# AVI SHPORER

### Research Scientist, MIT Kavli Institute

Somewhere, something incredible  is waiting to be known.
Carl Sagan

### Ongoing Projects and recent papers:

The LCO-K2 Warm Jupiters project - Studying the inflated gas giant exoplanet conundrum
Many of the short period gas giant exoplanets (aka hot Jupiters) have radii larger than theoretically expected. Although several explanations have been proposed none have completely solved this puzzle. One clue to understanding gas giant inflation is the empirical correlation between planet radius and stellar irradiation at the planet's orbit. While it is consistent with inflation due to increased stellar irradiation it does not identify the exact mechanism, and, correlation does not necessarily mean causation. That correlation is shown as a solid red line in the plot below, showing planet radius as a function of stellar irradiation for gas giants with a well measured radius and mass. The plot shows that due to low number statistics it is not clear how low in planet radius and irradiation the correlation continues and whether there are indeed no inflated gas giant below a certain irradiation level.
Through the LCO Transiting Exoplanet CHaracterization (TECH) group we have set out to detect transiting gas giants with an irradiation below about 108 erg s-1 cm-2, which we refer to as Warm Jupiters. The primary goal is to better characterize the radius-irradiation correlation. A second goal is to study orbital migration by measuring the orbital eccentricity of warm Jupiters. Due to their wider orbits warm Jupiters are not expected to be tidally circularized, and their eccentricity is a fingerprint of orbital migration processes. A third goal is to detect gas giants with lower equilibrium temperatures, for atmospheric studies, and with a weaker tidal interaction with the host star, for stellar obliquity studies.
The transiting Warm Jupiter candidates for this project are identified primarily in K2 data. Follow-up observations, to identify the candidates true nature, are done using LCO observational resources and resources available to group members.
The first two warm Jupiters discovered by this program were published in Shporer et al. 2017a. Another candidate turned out to be a brown dwarf, published in Bayliss et al. 2017, see also the AAS press conference about that discovery. Radial velocity follow-up Observations carried out as part of this program have also identified that 3 of the statistically validated K2 planets are in fact stellar binaries, published in Shporer et al. 2017b.

The astrophysics of visible-light orbital phase curves in the space age.

Shporer 2017, PASP Invited Review
The field of visible-light continuous time series photometry is now at its golden age, manifested by the continuum of past (CoRoT, Kepler), present (K2), and future (TESS, PLATO) space-based surveys delivering high precision data with a long baseline for a large number of stars. The availability of the high quality data has enabled astrophysical studies not possible before, including for example detailed asteroseismic investigations and the study of the exoplanet census including small planets. This has also allowed to study the minute photometric variability following the orbital motion in stellar binaries and star-planet systems which is the subject of this review. The orbital modulations are induced by a combination of gravitational and atmospheric processes, including the beaming effect, tidal ellipsoidal distortion, reflected light, and thermal emission. Therefore, the phase curve shape contains information about the companion's mass and atmospheric characteristics, making phase curves a useful astrophysical tool. For example, phase curves can be used to detect and measure the mass of short-period low-mass companions orbiting hot fast-rotating stars, out of reach of other detection methods. Another interesting application of phase curves is using the orbital phase modulations to look for non-transiting systems, which comprise the majority of stellar binary and star-planet systems. I discuss the science done with phase curves, the first results obtained so far, and the current difficulties and open questions related to this young and evolving subfield.

Transiting Exoplanets with LCO - The Network Awakens (LCO Key Project)
I am the PI of a Las Cumbres Observatory (LCO) Key Project aimed at observations of transiting exoplanets. The Key Project includes 3,582.5 hours over about 2 years (4 LCO semesters) from April 2017 to May 2019. About 52% of the time was awarded directly by LCO and 48% is expected to be contributed by LCO partners. The Key Project is divided into 4 parts, or observational activities: (1) Follow-up and confirmation of transiting warm Jupiters. (2) Observations of transits of systems showing transit timing variations (TTVs). (3) Follow-up and confirmation of transiting candidates orbiting bright and quiet stars, that once confirmed are suitable for detailed characterization. (4) Snapshot photometry, also called on/off photometry, to identify the variable object in transit survey data with wide PSFs where many stars could be blended together, namely KELT and TESS. More information about the Key Project is available here.

Radial velocity monitoring of Kepler heartbeat stars with Keck/HIRES

Primary collaborators: Jim Fuller, Kelly Hambleton, Susan Mullally. Heartbeat stars are an emerging class of eccentric binary stars with close periastron passages. The characteristic heartbeat signal evident in their light curves is produced by a combination of tidal distortion, reflection, and Doppler boosting near orbital periastron. Many heartbeat stars continue to oscillate after periastron and along the entire orbit, indicative of the tidal excitation of oscillation modes within one or both stars. These systems are among the most eccentric binaries known, and they constitute an exciting opportunity to observe tidal effects in action. We are carrying out a radial velocity monitoring of Kepler heartbeat stars using Keck/HIRES, in order to measure the orbit and characterize the two stars. Our sample currently includes over 30 systems, which is the largest sample of these unique systems where the orbit was measured with radial velocities. Our goal is to understand the formation and evolution of heartbeat stars, and to use them to study the processes of tidal dissipation and orbital migration. The physics learned from them will apply to many other astrophysical systems, such as high-eccentricity planet migration and eccentricity-induced mergers in triple systems.
The figure below shows on the left panel a Kepler phase folded light curve of a Heartbeat star (measurements in gray, smoothed light curve in red), along with the radial velocity orbit on the right panel (measurements in black, fitted orbital model in red). The orbital eccentricity of this binary systems is ≈0.8.
The first results from this project are published in Shporer et al. 2016, are presented on this website, and were featured in JPL News.

Kepler Beaming Binaries radial velocity follow-up with WIYN/Hydra

Primary collaborators: Keivan Stassun, Simchon Faigler, Tsevi Mazeh, Tabetha Boyajian
High-quality space-based time series photometry reveals the minute photometric modulations induced by orbital motion in short-period binary systems with stellar and substellar secondaries. Those modulations are induced by both gravitational and atmospheric processes. Gravitational processes include the beaming effect (aka Doppler boosting) and tidal ellipsoidal distortion, and the atmospheric processes include reflected light and thermal emission by the secondary atmosphere. Therefore, non-eclipsing systems are detectable using photometry alone. The availability of Kepler data for a large sample of stars combined with the sensitivity to non-eclipsing systems has the potential of transforming the Kepler survey into the equivalent of a radial velocity (RV) survey of a large sample of stars with a wide range of spectral type. This allows detecting intrinsically rare systems, where traditional approaches, e.g., RV and transit surveys, are highly inefficient. Those include systems where the companion is a brown-dwarf or a massive planet, or even a white dwarf. As this approach is still in its infancy, we are carrying out RV follow-up of Kepler photometric detections, to confirm the nature of the system and measure the orbit and the companion’s mass. In order to obtain a large number of RV measurements we are using the WIYN/Hydra multi-fiber spectrograph, where we used 29.5 nights so far during the 2014, 2015, and 2016 Kepler observing seasons to observe 5 Hydra one-degree diameter fields within the Kepler field. Our list of targets includes 131 Kepler beaming binary candidates, and we used additional fibers to observe 85 Kepler eclipsing binaries and 31 KOIs. A detailed comparison between the photometrically predicted companion’s mass and the mass measured through RVs will improve our understanding of this young approach, and will support similar projects using data from current and future space-based time series photometry missions including K2, TESS, and PLATO. Our primary long term goal is to characterize the so-called brown-dwarf desert - a paucity of short-period companions in the brown-dwarf mass range, orbiting G- and K-type stars - and explore how its characteristics depend on the primary spectral type.
The figure below shows figure detection of a non-eclipsing binary system in Kepler light curves (left; photometric measurements in gray and binned light curve in black), with the corresponding Hydra RV
confirmation (right; RV measurements in black). In both panels the measurements are folded on the orbital period and a model is overplotted in solid red line.

Time Variation of Kepler Transits Induced by Stellar Spots —

Primary collaborators: Tomer Holczer, Tsevi Mazeh
Some transiting planets discovered by the Kepler mission display transit timing variations (TTVs) induced by stellar spots that rotate on the visible hemisphere of their parent stars. An induced TTV can be observed when a planet crosses a spot and modifies the shape of the transit light curve, even if the time resolution of the data does not allow the detection of the crossing event itself. Our new approach can, in some cases, use the derived TTVs of a planet to distinguish between a prograde and
a retrograde planetary motion with respect to the stellar rotation. Assuming a single spot darker than the stellar disk, spot crossing by the planet can induce measured positive (negative) TTV, if the crossing occurs in the first (second) half of the transit. On the other hand, the motion of the spot toward (away from) the center of the stellar visible disk causes the stellar brightness to decrease (increase). Therefore, for a planet with prograde motion, the induced TTV is positive when the local slope of the stellar flux at the time of transit is negative, and vice versa. Thus, we can expect to observe a negative (positive) correlation between the TTVs and the photometric slopes for prograde (retrograde) motion. Detecting this correlation in Kepler transiting systems allows to distinguish between prograde and retrograde planetary motions, and a similar approach can be applied to eclipsing binaries. The figure on the left shows an example of a the detection of a negative correlation between TTV and local slope, for Kepler-17b. This suggests the system is in a prograde motion, as already identified in Desert et al. (2011, ApJS, 197, 14).
This project resulted so far in two publications:
Mazeh, Holczer, & Shporer 2015, ApJ, 800, 142 - Detailed description of the method.
Holczer, Shoprer, et al. 2015, ApJ, 807, 170 - Application of this method to KOIs.

### Previous work:

Studying atmosphere-dominated hot Jupiter Kepler phase curves:
• Evidence that inhomogeneous atmospheric reflection is common,
• Shporer & Hu 2015, AJ, 150, 112
• In this work we sought out to study hot Jupiter atmospheres through their Kepler (i.e. optical) phase curves. However, typically when a phase curve signal is identified it is a super position of several processes, including not only atmospheric processes (reflected light and thermal emission) but also gravitational processes like the beaming effect and tidal ellipsoidal distortion. Therefore, interpreting the Kepler phase curve requires simultaneous modeling of all processes which may introduce degeneracies. Therefore we studied 3 phase curves that show only an atmospheric signal while the gravitational processes are negligible. We model 2 of those light curves in this work (Kepler-12b and Kepler-41b), shown in the figure below, and a 3rd phase curve (Kepler-7b) was analyzed elsewhere. Somewhat surprisingly all 3 phase curves show a maximum at a later phase than the secondary eclipse phase. We interpret this as resulting from an asymmetric brightness distribution, in the optical, of the planets' atmosphere where the brightest region is shifted west of the substellar point. This is consistent with results based on IR phase curves of other systems. Since these 3 phase curves are the only ones where the atmospheric component is seen directly (without the gravitational signal) and they all show an asymmetric atmospheric brightness distribution, it suggests that a similar asymmetry exists in many other hot Jupiter atmospheres as well. We discuss the possible implications of such asymmetry.

• Atmospheric characterization of the hot Jupiter Kepler-13Ab,
Shporer et al. 2014, ApJ, 788, 92
Kepler-13Ab is a rare hot Jupiter as it orbits the hottest star currently known to host a hot Jupiter, hence it is one of the hottest known hot Jupiters. In a previous paper, Shporer et al. (2011), I measured its mass using the Kepler orbital phase curve. In this paper we set out to characterize it in more detail, especially its atmosphere. We measured the secondary eclipse in several different wavelengths: optical (Kepler), near-IR (P200/WIRC; although with a low S/N) and IR (Spitzer). The left panel below shows the Kepler secondary eclipse phased and binned light curve. The high S/N, of > 300, makes it one of the most significant measurements of an exoplanet secondary eclipse. The right panel shows how the various secondary eclipses constrain the geometric albedo (Ag) and brightness temperature (TD) parameter space. The combination of these constraints gives TEFF = 2,750 ± $\pm$160 K and Ag = 0.33 ± 0.06 for the day-side, confirming the expected high day-side temperature and identifying a high albedo. The host star has a visual binary companion 1.2 arcsec away, so they are fully blended in all our photometric data sets. We obtained a Keck/HIRES spectrum of each of the two stars and derived the flux ratio as a function of wavelength, which we then integrated to get the dilution factor for each of our data sets. The improved host star parameters we derived from the spectrum made it smaller and cooler than previously thought. This in turn led to a smaller planet radius based on the measured transit depth. This work also included an analysis of the Kepler full orbital phase curve, deriving a night-side brightness temperature and a refined planet mass through the beaming effect and the tidal ellipsoidal distortion of the host star. We also noticed that the secondary eclipse time is 34.6 ± 6.9 seconds earlier than expected based on the mid transit time and light-travel time across the orbit. All data included in this work is publicly available, either through dedicated archives or by request from the authors.

On using the beaming effect to measure spin-orbit alignment in stellar binaries with Sun-like components
Shporer et al. 2012, NewA, 17, 309
The beaming effect (aka Doppler boosting) is a minute effect that causes the observed flux of a source to vary following a radial velocity variation. In this work I applied the beaming effect to the stellar rotation during eclipses. As the eclipsing object crosses over the disk of the eclipsed star it blocks regions with different radial velocity, so due to the beaming effect it induces a photometric signal that is superimposed on the eclipse light curve. This signal is the photometric analog of the spectroscopic Rossiter-McLaughlin (RM) effect, and is sensitive to the sky-projected angle between the eclipsed star spin and the orbital angular momentum, commonly referred to as the spin-orbit angle. The figure shows the Photometric RM signal during eclipse for various spin-orbit angles ($\lambda$) in red, while the blue line shows the variation in the orbital beaming effect signal during eclipse, as the contribution from each of the stellar components changes due to the eclipse. The paper gives analytic approximations to the Photometric RM effect and discusses prospects of detecting it in Kepler data.

• Detection of KOI-13.01 using the photometric orbit
Shporer et al. 2011, AJ, 142, 195

The transiting planet candidate KOI-13.01 showed a clear variability along the orbit, i.e., a phase curve signal, showed in the figure to the right. Kepler data is marked as small gray dots, binned light curve is in black circles (marker size similar to error bars), and the gray thick line is the fitted model. Data within the transit (phase=0) and secondary eclipse (phase=0.5) is excluded. The variability along the orbital motion is composed of several components, shown in dashed line is the figure. Those include the beaming effect (B, red), the tidal ellipsoidal distortion (E, blue) and atmospheric reflection and thermal emission (R, green). By analyzing the phase curve I measured the planet's mass, thereby confirming its planetary nature since high precision radial velocities cannot be obtained for the hot early-type host star. This was the first time that an exoplanet mass was measured with the phase curve alone. I also showed that while excluding the transit and secondary eclipse data the planet can be detected using the phase curve modulations.

The impact of the convective blueshift effect on spectroscopic planetary transits
Shporer & Brown 2011, ApJ, 733, 30

The convective blueshift (CB) effect is a net blueshift emanating from the stellar surface, resulting from a larger contribution of rising hot and bright gas relative to the colder and darker sinking gas. Since the CB radial component varies across the stellar surface, the light blocked by the planet during a planetary transit has a varying radial velocity component. This in turn leads to an anomalous radial velocity signal during planetary transits which is superimposed on the Rossiter-McLaughlin (RM) effect radial velocity signal and the Keplerian orbital radial velocity signal. In this work we show that if not accounted for in the modeling of the in-transit radial velocity (aka spectroscopic transit) the CB effect can bias the estimate of the sky-projected spin-orbit angle, which is commonly measured with the RM effect. The figure on the right shows a model of a hot Jupiter orbiting a Sun-like star.  The radial velocity signal induced by the CB effect is shown in a blue line and that induced by the RM effect in a green line. The top panel shows only the CB effect signal, the middle panel shows the RM effect and the RM effect while adding the CB effect (black line). The bottom panel adds the Keplerian orbital signal (ORB; black dashed line) where the solid black line shows the combined signal from the orbit and the RM and CB effects.

A Ground-based measurement of the relativistic beaming effect in a detached double white dwarf binary
Shporer et al. 2010, ApJL, 725, 200

The beaming effect (aka Doppler boosting) is a minute effect that causes the observed flux of an object in a binary system to modulate following its radial velocity (RV) modulation. The photometric amplitude is linear in K/c, the RV amplitude divided by the velocity of light. Such small amplitude is usually too small to be measured with ground-based facilities. Although, similarly to many other physical processes, the beaming effect becomes much larger in a system composed of two compact objects. Here I measured the beaming effect using the LCOGT FTN 2.0m telescope in the double white dwarf (WD) system NLTT 11748. This was done following the discovery of eclipses in this system, making it the first eclipsing detached double WD system (Steinfadt, Kaplan, Shporer, et al. 2010). The figure on the right shows the phase folded light curve at the top, where different colors mark data from three different nights, and the binned phase folded light curve at the bottom, shifted for visibility. The solid black line is the fitted model. The measured beaming amplitude is 3.0(4) x 10-3 which compares well to the expected amplitude of 3.3(1) x 10-3. Other processes that might induce photometric modulations along the orbital motion (tidal ellipsoidal distortion, reflected light, heating) have a negligible amplitude in this system. This was one of the first measurements of this effect, the first measurement done from the ground, and the first to be done in a double WD system.

Ground-based multisite observations of two transits of HD 80606b
Shporer et al. 2010, ApJ, 722, 880

HD80606b is a unique transiting gas-giant as it has a 111 day eccentric orbit (e = 0.93). Observing a complete transit is challenging since it is rare due to the long period and it is 12 hours long so it cannot be observed from a single ground-based site. We carried out a multisite campaign including 10 observatories and 13 telescopes to observe two transits in 2009 and 2010. The phase folded transit light curve using all 13 data sets is shown below, where data sets are color codes according to the filter used. The fitted model for each filter is overplotted by a solid line and residuals are shown at the bottom with error bars. In this work I developed a method to combine partial transit light curves (some of them with no out-of-transit data), showing how coordinated multisite observations can result in high-quality transit light curves.

Photometric follow-up observations of the transiting Neptune-mass planet GJ 436b
Shporer et al. 2009, ApJ, 694, 1559

GJ 436b is the first exoplanet close in radius and mass to Neptune that was discovered to show transits (Gillon et al. 2007). In this work we collected 6 transit light curves from immediately after the discovery of transits (5 new + 1 already published), spanning one month. The figure on the right shows the 6 data sets (blue) overplotted by the fitted model (red solid line) and residuals are at the bottom of each panel (black). Data was obtained in 3 observatories using 4 telescopes. This work refined the system parameters and constrained any variation in the transit parameters (midtransit time and impact parameter) with time. The constraints on the impact parameter variation were consistent with a long-term variation, which if confirmed could have explained why transits were not identified previously in this system.

HAT-P-9b: A low-density planet transiting a moderately faint F star
Shporer et al. 2009, ApJ, 690, 1393

This is the discovery paper of the HAT-P-9b. It was identified as a transit candidate with HATNet instruments at FLWO and SMA. I carried out RV follow-up with the SOPHIE echelle spectrograph mounted on the OHP 1.93 m telescope. The RV orbit is shown in the top panel in the figure on the right (data in blue, fitted model in red). I also carried out photometric follow-up with the two telescopes at the Wise Observatory (1 m and 0.46 m) while additional transits were observed with the FLWO 1.2m, shown in the bottom panel in the figure on the right (data in blue, fitted model in red, and the label includes transit number, observatory and telescope name, and filter used). With a brightness of V = 12.3 mag, the host F-type star is at the faint end of transiting planet host stars discovered with similar instruments. The planet is inflated, with a radius of Rp = 1.40 RJup and a mass of Mp = 0.78 MJup, making a low mean planet density of 0.35 g cm-3.

Photometric analysis of the optical counterpart of the black hole HMXB M33 X-7
Shporer et al. 2007, A&A, 462, 1091

M33 X-7 is the second brightest X-ray source in the M33 galaxy, second only to its galactic nucleus. It is a high-mass X-ray binary (HMXB) showing eclipses in the X-ray every 3.45 days as the binary companion passes in front of the X-ray source (Pietsch et al. 2006). In this work we have identified the optical counterpart by detecting a star at the same position that shows variability in the visible light at the same period. As shown in the figure on the right the variability is detected in four optical bands: B and V (from the DIRECT project, Mochejska et al. 2000), and SDSS r' and i' (from the M33 CFHT survey, Hartman et al. 2006). The light curves show the familiar pattern of the tidal ellipsoidal distortion of the star by the nearby compact object (aka double peak shape). Using a simultaneous multi-band double-harmonic period analysis I confirmed the optical counterpart shows the same period as the X-ray eclipses. This detection of the optical counterpart opened the way to measuring its orbit through radial velocities and deriving the mass of the compact object dynamically. This was done by Orosz et al. 2007, who measured a mass of 70.0 ± 6.9 $M_\left\{\odot\right\}$ for the optical counterpart and 15.65 ± 1.45 $M_\left\{\odot\right\}$ for the compact objects, confirming it is indeed a stellar black hole in a binary system.

Photometric follow-up of the transiting planet WASP-1b
Shporer et al. 2007, MNRAS, 376, 1296

In this work I carried out photometric follow-up of the transiting planet WASP-1b, which is among the first transiting planets discovered using small-aperture wide-field instruments. The two transits included in this work were obtained with the Wise Observatory 1 m telescope within two weeks of the announcement of the planet's discovery by the WASP team, and this follow-up study was done in parallel to the one done by Charbonneau et al. (2007). The results presented in this paper included a refinement of the system parameters, including a more precise planet radius and transit ephemeris. The refined parameters showed that this planet has a large radius for its mass, leading to a low mean density, challenging theory of planet mass and radius. The figure below shows the two transit light curves obtained in this work (data in red, fitted model in blue, and residuals in black at the bottom), from October 4, 2006 (left panel) and October 9, 2006 (right panel).

Long-term V-band monitoring of the bright stars of M33 at the Wise Observatory
Shporer & Mazeh 2006, MNRAS, 370, 1429

In this paper we report the results of the Wise Observatory M33 variability search, which is a long-term V-band photometric survey conducted with the Wise Observatory 1 m telescope from 2000-2003. A total of 617 exposures of three fields were obtained in 95 nights. Light curves of 6418 objects were obtained and made publicly available. Of those, 290 were identified as variables, including cepheids, eclipsing binaries, periodic (unclassified), and non-periodic. Among those variables 127 were new variables, and for another 10 previously known non-periodic variables a period was identified. The identified periods range from 2.11 to almost 300 days. For 50 of the variables we have combined our data with that of the DIRECT data (Macri et al. 2001, AJ, 121, 870), obtaining light curves of up to 500 measurements with a time-span of about 7 years. The results of this survey are also presented at wise-obs.tau.ac.il/~shporer/m33.

CV (PDF)

Publications: