iPTF16geu: A multiply imaged, gravitationally lensed type Ia supernova

See allHide authors and affiliations

Science  21 Apr 2017:
Vol. 356, Issue 6335, pp. 291-295
DOI: 10.1126/science.aal2729

Multiple images of a type Ia supernova

General relativity indicates that any sufficiently massive object bends the path of light passing by it. This effect is known as gravitational lensing. Goobar et al. have identified a supernova that is strongly lensed by a foreground galaxy, causing it to be highly magnified and splitting the light into four separate images. What is more, it is a type Ia supernova, a well-studied variety with reliable properties that can be used to constrain models of the lensing. This distinctive object will enable cosmological measurements and can be used to probe the distribution of mass in the foreground galaxy.

Science, this issue p. 291


We report the discovery of a multiply imaged, gravitationally lensed type Ia supernova, iPTF16geu (SN 2016geu), at redshift z = 0.409. This phenomenon was identified because the light from the stellar explosion was magnified more than 50 times by the curvature of space around matter in an intervening galaxy. We used high-spatial-resolution observations to resolve four images of the lensed supernova, approximately 0.3 arc seconds from the center of the foreground galaxy. The observations probe a physical scale of ~1 kiloparsec, smaller than is typical in other studies of extragalactic gravitational lensing. The large magnification and symmetric image configuration imply close alignment between the lines of sight to the supernova and to the lens. The relative magnifications of the four images provide evidence for substructures in the lensing galaxy.

One of the foundations of Einstein’s theory of General Relativity is that matter curves the surrounding space-time. For the rare cases of nearly perfect alignment of an astronomical source, an intervening massive object, and the observer, multiple images of a single source can be detected—a phenomenon known as strong gravitational lensing.

Although many strongly lensed galaxies and quasars have been detected to date, it has proved extremely difficult to find multiply imaged lensed supernova (SN) explosions. Type Ia supernovae (SNe Ia) are particularly interesting sources because of their “standard-candle” nature. These explosions have nearly identical peak luminosity, which makes them excellent distance indicators in cosmology (1). For lensed SNe Ia, the standard-candle property allows the flux magnification to be estimated directly, independent of any model related to the lensing galaxy (2, 3). This removes two important degeneracies in gravitational lensing measurements: the mass-sheet degeneracy (4) and the source-plane degeneracy (5).

PS1-10afx, a lensed SN Ia at redshift z = 1.388 with a large amplification (μ ~ 30) where multiple images could have been expected, was reported several years ago (6). A foreground lens was later identified at z = 1.117 (7). At the time of the discovery, several interpretations were discussed, including a superluminous supernova (8). Because the lensed SN Ia hypothesis did not gain acceptance until long after the explosion had faded, no high-spatial-resolution imaging could be carried out in that case to verify the strong lensing nature of the system. Multiple images of another supernova, SN Refsdal (9), were discovered in a Hubble Space Telescope (HST) survey of the massive galaxy cluster MACS J1149.6+2223. As the source was identified as a core-collapse supernova, it could not be used to measure the lensing magnification directly.

Thanks to the well-known characteristics of their time-dependent brightness in optical and near-infrared (NIR) filters (the SN light curves), multiply imaged SNe Ia are also ideally suited for measuring time delays in the arrival of the images. This provides a direct probe of the Hubble constant, the cosmological parameter measuring the expansion rate of the universe (10), as well as leverage for studies of dark energy (11, 12), the cosmic constituent responsible for the accelerated expansion of the universe.

The intermediate Palomar Transient Factory (iPTF) searches the sky for new transient phenomena at optical wavelengths. It uses image differencing between repeated observations (13) with a large field-of-view camera (7.3 square degrees) at the 48-inch telescope (P48) at the Palomar Observatory (14). The first detection of iPTF16geu, with a statistical significance of five standard deviations (5σ), is from 5 September 2016. The new source was first recognized by a human scanner on 11 September (15). iPTF16geu (also known as SN 2016geu) was found near the center of the galaxy SDSS J210415.89-062024.7, at right ascension 21h04m15s.86 and declination –06°20′24″.5 (J2000 equinox).

Spectroscopic identification was carried out with the Spectral Energy Distribution (SED) Machine (16) at the Palomar 60-inch telescope (P60) on 2 October 2016, and iPTF16geu was found to be spectroscopically consistent with a normal SN Ia at z ≈ 0.4 (Fig. 1). Further spectroscopic observations from the Palomar 200-inch telescope (P200) and the 2.5-m Nordic Optical Telescope (NOT) were used to confirm the SN Ia identification and to establish the redshift of the host galaxy, from narrow sodium (Na i D) absorption lines, as z = 0.409. The P200 and NOT spectra also showed absorption features from the foreground lensing galaxy at z = 0.216. To estimate the velocity dispersion of the lensing galaxy, we fit two Gaussian functions with a common width to the Hα and [N ii] emission lines in the P200 spectrum in Fig. 1D. After taking the instrumental resolution into account, we measure σ = Embedded Image Å, corresponding to a velocity dispersion of σv = Embedded Image km s–1.

Fig. 1 Spectroscopic identification of iPTF16geu as a type Ia supernova and measurements of the redshifts of the SN host galaxy and the intervening lensing galaxy.

Measurements of the SN spectral energy distribution Fλ obtained with the P60, P200, and NOT telescopes are best fitted by a normal SN Ia spectral template. (A) Comparison with a nearby SN Ia, SN 2011fe, redshifted to z = 0.409 (green line) (22) at a similar rest-frame phase, expressed in units of days with respect to the time of the optical light curve maximum. The spectra also reveal narrow absorption and emission lines, marked by the dashed vertical lines, from which the redshifts of the lens (z = 0.216, blue lines) and SN host galaxy (z = 0.409, red lines) were determined. (B to D) Zoomed-in view in rest-frame wavelengths of the Ca ii H&K absorption features (B), the Na i D absorption features (C), and the Hα and [N ii] emission lines (D). The Hα and [N ii] emission lines at z = 0.216 were used to fit the velocity dispersion of matter in the lensing galaxy, σv = Embedded Image km s–1.

Photometric observations of iPTF16geu collected at P48 and with the SED Machine Rainbow Camera (RC) at P60 between 5 September and 13 October 2016 (Fig. 2) were used to estimate the peak flux and light curve properties of the SN with the SALT2 light curve fitting tool (17). The best-fit light curve template, also shown in Fig. 2, confirms that the observed light curve shapes are consistent with a SN Ia at z = 0.409. These fits also indicate some reddening of the supernova, which suggests that iPTF16geu suffers from moderate extinction by dust. This would produce dimming at optical wavelengths of 20 to 40%, with the largest losses in the g-band observations. Thanks to the standard-candle nature of SNe Ia, after correcting the peak magnitude for light curve properties (18, 19), the flux of the SN was found to be ~30 standard deviations brighter than expected for the measured redshift. This suggested that iPTF16geu was gravitationally lensed, and we estimated the lensing amplification to be μ ~ 52. Expressed in astronomical magnitudes, Δm = –4.3 ± 0.2 mag, where the uncertainty is dominated by the brightness dispersion of normal SNe Ia. Because the magnification is derived from comparing the observed brightness of iPTF16geu to other SNe Ia (20) within a narrow redshift range around z = 0.409, the measurement of the lensing magnification is independent of any assumptions on cosmology (e.g., the value of the Hubble constant or other cosmological parameters). The lensing magnification is also independent of a lens model, which is the only way to determine the magnification for almost all other strong lensing systems.

Fig. 2 Multicolor light curve of iPTF16geu showing that the supernova is 4.3 magnitudes (30 standard deviations) brighter than expected.

The magnitudes are measured with respect to time of maximum light (modified Julian date 57653.10) in the R-band at P48 and in the g-, r-, and i-bands with the SED Machine RC at P60. The filter transmission functions are shown in (24). The solid lines show the best-fitted SALT2 (17) model to the data. The dashed lines indicate the expected light curves at z = 0.409 (without lensing); the bands represent the standard deviation of the brightness distribution for SNe Ia. To fit the observed light curves, a brightness boost of 4.3 magnitudes is required.

The optical observations from Palomar, with a typical angular resolution (atmospheric seeing) of 2″, were insufficient to spatially resolve any multiple images that could result from the strong lensing nature of the system (Fig. 3A). We therefore obtained Ks-band (2.2 μm) observations from the European Southern Observatory (ESO) with the Nasmyth Adaptive Optics System Near-Infrared Imager and Spectrograph (NaCo) at the Very Large Telescope (VLT). An angular resolution of ~0.3″ [full width at half maximum (FWHM)] was obtained at the location of the target. Adaptive optics corrections of the seeing were performed using a natural bright star, ~30″ southeast of the SN location, indicated in Fig. 3 along with the SDSS pre-explosion image of the field (21).

Fig. 3 Image of the field of iPTF16geu showing the spatial resolution of the ground-based instruments used in this work.

Left: Pre-explosion multicolor image from SDSS indicating the bright natural guide star. (A) Position of the SN detection in the R-band at P48 (zoomed-in view near the galaxy SDSS J210415.89-062024.7). (B and C) Improved spatial resolution with the use of NGSAO (B) and LGSAO (C).

The near-IR image from VLT indicated that the structure was as expected in a strongly lensed system, with higher flux in the northeastern and southwestern regions of the system than in the center (Fig. 3B). Multiple images of the system were first resolved with observations from the Keck observatory at NIR wavelengths, using Laser Guide Star Adaptive Optics (LGSAO) with the OH-Suppressing Infrared Imaging Spectrograph (OSIRIS) instrument, which yielded an image quality of 0.07″ FWHM in the H-band centered at 1.6 μm (Fig. 3C).

Shown in Fig. 4 are LGSAO observations of iPTF16geu using the Near-Infrared Camera 2 (NIRC2) at the Keck telescope on 22 October and 5 November 2016, in the Ks-band and J-band (1.1 μm), respectively, and optical images obtained with HST on 25 October 2016. The HST observations were carried out through the F475W, F625W, and F814W filters, where the names correspond to the approximate location of the central wavelength in nanometers.

Fig. 4 High-spatial-resolution images from the Hubble Space Telescope and the Keck Observatory used to resolve the positions of the SN images, the partial Einstein ring of the host galaxy, and the intervening lensing galaxy.

(A to C) HST/WFC3 observations of iPTF16geu obtained on 25 October 2016 in the F475W, F625W, and F814W bands, respectively. The images reveal four point sources, except for F475W where SN images 3 and 4 are too faint. (D to F) NIR images obtained using adaptive optics–aided Keck observations in the J-, H-, and Ks-bands, respectively. All four SN images are clearly seen in the J-band (D). For the H- and Ks-band images, both the lensing galaxy at the center of the system and the lensed partial Einstein ring of the host galaxy are visible.

The observations exhibit four images of iPTF16geu, 0.26″ to 0.31″ from the center of the lensing galaxy, with nearly 90° azimuthal separations. The extended host galaxy, warped by the lens to form a partial Einstein ring, is brighter in the NIR spectrum relative to the observations through optical filters. Thus, the fainter individual SN images are poorly resolved for the observations with the longest wavelengths in Fig. 4. Furthermore, the SN Ia spectral energy distribution (redshifted to z = 0.4) peaks within the F625W and F814W filters [see, e.g., (22)]. Dimming by interstellar dust in the line of sight is roughly inversely proportional to wavelength in the optical and NIR spectra (23). The biggest impact from extinction by dust is therefore expected for the shortest wavelength, in F475W filter observations, where the two faintest SN images cannot be detected above the background light. The low-spatial-resolution light curves in Fig. 2 are dominated by the two brightest SN images, labeled 1 and 2 in Fig. 4D. The F625W-F814W magnitude difference (color) of the resolved images measured with HST indicates small differences in relative extinction among the SN images, except for image 4, which appears to have about two magnitudes of additional dimming in F814W.

Unaccounted dimming of light by scattering on dust grains in the line of sight would lead to an underestimation of the lensing amplification. If corrections for differential extinction in the intervening lensing galaxy among the SN images are included, the result is a wider range for the lensing magnification of iPTF16geu, between –4.1 and –4.8 mag (24).

The SN multiple-image positions in Fig. 3 were used to construct a lensing model, with an isothermal ellipsoid galaxy lens (25, 26) with ellipticity εe = 0.15 ± 0.07 and mass M = 1.70 (±0.06) × 1010 solar masses inside an ellipse with major axis of 1.13 kpc and minor axis of 0.97 kpc. Details of the lensing model are presented in (24). The lens model can be independently verified through comparisons between the model-predicted and observed velocity dispersion of the lensing galaxy. From the model we derive an estimate, Embedded Image = 156 ± 4 km s–1, in good agreement with the measured value of the velocity dispersion (Fig. 1D).

However, the adopted smooth isothermal ellipsoid lens model predicts brightness differences between the multiple SN images that are in disagreement with the observations. Including corrections for extinction in the resolved SN images in the F814W filter, we find large discrepancies between the model and measured magnitude differences for the multiple images of iPTF16geu: Embedded Image = (–0.3, –1.6, –1.5) mag for j = 2, 3, and 4, where the indices follow the numbering scheme adopted in Fig. 4. The observed discrepancy between the smooth model predictions for the SN images 1 and 2 compared to 3 and 4 (brighter by factors of 4 and 3, respectively) cannot be accounted for by time delays between the images, as they are predicted to be <35 hours (24). Graininess of the stellar distribution and dark matter sub-halos in the lens galaxy, in addition to the smooth mass profile, can cause variations to magnification without altering image locations. These milli- and microlensing effects (27, 28), small enough not to cause additional resolved image separations, offer a plausible explanation for the deviation from the smooth lens model.

Available forecasts for wide-field surveys (29) suggest that about one strongly lensed SN Ia could be expected in our survey, irrespective of redshift and magnification, with approximately a 30% chance of being in a quad configuration. For an average ellipticity of the lenses e = 0.3 (29), only about 1% of the lensed SNe are expected to have μ > ~50 (30). We have performed an independent rate estimate, with a somewhat simplified lensing simulation but including survey-specific parameters, and confirmed that the probability of detecting and classifying a highly magnified SN Ia like iPTF16geu does not exceed the few-percent level (24).

iPTF16geu appears to be a rather unlikely event, unless the actual rate of very magnified SNe is higher than anticipated—for example, if the contribution from lensing by any kind of substructures in galaxies is underestimated, or if we are otherwise lacking an adequate description of gravitational lensing at the ~1-kpc scale. The physical scale probed by the resolved images of iPTF16geu is comparable to the smallest of the 299 multiply imaged lensed systems in the Master Lens Database, (31). Using the standard-candle nature of SNe Ia, we can more easily detect strongly lensed systems with sub–arc second angular separations, allowing exploration of the bending of light at scales less than ~1 kpc—an otherwise challengingly small distance in studies of gravitational lensing (32). As demonstrated with iPTF16geu, discovered while still brightening with a modest-sized telescope and suboptimal atmospheric conditions, the locations of these rare systems can be identified in advance of extensive follow-up imaging at high spatial resolution.

Supplementary Materials

Materials and Methods

Figs. S1 to S3

Tables S1 to S5

References (3456)

References and Notes

  1. See supplementary materials.
  2. More information about the iPTF survey is available at
Acknowledgments: Supported by the Swedish National Science Council and the Swedish National Space Board (A.G., R.A., and E.M.); U.S. Department of Energy (DOE) grant DE-AC02-05CH11231 (P.E.N.); European Research Council grant 615929 (M.S.); NASA (S.R.K.); and NSF grant 1545949. The iPTF Swedish collaboration is funded through a grant from the Knut and Alice Wallenberg foundation. This research used resources of the National Energy Research Scientific Computing Center, supported by DOE contract DE-AC02-05CH11231. Some of the data presented here were obtained at the W. M. Keck Observatory, which is operated as a scientific partnership among the California Institute of Technology, the University of California, and NASA grant HST-GO-14862.002. The Observatory was made possible by the generous financial support of the W. M. Keck Foundation. Some of the data presented here were obtained with ALFOSC, which is provided by the Instituto de Astrofisica de Andalucia (IAA) under a joint agreement with the University of Copenhagen and NOTSA. The Space Telescope Science Data Analysis System (STSDAS) and the command language PyRAF are products of the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy (AURA) for NASA. Part of the processing was carried out off-line using the commercial software package MATLAB and Statistics Toolbox Release 2013a, The MathWorks Inc., Natick, MA. Photometric data used in this paper are available in tables S1, S2, and S5, spectroscopic data are available at public repository WISeREP (33) ( under the ID “SN 2016geu”; the positions of the SN images used in the lensing model are provided in table S4.
View Abstract


Navigate This Article