Cape Town, South Africa
Cape Town, South Africa

Time filter

Source Type

News Article | February 22, 2017
Site: www.eurekalert.org

Astronomers using the TRAPPIST-South telescope at ESO's La Silla Observatory, the Very Large Telescope (VLT) at Paranal and the NASA Spitzer Space Telescope, as well as other telescopes around the world [1], have now confirmed the existence of at least seven small planets orbiting the cool red dwarf star TRAPPIST-1 [2]. All the planets, labelled TRAPPIST-1b, c, d, e, f, g and h in order of increasing distance from their parent star, have sizes similar to Earth [3]. Dips in the star's light output caused by each of the seven planets passing in front of it (astronomy) -- events known as transits -- allowed the astronomers to infer information about their sizes, compositions and orbits [4]. They found that at least the inner six planets are comparable in both size and temperature to the Earth. Lead author Michaël Gillon of the STAR Institute at the University of Liège in Belgium is delighted by the findings: "This is an amazing planetary system -- not only because we have found so many planets, but because they are all surprisingly similar in size to the Earth!" With just 8% the mass of the Sun, TRAPPIST-1 is very small in stellar terms -- only marginally bigger than the planet Jupiter -- and though nearby in the constellation Aquarius (constellation) ) (The Water Carrier), it appears very dim. Astronomers expected that such dwarf stars might host many Earth-sized planets in tight orbits, making them promising targets in the hunt for extraterrestrial life, but TRAPPIST-1 is the first such system to be found. Co-author Amaury Triaud expands: "The energy output from dwarf stars like TRAPPIST-1 is much weaker than that of our Sun. Planets would need to be in far closer orbits than we see in the Solar System if there is to be surface water. Fortunately, it seems that this kind of compact configuration is just what we see around TRAPPIST-1!" The team determined that all the planets in the system are similar in size to Earth and Venus in the Solar System, or slightly smaller. The density measurements suggest that at least the innermost six are probably rocky in composition. The planetary orbits are not much larger than that of Jupiter's Galilean moon system, and much smaller than the orbit of Mercury in the Solar System. However, TRAPPIST-1's small size and low temperature mean that the energy input to its planets is similar to that received by the inner planets in our Solar System; TRAPPIST-1c, d and f receive similar amounts of energy to Venus, Earth and Mars, respectively. All seven planets discovered in the system could potentially have liquid water on their surfaces, though their orbital distances make some of them more likely candidates than others. Climate models suggest the innermost planets, TRAPPIST-1b, c and d, are probably too hot to support liquid water, except maybe on a small fraction of their surfaces. The orbital distance of the system's outermost planet, TRAPPIST-1h, is unconfirmed, though it is likely to be too distant and cold to harbour liquid water -- assuming no alternative heating processes are occurring [5]. TRAPPIST-1e, f, and g, however, represent the holy grail for planet-hunting astronomers, as they orbit in the star's habitable zone[6]. These new discoveries make the TRAPPIST-1 system a very important target for future study. The NASA/ESA Hubble Space Telescope is already being used to search for atmospheres around the planets and team member Emmanuël Jehin is excited about the future possibilities: "With the upcoming generation of telescopes, such as ESO's European Extremely Large Telescope and the NASA/ESA/CSA James Webb Space Telescope , we will soon be able to search for water and perhaps even evidence of life on these worlds." [1] As well as the NASA Spitzer Space Telescope , the team used many ground-based facilities: TRAPPIST-South at ESO's La Silla Observatory in Chile, HAWK-I on ESO's Very Large Telescope in Chile, TRAPPIST-North in Morocco, the 3.8-metre UKIRT in Hawaii, the 2-metre Liverpool and 4-metre William Herschel telescopes at La Palma in the Canary Islands, and the 1-metre SAAO telescope in South Africa. [2] TRAPPIST-South (the TRAnsiting Planets and PlanetesImals Small Telescope-South) is a Belgian 0.6-metre robotic telescope operated from the University of Liège and based at ESO's La Silla Observatory in Chile. It spends much of its time monitoring the light from around 60 of the nearest ultracool dwarf stars and brown dwarfs ("stars" which are not quite massive enough to initiate sustained nuclear fusion in their cores), looking for evidence of planetary transits. TRAPPIST-South, along with its twin TRAPPIST-North, are the forerunners to the SPECULOOS system, which is currently being installed at ESO's Paranal Observatory. [3] In early 2016, a team of astronomers, also led by Michaël Gillon announced the discovery of three planets orbiting TRAPPIST-1. They intensified their follow-up observations of the system mainly because of a remarkable triple transit that they observed with the HAWK-I instrument on the VLT. This transit showed clearly that at least one other unknown planet was orbiting the star. And that historic light curve shows for the first time three temperate Earth-sized planets, two of them in the habitable zone, passing in front of their star at the same time! [4] This is one of the main methods that astronomers use to identify the presence of a planet around a star. They look at the light coming from the star to see if some of the light is blocked as the planet passes in front of its host star on the line of sight to Earth -- it transits (astronomy) the star, as astronomers say. As the planet orbits around its star, we expect to see regular small dips in the light coming from the star as the planet moves in front of it. [5] Such processes could include tidal heating, whereby the gravitational pull of TRAPPIST-1 causes the planet to repeatedly deform, leading to inner frictional forces and the generation of heat. This process drives the active volcanism on Jupiter's moon Io. If TRAPPIST-1h has also retained a primordial hydrogen-rich atmosphere, the rate of heat loss could be very low. [6] This discovery also represents the largest known chain of exoplanets orbiting in near-resonance with each other. The astronomers carefully measured how long it takes for each planet in the system to complete one orbit around TRAPPIST-1 -- known as the revolution period -- and then calculated the ratio of each planet's period and that of its next more distant neighbour. The innermost six TRAPPIST-1 planets have period ratios with their neighbours that are very close to simple ratios, such as 5:3 or 3:2. This means that the planets most likely formed together further from their star, and have since moved inwards into their current configuration. If so, they could be low-density and volatile-rich worlds, suggesting an icy surface and/or an atmosphere. This research was presented in a paper entitled "Seven temperate terrestrial planets around the nearby ultracool dwarf star TRAPPIST-1", by M. Gillon et al., to appear in the journal Nature. The team is composed of M. Gillon (Université de Liège, Liège, Belgium), A. H. M. J. Triaud (Institute of Astronomy, Cambridge, UK), B.-O. Demory (University of Bern, Bern, Switzerland; Cavendish Laboratory, Cambridge, UK), E. Jehin (Université de Liège, Liège, Belgium), E. Agol (University of Washington, Seattle, USA; NASA Astrobiology Institute's Virtual Planetary Laboratory, Seattle, USA), K. M. Deck (California Institute of Technology, Pasadena, CA, USA), S. M. Lederer (NASA Johnson Space Center, Houston, USA), J. de Wit (Massachusetts Institute of Technology, Cambridge, MA, USA), A. Burdanov (Université de Liège, Liège, Belgium), J. G. Ingalls (California Institute of Technology, Pasadena, California, USA), E. Bolmont (University of Namur, Namur, Belgium; Laboratoire AIM Paris-Saclay, CEA/DRF - CNRS - Univ. Paris Diderot - IRFU/SAp, Centre de Saclay, France), J. Leconte (Univ. Bordeaux, Pessac, France), S. N. Raymond (Univ. Bordeaux, Pessac, France), F. Selsis (Univ. Bordeaux, Pessac, France), M. Turbet (Sorbonne Universités, Paris, France), K. Barkaoui (Oukaimeden Observatory, Marrakesh, Morocco), A. Burgasser (University of California, San Diego, California, USA), M. R. Burleigh (University of Leicester, Leicester, UK), S. J. Carey (California Institute of Technology, Pasadena, CA, USA), A. Chaushev (University of Leicester, UK), C. M. Copperwheat (Liverpool John Moores University, Liverpool, UK), L. Delrez (Université de Liège, Liège, Belgium; Cavendish Laboratory, Cambridge, UK), C. S. Fernandes (Université de Liège, Liège, Belgium), D. L. Holdsworth (University of Central Lancashire, Preston, UK), E. J. Kotze (South African Astronomical Observatory, Cape Town, South Africa), V. Van Grootel (Université de Liège, Liège, Belgium), Y. Almleaky (King Abdulaziz University, Jeddah, Saudi Arabia; King Abdullah Centre for Crescent Observations and Astronomy, Makkah Clock, Saudi Arabia), Z. Benkhaldoun (Oukaimeden Observatory, Marrakesh, Morocco), P. Magain (Université de Liège, Liège, Belgium), and D. Queloz (Cavendish Laboratory, Cambridge, UK; Astronomy Department, Geneva University, Switzerland). ESO is the foremost intergovernmental astronomy organisation in Europe and the world's most productive ground-based astronomical observatory by far. It is supported by 16 countries: Austria, Belgium, Brazil, the Czech Republic, Denmark, France, Finland, Germany, Italy, the Netherlands, Poland, Portugal, Spain, Sweden, Switzerland and the United Kingdom, along with the host state of Chile. ESO carries out an ambitious programme focused on the design, construction and operation of powerful ground-based observing facilities enabling astronomers to make important scientific discoveries. ESO also plays a leading role in promoting and organising cooperation in astronomical research. ESO operates three unique world-class observing sites in Chile: La Silla, Paranal and Chajnantor. At Paranal, ESO operates the Very Large Telescope, the world's most advanced visible-light astronomical observatory and two survey telescopes. VISTA works in the infrared and is the world's largest survey telescope and the VLT Survey Telescope is the largest telescope designed to exclusively survey the skies in visible light. ESO is a major partner in ALMA, the largest astronomical project in existence. And on Cerro Armazones, close to Paranal, ESO is building the 39-metre European Extremely Large Telescope, the E-ELT, which will become "the world's biggest eye on the sky".


In addition to the ground-based observations described in ref. 1, this work was based on 1,333 hours of new observations gathered from the ground with the 60-cm telescopes TRAPPIST-South (469 h) and TRAPPIST-North (202 h), the 8-m Very Large Telescope (3 h), the 4.2-m William Herschel Telescope (26 h), the 4-m UKIRT (25 h), the 2-m Liverpool Telescope (50 h), and the 1-m SAAO telescope (11 h), and from space with Spitzer (518 h). The new observations of the star gathered by the TRAPPIST-South1, 31, 32 60-cm telescope (La Silla Observatory, Chile) occurred on the nights of 29 December 2015 to 31 December 2015, and from 30 April 2016 to 11 October 2016. The observational strategy used was the same as that described in ref. 1 for previous TRAPPIST-South observations of the star. TRAPPIST-North33 is a new 60-cm robotic telescope installed in spring 2016 at Oukaïmeden Observatory in Morocco. It forms an instrumental project led by the University of Liège, in collaboration with the Cadi Ayyad University of Marrakesh, and is, like its southern twin TRAPPIST-South, totally dedicated to observations of exoplanet transits and small bodies of the Solar System. TRAPPIST-North observations of TRAPPIST-1 were performed from 1 June 2016 to 12 October 2016. Each run of observations consisted of 50-s exposures obtained with a thermoelectrically cooled 2k × 2k deep-depletion charge-coupled-device (CCD) camera (field of view of 19.8′ × 19.8′; image scale of 0.61″ per pixel). The observations used the same ‘I+z’ filter as for most of the TRAPPIST-South observations1. The new VLT/HAWK-I34 (Paranal Observatory, Chile) observations that revealed a triple transit of planets c, e and f (see main text and Extended Data Fig. 1) were performed during the night of 10 December 2015 to 11 December 2015, with the observational strategy described in ref. 1 (NB2090 filter), except that each exposure was composed of 18 integrations of 2 s. The 4-m telescope UKIRT (Mauna Kea, Hawaii) and its Wide-Field Camera (WFCAM), an infrared camera35, observed the star on 24 June, 16, 18, 29 and 30 July, and 1 August 2016. Here, too, the observational strategy was the same as used as in previous observations of the star1 (J filter; exposures of five integrations of 1 s). The 4.2-m William Herschel Telescope (La Palma, Canary Islands) observed the star for three nights in a row from 23 August 2016 to 25 August 2016 with its optical 2k × 4k auxiliary-port camera (ACAM)36, which has an illuminated circular field of view of diameter 8′ and an image scale of 0.25″ per pixel. The observations were performed in the Bessel I filter with exposure times of between 15 s and 23 s. Ten runs of observation of TRAPPIST-1 were performed by the robotic 2-m Liverpool Telescope between June and October 2016. These observations were obtained through a Sloan-z filter with the 4k × 4k IO:O CCD camera37 (field of view 10′ × 10′). A 2 × 2 binning scheme resulted in an image scale of 0.30″ per pixel. An exposure time of 20 s was used for all images. The 1-m telescope at the South African Astronomical Observatory (SAAO, Sutherland, South Africa) observed the star on the nights of 18 to 19 June 2016, 21 to 22 June 2016, and 2 to 3 July 2016. The observations consisted of 55-s exposures taken by the 1k × 1k Sutherland high-speed optical (SHOC) CCD camera38 (field of view 2.85′ × 2.85′) using a Sloan z filter and with a 4 × 4 binning, resulting in an image scale of 0.67″ per pixel. For all ground-based data, a standard pre-reduction (involving bias, dark, flat-field correction) was applied, and then the stellar fluxes were measured from the calibrated images using DAOPHOT aperture photometry software39. In a final stage, a selection of stable comparison stars was manually performed in order to obtain the most accurate differential photometry possible for TRAPPIST-1. The Spitzer Space Telescope observed TRAPPIST-1 using its Infrared Array Camera (IRAC) detector40 for 5.7 h on 21 February 2016, for 6.5 h on 3, 4, 7, 13, 15 and 18 March 2016, and continuously from 19 September 2016 to 10 October 2016. All of these observations were made at 4.5 μm in subarray mode (32 × 32 pixel windowing of the detector) with an exposure time of 1.92 s. The observations were made without dithering and in the pointing calibration and reference sensor (PCRS) peak-up mode41, which maximizes the accuracy in the position of the target on the detector so as to minimize the so-called pixel phase effect of IRAC indium antinomide arrays42. All of the Spitzer data were calibrated with the Spitzer pipeline S19.2.0, and delivered as cubes of 64 subarray images. Our photometric extraction was identical to that described in ref. 43. We used DAOPHOT to measure the fluxes by aperture photometry, and combined the measurements per cube of 64 images. The photometric errors were taken as the errors on the average flux measurements for each cube. The observations used here are summarized in Extended Data Table 1. The total photometric dataset—including the data in ref. 1—consists of 81,493 photometric measurements spread over 351 light curves. We converted each universal time (ut) of mid-exposure to the BJD time system44. We then performed an individual model selection for each light curve; tested a large range of models composed of a baseline model representing the flux variations correlated to variations of external parameters (for example, point-spread function size or position on the chip, time or airmass) as low-order (0 to 4) polynomial functions; and eventually added to this baseline model a transit model45 and/or a flare model (instantaneous flux increase followed by an exponential decrease) if a structure consistent in shape with these astrophysical signals was visible in the light curve (two flares were captured by Spitzer during its 20-day-monitoring campaign; see Fig. 1). The final model of each light curve was selected by minimization of the Bayesian information criterion (BIC)46. For all of the Spitzer light curves, we needed to include a linear or quadratic function of the x- and y-positions of the point-spread function (PSF) centre (as measured in the images by the fit of a two-dimensional gaussian profile) in the baseline model to account for the pixel phase effect42, 43, complemented in some light curves by a linear or quadratic function of the measured widths of the PSF in the x- and/or y-directions43. For each light curve presenting a transit-like structure whose existence was favoured by the BIC, we explored the posterior probability distribution function (PDF) of its parameters (width, depth, impact parameter and mid-transit timing) with an adaptive MCMC code1, 9. For the transits originating from the firmly confirmed planets b and c, we fixed the orbital period to the values in ref. 1. For the other transit-like structures, the orbital period was also a free parameter. As in ref. 1, we assumed circular orbits for the planets, and we assumed the normal distributions N(0.04, 0.082) dex, N(2,555, 852) K, N(0.082, 0.0112)M , and N(0.114, 0.0062)R as prior PDFs for the stellar metallicity, effective temperature, mass, and radius, respectively, on the basis of a priori knowledge of the stellar properties1, 47. We assumed a quadratic limb-darkening law for the star48, with coefficients interpolated for TRAPPIST-1 from the tables of ref. 49. Details of the MCMC analysis of each light curve are as in ref. 1. We used the resulting values for the timings of the transits to identify planetary candidates, by searching for periodicities and consistency between the derived transit shape parameters. Owing to the high precision and near-continuous nature of the photometry acquired by Spitzer in September and October 2016, this process allowed us to firmly identify the four new planets, d, e, f and g, with periods of 4.1 days, 6.1 days, 9.2 days and 12.3 days respectively (Extended Data Figs 2, 3). We then measured updated values for their transit timings through new MCMC analyses of their transit light curves, for which the orbital periods were fixed to the determined values. For the six planets b, c, d, e, f and g, we then performed a linear regression analysis of the measured transit timings, T , as a function of their epochs, E , to derive a transit ephemeris T  = T ( ± σT ) + E  × P ( ± σP), with T being the timing of a reference transit for which the epoch is arbitrarily set to 0, P being the orbital period, and σT and σP being their errors as deduced from the co-variance matrix (Table 1). For all planets, the residuals of the fit showed some significant deviation, indicating TTVs, which is unsurprising given the compactness of the system and the near-resonant chain formed by the six inner planets (see below). For a transit-like signal observed by Spitzer at BJD ~2,457,662.55 (Fig. 1), the significance of the detection (>10σ) was large enough to allow us to conclude that a seventh, outermost planet exists as well. This conclusion is based not only on the high significance of the signal and the consistency of its shape with one expected for a planetary transit, but also on the photometric stability of the star at 4.5 μm (outside of the frequent transits and the rare— about one per week—flares) as revealed by Spitzer (Fig. 1). In a final stage, we performed the global MCMC analysis of the 35 transits observed by Spitzer that is described in the main text. It consisted of two chains of 100,000 steps, whose convergence was successfully checked using the statistical test of ref. 50. The parameters derived from this analysis for the star and its planets are shown in Table 1. We used the TTV method10, 11 to estimate the masses of the TRAPPIST-1 planets. The continuous exchange of angular momentum between gravitationally interacting planets causes them to accelerate and decelerate along their orbits, making their transit times occur early or late compared with a Keplerian orbit14. All six inner TRAPPIST-1 planets exhibit transit timing variations owing to perturbations from their closest neighbours (Extended Data Fig. 4). The TTV signal for each planet is dominated primarily by interactions with adjacent planets, and these signals have the potential to be particularly large because each planet is near a mean motion resonance with its neighbours. As calculated from the present data, the TTV amplitudes range in magnitude from 2 min to more than 30 min. However, the distance of these pairs to exact resonances controls the amplitude and the period of the TTV signals and is not precisely pinned down by the present dataset. Moreover, the relatively short timeframe during which the transits have been monitored prevents an efficient sampling of the TTV oscillation frequencies for the different pairs of planets, defined by f(TTV) = n /P  − n /P , where P is the orbital period, n the mean motion, and i and j the planet indices10. We modelled TTVs using both numerical integrations (TTVFast51 and Mercury52) and analytical integrations (TTVFaster53) of a system of six gravitationally interacting, co-planar planets. TTVFaster is based on analytical approximations of TTVs derived using perturbation theory and includes all terms at first order in eccentricity. Furthermore, it includes only those perturbations to a planet from adjacent planets. To account for the 8/5 and 5/3 near-resonances in the system, we also included the dominant terms for these resonances, which appear at second and third order in the eccentricities. We determined these higher-order terms using the results of ref. 54. TTVFaster has the advantage that it is much faster to compute compared with n-body integrations. It is applicable for this system given the low eccentricities determined via TTV analysis (determined independently with n-body integrations and self-consistently with TTVFaster). We used two different minimization techniques: Levenberg–Marquardt55 and Nelder–Mead56. For the purpose of analysis, we used the 98 independent transit times for all six planets and 5 free parameters per planet (mass, orbital period, transit epoch and eccentricity vectors ecosω and esinω, with e being the eccentricity and ω the argument of periastron). We elected not to include the seventh planet, h, in the fit, because only a single transit has been observed and there is not yet an indication of detectable interactions with any of the inner planets. Likewise, we did not detect any perturbation that would require the inclusion of an additional, undetected non-transiting planet in the dynamical fit. The six-planet model provided a good fit to the existing data (Extended Data Fig. 4), and we found no compelling evidence for extending the present model complexity given the existing data. Our three independent analyses of the same set of transit timings revealed multiple, mildly inconsistent, solutions that fit the data equally well provided that non-circular orbits are allowed in the fit. It is likely that this solution degeneracy originates from the high dimensionality of the parameter space, combined with the limited constraints brought by the present dataset. The best-fit solution that we found—computed with Mercury52—has a chi-squared of 92 for 68 degrees of freedom, but involves non-negligible eccentricities (0.03 to 0.05) for all planets, probably jeopardizing the long-term stability of the system. In this context, we decided to present conservative estimates of the planets’ masses and upper limits for the eccentricities without favouring one of the three independent analyses. For each parameter, we considered as the 1σ lower/upper limits the smallest/largest values of the 1σ lower/upper limits of the three posterior PDFs, and the average of the two computed limits as the most representative value. The values and error bars computed for the planets’ masses and the 2σ upper limits for their orbital eccentricities are given in Table 1. Additional precise transit timings for all seven planets will be key in constraining further the planet masses and eccentricities and in isolating a unique, well defined, dynamical solution. We investigated the long-term evolution of the TRAPPIST-1 system using two n-body integration packages: Mercury52 and WHFAST57. We started from the orbital solution produced in Table 1, and integrated over 0.5 million years (Myr). This corresponds to roughly 100 million orbits for planet b. We repeated this procedure by sampling a number of solutions within the 1σ intervals of confidence. Most integrations resulted in the disruption of the system on a 0.5-Myr timescale. We then decided to use a statistical method that yields the probability of a system being stable for a given period of time, based on the planets’ mutual separations58. Using the masses and semi-major axes in Table 1, we calculated the separations between all adjacent pairs of planets in units of their mutual Hill spheres58. We found an average separation of 10.5 ± 1.9 (excluding planet h), where the uncertainty is the r.m.s. of the six mutual separations. We computed that TRAPPIST-1 has a 25% chance of suffering an instability over 1 Myr, and an 8.1% chance of surviving for 1 billion years (Gyr), in line with our n-body integrations. These results, obtained by two different methods, suggest that the TRAPPIST-1 system could be unstable over relatively short timescales. However, they do not take into account the proximity of the planets to their host star and the resulting strong tidal effects that might act to stabilize the system. We included tidal effects in an ameliorated version of the Mercury package59, 60, and found that they markedly enhance the system’s stability. However, the disruption is only postponed by tides in most simulations, and further investigations are needed in order to better understand the dynamics of the system. In general, the stability of the system appears to be very dependent on the assumptions of the orbital parameters and masses of the planets, and on the inclusion or exclusion of planet h and on its assumed orbital period and mass. It is also possible that other, still undetected, planets help to stabilize the system. The masses and exact eccentricities of the planets remain uncertain, and our results make it likely that only a very small number of orbital configurations lead to stable configurations. For instance, mean motion resonances can protect planetary systems over long timescales61. The system clearly exists, and it is unlikely that we are observing it just before its catastrophic disruption, so it is probably stable over a long timescale. These facts and the results of our dynamical simulations indicate that, given enough data, the very existence of the system should bring strong constraints on its components’ properties—their masses, orbital elements and tidal dissipation efficiencies, which are dependent on the planets’ compositions, mutual tidal effects of the planets, mutual inclinations, the orbit of planet h, the existence of other, maybe not transiting planets, and so on. The conversion of the ut times of the photometric measurements to the BJD system was performed using the online program created by J. Eastman and distributed at http://astroutils.astronomy.ohio-state.edu/time/utc2bjd.html. The MCMC software used to analyse the photometric data is a custom Fortran 90 code that can be obtained from M.G. on reasonable request. The n-body integration codes TTVFast, TTVFaster, and Mercury are freely available online at https://github.com/kdeck/TTVFast, https://github.com/ericagol/TTVFaster, and https://github.com/smirik/mercury. To realize Fig. 2a, we relied on TEPCAT, an online catalogue of transiting planets maintained by J. Southworth (http://www.astro.keele.ac.uk/jkt/tepcat/). The Spitzer data that support our findings are available from the Spitzer Heritage Archive database (http://sha.ipac.caltech.edu/applications/Spitzer/SHA). Source Data for Fig. 1 and Extended Data Figs 1, 2, 3, 4 are available online. The other datasets generated and/or analysed during the present study are available from M.G. on reasonable request.


News Article | February 15, 2017
Site: spaceref.com

An exotic binary star system 380 light-years away has been identified as an elusive white dwarf pulsar -- the first of its kind ever to be discovered in the universe -- thanks to research by the University of Warwick. Professors Tom Marsh and Boris Gänsicke of the University of Warwick's Astrophysics Group, with Dr. David Buckley from the South African Astronomical Observatory, have identified the star AR Scorpii (AR Sco) as the first white dwarf version of a pulsar -- objects found in the 1960s and associated with very different objects called neutron stars. The white dwarf pulsar has eluded astronomers for over half a century. AR Sco contains a rapidly spinning, burnt-out stellar remnant called a white dwarf, which lashes its neighbour -- a red dwarf -- with powerful beams of electrical particles and radiation, causing the entire system to brighten and fade dramatically twice every two minutes. The latest research establishes that the lash of energy from AR Sco is a focused 'beam,' emitting concentrated radiation in a single direction -- much like a particle accelerator -- something which is totally unique in the known universe. AR Sco lies in the constellation Scorpius, 380 light-years from Earth, a close neighbour in astronomical terms. The white dwarf in AR Sco is the size of Earth but 200,000 times more massive, and is in a 3.6 hour orbit with a cool star one third the mass of the Sun. With an electromagnetic field 100 million times more powerful than Earth, and spinning on a period just shy of two minutes, AR Sco produces lighthouse-like beams of radiation and particles, which lash across the face of the cool star, a red dwarf. As the researchers previously discovered, this powerful light house effect accelerates electrons in the atmosphere of the red dwarf to close to the speed of light, an effect never observed before in similar types of binary stars. The red dwarf is thus powered by the kinetic energy of its spinning neighbour. The distance between the two stars is around 1.4 million kilometres -- which is three times the distance between the Moon and the Earth. Professor Tom Marsh comments: "The new data show that AR Sco's light is highly polarised, showing that the magnetic field controls the emission of the entire system, and a dead ringer for similar behaviour seen from the more traditional neutron star pulsars." Professor Boris Gänsicke comments: "AR Sco is like a gigantic dynamo: a magnet, size of the Earth, with a field that is ~10,000 times stronger than any field we can produce in a laboratory, and it is rotating every two minutes. This generates an enormous electric current in the companion star, which then produces the variations in the light we detect." Reference: "Polarimetric Evidence of a White Dwarf Pulsar in the Binary System AR Scorpii," D. A. H. Buckley et al., 2017 Jan. 23, Nature Astronomy [http://www.nature.com/articles/s41550-016-0029, preprint: https://arxiv.org/abs/1612.03185]. Please follow SpaceRef on Twitter and Like us on Facebook.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2014

Kepler photometry of A stars shows that a considerable fraction (about 19 per cent) have a peculiar feature in the periodogram. This feature consists of a broad peak, thought to be due to differential rotation in a spotted star, and a sharp peak at slightly higher frequency. The pattern clearly involves some widespread stellar property and the sharp peak implies a strictly coherent periodicity. We investigate the possibility that the periodicity is due to rotation, pulsation or an orbital effect.We argue that neither rotation nor pulsation can provide a suitable, testable, explanation. We suggest that the sharp feature could be due to a planet in synchronous orbit around the rapidly rotating, spotted A star, not necessarily in transit. Spectroscopic observations of sufficient precision are required to falsify this hypothesis. © 2014 The Author Published by Oxford University Press on behalf of the Royal Astronomical Society.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2014

We consider the high mode density reported in the δ Scuti star HD 50844 observed by CoRoT. Using simulations, we find that extracting frequencies down to a given false alarm probability by means of successive pre-whitening leads to a gross overestimate of the number of frequencies in a star. This is due to blending of the peaks in the periodogram due to the finite duration of the time series. Pre-whitening is equivalent to adding a frequency to the data which is carefully chosen to interfere destructively with a given frequency in the data. Since the frequency extracted from a blended peak is not quite correct, the interference is not destructive with the result that many additional fictitious frequencies are added to the data. In data with very high signal-to-noise ratio, such as the CoRoT data, these spurious frequencies are highly significant. Continuous pre-whitening thus causes a cascade of spurious frequencies which leads to a much larger estimate of the mode density than is actually the case. The results reported for HD 50844 are consistent with this effect. Direct comparison of the power in the raw periodogram in this star with that in δ Scuti stars observed by Kepler shows that HD 50844 has a typical mode density. © 2014 The Author.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2014

A method is presented whereby the orbital parameters of a pulsating star in a binary ormultiple system can be determined using the time delay due to the changing distance of the pulsating star. The method differs from previously published methods in that a direct periodogram of the orbital frequency is derived using all possible information to extract the time delay from the stellar pulsations. The method is easy to use provided the pulsation frequencies are known. The method is tested using various simulations and applied to two stars which have been previously analysed in the literature. Application to 34 Kepler objects of interest which are also δ Sct variables resulted in the detection of five stars which are binaries. © 2014 The Author Published by Oxford University Press on behalf of the Royal Astronomical Society.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2013

Two years of Kepler data are used to investigate low-frequency variations in A-type stars. In about 875 (40 per cent) A-type stars, the periodogram shows a simple peak and its harmonic. If we assume that the photometric period is the period of rotation, we can derive the equatorial rotational velocity from a suitable radius estimate. It turns out that the distribution of equatorial velocities derived in this way is similar to the distribution of equatorial velocities of A-type main-sequence stars in the general field derived from spectroscopic line broadening, verifying our initial assumption. We suggest that the light variation is due to rotational modulation caused by starspots or some other corotating structure. In many stars the rotation peak in the periodogram has a characteristic shape which is not understood. The light amplitudes are highly variable.We deduce from the amplitude distribution that the sizes of starspots in A-type stars are similar to the largest sunspots. From the widths of the peaks in the periodogram we deduce that differential rotation in these stars is similar to that in the Sun. We find that the period-colour relationship used for gyrochronology in late-type stars extends to early F-type and probably late A-type stars as well. Flares in A-type stars have been recently detected. We add 13 additional A-type flare stars to this sample, which means that about 1.5 per cent of A-type stars in the Kepler field show flares. We conclude that A-type stars are active and, like cooler stars, have starspots and flares. © 2013 The Author. Published by Oxford University Press on behalf of the Royal Astronomical Society.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2012

We consider the possible use of combination frequencies seen in high-amplitude δ Scuti stars for mode identification. For this purpose, we extend a previous theory to obtain expressions for the relative amplitudes and phases of combination frequencies for the case where radius and surface normal variations are important. Although a term is present which is sensitive to the pulsation mode, this term is always combined with another one which depends on the physics of the outer layers. Whereas the terms can be separated in DA and DB white dwarfs using suitable assumptions, these assumptions are not valid in pulsating main-sequence stars. We also discuss the strong correlation of relative phase of the combination frequencies with frequency. We show that this correlation is a result of the non-sinusoidal nature of the dominant pulsation mode. © 2012 The Author Monthly Notices of the Royal Astronomical Society © 2012 RAS.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2012

Optical flares on early F- and A-type stars have never been observed with certainty. Inspection of several thousands of these stars in the Kepler public archives resulted in the discovery of flares in 25 G-type and 27 F-type stars. Because A-type stars are thought not to be active, the detection of flares on 19 A-type stars from a sample of nearly 2000 A stars is particularly noteworthy. The flares have relative intensities in the range 1-100 parts per thousand and typical durations of a few minutes to several hours. The mean interval between flares varies between 1 and 120days. We estimate the typical energy of flares to be around 10 35 erg in the F-type stars and about 10 36 erg in the A-type stars. Nearly all these stars vary at a low level with a period which is most likely the rotational period of the star. Comparison of the relative flare intensities with those in cool red stars observed by Kepler shows that flares in these stars, and certainly in the A-type stars, cannot easily be ascribed to cool flare-star companions. The huge energy released in the flares is difficult to understand. This is especially the case for A-type stars since these stars are thought to have very weak magnetic fields. The flare energy may possibly originate in magnetic reconnection of field lines between the primary star and a companion. © 2012 The Author Monthly Notices of the Royal Astronomical Society © 2012 RAS.


Balona L.A.,South African Astronomical Observatory
Monthly Notices of the Royal Astronomical Society | Year: 2011

We analyse the light curves of over 9000 A-F stars in the first public release of Kepler data and examine the noise properties in constant or nearly constant stars. For the A stars, we find a correlation between the excess power in certain frequency regions and the effective temperature which may be due to granulation. The majority of A-F stars vary with low frequencies (<5d-1) and low amplitudes (40-150 ppm). The low-frequency variation extends to the hottest A-type stars where the typical amplitude is about 40 ppm. We find that about 8 per cent of A8-A0 stars have light curves resembling those usually attributed to starspots in cool stars, including a few exhibiting travelling waves usually interpreted as differentially rotating starspots. A further 20 per cent of A stars have dominant low frequencies which are visible in the periodogram. If we assume that the dominant low frequency in A-F stars is the rotation frequency, we can calculate the distribution of equatorial rotational velocities given the stellar radii. The resulting distribution matches the distribution of equatorial rotational velocities in field stars of the same spectral type and luminosity class. However, the A8-A0 stars have an excess of slow rotators which can be explained as contamination from horizontal-branch stars. We conclude that the light variations in A-type stars may possibly be due to starspots or other corotating structures and that A-star atmospheres may not be quiescent as previously supposed. We also analyse low frequencies in Kepler A-type δ Sct stars which are too hot to be due to γ Dor pulsations. These do not appear to be due to simple combinations of high-frequency δ Sct modes. Unlike normal A-type stars, the dominant low frequency is close to twice the rotational frequency. In a significant proportion of δ Sct stars there is, in fact, a frequency of smaller amplitude at exactly half the dominant low frequency. There is clearly a quadrupole surface brightness distribution in a significant fraction of these stars, but the amplitudes seem to be too low to be explained as a proximity effect in a binary. © 2011 The Author Monthly Notices of the Royal Astronomical Society © 2011 RAS.

Loading South African Astronomical Observatory collaborators
Loading South African Astronomical Observatory collaborators