations.We note, of course, that our above discussion regard-ing the possible relationship between FRBs and dwarfgalaxies in general is based on a single data point of arepeating FRB, which may not be representative of thebroader FRB population (see Spitler et al. 2016; Scholzet al. 2016, for more details).We are very grateful to the staff of the Gemini Ob-servatory for their help and flexibility throughout thisprogram. We also thank R. F. Trainor and A. Delahayefor helpful discussions.Our work is based on observations obtained at theGemini Observatory (program GN-2016B-DD-2), whichis operated by the Association of Universities for Re-search in Astronomy, Inc., under a cooperative agree-ment with the NSF on behalf of the Gemini partner-ship: the National Science Foundation (United States),the National Research Council (Canada), CONICYT(Chile), Ministerio de Ciencia, Tecnolog´ıa e Innovaci´onProductiva (Argentina), and Minist´erio da Ciˆencia, Tec-nologia e Inova¸c˜ao (Brazil).This work has made use of data from the Euro-pean Space Agency (ESA) mission
Gaia ( ), processed by the Gaia
DataProcessing and Analysis Consortium (DPAC, ).Funding for the DPAC has been provided by national in-stitutions, in particular the institutions participating inthe
Gaia
Multilateral Agreement. This research madeuse of Astropy, a community-developed core Pythonpackage for Astronomy (Astropy Collaboration, 2013, ). S.P.T acknowledges support from a McGill Astro-physics postdoctoral fellowship. The research lead-ing to these results has received funding from the Eu-ropean Research Council (ERC) under the EuropeanUnion’s Seventh Framework Programme (FP7/2007-2013). C.G.B. and J.W.T.H. gratefully acknowledgefunding for this work from ERC Starting Grant DRAG-NET under contract number 337062. J.M.C., R.S.W.,and S.C. acknowledge prior support from the NationalScience Foundation through grants AST-1104617 andAST-1008213. This work was partially supported by theUniversity of California Lab Fees program under awardnumber LF-12-237863. The research leading to these re-sults has received funding from the European ResearchCouncil (ERC) under the European Unions SeventhFramework Programme (FP7/2007-2013). J.W.T.H. isan NWO Vidi Fellow. V.M.K. holds the Lorne Trottierand a Canada Research Chair and receives support froman NSERC Discovery Grant and Accelerator Supple-ment, from a R. Howard Webster Foundation Fellowshipfrom the Canadian Institute for Advanced Research (CI-FAR), and from the FRQNT Centre de Recherche en As-trophysique du Quebec. B.M. acknowledges support bythe Spanish Ministerio de Econom´ıa y Competitividad(MINECO/FEDER, UE) under grants AYA2013-47447-C3-1-P, AYA2016-76012-C3-1-P, and MDM-2014-0369of ICCUB (Unidad de Excelencia ‘Mar´ıa de Maeztu’).L.G.S. gratefully acknowledge financial support from theERC Starting Grant BEACON under contract number279702 and the Max Planck Society. Part of this re-search was carried out at the Jet Propulsion Laboratory,California Institute of Technology, under a contract withthe National Aeronautics and Space Administration.E.A.K.A. is supported by TOP1EW.14.105, which is fi-nanced by the Netherlands Organisation for ScientificResearch (NWO). M.A.M. is supported by NSF award
Facility:
Gemini:Gillett (GMOS)
Software:
ESO-MIDAS, astro-py, galfit, SExtractorREFERENCES